Early Childhood Caries in Cambodia - Minerva Access

i

Early Childhood Caries in Cambodia

Bathsheba Turton BDS MComDent (Otago)

A thesis submitted in total fulfilment of the requirements for the degree of

Doctor of Philosophy

21 September 2018

Melbourne Dental School,

Faculty of Medicine, Dentistry and Health Sciences,

The University of Melbourne.

ii

Abstract

Children in Cambodia have a severe burden of caries and data from the Cambodian national

oral health survey reflect that four in five 6-year-old children have one or more pulpally

involved teeth. One of the key features of dental caries in Cambodia is that most of the lesions

go untreated and most of the dental services are orientated towards those in an urban setting

and involve management of pain only. Global evidence on prevention of dental caries suggests

that interventions focusing on topical application of fluorides in the form of toothpastes or

varnishes and application of pit and fissure sealants (FS) might be the most effective

interventions for reducing caries experience.

Two groups of investigations were conducted; the first group of investigations were part of the

SEAL Cambodia project and the second group were part of the Cambodia Smile Project. The

aims of the SEAL Cambodia project were to evaluate two different protocols for the placement

of glass ionomer cement-based fissure sealants and the prevention of dental caries in the FPM

of 6 to 8 year old children in three provinces of Cambodia, then to investigate the caries

preventive effect of a refined protocol for GIC fissure protection in FPM of 6 to 8 year-old

children in Cambodia. The Cambodia Smile investigation aimed to describe the

epidemiological aspects of Early Childhood Caries in a Cambodian context, then investigate

the effectiveness of a pilot strategy for the reduction of dental caries experience in Cambodian

preschool children through an integrated primary health care model.

The original SEAL Cambodia project rendered a non-significant 10% preventive fraction for

new carious lesions on first permanent molars after 1-year. The modified protocol rendered a

90% preventive fraction at 1-year and 30% preventive fraction at 2-years.

The Cambodia Smile results affirmed existing evidence that Early Childhood Caries (ECC)

affected a large proportion of young children and that a package of oral health interventions

integrated with the routine vaccination schedule can render a 66% reduction in the severity of

dental caries among 2-year-old children.

This collection of studies represents the first efforts to build evidence around the reorientation

of dental services towards preventive therapies in a Cambodian context. Further investigation

is needed to better understand the social, structural and behavioural aspects of the ECC

phenomenon in Cambodia in order to better inform strategy. The SEAL Cambodia project

provides sufficient evidence for an appropriate clinical protocol for the application of FS in a

iii

Cambodian context. The Cambodia Smile pilot provides sufficient evidence to justify up-scale

and monitoring of a similar project to reduce the disease burden among a larger proportion of

Cambodian children.

iv

Dedication

To My Dad - The Dreamer

I’m not quite convinced that dreams ever come true. I think that dreams only ever come

through.

You might have had your head in the clouds, but when it comes down to it, I think that

you’ve got to be a little out of touch with reality in order to make a dream come through.

Maybe dreaming is about being open.

There was a farm and a motorbike and a factory, then you went to fix aeroplanes. There was a

shoe shop and a wee house with a yard then you went to study and preach. There was a

chapel and village and a bunch of cows and you brought the books of the world into our

house.

We didn’t know anything about Dentistry or Cambodia or PhDs but I think that the fact that

you were a dreamer made my dreams come through.

To My Mother, Kim Turton, who was the first person to introduce me to the concept of a PHD.

It’s true, my gear is ‘Piled High and Deep’ in her garage!

v

Declaration

This is to certify that:

i. The thesis comprises only my original work towards the PhD except where

indicated

ii. Due to acknowledgement has been made in the text to all other material used

iii. The thesis is less than 100,000 words in length, exclusive of tables, maps,

bibliographies and appendices

Bathsheba Jael Turton_____ _________ 21 September 2018

vi

Acknowledgements

First and foremost thank you to my supervisors Prof David Manton and Dr Felicity Crombie

who supported me through out, were always available, were excellent communicators and

without whom I wouldn’t have been able to realise the dream of participating in the world of

preventive dentistry in Cambodia

Assoc Prof Callum Durward – my advisor and the father of modern dentistry in Cambodia and

without whom, this research would not be possible. Yours are the shoulders that I stand on

when I want to gain a better view.

Phillip Sussex – who set up the SEAL protocol training materials. In addition, he inspired

critical thinking and made it possible to consider an at-scale intervention in a Cambodian

context.

The organisations and companies who sponsored the projects: GC Asia, Cam Kids, Ivolcar

Vivadent, and ANZSPD who provided support for projects.

Scholarships

Peter and Barbara Dennison who provided mentorship and support in interpreting my

experiences dealing with the problem of caries in Cambodia.

Dr Karen Sokal Gutierez who provided ongoing mentorship, friendship and support and helped

me to understand the situation better.

The Partnering organisations; International University Phnom Penh, Cambodia Dental

Association, Cambodia World Family, Buddhist Library Project, Australia-New Zealand

society of paediatric dentistry, Global Child Dental Fund and Cam Kids who cooperated to

make these projects possible. A special mention to One-2-One Cambodia who provided

administrative oversight for the projects.

Drs Tepirou Chher, and Sithan Hak from the Oral health Bureau who supported projects,

provided advice and continue to relentlessly advocate for better oral health for Cambodian

people.

Dr Sopharith Soeun who provided training for primary health care providers in the Cambodia

Smile intervention and is providing ongoing mentorship and training in the upscale phase.

Midwives in the health centres in Kampong Speu who made an impact for the Cambodia smile

intervention and were able to provide useful feedback that will shape dental prevention into the

future.

Ms Liong Sao who acted as a research assistant and who had the ability to solve problems

before I knew they were happening.

vii

Dental students of the Univeristiy of Puthisastra, International University, and University of

Health Sciences, Phnom Penh Cambodia. Dental students placed fissure sealants, collected and

entered data and made these projects possible.

For the SEAL Cambodia Project: Chen Pagna, Sok Povrath, Sok SereyPiseth, Sreang Rattanak

For the Cambodia smile cross-sectional study: Dr Yos Chantho, Dr Tak Ranouch, and Dr Soy

Rasy

For the Cambodia Smile intervention: Sieng Tida, Cham Roeun, Heng Chanlay, Sok Phirak,

Loy Sreylan, and Horn Vitou.

Laura Spero who helped with editing, endured my frustrations, and facilitated my problem

solving. You continue to inspire me.

viii

Table of Content

Abstract ..................................................................................................................................... ii

Dedication ................................................................................................................................ iv

Declaration................................................................................................................................ v

Acknowledgements ................................................................................................................. vi

Table of Content ................................................................................................................... viii

List of Tables ........................................................................................................................ xiii

List of figures .......................................................................................................................... xv

List of Appendices ................................................................................................................. xvi

Abbreviations ....................................................................................................................... xvii

Publications and Presentations ............................................................................................ xix

1.0 Chapter 1 – Literature Review .................................................................................... 2

1.1 Dental caries ....................................................................................................................... 2

1.1.1 The caries balance ......................................................................................................... 3

1.1.2 Observing and quantifying the signs of dental caries ................................................... 5

1.2 Early childhood caries ....................................................................................................... 7

1.2.1 Early childhood caries................................................................................................... 7

1.2.2 The maternal-child link to ECC .................................................................................... 9

1.2.3 Nursing habits and ECC.............................................................................................. 10

1.2.3.1 Breast feeding ...................................................................................................... 10

1.2.3.2 Bottle feeding ....................................................................................................... 13

1.2.4 The diet and ECC ........................................................................................................ 13

1.2.5 The bacteria and ECC ................................................................................................. 15

1.2.6 Host factors ................................................................................................................. 16

1.2.7 Socio-behavioural factors and risk modelling ............................................................ 17

1.3 Prevention of ECC ........................................................................................................... 19

ix

1.3.1 Maternal-child interventions ....................................................................................... 20

1.3.2 School based interventions ......................................................................................... 21

1.3.3 Multi-sectorial cooperation ......................................................................................... 21

1.4 Prevention of ECC – managing the diet ........................................................................ 22

1.5 Prevention of ECC – Fluorides and topical agents ....................................................... 22

1.5.1 Fluoridated toothpaste ................................................................................................. 23

1.5.2 Fluoride Varnish ......................................................................................................... 24

1.5.3 Other fluoride delivery systems .................................................................................. 25

1.5.4 Bioavailable calcium and phosphate substrates .......................................................... 26

1.5.5 Antibacterial agents for preventing ECC .................................................................... 27

1.6 Prevention of caries on First Permanent Molars (FPM) .............................................. 28

1.6.1 Pit and Fissure sealants ........................................................................................... 29

1. 7 The situation in Cambodia ............................................................................................. 30

1.7.1 General health experiences in Cambodia .................................................................... 30

1.7.2 Dental caries in Cambodia .......................................................................................... 31

1.7.3 Policies, politics, and practice in Cambodia ............................................................... 33

1.7.3.1 Community health centres in Cambodia .............................................................. 33

1.7.3.2 Dental nurses in Cambodia .................................................................................. 33

1.7.3.3 Traditional Dentists .............................................................................................. 34

1.7.3.4 University Trained Dentists ................................................................................. 34

1.7.3.5 The Oral Health Bureau ....................................................................................... 34

1.8 Summary of literature review ......................................................................................... 35

1.9 References Chapter 1 ....................................................................................................... 36

2.0 Chapter 2 – The SEAL Cambodia pilot studies ............................................................ 50

2.0.1 The ‘birth’ of the SEAL Cambodia Community Project ............................................ 51

2.1 Aim .................................................................................................................................... 53

2.2 Objective ........................................................................................................................... 53

x

2.3 Methods ............................................................................................................................. 54

2.3.1 Selection of teeth for FS ............................................................................................. 54

2.3.2 Pilot A – one year evaluation of the original cohort ................................................... 54

2.3.2.1 Clinical examination at baseline and one year follow-up .................................... 55

2.3.3 Pilot B - pilot study using a revised protocol .............................................................. 55

2.3.2.1 Clinical examination ............................................................................................ 56

2.3.4 Data collection and analysis........................................................................................ 56

2.4 Results ............................................................................................................................... 57

2.4.1 Pilot A ......................................................................................................................... 57

2.4.2 Pilot B ......................................................................................................................... 63

2.5 Discussion.......................................................................................................................... 67

2.5.1 Caries prevention in Pilot A ........................................................................................ 69

2.5.1.1 Alternatives to achieve caries prevention in FPM ............................................... 69

2.5.2 The retention of FS material in Pilot B ....................................................................... 70

2.5.2.1 The pattern of material loss.................................................................................. 71

2.5.2.2 High vs Low viscosity GIC.................................................................................. 73

2.6 Conclusion ........................................................................................................................ 73

2.7 References Chapter 2 ....................................................................................................... 74

3.0 Introduction – SEAL Cambodia evaluation of a modified protocol for placing Fuji

VII Fissure Sealants (FS)....................................................................................................... 78

3.1 Aim .................................................................................................................................... 79

3.2 Objectives.......................................................................................................................... 79

3.3 Methods ............................................................................................................................. 80

3.3.1 Clinical procedures ..................................................................................................... 80

3.3.2 Clinical examination and questionnaire ...................................................................... 81

3.3.3 Part 1 – Methods for One-year investigation .............................................................. 81

3.3.4 Part 2 – Two year analysis .......................................................................................... 81

xi

3.3.5 Data analysis ............................................................................................................... 82

3.4 Results ............................................................................................................................... 83

3.4.1 Part 1 – One year comparison of old and new cohorts. .............................................. 83

3.4.2 Part 2 – Two year evaluation of the modified protocol. ............................................. 89

3.5 Discussion.......................................................................................................................... 95

3.5.1 General findings .......................................................................................................... 95

3.5.2 Preventive benefit of the SEAL Cambodia intervention ............................................ 95

3.5.3 Protocol considerations ............................................................................................... 96

3.5.4 Self-reported oral symptoms ....................................................................................... 98

3.5.5 The theoretical argument for SEAL as a preventive strategy ..................................... 98

3.6 Conclusion ........................................................................................................................ 99

3.7 References Chapter 3 ..................................................................................................... 100

4.0 Chapter 4 – Cambodia Smile cross-sectional study. Introduction ............................ 101

4.1 Aim .................................................................................................................................. 103

4.2 Objective ......................................................................................................................... 103

4.3 Methods ........................................................................................................................... 104

4.3.1 General approach ...................................................................................................... 104

4.3.2 Participants ................................................................................................................ 104

4.3.3 Clinical Measures...................................................................................................... 105

4.3.4 The questionnaire ...................................................................................................... 105

4.3.5 Data analysis ............................................................................................................. 105

4.4 Results ............................................................................................................................. 107

4.5 Discussion........................................................................................................................ 118

4.5.1 Caries experience within the Cambodia Smile cross-sectional survey ..................... 118

4.5.2 ECC and nursing habits ............................................................................................ 119

4.5.3 ECC and oral hygiene practices in Cambodia .......................................................... 122

4.5.4 Other risk indicators for ECC ................................................................................... 122

xii

4.6 Conclusions ..................................................................................................................... 123


5.0 Chapter 5 – The Cambodia Smile intervention .......................................................... 130

5.1 Aims ................................................................................................................................. 131

5.2 Objective ......................................................................................................................... 131

5.3 Methods ........................................................................................................................... 132

5.3.1 Health Centre and Population Selection ................................................................... 132

5.3.3 Clinical Protocol for the intervention ....................................................................... 132

5.3.4 The questionnaire for the two-year follow-up .......................................................... 133

5.3.5 Clinical measures ...................................................................................................... 133

5.3.6 Data collection and analysis...................................................................................... 133

5.4 Results ............................................................................................................................. 135

5.5 Discussion........................................................................................................................ 142

5.5.1 Differences in Oral Health Related Quality of Life .................................................. 142

5.5.2 Differences in caries experience by participation in the intervention ....................... 143

5.6 Conclusions ..................................................................................................................... 145


6.0 Chapter 6 - Overview, limitations, and recommendations of the SEAL Cambodia and

Cambodia Smile projects .................................................................................................... 152

6.1 Summary of findings from the ‘SEAL Cambodia’ project ........................................ 152

6.2 Summary of findings from the Cambodia Smile project ........................................... 155

6.3 Conclusions ..................................................................................................................... 158


xiii

List of Tables

Table 1.1 – Case definitions of Early Childhood Caries (ECC) and Severe Early Childhood

Caries (sECC) ............................................................................................................................ 8

Table 1.2 Dental caries experience in Cambodia between 1990 and 2011 ............................. 32

Table 2.1 – Pilot A – Attrition analysis for the 1-year follow-up by gender and age.............. 57

Table 2.2 – Pilot A - Caries experience at baseline by group membership, gender and age. .. 58

Table 2.3 – Pilot A - Caries incidence and sealant retention by group membership ............... 59

Table 2.4 – Pilot A - Caries incidence in the intervention group by baseline caries experience

and age ..................................................................................................................................... 60

Table 2.5 – Pilot A - Number of retained sealants by caries experience and gender .............. 62

Table 2.6 – Pilot B – Gender and group membership after cases with caries FPM at baseline

were removed ........................................................................................................................... 64

Table 2.7 – Pilot B - Proportion of sealant retention by material type in the intervention group

…………………...................................................................................................................... 65

Table 2.8 – Pilot B - Proportion of Fuji VII® FS retained at one month and one year by group

membership .............................................................................................................................. 66

Table 3.1 – Part 1 – Attrition analysis for the One year follow-up by group membership and

caries experience. ..................................................................................................................... 83

Table 3.2 – Part 1 - Sociodemographic Characteristics of Participants. .................................. 84

Table 3.3 – Part 1 - Clinical Characteristics of participants by group membership. ............... 86

Table 3.4 – Part 1 - Caries preventive increment by group membership at one year. ............. 87

Table 3.5 – Part 1 - Logistic regression model for risk of developing new carious lesions based

on group membership. ............................................................................................................. 88

Table 3.6 – Part 2 - Characteristics of participants by gender, age, and group membership and

school. ...................................................................................................................................... 89

Table 3.7 – Part 2 - Clinical characteristics of participants according to sociodemographic

factors. ...................................................................................................................................... 90

Table 3.8 – Part 2 - Participant attrition analysis for the 2 year follow-up of the second cohort.

………………………………………………………………………………………...91

xiv

Table 3.9 – Part 2 - Proportion of children with new carious lesions on occlusal of FPM at 2-

years by group membership. .................................................................................................... 92

Table 3.10 – Part 2 - Proportion children with new cavitated occlusal and interproximal carious

lesions by group membership. ................................................................................................. 93

Table 3.11 – Part 2 - Logistic regression models showing odds ratio for the chance of

developing one or more occlusal lesions on FPM at two-years............................................... 94

Table 4.2 – Cambodia Smile Cross-sectional study - Socio-demographics characteristics by

caries experience of the child. ................................................................................................ 109

Table 4.3 – Cambodia Smile cross-sectional study - Maternal caries experience by Child caries

experience. ............................................................................................................................. 110

Table 4.4 – Cambodia Smile cross-sectional study - Questions on nursing habits and oral health

behaviours by responses from the primary-caregiver. ........................................................... 112

Table 4.5 – Cambodia Smile cross-sectional survey - Questions on dietary habits by responses

from the primary caregiver. ................................................................................................... 114

Table 4.6 – Cambodia Smile cross-sectional survey - Bivariate analysis of dental behaviours

by child caries experience. ..................................................................................................... 116

Table 4.7 – Cambodia Smile cross-sectional study - Binomial logistic regression model for

Socio-behavioural habits by presence of significant caries. .................................................. 117

Table 5.1 – Attrition analysis of intervention group at 1-year. .............................................. 135

Table 5.2 – Sociodemographic characteristics of participants at follow-up by group. .......... 136

Table 5.3 – Caries severity by number of contacts among the intervention group. .............. 137

Table 5.4 – Caries experience by sociodemographic characteristics. .................................... 138

Table 5.5 – Clinical characteristics by exposure to the intervention. .................................... 140

Table 5.6 – Differences within the Family Impact Scale and subscale impacts and mean scores

by group membership. ........................................................................................................... 141

xv

List of figures

Figure 1.1 - The caries imbalance .............................................................................................. 4

Figure 2.1 – Selection and follow-up of children in Pilot B .................................................... 63

xvi

List of Appendices

Appendix 1 – Consent SEAL Cambodia (Original)

Appendix 2 – Post operative information SEAL Cambodia

Appendix 3 – SEAL Baseline examination form

Appendix 4 – SEAL Follow-up examination form

Appendix 5 – Standard SEAL Form

Appendix 6 – The SEAL Cambodia original protocol

Appendix 7 – The SEAL Cambodia modified protocol

Appendix 8 – Examination forms and questionnaire for the Cambodia smile cross-sectional

survey

Appendix 9 – Clinical documentation and Questionnaires for the Cambodia Smile

Intervention

Appendix 10 – Publications

Appendix 11 – Ethical approval documentation

xvii

Abbreviations

AAPD American Association of Pediatric Dentistry

ANZSPD The Australia and New Zealand Society of Paediatric Dentistry

BPOC Basic Package of Oral Cares

CAST The Caries Assessment Spectrum and Treatment Index

CHC Community Health Centre

CHX Chlorhexadine

CNOHS Cambodia National Oral Health Survey

CPP -ACP Casein Phosophopeptide-Amorphous Calcium Phosphate

CWF Community water fluoridation

DMF/DMFT/dmf Decayed, Missing, and Filled Teeth Index.

ECC Early Childhood Caries

FIS Family Impact Scale

FP Fissure Protection

FPM First Permanent Molars

FS Fissure Sealants

FV Fluoride Varnish

GCDF Global Child Dental Fund

GIC Glass Ionomer Cement

HBM Human Breast Milk

ICC Intra Class Correlation

ICDAS International Caries Detection and Assessment System

LB Lactobacilli

LMIC Low-to Middle-income countries

MMP Matrix Metalloproteanise

MOEYS Ministry of Education Youth and Sport

MOH Ministry of Health

NGOs Non-Governmental Organisations

OHB Oral Health Bureau

OHE Oral Health Education including diet advice, oral hygiene instruction

and delivery of toothbrush with a supply of fluoridated toothpaste

OHRQoL Oral-Health-Related Quality-of-Life

PCG Primary care giver

PUFA Pulpally involved, Ulcerated, Fistula, and Abscess index

SD Standard Deviation

SDF Silver Diammine Fluoride

sECC severe - Early childhood caries

SES Socioeconomic Status

SM Streptococcus mutans

xviii

SPSS Statistical Package for the Social Sciences

UN United Nations

WHO World Health Organization

WSL White spot lesions

y years

χ2 Chi-squared test

xix

Publications and Presentations

This thesis is based partially on the following publications by the author and manuscripts

which are reproduced in Appendix 10.

1. Turton B J, Durward C S, Manton D (2015) Early Childhood Caries and Maternal

Caries Experience in a Convenience Sample of Cambodian Pre-schoolers. Pediatric

Dental Journal 25 (1), pp. 14-18.

2. Turton B J, Durward CS, Manton D, Bach K, Yos C (2015) Socio-behavioural risk

factors for early childhood caries (ECC) in Cambodian preschool Children European

Archives of Paediatric Dentistry. 17(2), pp. 97-105.

3. Tac N, Turton B, Durward C (2016) Retention of Glass Ionomer Cement fissure

sealant using a revised protocol Cambodian Dental Journal 12, 14-18.

4. Turton B and Durward C. Management of Early Childhood Caries – a Comparison of

Different Approaches. Thai Dental Public Health Journal 2017 Jan-Feb;22(suppl.),

pp. 67-77

Presentations

1. Turton, B.*, Durward, C., Bach, K., Manton, D. (2014) Seal Cambodia – 60,000

children over 3 years; IADR SEA 28TH Annual meeting, Kutching, Malaysia.

2. Souen, S., Turton, B., Durward, C. (2015) Integrating oral health and general health in

a Cambodian setting; IADR SEA 29th Annual meeting, Bali, Indonesia.

3. Tak, N., Turton, B., Durward, C. (2015) Seal Cambodia – A comparison of two

protocols; IADR SEA 29th Annual meeting, Bali, Indonesia.

4. Turton, B.*, Durward, C. (2015) Seal Cambodia – Placing fissure protection in a

community setting Successfully; FDI world Congress, Bangkok, Thailand.

5. Turton, B.*, (2015) Cambodia Smile – integrating oral health and general health in a

Cambodian setting. 2015 Global children’s nutrition and oral health symposium,

UCSF, California, USA.

6. Turton, B.*, Durward, C., Crombie, F., Manton D. SEAL Cambodia: Improved Caries

Prenvtion with a modified Protocol. (2016) . IADR General Session. Seoul, Republic

of Korea.

xx

7. Sreng, R., Turton, B., Chen, P., Sok S, Durward C. (2016) SEAL Cambodia – Retention

rates of ART Sealants according to provider type. IADR General Session. Seoul,

Republic of Korea.

8. Sopharith, S., Turton B.*, Durward, C., Crombie, F, Manton, D., (2016) Cambodia

Smile – Preliminary results from an early childhood caries intervention. IADR General

Session. Seoul, Republic of Korea.

9. Turton, B.*, Durward, C., Tac, N., & Manton, D. J. (2016). Seal Cambodia Caries

Prevention Using a Modified Protocol for School-based GIC Sealant Placement. 63rd

Congress of the European Organisation for Caries Research in Athens, Greece.

10. Turton B.*, (2017). Reorientation of Dental Services – Lessons Learnt From Cambodia.

Oral Health Improvement Strategies for Asia (IMUOHS): Online Conference. Kuala

Lumpur, Malaysia.

11. Turton B.,* (2017). Management of ECC, Lessons Learnt from Cambodia. 9th Asian

Conference for Oral Health Promotion in School Children. Siem Reap, Cambodia.

12. Turton, B.*, Durward, C., Soeun, S., Crombie, F., Manton, D., (2017). Cambodia Smile

– 2-year follow-up of a community based Early Childhood Caries (ECC) intervention.

Congress of the International Association of Paediatric Dentistry. Santiago, Chile.

13. Turton, B.*, Crombie, F., Durward, C., Manton D. (2018). Seal Cambodia – a tale of

two protocols. ANZSPD, Brisbane, Australia.

14. Turton, B.*, Durward, C., Soeun S., Crombie, F*., Manton, D. (2018). Cambodia Smile

– Caries Prevention by primary health care providers. ORCA, Copenhagen, Denmark

1

Chapter 1

Literature Review

Published in part as: Turton B and Durward C. Management of Early Childhood Caries

– a Comparison of Different Approaches. Thai Dental Public Health Journal 2017 Jan-

Feb;22(suppl.):67-77

Presented in part at the 9th Asian Conference for Oral Health Promotion in School

Children. (2017) Siem Reap, Cambodia: Turton B. Management of ECC, Lessons Learnt

from Cambodia.

2

1.0 Chapter 1 – Literature Review

Introduction

Dental caries remains the most prevalent disease in the world, accounting for a considerable

health burden in both developed and developing countries (Kassebaum 2015). Dental caries is

a multifactorial disease involving an interplay between the bacteria, the host, and the diet, with

free-sugars recently being given emphasis as the most potent initiator and driver of the caries

process (Sheiham & James 2015). In light of this, the globalisation of food sources and the

nutrition transition towards highly processed and high added-sugar foods has been

contemporaneous with the plateauing of improvements in caries rates in developed countries

and increases in dental caries experience among at-risk populations. Countries in the South

East Asian Region are thought to have the most severe caries experience worldwide and

Cambodia has been reported as having the most severe experience of Early Childhood Caries

(ECC) in that region (Duangthip et al. 2017).

Dental caries is classified by the World Health Organization (WHO) as a non-communicable

disease and has a multidirectional relationship with other common conditions such as under-

nutrition (Hooley et al. 2012a). This literature review aims to describe the process of dental

caries, particularly in the context of the pre-school age-group. In addition, the current best

practices for prevention and management of dental caries in a Cambodian context will be

considered.

1.1 Dental caries

Before examining the risk factors associated with ECC, it is appropriate to consider the caries

process first. Dental caries is a behaviourally driven bacterially-based process that can lead to

the irreversible loss of mineralised tooth structure (Fejerskov 1997) and it is the most prevalent

disease in humans (Allukian 2000; Frencken et al. 2012). Although world-wide estimates may

be somewhat out-dated, dental caries affects 60 - 90% of children and almost all adults

(Petersen et al. 2005; Marcenes et al. 2013). Also of concern is that the prevalence and severity

of dental caries appears to be increasing for those in Low and Middle Income Countries (LMIC)

and that the progressive decline in the dental caries experience amongst individuals from

developed countries appears to have plateaued (Petersen et al. 2005; Kassebaum et al. 2015).

Furthermore, when caries experience is broken down by socio-demographic characteristics

within countries, it is possible to see that individuals in the younger age groups and those in

3

disadvantaged minority groups, such as indigenous peoples, are most at risk (Poon et al. 2015;

Schwendicke et al. 2015). The reason these groups are experiencing a disproportionate level of

disease experience could be that the lifestyle and environment in the respective settings may

modify the balance of variables and interventions that contribute to the caries process. (Tiwari

et al. 2018)

1.1.1 The caries balance

The concept of a caries balance (Figure 1) recognises that a carious lesion exists in a dynamic

state where-by protective factors and pathological factors interact (Featherstone 1999). When

the pathological factors are dominating, net mineral loss occurs. In the situation that the system

fails to correct itself (with protective factors) then cavitation may occur at the tooth surface

once sufficient mineral is lost from the lesion (Kidd & Fejerskov 2004; Featherstone 2004).

The pathological factors relate to the oral microbiome, the diet and the host (individual) in the

context of time. With regards to the microbiome, it is important to recognise that the

microorganisms along with by-products and supporting structures on the surface of the tooth

exist within and make up a biofilm. Furthermore, the presence of specific bacteria and bacterial

by-products within that biofilm will alter the environment at the surface of the tooth. Various

bacteria have been implicated in the caries process such as Streptococcus mutans (SM),

Streptococcus sobrinus, Actinomyces, Bifidobacteria and Lactobacilli (LB) (Takahashi 2015).

Almost all human beings carry these organisms; it is not specific species that are responsible

for the caries process but rather a dysbiotic ecological community of bacteria (Marsh 2004).

When the biofilm has a higher proportion of acid producing (acidogenic) and acid resistant

(aciduric) bacteria, combined with insufficient calcium and phosphate ions in the biofilm

solution at the tooth surface to maintain saturation with respect to tooth mineral, then net

mineral loss will occur. When organic acids are produced by the bacteria the hydrogen ions

will dissociate and diffuse into the dental enamel, and in doing so, dissolve the mineral content

of the tooth to eventually produce the clinically detectable signs of a carious lesion.

4

Figure 1.1 –The caries imbalancea

aFeatherstone (Featherstone 2004).

Some species of bacteria are more acidogenic and aciduric than others. These species thrive in

an environment with a high frequency of exposure to fermentable carbohydrates, specifically

free-sugars (WHO 2015). When fermentable carbohydrates are provided as a substrate to the

biofilm then primarily lactic acid is produced, which can rapidly lower pH and dissolve

mineral. In that situation, bacteria which can withstand (and flourish in) an acidic environment

will become the dominant species in an oral environment rich in fermentable carbohydrates. In

other words; the diet will place selective pressures on the biofilm to favour species with

particular characteristics (Ruby & Goldner 2007).

The ability of a food or drink to promote the caries process is termed ‘cariogenicity’

(Tahmassebi et al. 2006). It is both the quantity and the frequency of exposure to fermentable

carbohydrates that triggers this transformation. Regular snacking has been associated with a

higher caries burden (Feldens et al. 2012). In addition, there is evidence at the population level

of a dose-response effect of sugar ingestion to dental caries (Moynihan 2005; Sheiham & James

2015). One of the key protective mechanisms from the host is saliva, which has an ability to

help food bolus formation and subsequent oral clearance, buffer acid, and to provide a reservoir

5

of protein-stabilised supersaturated (with respect to tooth mineral) concentrations of calcium

and phosphate that facilitates remineralisation among other functions (Featherstone 2006).

In a normal scenario and given enough time, the effect of the bacterially-derived acid at the

time of eating can be balanced by the saliva-based protective mechanisms within the mouth; if

the health of the host (individual) is compromised, decreasing salivary quality and/or quantity,

then these protective mechanisms may not be operating optimally, and the caries balance will

tip towards net mineral loss. The bicarbonate system is the major salivary mechanism for acid

buffering in addition to components such as phosphate, urea and amphoteric proteins and

enzymes. Maintenance of a neutral pH is important for sustaining super-saturation of

bioavailable minerals with respect to tooth mineral at the ionic concentrations present in the

mouth (Humphrey & Williamson 2001).

The ability of saliva to function both as an acid buffer and as a mineral reservoir compensates

for the fluctuations in pH. This means that the pH can decrease without violating the critical

pH, the lowest pH at which the solution is saturated with respect to tooth mineral. The critical

pH fluctuates depending the amount of calcium and phosphate available in the biofilm solution.

At a lower pH then a greater concentration of calcium and phosphate is required in order to

prevent net mineral loss from occurring (Dawes 2003). In the situation that the host has an

underlying systemic health condition, then the saliva may be of poor quality or low quantity

and thereby be unable to prevent mineral loss from dental enamel. That is to say, for a

compromised host the pathological factors are more likely to dominate, and the signs of dental

caries will appear in areas below undisturbed cariogenic biofilm (Featherstone 2004).

1.1.2 Observing and quantifying the signs of dental caries

The signs of dental caries may be observed along a spectrum of changes that are first detected

clinically as a white spot lesions of enamel (WSL) on a desiccated enamel surface, advancing

to enamel surface roughness, cavitation and eventually infection of the dental pulp by bacteria

(Kidd & Fejerskov 2004). It is important that indices used to measure the burden of disease are

able to capture the full spectrum of disease presentation and there have been a number of

attempts to achieve this. The Decayed, Missing and Filled (DMF) index is the most common

and is recommended by the World Health Organisation because of its ease of use (WHO 2013).

The main limitation of the DMF index is that it only captures the presence of a carious lesion

when surface cavitation is present (Broadbent & Thomson, 2005). To address this issue the

International Caries Detection and Assessment System (ICDAS) was developed to detect and

6

record the signs of dental caries from an early WSL (only detectable when the tooth is dry)

through to gross loss of hard tooth structure (Ismail et al. 2007). The disadvantage of this

system it that it requires optimal tooth cleaning and drying and illumination to detect the earliest

lesion which limits its use in a field environment, although some variations of ICDAS have

been created to address this. For example, the Caries Assessment Spectrum and Treatment

(CAST) index was developed which records the signs of caries at the stage in which a carious

lesion is visible on a wet surface. The data generated by CAST can be easily manipulated to

generate DMF data for cross comparison to historical datasets (Frencken et al. 2011). However,

the uptake and use of CAST has been limited.

Although it is important to detect the early consequences of the caries process, in some settings

a large proportion of carious lesions go untreated, leading to a high prevalence of pulpally

involved teeth. The Philippines is one of these settings and it was here that the Pulpally-

involved Ulcerated Fistula and Abscess (PUFA) index was developed (Monse et al. 2010). This

index can be used to effectively quantify the severity of the effect of carious lesions on the pulp

and has been useful in examining the relationship between severe dental caries and malnutrition

(Benzian et al. 2011). The main disadvantage of the PUFA index is that it is best used in

addition to one of the other indices rather than independently, and so it may represent an

additional administrative burden on examiners.

One of the other challenges in classifying carious lesions is to recognise the differences in

disease progression between the primary and the permanent dentitions. Regardless of the index

used, each index requires that the primary and permanent dentitions are to be reported

separately. The reason why this is important is that the two dentitions differ in both anatomy

and by their presence in the mouth during the evolving stages of growth and development

(WHO 2013). This is particularly the case for preschool children who can demonstrate a rapidly

progressing and severe expression of dental caries due to their unique dietary and oral hygiene

habits (Douglas et al. 2001).

7

1.2 Early childhood caries

1.2.1 Early childhood caries

Preschool children represent a unique group in that their dental caries disease experience can

change rapidly (Douglass et al. 2001) and that they are in the process of evolving through many

developmental milestones. Although the caries process is the same for both adults and children,

when a young child suffers from ECC (defined in Table 1) the process can happen more quickly

and the experience of ECC is often considered as a subset of caries within the general caries

disease spectrum. The reasons that the disease might progress rapidly are related to a

combination of the unique behavioural context of a growing child and tooth morphology

(Ismail 2003). The other unique aspect of this disease experience is that the symptoms of the

disease have a profound impact on quality of life for both the child and those around them. The

impacts of the disease are not just isolated to the affected individual (Martins-Júnior et al.

2013).

To complicate matters further, there is a lack of consistency with the language used to describe

the caries experience of young children. This may be because different socio-behavioural

features tend to be implicit in each setting (Dye et al. 2015; Harris et al. 2004). The result has

been terms such as ‘baby bottle tooth decay’ or ‘nursing caries’ (Ismail & Sohn 1999). The

problem with these terms is that they do not capture the degree of severity amongst different

children or groups. Also, from a conceptual point of view, the aetiologically specific terms

(philosophically) could block consideration of other aetiological factors, which are important

when trying to understand the disease in various settings (Milnes 1996).

Quantifying dental caries amongst different populations has often been achieved using

indicator age groups as defined by the WHO, however, these fail to examine the disease

experience in those below six years-of-age; whereas open lesions can develop before the child

is two years-of-age (Douglass et al. 2001). Estimates of world-wide caries experience suggest

that the prevalence and severity of the disease has been reduced considerably over the last 50

to 100-years. However, for those individuals in LMIC and for those in compromised

socioeconomic settings, dental caries experience has increased (Pitts et al. 2011; Schwendicke

et al. 2015; Willems et al. 2005). Global data on the prevalence of dental caries in preschool

children in developed countries is as low as 1% to as high as 50%; and the prevalence of dental

caries in preschool children in developing countries is as high as 70% (Milnes 1996; Petersen

et al. 2005). This contrasts what has been reported in countries such as the Philippines (Monse

8

et al. 2010), Cambodia (Todd et al. 1994; Turton et al. 2015), and Brazil (Chaffee et al. 2013)

where it is estimated that nearly 100% of preschool children have cavitated carious lesions.

The same severe disease burden is often observed for indigenous groups within developed

settings (Hsieh et al. 2014; Peressini et al. 2004; Schroth et al. 2009; Ministry of Health, New

Zealand 2011; Hallett & O’Rourke 2003). Further estimates have suggested that the majority

of dental needs occur in a minority of the most vulnerable groups of children (Mouradian 2001;

Schwendicke et al. 2015; Moffat et al. 2017).

Aside from the differences in dental caries distribution between different populations, the other

issue that is not clear is whether dental caries in preschool children represents one single disease

paradigm or whether there are multiple disease patterns present. For instance, it has been

reported that the most common area for carious lesions to develop in young children is on the

buccal surfaces of the maxillary anterior teeth (Saleem et al. 2015; Sowole et al. 2007). Also,

there is some discussion about whether the second primary molars will develop carious lesions

to a greater extent because they have deeper pits and fissures (Sowole et al. 2007). It is likely

that one influencing factor in the difference in caries experience by tooth type would be because

the diet is different at the differing ages of eruption of specific teeth (Douglass et al. 2001;

Hallett & O’Rourke 2006).

This highlights the need for diagnostic criteria that are age dependant and site specific. The

American Association of Paediatric Dentistry (AAPD) has partly addressed this when it

presented a set of diagnostic criteria for both ECC and Severe-ECC (sECC) (Table 1).

Throughout the remainder of the present review ECC and sECC will be defined by those

criteria laid out in Table 1.

Table 1.1 – Case definitions of Early Childhood Caries (ECC) and Severe Early

Childhood Caries (sECC)a

Age (Months) ECC sECC

<12 1 or more dmf surfacesb 1 or more smooth dmf surfaces

12-23 1 or more dmf surfaces 1 or more smooth dmf surfaces

36-47 1 or more dmf surfaces 1 or more cavitated, filled, or missing (due to

caries) smooth surfaces in the primary

maxillary anterior teeth OR dmfs score > 4



maxillary anterior teeth OR dmfs score > 5



maxillary anterior teeth OR dmfs score > 6 aadapted from Drury et al. 1999.

9

bAny carious lesion, non-cavitated or cavitated, missing due to caries or filled surface. Includes primary teeth

only.

Another consideration when describing carious lesions is the impacts which are not captured

by counting teeth. Diagnostic criteria and clinical measures are helpful, but they do not capture

the effect of experience of the disease on the individual completely (Locker & Allen, 2007)

(McGrath et al. 2004). Carious lesions in preschool children have implications for further caries

experience (Broadbent et al. 2008), growth and development (including speech), and the ability

of that child to perform their social role in society (Schroth et al. 2009; Alkarimi et al. 2012;

Li et al. 2015; Hooley et al. 2012; Sheiham 2006; Ayhan et al. 1996; Broder 2007).

Dental caries is, in most individuals, preventable (Allukian 2000; Petersen 2003; Watt et al.

2015) and so all potential impacts could be avoided. In light of this, one of the ways in which

the preschool group differs from other age groups is that some of the risk behaviours associated

with dental caries are, to a large extent, dependent on the caregiver (Mouradian 2001). For

more than 25-years there have been active calls to focus on the relationship between mothers

and maternal and child oral health services in order to reduce the burden of dental caries at the

population level (Frazier & Horowitz 1990).

1.2.2 The maternal-child link to ECC

Although there is a strong evidence-base for associations between maternal and child oral

health, the causal relationship is less clear (Abiola et al. 2009). Some authors have gone as far

as to describe both direct and indirect causal relationships (Okada et al. 2002) and it may be

more appropriate to consider a multi-factorial model in which the characteristics of the mother

influence known protective and pathological factors in the child (Featherstone 2004). When

maternal characteristics are considered in the context of the traditional causal triad of dental

caries (host, bacteria and diet) then the complexity of the relationship can be observed (Selwitz

et al. 2007).

From a psycho-social point of view, the resistance of the host (child) to pathological factors is

related to maternal nutritional status (Black et al. 2008) and psychological maternal stress

(Tang et al. 2005). In addition, the child’s diet is also influenced by maternal stress and

maternal oral health literacy independently of socioeconomic status which also has an effect

(Divaris et al. 2011; Vann et al. 2010). The description of these interactions does not take into

account the expression of cariogenic bacteria in a young child’s mouth which has been

10

associated with high maternal salivary counts of S. mutans and Lactobacilli (as indicators of a

cariogenic diet) (Chaffee et al. 2013), preterm low birth weight (Boggess & Edelstein 2006),

and maternal oral hygiene practices (Chaffee et al. 2013). To add to the complexity, the

relationship between maternal characteristics and oral hygiene practices in young children are

further associated with parental stress (Finlayson 2007; Menon et al. 2000), parental attitudes

(Adeniyi et al. 2009; Tang et al. 2005), maternal oral health literacy, maternal age, maternal

education, and maternal location of residence (Abiola Adeniyi et al. 2009).

In the absence of a direct causative pathways (Okada et al. 2002) and standardised collection

of data (Kuthy 1997) it can be challenging to make a clear statement about how maternal

characteristics influence child oral health outcomes. Despite the multifactorial relationship, the

evidence of an association between caries experience in the child and caries experience in the

mother (Hooley et al. 2012b). Furthermore, it has been demonstrated that maternal caries

experience, particularly untreated lesions, at the time of birth can predict the caries experience

of their respective offspring into adulthood (Shearer et al. 2011; Harris et al. 2004; Smith et al.

2002). It is likely that both mother and child are exposed to a majority of the same protective

and pathological factors and so the presentation of carious lesions will depend upon how the

caries balance is managed in both individuals.

1.2.3 Nursing habits and ECC

Balancing protective and pathological factors is pertinent in the debate about optimal nursing

habits. One of the key issues for debate is the contribution of human breast milk (HBM) to the

dental caries process. Nursing and early childhood feeding practices are culturally bound, the

evidence for this is seen in the way that nursing habits vary widely in different settings

(Rasbridge 1995) and the way in which mothers wish to conform to social norms for nursing

in order to avoid being seen as a ‘bad mother’ (Desclaux 2009). Taking into account cultural

variation, the understanding of the relationship between dental caries and nursing habits is

distorted by a lack of investigations into breast-feeding habits and associated dietary

components in a developing world setting, and a lack of consistency with regards to reporting

such habits (Chaffee et al. 2014; Valaitis et al. 1999; Feldens et al. 2012). Despite this deficit,

some aspects of nursing are universally accepted, and this is important to consider because

there are many stakeholders in the debate and health professionals should deliver a consistent

message (Fewtrell et al. 2007; WHO, 2003).

1.2.3.1 Breast feeding

11

The present recommendation from WHO is that children should be exclusively breast-fed for

the first six months-of-life. It is widely accepted that exclusive breast-feeding during the first

six months of life can reduce the risk of gastroenteritis, infections, asthma, atopic disease and

diabetes. All of these benefits are thought to be achieved by the complex and unique

combination of micronutrients, proteins, immune complexes and carbohydrates (Marriott et al.

2012; Ribeiro & Ribeiro 2004). This combination is distinct from other processed milk

substrates such as cow’s milk, (Lönnerdal 2003) which, in comparison, has a higher buffering

capacity and lower carbohydrate content (Bowen & Lawrence 2005). In other words, breast

milk contains substrates which are capable of nourishing a biofilm rich in acidogenic and

acidoduric bacteria. Whether HBM has the right characteristics to initiate that pathological

transformation of the biofilm is in question. It has been argued that the amount of bio-available

calcium and phosphate present in human breast milk can prevent the biofilm from reaching a

critical pH with respect to tooth mineral at which net mineral loss occurs (Erickson & Mazhari

1999). This is the basis for the hypothesis that HBM has no cariogenic potential provided that

it is the only source of carbohydrate (Nunes et al. 2012). A recent study tested the idea by

exposing children with and without ECC to HBM then measuring saliva and biofilm pH; the

study found that the HBM did not lead to a change in pH (Neves et al. 2016).

In the developing world there has been a recent increase in the availability and consumption of

refined sugars (Chaffee & Cheng 2014). In addition, sugar is being added to almost all

processed foods and those processed foods have become more accessible than traditional

(unprocessed) food, this phenomenon limits consumer choice (Ludwig & Nestle, 2008). The

universal availability of refined sugars could modify the relationship between HBM and the

dental caries process. When the context of complimentary feeding is considered, if all of the

foods added to the diet of the young child include refined sugar then the biofilm could become

highly cariogenic. When the transformed cariogenic biofilm is exposed to HBM the metabolic

bi-products will be highly acidic (Chaffee et al. 2014). These differences in the characteristics

of the biofilm across different settings (i.e. settings where complimentary foods are low or high

in sugar) could also explain why some investigations report breastfeeding to be a protective

behaviour (Peterson et al. 2003; Roberts et al. 1994; Silver et al. 1992; Vignaraja et al. 1992)

while others report breastfeeding as a risk indicator (Nunes 2012, Chaffee 2014). It could be

theorised that the role that HBM plays in ECC is inconsistent across different settings (Valaitis

et al. 1999) and that it is often confounded by other socio-behavioural aspects (Erickson &

Mazhari 1999; Harris et al. 2004).

12

Widespread confounding has made it difficult to reach a consensus about the relationship

between HBM and the dental caries process. While it is accepted that exclusive breast-feeding

should occur over the first six months of life (WHO, 2003), there is currently no consensus on

what age breast-feeding should stop and how that might influence the ‘caries equation’

(Chaffee et al. 2014). Some from the dental community have suggested that breast-feeding

should stop as soon as the child is able to drink from a cup (Ribeiro & Ribeiro, 2004) while

others from the nursing community have said that there is no right time and that mothers should

be encouraged to breast-feed for as long as they can (Valaitis et al. 1999). The right time to

stop may differ depending on the individual and social contexts (Salone et al. 2013).

This lack of clarity in terms of evidence may be due to the fact that prolonged breast-feeding

is not well defined. For some countries it is uncommon to breast-feed past 6 or 12 months

(Hallett & O’Rourke 2003; Peres et al. 2017) while in other countries it is normal to breast-

feed past the age of two years (Chaffee et al. 2014). In those countries in which breast feeding

commonly stops before the first year-of-age, HBM does not appear to contribute to detectible

signs of carious lesions (Erickson & Mazhari 1999). In contrast, a setting where children sleep

with their parents, where night time ‘on demand’ suckling occurs, and where the child is breast-

fed past the age of two years, then HBM has been implicated as a risk factor for sECC (Chaffee

et al. 2014; Thitasomakul et al. 2009; van Palenstein Helderman et al. 2006; Serwint et al.

1993). Recently, there have been more studies which identify higher risk of ECC associated

with breast-feeding past 24 months (Avila et al. 2015; Wong et al. 2017; Feldens et al. 2012;

Peres et al. 2017).

The challenge in defining nursing habits lies in the fact that not all children are exclusively

breast-fed during the first six months of their life. Non-exclusive breast feeding differs across

different settings, even within the same region. For example, the prevalence of exclusive

breast-feeding up to six months for Vietnam is estimated to be 15.5%, for Timor Leste 30.7%,

for Philippines 33.7%, for Indonesia 38.9% and for Cambodia 60.1% (Khitdee 2017).

Furthermore, exclusive breast-feeding can differ within families and communities. Children

are less likely to be exclusively breast-fed during the first six months if they were the first born,

have a working mother, or have an ‘older’ mother (Senarath et al. 2010). The effect that a

particular nursing habit might have on dental caries is dependent on the day-to-day dietary and

oral hygiene behaviours, as well as the age of the child (Chaffee & Cheng 2014). Furthermore,

social indicators are associated with both caries and variations in food profiles (Gatica et al.

2012).

13

1.2.3.2 Bottle feeding

The bottle is the main viable alternative to breast-feeding for delivery of nutrition to the child

during infancy. The timing, content, and frequency of drinking are key aspects when describing

the risk of bottle-feeding in the ECC process. Children who are allowed to sip on their bottle

(carrying cariogenic contents) for a prolonged period of time, particularly at night time while

they are sleeping, have a high risk of ECC (Gussy et al. 2006; Hallett & O’Rourke 2003;

Hooley 2012b; Peltzer & Mongkolchati 2015). The other aspect to consider is whether there

are long-term health impacts (chronic disease, learning outcomes, growth and development)

associated with bottle-feeding as opposed to breast-feeding. It was recently reported that in a

middle-class context within a developed country, once fixed family effects are taken into

account, there are no adverse health outcomes associated with bottle-feeding (Colen & Ramey

2014). This statement draws the reader back to the idea that the risk of bottle-feeding

contributing to unfavourable oral health outcomes depends upon the way that children are

bottle-fed rather than the modality itself.

1.2.4 The diet and ECC

The transition of the child from a liquid diet towards a solid diet is fraught with challenges, and

one of those is the addition of free sugars to the diet. In many settings, various forms of sugar

are being added to an infant’s food (Thitasomakul et al. 2009), the infant’s bottle or even to the

breast milk (Ribeiro & Ribeiro 2004). This is a significant problem because of the effect that

sugar has in: (1) transforming the biofilm into a cariogenic biofilm, with a low pH and under-

saturation with respect to tooth mineral (Leme et al. 2006); (2) ‘recalibrating’ the child towards

favouring sweet foods; and (3) contributing to obesity and diabetes, which are reaching

epidemic levels in young children across the globe (Ka & Ca, 2012; Lustig et al. 2012; Moodie

et al. 2013). These problems are amplified by the world-wide increase in the gross consumption

of sugar, especially in low income groups, where those in the lowest income quintile residing

in urban areas increased the proportion of caloric sweetener within total carbohydrates from

10.0% to 27.7% between the years 1962 and 2000 (Popkin & Nielsen 2003; Ismail et al. 1997;

Drewnowski, 2000).

In terms of sugar, not all sugars are equal in their ability to contribute to disease processes and

the recent release of the WHO recommendations, free sugars are major contributors to dental

caries and chronic disease. Free sugars are “all monosaccharides and disaccharides added to

foods by the manufacturer, cook or consumer, plus sugars naturally present in honey, syrups

14

and fruit juices” (WHO 2015). Free sugars have been shown to contribute to dental caries in

excess of other sugar groups such as intrinsic sugars, which are those sugars naturally

incorporated within the structure of intact grains, fruits and vegetables. Free sugars easily

diffuse into the biofilm and can be potent initiators of the caries process (Sheiham & James

2015).

One of the most potent vehicles for delivery of free sugars are sugary liquids which include

sugary medicines, beverages with added sugar, fruit juices (in their natural state without

additives), and soft drinks (Hallett & O’Rourke 2003). The reason sugary drinks are such potent

mediators of dental caries is both because of the high quantity of sugar contained within the

drink, and also because it may be consumed from a vessel other than a cup (such as a bottle)

thus continuously delivering sugar to the mouth and biofilm over a prolonged period of time

(Tahmassebi et al. 2006). When a child is able to expose the oral environment to a sugary drink

for many hours during the course of a day (e.g. through prolonged sipping), the ability of saliva

to neutralise acid will be compromised and the biofilm will become more acidogenic (Wan et

al. 2003). The mineral loss that occurs as a result is more rapid for immature enamel such as

that found in an infant’s mouth (Tahmassebi et al. 2006). The evidence for the relationship

between free sugars and more severe caries experiences is seen in the association between

caries and consumption of soft drinks between meals, consumption of table sugars, and

consumption of sugary desserts (Ismail et al. 1997; Sheiham & James 2015).

The presence of sugars in any particular food can be said to increase the potential of that

substrate to cause dental caries; however the actual cariogenicity of one food in comparison to

another can vary. There are a number of characteristics that make a particular food cariogenic,

such as retentiveness and chemical composition which has led to the formulation of a

cariogenicity index (Chaffee & Cheng 2014; Evans et al. 2013). One of the problems with

determining the potential of a food to cause dental caries is that the effect of a food on the

caries balance is moderated by the frequency and mode of delivery. There are a multitude of

different variables that may not always be captured by a simple measure of presence or absence

of a substrate in the diet; furthermore, other socio-behavioural factors may modify the way that

a substrate interacts or is delivered to the oral environment (Gatica et al. 2012). This means

that in certain settings, a food consumed in a particular way may have a higher cariogenic

potential than the same food consumed in a different setting in a different way.

15

1.2.5 The bacteria and ECC

A further aspect contributing to the complexity of the caries process involves the characteristics

of the biofilm and how they will moderate the impact of a cariogenic substrate on the caries

balance. Although SM has been established as an aetiological factor in the caries process, the

presence or absence of SM is not a sufficiently sensitive predictor for the presence of clinical

disease (Harris et al. 2004). To address this, SM data are often presented in two ways; bacterial

counts from saliva, and bacterial counts from the plaque biofilm. Using these techniques, once

cariogenic bacteria reach a certain number within the biofilm, then following exposure to a

fermentable carbohydrate rapid demineralisation can occur (Klock & Krasse 1979; Li &

Caufield 1995). In other words, when a biofilm high in cariogenic bacteria is exposed to a

cariogenic substrate then rapid production of acid will occur, but if a healthy biofilm (where

acidogenic and acidoduric bacteria have not been favoured) is exposed to the same substrate

then the rate of demineralisation will be slower, because not as much acidic by-product will be

produced (Ruby & Goldner 2007).

There is much debate over which specific species within the biofilm are primary agents for

propagating the caries process. In light of this, efforts have been made to characterise the

biofilm by looking at particular strains of bacteria present: common strains have been shown

within races and families, particularly between mothers and their children, and between

siblings and among social groups (Li & Caufield 1995; Mitchell et al. 2009). This commonality

supports the hypothesis that infants receive the bacteria from members of their family and

social community and so the question of mode of transmission has been raised (Berkowitz

2003). Along with the debate around transmission there are questions concerning common

behaviours within a social community leading to selective pressures for certain strains of

bacteria to be prominent. Shared community behaviours can lead to a commonality in the

prominent bacteria as has been observed by some (Yang et al. 2012). This appears to be

consistent with the finding that if members in a child’s community (such as a mother or a

brother) are carrying cariogenic bacteria, then that child is more likely to have detectable levels

of those bacteria at an earlier age (Wan et al. 2003). This could also be a confounder in the

literature that shows that detectable SM has been associated with behaviours such as kissing

the infant on the lips, sharing eating utensils or pre-chewing food (Pattanaporn 2012; Li &

Caufield 1995; Wan et al. 2003). Again, selective pressures shared by a family group (driven

by a shared cariogenic diet) could have also lead to the association between the proportion of

16

SM or LB in the mother’s saliva, and the presence of sECC (Chaffee et al. 2013; Tinanoff et

al. 2002).

The detection of SM in the oral cavity is possible before tooth eruption and some studies show

the age of detection is associated with sECC (Wan et al. 2001). This finding has been used in

the debate between an infective theory versus an ecological theory; the infective theory was

popular before more sophisticated means of detecting microbial species were employed. Now,

with methods such as PCR and m16RNA, it is known that a non-shedding surface (e.g. a tooth)

is not required for the presence of SM (Nyvad et al. 2013). These new analytical methods

helped support the second theory, which proposes that ECC is propagated by the continuous

presence of potentially cariogenic bacteria and that the biofilm can achieve cariogenic potential

at any stage (Wan et al. 2003). Also, taking into account the near universal presence of SM

(Harris et al. 2004; Li & Caufield 1995; Tinanoff et al. 2002), when the biofilm is exposed to

free sugars (fermentable carbohydrates) with high frequency and oral persistence, then it will

transform and net mineral loss can subsequently occur. This process is independent from the

age of the host, or the time of eruption (Harris et al. 2004; Wan et al. 2001). In that scenario,

and in the context of the described process, the clinical signs of disease will be observed on the

teeth. If the biofilm is transformed (because of the cariogenic diet) at the time that newly

erupted teeth with immature enamel are present then it is likely that caries will progress more

rapidly (Berkowitz 2003). If the biofilm is dysbiotic then more cariogenic bacteria will

dominate and SM will be detectable as a risk indicator.

1.2.6 Host factors

The expression of dental caries and its clinical signs will be modified by the characteristics of

the host. Although biochemical changes occur at the surface of the tooth it is important to

consider the caries process in the context of the individual and the interplay of all the related

body systems. Two of the key mechanisms by which host factors modify the caries process are

the excretion of saliva, the thickness and mineral density of the enamel and the shape of the

tooth (tooth morphology) (Fejerskov & Kidd 2009). Although salivary dysfunction is unusual

in young children, when saliva has poor quality or is of insufficient quantity, then the

antibacterial defences are reduced, acid and calcium buffering is decreased and substrate is

cleared less efficiently from the mouth. This places the child at greater risk of caries,

particularly if the diet is high in free-sugars. Medications, medical care (especially

17

radiotherapy) and immuno-suppressive diseases can affect the saliva and subsequently tip the

caries balance away from health (Featherstone 2004; Ribeiro et al. 2013).

In addition to saliva, the status of the ‘host’ can also be compromised if there is a disturbance

in the development of the dental enamel. Such a disturbance can result in dental enamel that is

low in mineral content (hypomineralisation) or decreased in thickness (hypoplasia) with

various clinical presentations (Elfrink et al. 2014). Enamel formation (amelogenesis) happens

during the early stages of growth and development from the second trimester of gestation until

late childhood, and the process can be interrupted or compromised by febrile disease,

medications, environmental pollutants (Ribeiro et al. 2013; and malnutrition (Alvarez 1995;

Fonseca, 2017; Silva et al. 2016). In the situation that amelogenesis is compromised leading to

hypomineralised enamel then the enamel may be more soluble in response to pH changes

within the biofilm (Ghanim et al. 2013). If amelogenesis results in tooth morphology that is

plaque retentive, such as in the case of deep pits and fissures, or in the presence of enamel

hypoplasia, then the biofilm may have a physical niche and not be disturbed on a regular basis

by tooth-brushing, however, a cariogenic state needs to be present (Fejerskov & Kidd, 2009).

While the environmental factors described above can influence the morphology of tooth

structure, there is also evidence that genetics can play a significant role in the expression of

caries patterns. Some genotypes associated with tooth formation have been identified as either

increasing the risk of or protecting from carious lesion formation (Abbasoglu 2015; Olszowski,

2015). Some estimates report that caries expression is up to 40% hereditary (Galton 2012),

while others suggest that the genetic components influence the caries pattern (Shaffer 2009).

Should the genotype play a significant role in moderating caries expression, then this could be

an important target for therapeutic intervention at a molecular or genetic level in the future.

1.2.7 Socio-behavioural factors and risk modelling

Although compromised tooth morphology increases caries risk, the actual clinical presentation

of dental caries is not primarily dependent on the shape of the tooth but rather the interaction

between the tooth surface and the social and behavioural habits of the individual (Harris et al.

2004; Frencken et al. 2012). The socio-behavioural factors represent the interlacing of all of

the different protective and pathological factors, the combination of which is unique to an

individual (and even a tooth surface) but may show trends within communities (Harris et al.

2004). In recognition of this, many efforts have been made to build predictive models to

identify those at a higher risk of developing sECC; however, the models do not appear to be

18

applicable between one population and the next. For example, the USA first started to consume

high quantities of ‘sugar swetened soft drinks in the 1940’s. During this period soda was being

consumed at just one-to-two servings-per-day and only by those of higher socioeconomic

background. A contemporaneous study reported that sugar sweetened beverages, despite their

sugar content, were not a risk indicator for caries experience (Bibby 1955). It could have been

that those who could afford to buy soda were also more likely to use fluoride toothpaste which

helped to maintain the caries balance (Tahmassebi et al. 2006). In contrast, consumption of

sugary drinks appears to be an important predictor for dental caries risk in 2015 in Vietnam,

where there has been a 109% increase in the consumption of sugar sweetened beverages from

2009 to 2014 (Silver, 2015.). In Vietnam oral hygiene practices are not universal, which

probably also increases the caries risk (Do et al. 2009). Although there has not been a national

oral health survey in Vietnam since the year 1999 it is thought that caries experience is

increasing with the increased access to sugar (Do et al. 2009).

This highlights the challenging scenario around preventing and managing ECC and the way in

which protective factors and pathological factors are moderated by an individual’s behaviour

and their environment. That tension has led to some debate as to whether dental caries can be

considered a communicable disease or whether it is a non-communicable disease (Tyas et al.

2000). WHO lists dental caries as a non-communicable disease and there have been recent

efforts to apply the strategies for managing chronic diseases to managing ECC (Edelstein &

Ng 2015). It is clear that dental caries is dependent upon commensal bacteria and dependent

upon exposure to fermentable carbohydrates (free-sugars) but whether the clinical signs of the

disease are observed will depend largely upon the drivers of the caries process and how they

exist in the context of an individual lifestyle (Fejerskov & Kidd 2009; Featherstone 2004;

Harris et al. 2004). Nearly all human beings carry the virulent strains of SM and all humans

consume carbohydrates, but not all humans will develop the clinical signs of dental caries until

the socio-behavioural aspects meet a threshold that tips the caries balance towards pathology

(Petersen et al. 2005; Wan et al. 2003). Whether or not dental caries develops depends upon

the presence of several common risk factors that also determine the presence of other non-

communicable diseases such as diabetes and cardiac disease. For this reason, from the

perspective of population health, dental caries is considered to be a non-communicable disease

(Tyas et al. 2000) and strategies for managing chronic diseases can be applied to the ECC

problem (Edelstein & Ng 2015).

19

Understanding the aetiology of ECC is important for the effective distribution of resources for

prevention and reduction of the burden of disease across a population. To date, the most

universal predictor of future dental caries experience is past dental caries experience; but that

is not useful for an infant who does not yet have teeth in the mouth (Fontana, 2015). In saying

that, once the teeth have erupted then the early signs of dental caries can be observed at a point

in which the damage is reversible. The early signs of ECC are WSL and when observed they

can predict future disease experience and indicate a child who may require further support to

stay cavity free.

1.3 Prevention of ECC

Dental caries (and ECC) is a multifactorial disease and so even with the appropriate application

of preventive dental products, at the most optimal frequency, ECC may still occur if the social

and behavioural determinants of the disease are not addressed. Traditionally, the mantra of oral

health education involved telling patients that they should see their dentists regularly, that they

should brush their teeth two times-per-day and that they should stop eating sugar. These ideas

have created some unrealistic expectations around the power of knowledge to result in

behavioural change and those expectations are thought to have resulted in victim blaming

(Frazier & Horowitz 1990). Oral health education has been shown to be effective in producing

a long-term increase in dental knowledge, but it has not been shown to have a long-term impact

on behaviour change or a reduction in caries risk or changes in diet (Sheiham & Watt 2000).

The nuance in the difference between oral health education and oral health promotion is

important, as the desired outcome of both is a change in behaviour and subsequently a lower

risk of disease. Oral health promotion involves transforming knowledge into action rather than

(just) the transfer of information (Pine 2013). To achieve this, oral health promotion is

something that needs to be interlaced with the social determinants of the disease (Chaffee &

Cheng 2014). The conventions of health promotion describe prevention in terms of primary

prevention which is “action taken prior to the onset of disease”, secondary prevention “as the

application of available measures to detect early departures from health and to introduce

appropriate treatment and interventions” and tertiary prevention as measures to reduce or

eliminate long-term impairments and disabilities, minimising suffering caused by existing

departures from good health and to promote patient’s adjustments to his/her condition”

(Mbawala et al. 2015).

20

Although there has been limited success, primary prevention through health promotion is

essential given the nature of the disease process, which is very hard to reverse once a lesion

has been initiated. Primary prevention should work on controlling the primary drivers of dental

caries, the diet and removing plaque from the teeth (oral hygiene). Secondary prevention,

through products such as topical fluorides, and tertiary prevention, through placement of dental

restorations, might slow the disease process, but ultimately that person will still be at risk of

the disease unless primary prevention has been successful (Litt et al. 1995). For primary

prevention to be successful it needs to be integrated within a variety of different settings and

the intervention must happen as early in the disease process as possible (Steffensen, 1990).

Some authors suggest that intervention should occur in an easily controlled environment such

as a school setting; others suggest that intervention should occur during the first two years of

life (Gussy et al. 2006; Vanobbergen et al. 2001; Schröder & Granath 1983); others advocate

for intervention during the first year of life (Rozier et al. 2003; Schroth et al. 2009); and still

others recommend focusing on the mother during pregnancy (Chaffee et al. 2013; Wigen &

Wang 2012). In any case, the intervention must be co-ordinated across all health disciplines

(Watt 2007) and the two most popular models for doing that are through maternal-child

services or through school institutions.

1.3.1 Maternal-child interventions

There is a need for consistency across health disciplines and given that the infant is largely

dependent on the adults in their life for the food that they consume and hygiene practices, the

mother is often the chosen target for such interventions. A number of interventions including

motivational interviewing for the mother have been reported to be effective in behaviour

change (Harrison et al. 2007; Kowash et al. 2000; Tiwari et al. 2015) as well as reducing the

proportion of children with detectible SM, which implies a less cariogenic biofilm (Jameisonet

al. 2018; Mitchell et al. 2009). Unfortunately, maternal-child interventions are not without

barriers such as lack of funding, the attitudes and behaviours of those mothers who might need

it most, the location and geographical barriers associated with those most at risk, the

knowledge, skills and enthusiasm of health providers in performing motivational interviewing,

and the failure of the health system to be patient-centred, making it acceptable from the ‘health

consumers’ point of view (Steffensen 1990). Despite the challenges, maternal-child

interventions appear to be a promising target in addressing the social determinants of disease

(Medeiros et al. 2015).

21

1.3.2 School based interventions

The key limitation around targeting services at preschool children is that accessing children in

a community setting can be logistically challenging. Focusing an intervention on an institution

where children and families regularly gather has the benefit of being logistically

straightforward and, in addition, an institution such as a school is a contained environment

which can be influenced and altered. There are a number of school-based interventions that

have been shown to reduce caries experience; these include the ‘Irene donut’ in Indonesia

(Adyatmaka et al. 2017) and the ‘Fit For Schools’ program in Cambodia, Vietnam, Laos, and

the Philippines (Duijster et al. 2017). Modifying an environment that a child participates in

alters their perception of what is ‘normal’ and has a higher chance of altering behaviour (Kay

& Locker 1996). Modifying the school environment allows for modification of multiple

determinants of dental caries all in one location. A school represents a place where there is an

intersection of multiple sectors of the community in a single environment.

1.3.3 Multi-sectorial cooperation

Many determinants for dental caries are shared among other non-communicable diseases and

efforts should be directed at common risk factors. In general, the common risk factors are listed

as smoking, alcohol ingestion, diet, exercise, and stress management. By addressing some of

these determinants which are related to oral health, there could also be a reduction in diabetes

and other non-communicable disease and duplication of effort can be avoided (Sheiham &

Watt 2000). To facilitate this, strong multi-sectorial relationships need to be fostered (Frazier

& Horowitz 1990). With respect to dental caries, early risk identification and subsequent early

intervention is important. Typically, the first contact after birth of an infant with the healthcare

system is with a primary health care worker who is administering vaccinations, and therefore

they are well placed to intervene (Schroth et al. 2009; Rozier et al. 2003; Douglass et al. 2001;

Litt et al. 1995; Schröder & Granath 1983).

Dental interventions, such as placement of Fluoride Varnish (FV), for preschool children can

be carried out successfully by non-dental professionals in a number of different settings (Isong

et al. 2011; Mouradian 2001; Slade et al. 2011; Smith & Riedford 2013; Steffensen 1990). This

demonstrates the opportunity to avoid duplication of effort and a possible opportunity for cross-

sector co-operation. In saying that, barriers exist, such as primary health care workers already

being burdened with multiple tasks (Slade et al. 2007). To manage the barriers for such

integration, it may be ideal to adopt a ‘three-pronged’ approach by engaging (first) the

22

community, (second) the health professionals and preschool teachers, and (third) the preschool

children (Cariño et al. 2003).

Reducing the barriers to accessing oral health care can lead to improvements in oral health. A

notable barrier can be whether the service is provided in a way that is accessible and culturally

appropriate (Schluter et al. 2007; Schroth et al. 2009) with services reoriented to meet the

needs of the target group (Albino & Tiwari 2016; Jamieson et al. 2018). In the case of ECC,

preschool children are usually reliant on their caregivers to first perceive the need and then

second to access care. Often there is no perceived need until the clinical manifestations of

dental caries are severe and primary prevention or even secondary prevention is no longer

feasible (Litt et al. 1995; Nowjack-Raymer & Gift 1990). A well-structured service that is

affordable and accessible will be used by preschool children and their caregivers and ultimately

lead to improvements in health outcomes (Gross–Panico & Freeman 2012; Binkley et al. 2005).

1.4 Prevention of ECC – managing the diet

One area that is particularly difficult to influence is that of diet; especially infant diets during

the transition period from a liquid diet to a solid diet. Many different health disciplines have an

interest in this aspect of life; paediatricians know that appropriate food at this age will lead to

better growth and development (Darapheak et al. 2013), dietitians recommend that a good

variety of food at this age will lead to a varied and healthy diet later on (Arimond & Ruel 2004),

and dentists know that if the frequency of exposure to free sugars is unfavourable then sECC

will result (Feldens et al. 2012). There is consensus around the benefits of decreased

consumption of free sugars and around the importance of introducing a variety of nutritious

foods (WHO, 2015).

1.5 Prevention of ECC – Fluorides and topical agents

The premier tool for secondary prevention of dental caries is the use of fluoridated products

and there is good evidence for the use of dental products containing fluoride in reducing the

caries burden (Featherstone 1999; Petersen & Lennon 2004). There are three mechanisms

through which caries prevention with fluoride is achieved (1) enhanced remineralisation

through modification of the ionic saturation within the biofilm fluid with relation to fluorapatite

(2) reduced demineralisation because the fluorapatite that is formed in the enamel of the tooth

after exposure is less soluble than the native impure hydroxyapatite, and (3) the antibacterial

23

activity in which high concentration fluoride inhibits cariogenic bacterial metabolism

(Fejerskov 2004)

The benefits of fluoride were recognised in the US Surgeon General’s report on reducing

caries; (Evans & Kleinman 2000). That report called for supervised tooth-brushing, systemic

fluoride supplementation, use of FV and gels, and intensive patient counselling (Allukian 2000;

Tinanoff & Reisine 2009). The WHO identifies access to fluoride as an important public health

priority for reducing disease burden (Petersen & Ogawa, 2016). Other authors have highlighted

the need for combined approaches using more than one fluoride source (Cariño et al. 2003;

Holve 2008; Schroth et al. 2009) and still others have laid out distinct guidelines for prevention

of ECC (Gussy et al. 2006). It is clear that more than one strategy is needed to reduce the

burden of dental caries and that use of dental products is just one of these options (Frencken et

al. 2012).

1.5.1 Fluoridated toothpaste

The WHO promotes fluoridated toothpaste in its Basic Package of Oral Care (BPOC) in which

it lists, among other things, access to low cost fluoride containing vehicles as being important

(Petersen 2003). In many cases, low cost fluoride has taken the form of fluoridated toothpaste

and so it is important to consider the common arguments around the use of fluoridated

toothpaste. The main risk associated with fluoride toothpastes in young children is that of dental

fluorosis, and so the concentration of fluoride in the toothpaste must balance the risk of dental

caries against the risk of dental fluorosis (Wong et al. 2011). The use of a higher concentration

fluoride toothpaste is well justified in a high-risk patient as there is evidence that a higher

concentration of fluoride in toothpaste will have a greater preventive effect compared with a

lower concentration (Marinho et al. 2003; Wright et al. 2014; dos Santos et al. 2013).

The research investigating the use of toothpaste in the preschool age-group has focused on the

method and the frequency of administration (Hsieh et al. 2014). Ingesting more than a ‘pea

sized’ amount of 1000 ppm toothpaste has been associated with mild fluorosis in some studies

(Wright et al. 2014). Ideally the toothpaste should be administered in a supervised setting

(Macintosh et al. 2010; Hölttä & Alaluusua 1992; Rong, et al. 2003), two times-per-day, and

the toothpaste slurry should be in the mouth for at least two minutes and expectorated without

rinsing or eating anything for two hours after brushing (Kaneilis 2000; Sjögren et al. 1995).

Unfortunately, as mentioned above, oral hygiene behaviours are associated with the socio-

economic characteristics of the child and some groups are unable to undertake these habits.

24

1.5.2 Fluoride Varnish

In the situation that a child or family are unable to undertake effective and regular oral hygiene

practices, or where there is additional caries risk, it may be beneficial to supplement with

professionally applied fluorides. Of the professionally applied fluorides, Fluoride Varnish (FV)

has been the most explored therapy with systematic reviews determining preventive fractions

between 5% and 63% (Mishara 2017). FV is effective for several reasons. Firstly, because it is

carried in a suspension that adheres to the tooth, the tooth surface is exposed to a high

concentration of fluoride ions for an extended period compared to other topical products; that

predictably facilitates the formation of fluorapatite within tooth enamel which is more resistant

to demineralisation than hydroxyapatite (Cochrane & Reynolds 2012). In addition, the fluoride

ions enhance remineralisation and inhibit biofilm production by inhibiting bacterial

metabolism (Fejerskov 2004).

Ideally, FV would be applied three to four times-per-year for optimal preventive effect in

preschool children and a dose response relationship has been demonstrated where by multiple

applications are more effective at reducing dental caries (Carvalho et al. 2010; Weintraub et al.

2006). At population level, the measured impact of FV is that it appears to slow the progression

of the lesion rather than stopping its initiation (Holve 2008) and it has been shown to be able

to reverse early (non-cavitated) enamel lesions in the presence of bioavailable calcium and

phosphate (Bawden 1998). This is achieved by increased mineral density in the outer surface

rather than in the deeper layers of the white spot lesion (Pithon et al. 2015). An antibacterial

effect has been demonstrated (Lippert et al. 2014), although a recent study demonstrated that

the type of microflora present does not change following FV therapy (Anderson et al. 2016).

Despite early promise for the use of FV therapy in the preschool age group, a number of studies

in the last 10-years have produced results which might suggest limited benefits. Of the studies

that report FV to be ineffective, some are underpowered (Memarpour 2015; Oliveira 2014);

other studies are in populations with a prevalence of ECC below 40% at around 2 years-of-age

(Agouropoulos et al. 2014; Anderson et al. 2016; Divaris et al. 2013; Jiang et al. 2014; Oliveira

et al. 2014). Some researchers recruited only participants who had already achieved 2 to 3

years-of-age with no clinical signs of dental caries (Muñoz-Millán et al. 2018; Tickle et al.

2017). Still other researchers have reported issues with FV being applied on a conditional basis

whereby clinicians didn’t always apply FV where it is needed or FV was only applied in

perceived high-risk cases, potentially biasing the measured outcome of the therapy at a

25

population level. If only those children who are very obviously at a high risk of developing

new carious lesions receive FV therapy, then other high and medium risk children who did not

receive FV therapy are still less likely to develop lesions based on the socio-behavioural

differences rather than being based on the benefits (or lack there-of) of the FV therapy,

(Fontana 2018). With this in mind, it seems possible that application of FV to lower caries risk

populations (prevalence <40% at 2-years) without systematic criteria for application is unlikely

to adjust the prevalence of carious lesions in preschool children. Despite this, it is likely that

even when the FV fails to prevent the initiation of a lesion, it will still slow the progression of

an existing lesion (Tickle et al. 2017).

1.5.3 Other fluoride delivery systems

Fluoride Varnish, along with other fluoride products, works in part by inhibiting bacterial

metabolism whilst present, enhancing remineralisation and reducing demineralisation. These

effects are primarily mediated by the presence of fluoride at the surface of the teeth rather than

by incorporation of fluoride into the enamel of developing teeth (Featherstone 1999; Tyas et

al. 2000). There are other ways of applying fluoride topically such as fluoride rinses and

fluoride gels; they both present the risk of ingestion and have not been shown to be as effective

as FV (Wong et al. 2011; Petersen & Lennon 2004).

It is clear that the vast majority of the benefits of fluoride are achieved topically, although there

is well documented success in administration of fluoride systematically. Fluoride drops or

fluoride tablets have been shown to reduce caries in high risk populations (Marya et al. 2012;

Featherstone 1999). In contrast to administration of fluoride tablets, Community Water

Fluoridation (CWF) has the advantage of not requiring any behaviour change for success and

when delivering it across a population, it can reduce inequalities between the rich and the poor

(Sisson, 2007). The economic benefit of CWF has been estimated in the most recent analysis

to realise a nine times cost benefit compared with restorative care (Moore et al. 2017).

While CWF is ingested, most of the benefit is achieved through the topical effect of fluoride at

the tooth surface and there is evidence that water with 0.7-1.0 ppm fluoride reaches therapeutic

levels in the oral cavity (Featherstone 1999). The benefits of systemic ingestion of fluoride on

an erupted tooth are less clear and some studies have demonstrated benefit in that scenario

(Cho et al 2014; Newbrun 2007; Buzalaf et al. 2011). The most recent systematic review found

a 35% reduction in the permanent dentition and a 26% reduction in carious lesions in the

26

primary dentition (Iheozor-Ejiofor et al. 2015). However, this review likely underestimated the

real benefits due to restrictive review criteria (Rugg-Gunn 2016).

That there is a large body of evidence for the benefits of CWF and that the therapy has been

demonstrated to reduce inequalities are two are the considerations that have led to CWF being

considered as one of the 10 great public health achievements of the 20th century (Rugg-Gunn

et al. 2016). In spite of this, CWF has been subject to allegations of harm as a result of ‘mass

medication’. There is evidence that CWF benefits at-risk communities the most; however, even

those in higher socioeconomic groups benefit. There is also evidence that there could be a 12%

increase in enamel defects (Iheozor-Ejiofor et al. 2015): some evidence suggests that the

particular type of enamel defect that increases is a diffuse opacity rather than those defects of

more concerning appearance (MacKay & Thomson 2005). Multiple systematic reviews have

demonstrated that CWF at therapeutic levels is not associated with adverse health outcomes

(Broadbent et al. 2015). Although it is clear that CWF and fluoride tablets benefit the

community, the problem is that high risk populations in a developing world setting do not

always have access to reticulated water systems and they may not be able to access fluoride

tablets which may represent a burden in terms of behaviour change and finances.

1.5.4 Bioavailable calcium and phosphate substrates

Although fluoride has long been the mainstay of preventive dentistry, it has limitations;

namely, the failure of most commercial products to provide bioavailable calcium and phosphate

ions to rebuild the enamel matrix and the failure to replace mineral at a subsurface level

(Cochrane et al. 2010). Although fluoride can enhance remineralisation, if there are no free

calcium and phosphate ions available in solution at the surface of the tooth then

remineralisation cannot progress after the salivary sourced supply is exhausted (Cochrane &

Reynolds 2012). Early attempts to address this deficit involved the delivery of non-soluble

calcium and phosphate and there was little success (Cochrane et al. 2010). The situation

changed when calcium and phosphate ions were delivered along with stabilising peptides that

could ensure that the appropriate substrate was bioavailable for remineralisation. One of the

more effective of such systems include Casein phosophopeptide-amorphous calcium phosphate

(CPP-ACP); this system stabilises the calcium and phosphate ions together in small clusters

(~2 nm diameter) which allows for the ability to create significantly increased concentrations

of calcium and phosphate, more than six times higher than in saliva. The presence of

bioavailable calcium and phosphate enhances remineralisation at the tooth surface (Cochrane

27

& Reynolds 2012). This remineralisation can be further enhanced by the presence of fluoride,

which has been shown to have an additive effect (Cochrane & Reynolds 2012; Nongonierma

& FitzGerald 2012). The in vitro studies for such products are able to demonstrate large

changes in mineral content within dental enamel and in vivo studies are increasing in number

(Nongonierma & FitzGerald 2012). Recent studies in preschool populations show positive

results after the use of products containing CPP-ACP; however, those studies demonstrated

some limitations in terms of cost and compliance which could compromise application at a

larger scale and for more disadvantaged populations (Plonka et al. 2013; Pukallus et al. 2013).

1.5.5 Antibacterial agents for preventing ECC

Aside from high concentration fluoride products, antibacterial treatments have been tested

regarding their efficacy in reducing ECC, including the use of povidone iodine, xylitol,

chlorhexidine (CHX) products and silver compounds. Small pilot samples have shown some

promising indications of a reduced risk of enamel demineralisation associated with application

of povidone iodine every two months (Lopez et al. 2002); however, swabbing the mouth with

povidone iodine at the end of dental treatment under general anaesthetic was not shown to be

effective in reducing future caries increment (Twetman 2008). Recent systematic analysis has

been able to demonstrate a moderate caries reduction after regular exposure to xylitol in the

form of chewing gums or lozenges. The caries reduction is thought to be achieved because the

xylitol, as a sugar substitute, is not able to be metabolised into acid thereby driving a less

acidogenic biofilm and potentially a more diverse biofilm (Marghalani et al. 2017). The results

of this therapy are still inconclusive.

The studies testing CHX have been split between application to the mother and application to

the child; they are also split between CHX varnishes and CHX rinses. One of the limitations of

CHX is that it is highly positively charged which means that it binds to the outer surface of a

biofilm rather than penetrating. It has been demonstrated that CHX does have an effect on SM

and there is mixed evidence about the ability of CHX to inhibit the transmission of SM from

mother to child (Twetman 2008). It is possible that the inconsistencies in terms of when the

mother receives the CHX therapy may have led to inconsistencies in results. In any case,

although CHX has been able to reduce SM levels in mothers, this reduction is transient and not

maintained long-term (Chaffee et al. 2013). In practice, that means that a transient reduction in

SM levels in the mother does not necessarily change the biofilm characteristics of the mother

or of the child (Gussy et al. 2006). There is some evidence that the application of CHX

28

varnishes for the child 2 to 4-times per year may reduce ECC when combined with professional

visits (Gisselsson et al. 2005). However, the evidence for CHX gels and rinses to prevent ECC

is lacking, likely because it does not address the drivers of the caries process (Tyas et al. 2000).

Another possibility for the benefits of CHX is in terms of Matrix Metalloproteinase (MMP)

inhibition leading to preservation of dentine structures (Breschi et al. 2010; Tjäderhane et al.

2013) In light of the current evidence, CHX may be effective as an adjunct therapy in an overall

preventive program, however, products containing fluoride and/or bio-available calcium and

phosphate are still the mainstay of ECC prevention.

The final group of antibacterial agents that have been widely studied, are silver compounds

such as silver nitrate (with sodium fluoride) and silver diammine fluoride (SDF). These agents

are thought to act both through the previously stated actions of fluoride and through MMP

inhibition (Zhao et al. 2018). Although there is some evidence that the biofilm activity is altered

after the tooth surface is treated by SDF, recent evidence suggests that the abundance of caries-

associated bacteria that are present on the surface of SDF treated carious lesions remains

unchanged (Milgrom 2018). While SDF has largely been examined for its ability to arrest the

progression of cavitated carious lesions (Contreras et al. 2017), there is also some evidence that

it will prevent lesion cavitation in First Permanent Molars (FPM) (Zhi et al. 2012). SDF appears

to be most useful in the management of cavitated lesions, rather than earlier (ICDAS 1 & 2)

lesions, where its affect is similar to that of FV (Duangthip et al. 2017).

1.6 Prevention of caries on First Permanent Molars (FPM)

The FPM represents the transition from a complete primary dentition into the mixed and

permanent dentition and a new chance to achieve a healthy and functional dentition over time.

There is some logic to support the targeting of FPM as the most susceptible tooth in the

permanent dentition and as a strategy for producing a higher proportion of adolescent children

with a sound dentition. The key methods for preventing cavitated carious lesions in FPM are

(a) the use of pit and fissure sealants (Ahovuo-Saloranta et al. 2017), (b) the use of the ‘cross

brushing’ technique with fluoride toothpaste (Braga et al. 2009; Frazão 2011), (c) the

application of FV and (d) the placement of SDF (Zhi et al. 2012). The therapy with the most

evidence is the use of pit and fissure sealants; however, some questions remain around the most

effective materials and the preventive benefits among populations with varying caries

experience (Ahovuo-Saloranta et al. 2017).

29

1.6.1 Pit and Fissure sealants

The general consensus suggests that fissure sealing or protection (with resin or GIC-based

products) placed on sound or non-cavitated carious lesions will prevent the progression or

initiation of carious lesions in pits and fissures (Wright et al. 2016). The mechanisms by which

resin-based fissure sealing works are by mechanical blockage of plaque retentive features and

creation of an easily cleansable surface where the bacteria in the base of the lesion is isolated

from the substrate. In the case of GIC fissure protection there is the mechanical protection of

the pits and fissures, as well as the exchange of fluoride ions at the material-tooth interface,

which, in vitro, has been demonstrated to create an enamel surface with increased fluorapatite

which is more resistant to acid dissolution following loss of the GIC material (Chow & Vogel

2001). Some guidelines describe the preferred timing of placement of a fissure sealant to be

within the first year after tooth eruption. After a tooth has been present in the mouth for 2-years

and remained healthy, some then suggest that there is limited benefit to be gained by placing a

sealant in a pit or fissure system provided that there is no change in caries risk behaviours

(Wright et al. 2016). Therefore, pit and fissure sealants do not address the factors that drive the

caries process and they do not change the characteristics associated with caries risk.

While there are some conventions around timing and risk criteria for placement of fissure

sealants, there are ongoing debates around which materials might perform better over time. It

is clear that there are differences in retention according to material type, with resin sealants

being retained longer; however, prevention of carious lesions in pits and fissures appears to be

comparable between resin and GIC sealants at 24 months (Zhang et al. 2013; Yeogepal 2015;

Ahovuo-Saloranta 2017). The evidence demonstrating preventive benefit for time periods

longer than 24 months is sparse, but studies do favour a similar conclusion (Ahovuo-Saloranta

2017) while long term studies strongly favour resin fissure sealant materials (Simonsen 2011).

The limitations in evidence come from the fact that the quality of studies using GIC materials

appears to be lower than for studies using resin-based materials; therefore, it is difficult to draw

strong conclusions (Ahovuo-Saloranta et al. 2017).

One of the key debates around choice of materials is whether retention of the material is needed

in order to achieve caries prevention. It is clear that resin sealants have higher retention rates

and that resin sealants rely on retention as the mechanism for effective prevention. In contrast,

while GIC sealants may be lost, remnants might remain at a microscopic level deep inside the

pit or fissure and that this might provide the preventive effect. Some microscopic imaging has

30

suggested that this might be the case (Hu et al. 2016). GIC forms a calcium bond at the tooth

surface and material is lost due to wear and cohesive failure which can leave remnants of GIC

attached to the tooth surface. The loss of GIC material from a caries-stable tooth surface could

also be a potential advantage for a patient who would then not be required to pay to maintain

that material in place as is the case for resin sealants.

Within the two main types of materials (resin composite or GlC) there are also variations of

the materials such as different generations of resins, high or low viscosity, or chemical and

light cured materials, and also some materials that are a mixture of GIC and resin (e.g.

Compomers and resin modified GICs); at this stage there are no clear additional benefits from

using these hybrid materials, since the body of evidence is still very limited (Ahovuo-Saloranta

et al. 2017).

1. 7 The situation in Cambodia

A number of differences exist between the Cambodian environment and that of countries where

the majority of research on caries prevention has been carried out. Cambodia is a developing

country located in the heart of South East Asia with a coast line to the Gulf of Thailand.

Cambodia has a rich history with its height of prosperity between the ninth and fifteenth

centuries; since then Cambodia has battled border disputes with Vietnam and Thailand,

colonialism by the French, and most recently a civil war and the genocidal Khmer Rouge

regime. In the early 1990’s the United Nations (UN) entered Cambodia in addition to numerous

Non-Governmental Organisations (NGO). In present day Cambodia, the presence of the UN is

much less obvious and there are more NGO’s per capita than any other nation on earth. Despite

this, an estimated 24% of the population remain below the poverty line and the government

continues to allow foreign interests to pillage the natural resources under the eyes of

government leaders who have been in power for nearly 30-years (Brinkley 2011). On the

positive side, Cambodia has been relatively politically stable in recent decades and economic

growth has been consistently strong; and more of the population have joined the growing

middle class, which is mostly located in Phnom Penh (Strangio 2014).

1.7.1 General health experiences in Cambodia

Consistent with economic development comes a transition from a traditional diet towards a diet

high in sugar and ultra-processed foods (Drewnowski, 2000). Along with this is a stark rise in

non-communicable diseases such as diabetes and hypertension (WHO 2017) with little

31

infrastructure to deal with chronic diseases and their consequences (WHO 2012). There have

been some improvements in nutrition as exemplified by the rate of stunting which went from

more than half of 5 year-old children down to two in five over the last 15-years; however,

progress seems to have plateaued in the last five years. There have also been considerable

improvements in the infant mortality rate which has decreased from 95 down to 28 per 1000

live births between the years 2000 and 2014 (National Institute of Statistics 2015).

In light of the consistently high rates of under-nutrition of Cambodian children, recent

interventions have focused on infant feeding practices. Recently the WHO has identified that

there have been some advances made in exclusive breast-feeding during the first six months in

Cambodia; however, complimentary feeding practices are less than optimal (Helen Keller

International 2015). Sugar is being introduced into the diet early, while meat and vegetables

are being introduced relatively late (Marriott et al. 2012; Desclaux & Alfieri 2009), and only

one-in-three Cambodian children are receiving the ‘minimally acceptable diet’ (National

Institute of Statistics 2015). Furthermore, a lack of variety in the diet has been associated with

high rates of stunting in Cambodian children (Darapheak et al. 2013). The effect of these diets

is also seen in the high rate of iron deficiency anaemia (more than half) and vitamin A

deficiency (two in five), which are among the highest rates in the world (Black et al. 2008;

National Institute of Statistics 2015).

1.7.2 Dental caries in Cambodia

Along with high rates of under-nutrition, Cambodia also has a high caries burden with nearly

all (97%) of children who are 6 years-of-age having cavitated, untreated carious lesions. This

is not surprising considering the low rates of tooth-brushing and the early introduction of

dietary sugar (Chher et al. 2016) along with reported on-demand feeding practices (Chher

2016; Turton 2015; Rasbridge & Kulig 1995). Furthermore, it is interesting to note that the

severity of caries for those in the younger age groups is much higher than that in the older age

groups who experienced childhood before the nutrition transition (Table 2). Not only is the

caries experience more severe for the younger age-groups but it can also be observed that there

has been an increase in caries experience between 1990 when the first sample was taken and

2011 when the second sample was taken.

32

Table 1.2 Dental caries experience in Cambodia between 1990 and 2011a

Age

group Year n Prevalence

Mean number of DMFT

DT MT FT DMFT

12

1990 288 79.5 1.5 0.1 0.0 1.6

2011 272 78.3 3.4 0.1 0.1 3.5

15

1990 265 69.1 1.6 0.1 0.0 1.7

2011 230 80.4 3.9 0.1 0.3 4.2

35-44

1990 270 89.5 3.4 2.5 0.5 6.4

2011 408 80.4 3.8 1.3 0.5 5.6

65-74 1990 253 93.3 3.3 13.0 0.1 16.4

60-84 2011 397 75.8 3.8 4.1 0.1 8.1

ataken from Chher et al. 2016.

In Cambodia, dental caries experience is almost ubiquitous, and lesions largely go untreated,

subsequently contributing to large impacts on Oral-Health-Related Quality-of-Life (OHRQoL)

(Turton et al. 2015) that cannot be improved with transient access to dental care (Turton et al.

2015). A recent study that investigated dental caries experience in an NGO-supported

community demonstrated that one in three children manifested the signs of dental caries at one

year-of-age, and that three in four manifested signs at two years-of-age (Bach 2014).

Furthermore, the research demonstrated that one of the major issues was inappropriate feeding

practices during the transitional feeding stage when new (highly fermentable) solid foods were

added to the diet (Bach 2014). This demonstrates that the caries experience in Cambodia is

severe, that it largely goes untreated and that the signs of the disease are detectable early in life.

33

1.7.3 Policies, politics, and practice in Cambodia

Untreated carious lesions are just one of the challenges associated with living or working in

Cambodia. In response to this, multiple local Cambodian NGO’s have attempted to meet the

dental needs of the population by providing dental care, often in a mobile dental setting. In the

scenario of transient mobile dental visits by foreign and local dentists and dental students, there

are a number of issues that could undermine the sustained management of the disease. Some

authors have identified the following mechanisms: lack of coverage and sustainability;

inappropriate volunteer actions; lack of evidence-based volunteer actions; lack of

accountability of volunteer and NGO actions; and lack of integration with and devaluation of

existing local healthcare systems and it’s workers (Holmgren 2011). There is a place for foreign

and local volunteers to assist in disease management; however, this must be with careful

consideration of the local health care systems, local health professionals and local cultural

characteristics.

1.7.3.1 Community health centres in Cambodia

In theory, all Cambodians have access to a Community Health Centre (CHC) at which they can

receive maternal care as well as vaccinations and vitamin supplementation for preschool

children. However, the rates of access to care vary from 58% for children of mothers with no

education, and 70% for children of educated mothers. It is interesting to note that the staff at

these CHCs may be on a wage less than US$100.00 per month depending on the region and

that their wage is supplemented by incentive payments to conduct particular interventions such

as vaccinations (Matsuoka et al. 2013). Anecdotal evidence also suggests that payment is

sometimes sought from the patients by the health care workers, and that patients may be

encouraged to seek treatment at the home of the primary health care provider where they can

charge a fee-for-service.

1.7.3.2 Dental nurses in Cambodia

Despite the challenges, Cambodia has produced one innovative response to the oral health care

situation through the creation of the ‘dental nurse’ scope of practice. Dental nurses undertake

a three-year associate degree program with education divided between primary health care

nursing and basic dental interventions (Glass Ionomer Cement (GIC) sealants, fluoride

applications, oral health education, local anaesthesia, scaling, Atraumatic Restorative

Technique (ART) restorations and extractions). Dental nurses are mostly employed at CHCs

34

in rural areas and can provide the BPOC as advocated by the WHO. Since dental nurses were

permitted to open their own private practices in 2016, and since most graduates are now not

employed by the Ministry of Health, their role in providing oral health promotion services in

the community is limited. There are now four dental nurse schools and most students come

from Phnom Penh or provincial capitals. One of the issues for those working for the provincial

health departments has been that the dental nurses are poorly resourced and supported, and

without the necessary materials or instruments they have been unable to fulfil this role (Chher

et al. 2009). Also, the evidence from the Cambodian National Oral Health Survey (CNOHS)

suggests that this is not a service that is frequented by the wider population.

1.7.3.3 Traditional Dentists

Perhaps, the most common practitioner that a mother and her child might seek dental treatment

from is the traditional dentist (Oral Health Bureau 2014), who practice dentistry usually out of

their homes, based on training received in an apprentice-like system where skills are passed

down, usually within families. In July 2015 there was a report that 200 such practices had been

closed down because they were not registered; however, it is likely that they later re-opened by

paying a registered university-trained dentist to use their name to register the clinic (Jackson et

al. 2015). This is worrying because one study showed that less than one in four of these

practices had appropriate sterilising procedures (Thim & Durward, 2016). Recently there has

been a shift at the Ministry of Health level that is looking to support the existing dental council

to regulate the practice of dentistry along with the other medical professions (USAID ASSIST

Project, 2014) and that might provide a pathway for these traditional dentists to either

demonstrate their competence or to stop any new traditional dental practices from opening.

1.7.3.4 University Trained Dentists

Part of the reason why traditional dentists are still thriving in Cambodia may relate to the very

low number of university trained dentists working in rural Cambodia, and the higher fees

charged by university-trained dentists. Following the fall of the Khmer Rouge, there were only

34 university-trained dentists left. Although the profession has been rebuilt over the last three

decades, generally, new graduate dentists prefer to practice in the larger urban centres

(Durward et al. 2016). The lure of the city is potent, and there is a growing dental tourism

industry (Nguon & Kay 2008). Consequently, conventional dental services are largely

unobtainable by the rural preschool child who is suffering from a severe disease burden.

1.7.3.5 The Oral Health Bureau

35

The Oral Health Bureau (OHB) sits within the Department of Preventive Medicine of the

Ministry of Health. The role of the OHB is to advise the Ministry of Health on policy and

practice for maintaining oral health at population level. The OHB has explicitly recognised that

that dental services are generally not accessible to those in need (Chher et al. 2016). There is

an acceptance that it is not possible to build an extensive network of dental services and even

if there could be, it would not address the underlying determinants of the disease and that

children would continue to manifest the signs of dental caries. Recommendations issued by the

OHB included working towards better access to low cost fluoride, focusing on preventing

dental caries from birth by targeting mothers and children, and setting up demonstration

programs to assess the cost-effectiveness of alternative preventive programs (Oral Health

Bureau 2014).

1.8 Summary of literature review

Addressing the issue of dental caries among young children at population level can be

challenging due to the social and structural barriers that exist in Cambodia. Despite this,

addressing ECC is a key public health issue since the disease is almost universal and has wide

ranging impacts for the individual children and for those around them. Children in Cambodia

have a severe experience of dental caries along with a number of other chronic health

conditions. The socio-political landscape for providing oral health care in Cambodia is

challenging and there is a need for better support and integration with the existing health

infrastructure. Although, a number of effective preventive modalities are available to address

the problem of ECC, the Oral Health Bureau recognises that more research is required to

demonstrate the feasibility and the effectiveness of such prevention-oriented programs in a

Cambodian setting (Chher et al. 2016).

36

1.9 References Chapter 1

Abiola Adeniyi, A., Eyitope Ogunbodede, O., Sonny Jeboda, O., & Morenike Folayan, O.

(2009). Do maternal factors influence the dental health status of Nigerian pre-school children?

International Journal of Paediatric Dentistry, 19(6), 448–454.

Adyatmaka, I, Adyatmaka, A, & Bachtiar, A. (2017). Dental Immunisation - Urgency to build

empowerment in community. Pennsylvania, USA: AEGIS Publications, LLC.

Agouropoulos, A., Twetman, S., Pandis, N., Kavvadia, K., & Papagiannoulis, L. (2014).

Caries-preventive effectiveness of fluoride varnish as adjunct to oral health promotion and

supervised tooth-brushing in preschool children: A double-blind randomized controlled trial.

Journal of Dentistry; Oxford, 42(10), 1277–1283.

https://doi.org/http://dx.doi.org.ezp.lib.unimelb.edu.au/10.1016/j.jdent.2014.07.020

Ahovuo-Saloranta, A., Forss, H., Walsh, T., Nordblad, A., Mäkelä, M., & Worthington, H. V.

(2017). Pit and fissure sealants for preventing dental decay in permanent teeth. In Cochrane

Database of Systematic Reviews. John Wiley & Sons, Ltd. Retrieved from

http://onlinelibrary.wiley.com.ezp.lib.unimelb.edu.au/doi/10.1002/14651858.CD001830.pub5

/abstract

Albino, J., & Tiwari, T. (2015). Preventing Childhood Caries: A Review of Recent Behavioral

Research. Journal of Dental Research. https://doi.org/10.1177/0022034515609034

Allukian Jr, M. (2000). The neglected epidemic and the surgeon general’s report: a call to

action for better oral health. American Journal of Public Health, 90(6), 843.

Alvarez, J. O. (1995). Nutrition, tooth development, and dental caries. The American Journal

of Clinical Nutrition, 61(2), 410S – 416S.

Anderson, M., Dahllöf, G., Twetman, S., Jansson, L., Bergenlid, A.-C., & Grindefjord, M.

(2016). Effectiveness of Early Preventive Intervention with Semiannual Fluoride Varnish

Application in Toddlers Living in High-Risk Areas: A Stratified Cluster-Randomized

Controlled Trial. Caries Research, 50(1), 17–23. https://doi.org/10.1159/000442675

Anderson, M., Grindefjord, M., Dahllöf, G., Dahlén, G., & Twetman, S. (2016). Oral

microflora in preschool children attending a fluoride varnish program: a cross-sectional study.

BMC Oral Health, 16(1), 130. https://doi.org/10.1186/s12903-016-0325-6

Arimond, M., & Ruel, M. T. (2004). Dietary diversity is associated with child nutritional status:

evidence from 11 demographic and health surveys. The Journal of Nutrition, 134(10), 2579–

2585.

Avila, W. M., Pordeus, I. A., Paiva, S. M., & Martins, C. C. (2015). Breast and Bottle Feeding

as Risk Factors for Dental Caries: A Systematic Review and Meta-Analysis. PLoS One; San

Francisco, 10(11), e0142922.

https://doi.org/http://dx.doi.org.ezp.lib.unimelb.edu.au/10.1371/journal.pone.0142922

Bach K. (2014) The Oral Health of Cambodian Pre-School-Aged Children. DCD Thesis.

Faculty of Medicine, Dentistry and Health Science, The University of Melbourne.

https://doi.org/http:/dx.doi.org.ezp.lib.unimelb.edu.au/10.1371/journal.pone.0142922

37

Bawden, J. W. (1998). Fluoride varnish: a useful new tool for public health dentistry. Journal

of Public Health Dentistry, 58(4), 266–269.

Benzian, H., Monse, B., Heinrich-Weltzien, R., Hobdell, M., Mulder, J., & van Palenstein

Helderman, W. (2011). Untreated severe dental decay: a neglected determinant of low Body

Mass Index in 12-year-old Filipino children. BMC Public Health, 11(1), 558.

Berkowitz, R. J. (2003). Causes, treatment and prevention of early childhood caries: a

microbiologic perspective. Journal of the Canadian Dental Association, 69(5), 304–307.

Bibby, B. G. (1955). Effect of sugar content of foodstuffs on their caries-producing

potentialities. The Journal of the American Dental Association, 51(3), 293–306.

https://doi.org/10.14219/jada.archive.1955.0186

Black, R. E., Allen, L. H., Bhutta, Z. A., Caulfield, L. E., De Onis, M., Ezzati, M. (2008).

Maternal and child undernutrition: global and regional exposures and health consequences. The

Lancet, 371(9608), 243–260.

Boggess, K. A., & Edelstein, B. L. (2006). Oral health in women during preconception and

pregnancy: implications for birth outcomes and infant oral health. Maternal and Child Health

Journal, 10(1), 169–174.

Bowen, W. H., & Lawrence, R. A. (2005). Comparison of the cariogenicity of cola, honey,

cow milk, human milk, and sucrose. Pediatrics, 116(4), 921–926.

Braga, M. M., Mendes, F. M., De Benedetto, M. S., & Imparato, J. C. P. (2009). Effect of silver

diammine fluoride on incipient caries lesions in erupting permanent first molars: a pilot study.

Journal of Dentistry for Children, 76(1), 28–33.

Breschi, L., Mazzoni, A., Nato, F., Carrilho, M., Visintini, E., Tjäderhane, L., Pashley, D. H.

(2010). Chlorhexidine stabilizes the adhesive interface: A 2-year in vitro study. Dental

Materials, 26(4), 320–325. https://doi.org/10.1016/j.dental.2009.11.153

Brinkley, J. (2011). Cambodia’s curse. Foreign Affairs, 88. Retrieved from

https://www.foreignaffairs.org/articles/cambodia/2009-03-01/cambodias-curse

Broadbent, J. M., & Thomson, W. M. (2005). For debate: problems with the DMF index

pertinent to dental caries data analysis. Community Dentistry and Oral Epidemiology, 33(6),

400–409.

Buzalaf, M.A.R., Pessan, J.P., Honório, H.M. and Ten Cate, J.M., 2011. Mechanisms of action

of fluoride for caries control. In Fluoride and the oral environment (Vol. 22, pp. 97-114).

Karger Publishers.

Cariño, K. M. G., Shinada, K., & Kawaguchi, Y. (2003). Early childhood caries in northern

Philippines. Community Dentistry and Oral Epidemiology, 31(2), 81–89.

Carvalho, D. M., Salazar, M., Oliveira, B. H. de, & Coutinho, E. S. F. (2010). Fluoride

varnishes and decrease in caries incidence in preschool children: a systematic review. Revista

Brasileira de Epidemiologia, 13(1), 139–149.

38

Chaffee, B. W., & Cheng, A. (2014). Global Research Trends on Early-Life Feeding Practices

and Early Childhood Caries: A Systematic Review. Journal of Oral Diseases, 2014. Retrieved

from http://www.hindawi.com/journals/jod/2014/675658/abs/

Chaffee, B. W., Feldens, C. A., & Vítolo, M. R. (2014). Association of long-duration

breastfeeding and dental caries estimated with marginal structural models. Annals of

Epidemiology, 24(6), 448–454.

Chaffee, B. W., Gansky, S. A., Weintraub, J. A., Featherstone, J. D. B., & Ramos-Gomez, F.

J. (2013). Maternal oral bacterial levels predict early childhood caries development. Journal of

Dental Research, 93(3), 238-244.

Chher, T., Hak, S., Courtel, F., & Durward, C. (2009). Improving the provision of the Basic

Package of Oral Care (BPOC) in Cambodia. International Dental Journal, 59(1), 47–52.

Chher T, Turton B, Hak S, Beltran E, Courtel F, Durward C, & Hobdell M. (2016). Dental

caries experience in Cambodia: Findings from the 2011 Cambodia National Oral Health

Survey. Journal of International Oral Health 1(1), 1-16.

Cho, H.J., Jin, B.H., Park, D.Y., Jung, S.H., Lee, H.S., Paik, D.I. and Bae, K.H., 2014. Systemic

effect of water fluoridation on dental caries prevalence. Community Dentistry and Oral

Epidemiology, 42(4), pp.341-348.

Chow, L. C., & Vogel, G. L. (2001). Enhancing remineralization. Operative Dentistry, pp 27–

38.

Cochrane, N. J., Cai, F., Huq, N. L., Burrow, M. F., & Reynolds, E. C. (2010). New approaches

to enhanced remineralization of tooth enamel. Journal of Dental Research, 89(11), 1187–1197.

Cochrane, N. J., & Reynolds, E. C. (2012). Calcium phosphopeptides—mechanisms of action

and evidence for clinical efficacy. Advances in Dental Research, 24(2), 41–47.

Colen, C. G., & Ramey, D. M. (2014). Is breast truly best? Estimating the effects of

breastfeeding on long-term child health and wellbeing in the United States using sibling

comparisons. Social Science & Medicine, 109, 55–65.

Contreras, V., Toro, M. J., Elías-Boneta, A. R., & Encarnación-Burgos, A. (2017).

Effectiveness of silver diamine fluoride in caries prevention and arrest: a systematic literature

review. General Dentistry. Retrieved from

http://www.agd.org/media/391570/GenDent_MJ17_Contreras.pdf

Darapheak, C., Takano, T., Kizuki, M., Nakamura, K., & Seino, K. (2013). Consumption of

animal source foods and dietary diversity reduce stunting in children in Cambodia.

International Archives of Medicine, 6(1), 29.

Desclaux, A., & Alfieri, C. (2009). Counseling and choosing between infant-feeding options:

overall limits and local interpretations by health care providers and women living with HIV in

resource-poor countries (Burkina Faso, Cambodia, Cameroon). Social Science & Medicine,

69(6), 821–829.

39

Divaris, K., Lee, J. Y., Baker, A. D., & Vann Jr, W. F. (2011). The relationship of oral health

literacy with oral health-related quality of life in a multi-racial sample of low-income female

caregivers. Health Qual Life Outcomes, 9(108), 1172–1186.

Divaris, K., Preisser, J. S., & Slade, G. D. (2013). Surface-Specific Efficacy of Fluoride

Varnish in Caries Prevention in the Primary Dentition: Results of a Community Randomized

Clinical Trial. Caries Research, 47(1), 78–87. https://doi.org/10.1159/000344015

Do, L. G., Spencer, A. J., Roberts-Thomson, K., Trinh, H. D., & Nguyen, T. T. (2009). Oral

health status of Vietnamese children: findings from the National Oral Health Survey of

Vietnam 1999. Asia-Pacific Journal of Public Health. Retrieved from

http://aph.sagepub.com/content/early/2009/07/02/1010539509340047.abstract

dos Santos, A. P. P., Nadanovsky, P., & de Oliveira, B. H. (2013). A systematic review and

meta-analysis of the effects of fluoride toothpastes on the prevention of dental caries in the

primary dentition of preschool children. Community Dentistry and Oral Epidemiology, 41(1),

1–12. https://doi.org/10.1111/j.1600-0528.2012.00708.x

Drewnowski, A. (2000). Nutrition transition and global dietary trends. Nutrition, 16(7–8), 486–

487. https://doi.org/10.1016/S0899-9007(00)00295-1

Duangthip, D., Gao, S. S., Lo, E. C. M., & Chu, C. H. (2017). Early childhood caries among

5-to 6-year-old children in Southeast Asia. International Dental Journal, 67(2), 98–106.

Duangthip, D., Wong, M. C. M., Chu, C. H., & Lo, E. C. M. (2017). Caries arrest by topical

fluorides in preschool children: 30-month results. Journal of Dentistry 1(70), 74-79.

Duijster, D., Monse, B., Dimaisip-Nabuab, J., Djuharnoko, P., Heinrich-Weltzien, R., Hobdell,

M., … Benzian, H. (2017). “Fit for school” – a school-based water, sanitation and hygiene

programme to improve child health: Results from a longitudinal study in Cambodia, Indonesia

and Lao PDR. BMC Public Health, 17, 302. https://doi.org/10.1186/s12889-017-4203-1

Thim S, Durward C. (2016). Traditional dentists practices in Siem Reap Cambodia Dental

Journal 6, 6-12.

Edelstein, B. L., & Ng, M. W. (2015). Chronic Disease Management Strategies of Early

Childhood Caries: Support from the Medical and Dental Literature. Pediatric Dentistry, 37(3),

281–287.

Elfrink, M. E., Moll, H. A., Kiefte-de Jong, J. C., Jaddoe, V. W., Hofman, A., Jacob, M., &

Veerkamp, J. S. (2014). Pre-and postnatal determinants of deciduous molar hypomineralisation

in 6-year-old children. The generation R study. Retrieved from

http://dx.plos.org/10.1371/journal.pone.0091057

Erickson, P. R., & Mazhari, E. (1999). Investigation of the role of human breast milk in caries

development. Pediatric Dentistry, 21, 86–90.

Evans, E. W., Hayes, C., Palmer, C. A., Bermudez, O. I., Naumova, E. N., Cohen, S. A., &

Must, A. (2013). Development of a pediatric cariogenicity index. Journal of Public Health

Dentistry, 73(3), 179–186. https://doi.org/10.1111/jphd.12009

https://doi.org/10.1111/jphd.12009

40

Evans, C.A. and Kleinman, D.V., 2000. The Surgeon General's report on America's oral health:

opportunities for the dental profession. The Journal of the American Dental Association,

131(12), pp.1721-1728.

Featherstone, J. D. (1999). Prevention and reversal of dental caries: role of low level fluoride.

Community Dentistry and Oral Epidemiology, 27(1), 31–40.

Featherstone, J. D. B. (2004). The continuum of dental caries—evidence for a dynamic disease

process. Journal of Dental Research, 83(suppl 1), C39–C42.

Fejerskov, O. (2004). Changing Paradigms in Concepts on Dental Caries: Consequences for

Oral Health Care. Caries Research, 38(3), 182–191.

Fejerskov, O., & Kidd, E. (2009). Dental caries: the disease and its clinical management. John

Wiley & Sons. Retrieved from

https://books.google.com.au/books?hl=en&lr=&id=fZfXWhSmG1UC&oi=fnd&pg=PR1&dq

=kidd+caries+cause&ots=OwH7O70sEI&sig=fjXX1b3C8PDUf8WgPAZOmEH0Fx0

Feldens, C. A., Rodrigues, P. H., de Anastácio, G., Vítolo, M. R., & Chaffee, B. W. (n.d.).

Feeding frequency in infancy and dental caries in childhood: a prospective cohort study.

International Dental Journal. 68(2), 113-121.

Feldens, C. A., Rodrigues, P. H., Rauber, F., Chaffee, B. W., & Vitolo, M. R. (2012). Food

expenditures, cariogenic dietary practices and childhood dental caries in southern Brazil.

Caries Research, 47(5), 373–381.

Fewtrell, M. S., Morgan, J. B., Duggan, C., Gunnlaugsson, G., Hibberd, P. L., Lucas, A., &

Kleinman, R. E. (2007). Optimal duration of exclusive breastfeeding: what is the evidence to

support current recommendations? The American Journal of Clinical Nutrition, 85(2), 635S –

638S.

Finlayson, T. L., Siefert, K., Ismail, A. I., & Sohn, W. (2007). Psychosocial factors and early

childhood caries among low-income African–American children in Detroit. Community

Dentistry and Oral Epidemiology, 35(6), 439–448.

Fonseca, M. A. da. (2017). Malnutrition and Oral Health in Children. Current Oral Health

Reports, 4(2), 92–96. https://doi.org/10.1007/s40496-017-0130-6

Fontana, M. (2015). The Clinical, Environmental, and Behavioral Factors That Foster Early

Childhood Caries: Evidence for Caries Risk Assessment. Pediatric Dentistry, 37(3), 217–225.

Fontana, M., Eckert, G.J., Keels, M.A., Jackson, R., Katz, B., Levy, B.T. and Levy, S.M., 2018.

Fluoride Use in Health Care Settings: Association with Children’s Caries Risk. Advances in

Dental Research, 29(1), pp.24-34.

Frazão, P. (2011). Effectiveness of the bucco-lingual technique within a school-based

supervised tooth-brushing program on preventing caries: a randomized controlled trial. BMC

Oral Health, 11(1), 11.

Frencken, J. E., de Amorim, R. G., Faber, J., & Leal, S. C. (2011). The Caries Assessment

Spectrum and Treatment (CAST) index: rational and development. International Dental

Journal, 61(3), 117–123.

41

Frencken, J. E., Peters, M. C., Manton, D. J., Leal, S. C., Gordan, V. V., & Eden, E. (2012).

Minimal intervention dentistry for managing dental caries–a review. International Dental

Journal, 62(5), 223–243.

Gatica, G., Barros, A.J., Madruga, S., Matijasevich, A. and Santos, I.S., 2012. Food intake

profiles of children aged 12, 24 and 48 months from the 2004 Pelotas (Brazil) birth cohort: an

exploratory analysis using principal components. International Journal of Behavioral Nutrition

and Physical Activity, 9(1), p.43.

Ghanim, A., Manton, D., Marino, R., Morgan, M., & Bailey, D. (2013). Prevalence of

demarcated hypomineralisation defects in second primary molars in Iraqi children.

International Journal of Paediatric Dentistry, 23(1), 48–55.

Gisselsson, H., Emilson, C.-G., Birkhed, D., & Björn, A.-L. (2005). Approximal caries

increment in two cohorts of schoolchildren after discontinuation of a professional flossing

program with chlorhexidine gel. Caries Research, 39(5), 350–356.

Gussy, M. G., Waters, E. G., Walsh, O., & Kilpatrick, N. M. (2006). Early childhood caries:

current evidence for aetiology and prevention. Journal of Paediatrics and Child Health, 42(1-

2), 37–43.

Hallett, K. B., & O’Rourke, P. K. (2003). Social and behavioural determinants of early

childhood caries. Australian Dental Journal, 48(1), 27–33.

Harrison, R., Benton, T., Everson-Stewart, S., & Weinstein, P. (2007). Effect of Motivational

Interviewing on Rates of Early Childhood Caries: A Randomized Trial. Pediatric Dentistry,

29(1), 16–22.

Harris, R., Nicoll, A. D., Adair, P. M., & Pine, C. M. (2004). Risk factors for dental caries in

young children: a systematic review of the literature. Community Dental Health, 21(1), 71–85.

Helen Keller International. (2015). Assessment of Promotion of Foods Consumed by Infants

and Young Children in Phnom Penh: Assessment and Research on Child Feeding (ARCH) –

Cambodia Country Report. Retrieved September 30, 2016, from http://archnutrition.org/wp-

content/uploads/2016/07/Assessment-of-Promotion-of-Foods-Consumed-by-Infants-and-

Young-Children-in-Phnom-Penh-2.pdf

Hölttä, P., & Alaluusua, S. (1992). Effect of supervised use of a fluoride toothpaste on caries

incidence in pre-school children. International Journal of Paediatric Dentistry, 2(3), 145–149.

Holve, S. (2008). An observational study of the association of fluoride varnish applied during

well child visits and the prevention of early childhood caries in American Indian children.

Maternal and Child Health Journal, 12(1), 64–67.

Hooley, M., Skouteris, H., Boganin, C., Satur, J., & Kilpatrick, N. (2012a). Body mass index

and dental caries in children and adolescents: a systematic review of literature published 2004

to 2011. Syst Rev, 1(1), 57.

Hooley, M., Skouteris, H., Boganin, C., Satur, J., & Kilpatrick, N. (2012b). Parental influence

and the development of dental caries in children aged 0–6 years: a systematic review of the

literature. Journal of Dentistry, 40(11), 873–885.

42

Hsieh, H.-J., Huang, S.-T., Tsai, C.-C., & Hsiao, S.-Y. (2014). Tooth-brushing habits and risk

indicators of severe early childhood caries among aboriginal Taiwanese. Asia-Pacific Journal

of Public Health, 26(3), 238–247.

Hu, X., Zhang, W., Fan, M., Mulder, J., & Frencken, J. E. (2016). Frequency of remnants of

sealants left behind in pits and fissures of occlusal surfaces after 2 and 3 years. Clinical Oral

Investigations, 1–7. https://doi.org/10.1007/s00784-016-1766-7

Iheozor-Ejiofor, Z., Worthington, H.V., Walsh, T., O'Malley, L., Clarkson, J.E., Macey, R.,

Alam, R., Tugwell, P., Welch, V. and Glenny, A.M., 2015. Cochrane Review of Water

Fluoridation. Fluoride, 48(3), p.269.

Ismail, A. I. (2003). Determinants of health in children and the problem of early childhood

caries. Pediatric Dentistry, 25(4), 328–333.

Ismail, A. I., Sohn, W., Tellez, M., Amaya, A., Sen, A., Hasson, H., & Pitts, N. B. (2007). The

International Caries Detection and Assessment System (ICDAS): an integrated system for

measuring dental caries. Community Dentistry and Oral Epidemiology, 35(3), 170–178.

Ismail, A. I., Tanzer, J. M., & Dingle, J. L. (1997). Current trends of sugar consumption in

developing societies. Community Dentistry and Oral Epidemiology, 25(6), 438–443.

Jackson Will, Kaliyann Thia, Moung Vandy, Putting the bite on ‘traditional dentists’, 27 March

2015, Phnom Penh Post: https://www.phnompenhpost.com/post-weekend/putting-bite-

traditional-dentists

Jamieson, L., Smithers, L., Hedges, J., Parker, E., Mills, H., Kapellas, K., Lawrence, H.P.,

Broughton, J.R. and Ju, X., 2018. Dental Disease Outcomes Following a 2-Year Oral Health

Promotion Program for Australian Aboriginal Children and Their Families: A 2-Arm Parallel,

Single-blind, Randomised Controlled Trial. EClinicalMedicine, 1, 43-50.

Jiang, E. M., Lo, E. C. M., Chu, C. H., & Wong, M. C. M. (2014). Prevention of early childhood

caries (ECC) through parental tooth-brushing training and fluoride varnish application: A 24-

month randomized controlled trial. Journal of Dentistry; Oxford, 42(12), 1543–1550.

https://doi.org/http://dx.doi.org.ezp.lib.unimelb.edu.au/10.1016/j.jdent.2014.10.002

Kaneilis, M. J. (2000). Caries risk assessment and prevention: strategies for Head Start, Early

Head Start, and WIC. Journal of Public Health Dentistry, 60(3), 210–217.

Ka, S., & Ca, P. (2012). Childhood dental caries and childhood obesity. Different problems

with overlapping causes. American Journal of Dentistry, 25(1), 59–64.

Kassebaum, N. J., Bernabé, E., Dahiya, M., Bhandari, B., Murray, C. J. L., & Marcenes, W.

(2015). Global Burden of Untreated Caries A Systematic Review and Metaregression. Journal

of Dental Research, 0022034515573272.

Kay, E. J., & Locker, D. (1996). Is dental health education effective? A systematic review of

current evidence. Community Dentistry and Oral Epidemiology, 24(4), 231–235.

https://doi.org/10.1111/j.1600-0528.1996.tb00850.x

Khitdee, C. (2017). The Epidemiology of Early Childhood Caries. Thai Dental Public Health

Journal, 22(supp 1).

https://doi.org/10.1007/s00784-016-1766-7

https://www.phnompenhpost.com/post-weekend/putting-bite-traditional-dentists

https://www.phnompenhpost.com/post-weekend/putting-bite-traditional-dentists

43

Kidd, E. A. M., & Fejerskov, O. (2004). What constitutes dental caries? Histopathology of

carious enamel and dentin related to the action of cariogenic biofilms. Journal of Dental

Research, 83(suppl 1), C35–C38.

Klock, B., & Krasse, B. (1979). A comparison between different methods for prediction of

caries activity. European Journal of Oral Sciences, 87(2), 129–139.

https://doi.org/10.1111/j.1600-0722.1979.tb00664.x

Kowash, M. B., Pinfield, A., Smith, J., & Curzon, M. E. J. (2000). Dental health education:

effectiveness on oral health of a long-term health education programme for mothers with young

children. British Dental Journal, 188(4), 201–205.

K Pattanaporn, P. S. (2012). Mode of delivery, mutans streptococci colonization, and early

childhood caries in three- to five-year-old Thai children. Community Dentistry and Oral

Epidemiology, 41(3). https://doi.org/10.1111/cdoe.12013

Kuthy, R. A., Siegal, M. D., & Wulf, C. A. (1997). Establishing maternal and child health data

collection priorities for state and local oral health programs. Journal of Public Health Dentistry,

57(4), 197–205.

Leme, A. P., Koo, H., Bellato, C. M., Bedi, G., & Cury, J. A. (2006). The role of sucrose in

cariogenic dental biofilm formation—new insight. Journal of Dental Research, 85(10), 878–

887.

Lippert, F., Hara, A. T., Martinez-Mier, E. A., & Zero, D. T. (2014). Laboratory Investigations

Into the Potential Anticaries Efficacy of Fluoride Varnishes. Pediatric Dentistry, 36(4), 291–

295.

Li, Y., & Caufield, P. W. (1995). The fidelity of initial acquisition of mutans streptococci by

infants from their mothers. Journal of Dental Research, 74(2), 681–685.

Lönnerdal, B. (2003). Nutritional and physiologic significance of human milk proteins. The

American Journal of Clinical Nutrition, 77(6), 1537S – 1543S.

Lopez, L., Berkowitz, R., Spiekerman, C., & Weinstein, P. (2002). Topical antimicrobial

therapy in the prevention of early childhood caries: a follow-up report. Pediatric Dentistry,

24(3), 204–206.

Ludwig, D. S., & Nestle, M. (2008). Can the food industry play a constructive role in the

obesity epidemic? Journal of American Medical Association, 300(15), 1808–1811.

Lustig, R. H., Schmidt, L. A., & Brindis, C. D. (2012). Public health: The toxic truth about

sugar. Nature, 482(7383), 27–29.

Macintosh, A. C., Schroth, R. J., Edwards, J., Harms, L., Mellon, B., & Moffatt, M. (2010).

The impact of community workshops on improving early childhood oral health knowledge.

Pediatric Dentistry, 32(2), 110–117.

Mackay, T.D. and Thomson, W., 2005. Enamel defects and dental caries among Southland

children. New Zealand Dental Journal, 101(2), pp.35-43.

44

Marcenes, W., Kassebaum, N. J., Bernabé, E., Flaxman, A., Naghavi, M., Lopez, A., &

Murray, C. J. L. (2013). Global Burden of Oral Conditions in 1990-2010 A Systematic

Analysis. Journal of Dental Research, 0022034513490168.

Marghalani, A. A., Guinto, E., Phan, M., Dhar, V., & Tinanoff, N. (2017). Effectiveness of

Xylitol in Reducing Dental Caries in Children. Pediatric Dentistry, 39(2), 103–110.

Marinho, V. C., Higgins, J., Logan, S., & Sheiham, A. (2003). Fluoride toothpastes for

preventing dental caries in children and adolescents. The Cochrane Library. Retrieved from

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD002278/full

Marriott, B. P., White, A., Hadden, L., Davies, J. C., & Wallingford, J. C. (2012). World Health

Organization (WHO) infant and young child feeding indicators: associations with growth

measures in 14 low-income countries. Maternal & Child Nutrition, 8(3), 354–370.

https://doi.org/10.1111/j.1740-8709.2011.00380.x

Martins-Júnior, P. A., Vieira-Andrade, R. G., Corrêa-Faria, P., Oliveira-Ferreira, F., Marques,

L. S., & Ramos-Jorge, M. L. (2013). Impact of early childhood caries on the oral health-related

quality of life of preschool children and their parents. Caries Research, 47(3), 211–218.

Marya, C. M., Ashokkumar, B. R., Dhingra, S., Dahiya, V., & Gupta, A. (2012). Exposure to

High-Fluoride Drinking Water and Risk of Dental Caries and Dental Fluorosis in Haryana,

India. Asia-Pacific Journal of Public Health, 1010539512460270.

Matsuoka, S., Obara, H., Nagai, M., Murakami, H., & Lon, R. C. (2013). Performance-based

financing with GAVI health system strengthening funding in rural Cambodia: a brief

assessment of the impact. Health Policy and Planning, czt030.

https://doi.org/10.1093/heapol/czt030

Medeiros, P. B., Otero, S. A., Frencken, J. E., Bronkhorst, E. M., & Leal, S. C. (2015).

Effectiveness of an oral health program for mothers and their infants. International Journal of

Paediatric Dentistry, 25(1), 29–34.

Memarpour M, Fakhraei E, Dadaein S, Vossoughi M. Efficacy of Fluoride Varnish and Casein

Phosphopeptide-Amorphous Calcium Phosphate for Remineralization of Primary Teeth: A

Randomized Clinical Trial. Med Princ Pract. 2015 May;24(3):231–7. Menon. (n.d.). Parental

stress as a predictor of early childhood caries among preschool children in India. International

Journal of Paediatric Dentistry, 23(3), 160–165. https://doi.org/10.1111/j.1365-

263X.2012.01238.x

Ministry of Health. 2010. Our Oral Health: Key findings of the 2009 New Zealand Oral Health

Survey. Wellington: Ministry of Health

Mitchell, S. C., Ruby, J. D., Moser, S., Momeni, S., Smith, A., Osgood, R., Childers, N. (2009).

Maternal transmission of mutans streptococci in severe-early childhood caries. Pediatric

Dentistry, 31(3), 193.

Moffat, S. M., Foster Page, L. A., & Thomson, W. M. (2017). New Zealand’s School Dental

Service over the Decades: Its Response to Social, Political, and Economic Influences, and the

Effect on Oral Health Inequalities. Frontiers in Public Health, 5, 177.

https://doi.org/10.3389/fpubh.2017.00177

https://doi.org/10.1111/j.1365-263X.2012.01238.x

https://doi.org/10.1111/j.1365-263X.2012.01238.x

45

Monse, B., Heinrich-Weltzien, R., Benzian, H., Holmgren, C., & van Palenstein Helderman,

W. (2010). PUFA–An index of clinical consequences of untreated dental caries. Community

Dentistry and Oral Epidemiology, 38(1), 77–82.

Moodie, R., Stuckler, D., Monteiro, C., Sheron, N., Neal, B., Thamarangsi, T., (2013). Profits

and pandemics: prevention of harmful effects of tobacco, alcohol, and ultra-processed food and

drink industries. The Lancet, 381(9867), 670–679.

Moore, D., Poynton, M., Broadbent, J. M., & Thomson, W. M. (2017). The costs and benefits

of water fluoridation in NZ. BMC Oral Health, 17(1), 134. https://doi.org/10.1186/s12903-

017-0433-y

Muñoz-Millán, P., Zaror, C., Espinoza-Espinoza, G., Vergara-Gonzalez, C., Muñoz, S., Atala-

Acevedo, C., & Martínez-Zapata, M. J. (2018). Effectiveness of fluoride varnish in preventing

early childhood caries in rural areas without access to fluoridated drinking water: A randomized

control trial. Community Dentistry and Oral Epidemiology, 46(1), 63–69.

https://doi.org/10.1111/cdoe.12330

National Institute of Statistics, Directorate General for Health, and ICF International (2015).

Cambodia Demographic and Health Survey 2014. Phnom Penh, Cambodia, and Rockville,

Maryland, USA: National Institute of Statistics, Directorate General for Health, and ICF

International.

Neves, P. A. M., Ribeiro, C. C. C., Tenuta, L. M. A., Leitão, T. J., Monteiro-Neto, V., Nunes,

A. M. M., & Cury, J. A. (2016). Breastfeeding, Dental Biofilm Acidogenicity, and Early

Childhood Caries. Caries Research, 50(3), 319–324. https://doi.org/10.1159/000445910

Newbrun, E., 2004. Systemic benefits of fluoride and fluoridation. Journal of Public Health

Dentistry, 64, pp.35-39.

Nguon S and Kay K (2008); Cambodia Opens Wide for Dental Tourism Phnom Penh Post

(http://www.phnompenhpost.com/national/cambodia-opens-wide-dental-tourism accessed on

15 October 2015)

Nongonierma, A. B., & FitzGerald, R. J. (2012). Biofunctional properties of

caseinophosphopeptides in the oral cavity. Caries Research, 46(3), 234–267.

Nunes, A. M. M., Alves, C. M. C., Borba de Araújo, F., Ortiz, T. M. L., Ribeiro, M. R. C.,

Silva, A. A. M. da, & Ribeiro, C. C. C. (2012). Association between prolonged breast-feeding

and early childhood caries: a hierarchical approach. Community Dentistry and Oral

Epidemiology, 40(6), 542–549. https://doi.org/10.1111/j.1600-0528.2012.00703.x

Nyvad, B., Crielaard, W., Mira, A., Takahashi, N., & Beighton, D. (2013). Dental Caries from

a Molecular Microbiological Perspective. Caries Research, 47(2), 89–102.

https://doi.org/http://dx.doi.org.ezp.lib.unimelb.edu.au/10.1159/000345367

Okada, M., Kawamura, M., Kaihara, Y., Matsuzaki, Y., Kuwahara, S., Ishidori, H., & Miura,

K. (2002). Influence of parents’ oral health behaviour on oral health status of their school

children: an exploratory study employing a causal modelling technique. International Journal

of Paediatric Dentistry, 12(2), 101–108. https://doi.org/10.1046/j.1365-263X.2002.00338.x

https://doi.org/10.1111/cdoe.12330

https://doi.org/10.1159/000445910

http://www.phnompenhpost.com/national/cambodia-opens-wide-dental-tourism

46

Oliveira, B. H., Salazar, M., Carvalho, D. M., Falcão, A., Campos, K., & Nadanovsky, P.

(2014). Biannual Fluoride Varnish Applications and Caries Incidence in Preschoolers: A 24-

month Follow-Up Randomized Placebo-Controlled Clinical Trial. Caries Research, 48(3),

228–236. https://doi.org/10.1159/000356863

Oral health Bureau, Department of Preventive Medicine, Ministry of Health Cambodia (2014)

Final report of the 2011 Cambodia National Oral Health Survey

Peltzer, K., & Mongkolchati, A. (2015). Severe early childhood caries and social determinants

in three-year-old children from Northern Thailand: a birth cohort study. BMC Oral Health,

15(1), 108.

Peres, K. G., Nascimento, G. G., Peres, M. A., Mittinty, M. N., Demarco, F. F., Santos, I. S.,

Barros, A. J. D. (2017). Impact of Prolonged Breastfeeding on Dental Caries: A Population-

Based Birth Cohort Study. Pediatrics, 140(1), e20162943. https://doi.org/10.1542/peds.2016-

2943

Petersen, P. E. (2003). The World Oral Health Report 2003: continuous improvement of oral

health in the 21st century–the approach of the WHO Global Oral Health Programme.

Community Dentistry and Oral Epidemiology, 31(s1), 3–24.

Petersen, P. E., Bourgeois, D., Ogawa, H., Estupinan-Day, S., & Ndiaye, C. (2005). The global

burden of oral diseases and risks to oral health. Bulletin Of The World Health Organization,

83(9), 661–669.

Petersen, P. E., & Lennon, M. A. (2004). Effective use of fluorides for the prevention of dental

caries in the 21st century: the WHO approach. Community Dentistry and Oral Epidemiology,

32(5), 319–321.

Petersen, P. E., & Ogawa, H. (2016). Prevention of dental caries through the use of fluoride–

the WHO approach. Community Dental Health, 33(2), 66–68.

Pithon, M. M., dos Santos, M. J., Andrade, C. S., Leão Filho, J. C. B., Braz, A. K. S., de Araujo,

R. E., Maia, L. C. (2015). Effectiveness of varnish with CPP–ACP in prevention of caries

lesions around orthodontic brackets: an OCT evaluation. The European Journal of

Orthodontics, 37(2), 177–182.

Plonka, K. A., Pukallus, M. L., Holcombe, T. F., Barnett, A. G., Walsh, L. J., & Seow, W. K.

(2013). A randomized controlled clinical trial comparing a remineralizing paste with an

antibacterial gel to prevent early childhood caries. Pediatric Dentistry, 35(1), e8–e12.

Popkin, B. M., & Nielsen, S. J. (2003). The Sweetening of the World’s Diet. Obesity Research,

11(11), 1325–1332. https://doi.org/10.1038/oby.2003.179

Pukallus, M. L., Plonka, K. A., Holcombe, T. F., Barnett, A. G., Walsh, L. J., & Seow, W. K.

(2013). A Randomized Controlled Trial of a 10 Percent CPP-ACP Cream to Reduce Mutans

Streptococci Colonization. Pediatric Dentistry, 35(7), 550–555.

Quinonez, R. B., Keels, M. A., Vann Jr, W. F., McIver, F. T., Heller, K., & Whitt, J. K. (2000).

Early childhood caries: analysis of psychosocial and biological factors in a high-risk

population. Caries Research, 35(5), 376–383.

https://doi.org/10.1159/000356863

47

Rasbridge, L. A., & Kulig, J. C. (1995). Infant feeding among Cambodian refugees. MCN: The

American Journal of Maternal/Child Nursing, 20(4), 213–218.

Ribeiro, N. M., & Ribeiro, M. A. (2004). Breastfeeding and early childhood caries: a critical

review. Jornal de Pediatria, 80(5), s199–s210.

Ribeiro, T. R., Dria, K. J., de Carvalho, C. B. M., Monteiro, A. J., Fonteles, M. C., de Moraes

Carvalho, K., & Fonteles, C. S. R. (2013). Salivary peptide profile and its association with

early childhood caries. International Journal of Paediatric Dentistry, 23(3), 225–234.

https://doi.org/10.1111/j.1365-263X.2012.01258.x

Rong, W. S., Bian, J. Y., Wang, W. J., & De Wang, J. (2003). Effectiveness of an oral health

education and caries prevention program in kindergartens in China. Community Dentistry and

Oral Epidemiology, 31(6), 412–416.

Ruby, J., & Goldner, M. (2007). Nature of symbiosis in oral disease. Journal of Dental

Research, 86(1), 8–11.

Rugg-Gunn, A. J., Spencer, A. J., Whelton, H. P., Jones, C., Beal, J. F., Castle, P., Lennon, M.

A. (2016). Critique of the review of’Water fluoridation for the prevention of dental caries’

published by the Cochrane Collaboration in 2015. British Dental Journal, 220(7), 335–340.

Salone, L. R., Vann, W. F., & Dee, D. L. (2013). Breastfeeding: an overview of oral and general

health benefits. The Journal of the American Dental Association, 144(2), 143–151.

Schluter, P. J., Durward, C., Cartwright, S., & Paterson, J. (2007). Maternal Self-Report of Oral

Health in 4-Year-Old Pacific Children from South Auckland, New Zealand: Findings from the

Pacific Islands Families Study. Journal of Public Health Dentistry, 67(2), 69–77.

Schroth, R. J., Harrison, R. L., & Moffatt, M. E. (2009). Oral health of indigenous children and

the influence of early childhood caries on childhood health and well-being. Pediatric Clinics

of North America, 56(6), 1481–1499.

Selwitz, R. H., Ismail, A. I., & Pitts, N. B. (2007). Dental caries. The Lancet, 369(9555), 51–

59.

Senarath, U., Dibley, M. J., & Agho, K. E. (2010). Factors associated with nonexclusive

breastfeeding in five East and Southeast Asian countries: a multilevel analysis. Journal of

Human Lactation. Retrieved from

http://jhl.sagepub.com/content/early/2010/01/28/0890334409357562.abstract

Serwint, J. R., Mungo, R., Negrete, V. F., Duggan, A. K., & Korsch, B. M. (1993). Child-

rearing practices and nursing caries. Pediatrics, 92(2), 233–237.

Shearer, D. M., Thomson, W. M., Broadbent, J. M., & Poulton, R. (2011). Maternal oral health

predicts their children’s caries experience in adulthood. Journal of Dental Research, 90(5),

672–677.

Sheiham, A., & James, W. P. T. (2015). Diet and Dental Caries The Pivotal Role of Free Sugars

Reemphasized. Journal of Dental Research, 0022034515590377.

https://doi.org/10.1177/0022034515590377

https://doi.org/10.1177/0022034515590377

48

Silva, M.J., Scurrah, K.J., Craig, J.M., Manton, D.J. and Kilpatrick, N., 2016. Etiology of molar

incisor hypomineralization–A systematic review. Community Dentistry and Oral

Epidemiology, 44(4), pp.342-353.

Silver, M. (2015). Guess Which Country Has The Biggest Increase In Soda Drinking. Retrieved

June 25, 2015, from http://www.npr.org/sections/goatsandsoda/2015/06/19/415223346/guess-

which-country-has-the-biggest-increase-in-soda-drinking

Simonsen, R.J. and Neal, R.C., 2011. A review of the clinical application and performance of

pit and fissure sealants. Australian Dental Journal, 56, pp.45-58.

Sisson, K. L. (2007). Theoretical explanations for social inequalities in oral health. Community

Dentistry and Oral Epidemiology, 35(2), 81–88. https://doi.org/10.1111/j.1600-

0528.2007.00354.x

Sjögren, K., Birkhed, D., & Rangmar, B. (1995). Effect of a modified toothpaste technique on

approximal caries in preschool children. Caries Research, 29(6), 435–441.

Smith, R. E., Badner, V. M., Morse, D. E., & Freeman, K. (2002). Maternal risk indicators for

childhood caries in an inner city population. Community Dentistry and Oral Epidemiology,

30(3), 176–181. https://doi.org/10.1034/j.1600-0528.2002.300303.x

Steffensen, J. E. (1990). Literature and concept review: issues in maternal and child oral health.

Journal of Public Health Dentistry, 50(6), 358–369.

Strangio, S. (2014). Hun Sen’s Cambodia. Yale University Press. Retrieved from

https://books.google.com.au/books?hl=en&lr=&id=xRY2BQAAQBAJ&oi=fnd&pg=PR1&d

q=Hun+sens+cambodia&ots=ecRAws6gtg&sig=jEMPiy2aHhZ-DjDc3hcKFF-LZkw

Tac N, Turton B, Durward C (2016) Retention of Glass Ionomer Cement tissure sealant using

a revised protocol Cambodian Dental Journal 12, 14-18.

Tahmassebi, J. F., Duggal, M. S., Malik-Kotru, G., & Curzon, M. E. J. (2006). Soft drinks and

dental health: a review of the current literature. Journal of Dentistry, 34(1), 2–11.

Tang, C., Quinonez, R. B., Hallett, K., Lee, J. Y., & Kenneth Whitt, J. (2005). Examining the

association between parenting stress and the development of early childhood caries.

Community Dentistry and Oral Epidemiology, 33(6), 454–460.

Thitasomakul, S., Piwat, S., Thearmontree, A., Chankanka, O., Pithpornchaiyakul, W., &

Madyusoh, S. (2009). Risks for early childhood caries analyzed by negative binomial models.

Journal of Dental Research, 88(2), 137–141.

Tickle, M., O’Neill, C., Donaldson, M., Birch, S., Noble, S., Killough, S. Verghis, R. (2017).

A randomized controlled trial of caries prevention in dental practice. Journal of Dental

Research, 0022034517702330.

Tinanoff, N., Kanellis, M. J., & Vargas, C. M. (2002). Current understanding of the

epidemiology, mechanisms, and prevention of dental caries in preschool children. Pediatric

Dentistry, 24(6), 543–551.

https://books.google.com.au/books?hl=en&lr=&id=xRY2BQAAQBAJ&oi=fnd&pg=PR1&dq=Hun+sens+cambodia&ots=ecRAws6gtg&sig=jEMPiy2aHhZ-DjDc3hcKFF-LZkw

https://books.google.com.au/books?hl=en&lr=&id=xRY2BQAAQBAJ&oi=fnd&pg=PR1&dq=Hun+sens+cambodia&ots=ecRAws6gtg&sig=jEMPiy2aHhZ-DjDc3hcKFF-LZkw

49

Tinanoff, N. and Reisine, S., 2009. Update on early childhood caries since the Surgeon

General's Report. Academic pediatrics, 9(6), pp.396-403.

Tjäderhane, L., Nascimento, F. D., Breschi, L., Mazzoni, A., Tersariol, I. L. S., Geraldeli, S.,

Pashley, D. H. (2013). Optimizing dentin bond durability: Control of collagen degradation by

matrix metalloproteinases and cysteine cathepsins. Dental Materials, 29(1), 116–135.

https://doi.org/10.1016/j.dental.2012.08.004

Turton, B. J., Thomson, W. M., Foster Page, L. A., Saub, R., & Ishak, A. R. (2015).

Responsiveness of the Child Perceptions Questionnaire 11–14 for Cambodian children

undergoing basic dental care. International Journal of Paediatric Dentistry, 25(1), 2–8.

Turton, B. J., Thomson, W. M., Page, L. A. F., Saub, R. B., & Razak, I. A. (2015). Validation

of an Oral Health–Related Quality of Life Measure for Cambodian Children. Asia-Pacific

Journal of Public Health, 27(2), NP2339–NP2349.

Twetman, S. (2008). Prevention of Early Childhood Caries (ECC) Review of literature

published 1998–2007. European Archives of Paediatric Dentistry, 9(1), 12–18.

Tyas, M. J., Anusavice, K. J., Frencken, J. E., & Mount, G. J. (2000). Minimal intervention

dentistry—a review*. International Dental Journal, 50(1), 1–12.

USAID Assist Project. (2014) Strengthening health professionals regulation in Cambodia.

[ONLINE] Available from: https://www.usaidassist.org/sites/assist/files

/strengthening_health_professions_regulation_in_cambodia_-_fact_sheet_30_sept_2014_

english_0.pdf [accessed on 15 October 2015]

Valaitis, R., Hesch, R., Passarelli, C., Sheehan, D., & Sinton, J. (1999). A systematic review

of the relationship between breastfeeding and early childhood caries. Canadian Journal of

Public Health= Revue Canadienne de Sante Publique, 91(6), 411–417.

Vann, W. F., Lee, J. Y., Baker, D., & Divaris, K. (2010). Oral Health Literacy among Female

Caregivers Impact on Oral Health Outcomes in Early Childhood. Journal of Dental Research,

89(12), 1395–1400.

van Palenstein Helderman, W. H., Soe, W., & Van’t Hof, M. A. (2006). Risk factors of early

childhood caries in a Southeast Asian population. Journal of Dental Research, 85(1), 85–88.

Wan, A. K. L., Seow, W. K., Purdie, D. M., Bird, P. S., Walsh, L. J., & Tudehope, D. I. (2001).

Oral colonization of Streptococcus mutans in six-month-old predentate infants. Journal of

Dental Research, 80(12), 2060–2065.

Wan, A. K. L., Seow, W. K., Purdie, D. M., Bird, P. S., Walsh, L. J., & Tudehope, D. I. (2003).

A longitudinal study of Streptococcus mutans colonization in infants after tooth eruption.

Journal of Dental Research, 82(7), 504–508.

Weintraub, J. A., Ramos-Gomez, F., Jue, B., Shain, S., Hoover, C. I., Featherstone, J. D. B., &

Gansky, S. A. (2006). Fluoride Varnish Efficacy in Preventing Early Childhood Caries. Journal

of Dental Research, 85(2), 172–176. https://doi.org/10.1177/154405910608500211

World Health Organization (2013) Oral health surveys: basic methods - 5th edition. Sao Paulo,

Brazil, University of Sao Paulo.

50

World Health Organisation. (2004) International immunisation framework. [Online] Available

from: www.who.int/immunization/.../NUVI_Impact_on_IS_H_FrameworkRefs. [Accessed 5

Sept 2015]

World Health Organization. (2012) Health service delivery profile Cambodia. [Online]

Available from: http://www.wpro.who.int/health_services/

service_delivery_profile_cambodia.pdf. [Accessed 12 March 2018]

World Health Organization (2015) Guideline: Sugars intake for adults and children. [Online]

Available from: http://apps.who.int/iris/handle/10665/149782 [Accessed 17 May 2016]

World Health Organization (2003) Global strategy for infant and young child feeding. [Online]

Available from:

https://books.google.com/books?hl=en&lr=&id=biABXOXrajYC&oi=fnd&pg=PR4&dq=wh

o+reccomendation+infant+nutrition%5C&ots=1lC0e8b_RI&sig=JIuxNFypdzeuYUvC03q-

sJ-4now [Accessed 7 May 2018]

World Health Organization (2017) Noncommunicable diseases: progress monitor 2017.

Geneva: World Health Organization.

Wong, M., Glenny, A.-M., Tsang, B. W., Lo, E., Worthington, H. V., & Marinho, V. C. (2011).

Cochrane review: Topical fluoride as a cause of dental fluorosis in children. Evidence-Based

Child Health: A Cochrane Review Journal, 6(2), 388–439.

Wong, P. D., Birken, C. S., Parkin, P. C., Venu, I., Chen, Y., Schroth, R. J., & Maguire, J. L.

(2017). Total Breast-Feeding Duration and Dental Caries in Healthy Urban Children. Academic

Pediatrics, 17(3), 310–315. https://doi.org/10.1016/j.acap.2016.10.021

Wright, J. T., Hanson, N., Ristic, H., Whall, C. W., Estrich, C. G., & Zentz, R. R. (2014).

Fluoride toothpaste efficacy and safety in children younger than 6 years: a systematic review.

The Journal of the American Dental Association, 145(2), 182–189.

Wright, J. T., Tampi, M. P., Graham, L., Estrich, C., Crall, J. J., Fontana, M., (2016). Sealants

for preventing and arresting pit-and-fissure occlusal caries in primary and permanent molars:

a systematic review of randomized controlled trials—a report of the American Dental

Association and the American Academy of Pediatric Dentistry. The Journal of the American

Dental Association, 147(8), 631–645.

Yang, F., Zeng, X., Ning, K., Liu, K.-L., Lo, C.-C., Wang, W., Chang, X. (2012). Saliva

microbiomes distinguish caries-active from healthy human populations. The ISME Journal,

6(1), 1.

Zhao, I.S., Gao, S.S., Hiraishi, N., Burrow, M.F., Duangthip, D., Mei, M.L., Lo, E.C.M. and

Chu, C.H., 2018. Mechanisms of silver diamine fluoride on arresting caries: a literature review.

International Dental Journal, 68(2), pp.67-76.

Zhi, Q. H., Lo, E. C. M., & Lin, H. C. (2012). Randomized clinical trial on effectiveness of

silver diamine fluoride and glass ionomer in arresting dentine caries in preschool children.

Journal of Dentistry, 40(11), 962–967.

http://www.who.int/immunization/.../NUVI_Impact_on_IS_H_FrameworkRefs

http://www.wpro.who.int/health_services/%20service_delivery_profile_cambodia.pdf

http://www.wpro.who.int/health_services/%20service_delivery_profile_cambodia.pdf

http://apps.who.int/iris/handle/10665/149782

https://books.google.com/books?hl=en&lr=&id=biABXOXrajYC&oi=fnd&pg=PR4&dq=who+reccomendation+infant+nutrition%5C&ots=1lC0e8b_RI&sig=JIuxNFypdzeuYUvC03q-sJ-4now



48

Chapter 2

A GIC Fissure Protection intervention in Cambodia – 12 month

results and Pilot Study

Published in part as: Tac N, Turton B, Durward C (2016) Retention of Glass Ionomer

Cement tissue sealant using a revised protocol Cambodian Dental Journal 12, 14-18.

Presented in part as:

Turton B, Durward C, Bach K, Manton D. (2014) Seal Cambodia – 60,000 children over 3

years; IADR SEA 28TH Annual meeting, Kutching, Malaysia.

Tak N, Turton B, Durward C. (2015) Seal Cambodia – A comparison of two protocols; IADR

SEA 29th Annual meeting, Bali, Indonesia.

Turton B, Durward C. (2015) Seal Cambodia – Placing fissure protection in a community

setting successfully; FDI World Congress, Bangkok, Thailand.

49

Abstract

Objective: To compare two different protocols for the placement of hand-mixed GIC Fissure

Protection (FP) placed on FPM of 6 to 8 year-old Cambodian children as part of the SEAL

Cambodia Project.

Methods: The study had two pilots: Pilot A, a one-year evaluation of the original cohort and

protocol; and Pilot B a one-year evaluation of a new protocol. Pilot A included a cohort of

children who received FP using the original protocol (Group A), along with a matched control

group (Group B) who did not receive FP. Pilot B involved two convenience samples of children

who were recruited based on the routine schedule of schools to be sealed as part of the wider

Seal Cambodia Project. One group received FP using the modified protocol (Group C) while

the other received FP using the original protocol (Group D). The modified protocol involved

control of the temperature of the material, the timing of mixing, and placement.

Results: Pilot A achieved a 61% follow-up rate with 26.8% retention and 10% preventive

increment at 1-year. Pilot B achieved an 86% follow-up rate, with 67.1% retention rate at 6-

months.

Conclusion: The protocol associated with Pilot B achieved a higher sealant retention rate and

should be adopted at scale across the SEAL Cambodia project.

50

2.0 Chapter 2 – The SEAL Cambodia pilot studies

Dental caries experience in Cambodia is severe and the services available to manage the burden

of disease are limited. The very low ‘filled’ (<1%) component of the dmft/DMFT reported in

the 2011 CNOHS reported that most Cambodian children had not received prior dental

treatment (Chher et al. 2016). Therefore, if a carious lesion develops, it is likely that no

operative treatment would be received, and eventually the tooth (often associated with pain and

infection) could require extraction. It is clear that the drivers of the severe caries experience in

Cambodia are behavioural and mediated through late oral hygiene practices and unfavourable,

high added sugar diets (Chher et al. 2016). Although the aetiological factors of the caries

process might be shared among other high caries groups, few interventional studies in settings

similar to Cambodia have been published. Therefore, it is important to establish evidence for

the most effective and efficient preventive techniques in a Cambodian environment (Chher et

al. 2009). This involves selection of the best techniques as well as the most appropriate

logistical models for delivery. These concepts are influenced by the limited resources and the

ubiquitous distribution of disease across the population.

In the case of Cambodia, dental caries affects the majority of young children, with 97% of 6

year-old children having one or more cavitated lesions, and 86% having one or more pulpally

involved teeth. The disease is more severe in urban areas, especially the Phnom Penh and

Kampot provinces (Chher et al. 2016). When considering a population-wide intervention, the

general options are to consider either a targeted approach or a whole population approach.

Targeted approaches are most appropriate when the burden of disease sits within limited

subsections of the community, such is the case in more developed settings, and documented in

New Zealand and Australia with the lift-the-lip campaign, Scotland with the Child Smile

project, and the USA with the well child program (Chia et al. 2015; Lam et al. 2012;

Macpherson et al. Anopa et al. 2013; Holve, 2008).

In Cambodia, the near universal caries experience justifies a ‘whole population’ approach. In

terms of the methods available to prevent dental caries, the best evidence supports tooth-

brushing with fluoride toothpaste, placement of fissure sealants, and professional topical

fluoride placement. Daily tooth-brushing with fluoride toothpaste can reduce dental caries

incidence by between 20-40% (Marinho et al. 2003). Fissure protection with both glass

ionomer cement and resin-based sealants has been reported to prevent new carious lesions on

FPM by up to 80% over 3 y (Ahovuo-Saloranta et al. 2016). Studies on fluoride varnish are

51

extensive and systematic analysis suggests that applications 2 to 3 times-per-year can reduce

caries experience by up to 40% (Agouropoulos et al. 2014). Each of these approaches have

implications for the structure of how services are delivered.

2.0.1 The ‘birth’ of the SEAL Cambodia Community Project

In 2011 the Global Child Dental Fund (GCDF), an international charity focused on oral health,

initiated a meeting of local and international dental specialists to consider which technologies

and which organisational structures might be appropriate for addressing the high rate of dental

caries in Cambodian children. There was consensus to move away from extraction-based

‘bottom of the cliff’ interventions, and towards “upstream” interventions to prevent carious

lesions developing. The GCDF Taskforce was formed in Cambodia, which included oral health

stakeholders from the Ministry of Health, The Cambodian Dental Association, Non-

Governmental Organisations (NGOs), and local and international universities, and gave its

strong support to this innovative child oral health project. Out of that taskforce the ‘SEAL

Cambodia’ Community Project was born. The rationale of SEAL Cambodia was that if most

of the new lesions in the permanent dentition occurring between 6 to 8 years-of-age were

occurring on FPM, then targeting those teeth would reduce the overall burden of caries during

the primary school years and into adolescence. The delivery of services in a school setting

rather than a community setting was deemed to be more cost-effective and targeted. And along

with the logistical benefit, it also offered a way to reorientate local and visiting volunteers to

participate in a dental intervention which was focused on upstream intervention such as

prevention.

The GCDF Task force set out a plan to engage local and international partners in providing a

proven preventive intervention for young children, which could serve as a demonstration model

of care in a developing world setting. A target was set for placing FP on the FPM of 60,000 6

to 8 year-old children with GIC over a three year period. Sponsorship for the project was

obtained from GCDF, CamKids (the Cambodian Children’s Charity) and GC Asia (which also

donated the GIC sealant material Fuji VII). The project was accepted and approved by the

Ministry of Health (MOH), and the Ministry of Education, Youth and Sport (MOEYS). The

Australia and New Zealand Society of Paediatric Dentistry (ANZSPD) also made a donation

in the first year enabling the purchase of a tuk tuk (motorcycle pulled passenger trailer) to

transport local and overseas volunteers to the schools around Phnom Penh to provide FP.

52

Although there is a lot of evidence to support the use of GIC fissure sealants, no research is

available for the efficacy of FS in a population with such extreme caries experience as

Cambodia. The ART technique was chosen for placement of FS and most studies that

investigated the use of that technique used high viscosity GIC such as Fuji IX® (GC Corp,

Tokyo, Japan) or Ketac Fil® (3MTM ESPETM) as recommended in the original paper (de

Amorim et al. 2012). In contrast, SEAL Cambodia planned a low viscosity GIC (Fuji VII®, GC

Corp, Tokyo, Japan) which was designed specifically for fissure protection. Fuji VII® releases

more fluoride than higher viscosity GIC materials, is said to flow more readily into fissure

systems and is more moisture tolerant, making it more suitable for placement on partially

erupted teeth (Ganesh & Tandon 2007). The disadvantage is that Fuji VII® has a lower

compressive strength, although, this is less important in a sealant scenario. Another concern

around using a low viscosity GIC was that the Fuji III® precursor to Fuji VII® showed very

limited retention with just one in six sealants being retained over two years in previous studies

(Fross et al. 1994; Poulsen et al. 2001). Given that individuals targeted by the SEAL Cambodia

project were a distinctive population with an extreme caries experience and almost no access

to dental care, and that there were some special considerations around material choice, it was

prudent to closely monitor clinical outcomes including both caries prevention and retention of

sealant material order to ensure that the most reasonable outcomes were being achieved.

53

2.1 Aim

To evaluate two different protocols for the placement of FS and the prevention of dental caries

in the FPM of 6 to 8 year-old children in three provinces of Cambodia.

2.2 Objective

To evaluate the caries preventive effect and retention, respectively, of FS placed using different

protocols in two pilot samples taken from within the SEAL Cambodia Project.

54

2.3 Methods

The study reports on two pilot studies that are quasi-randomised control trials: Pilot A, a one

year evaluation of the original cohort and protocol; and Pilot B, a one year evaluation of a new

protocol. Each pilot examined a different clinical protocol to assess: caries prevention in the

case of Pilot A; and improvement in the retention in the case of Pilot B. All clinical

examinations and clinical procedures were conducted in a school yard setting with battery

powered lights and children in a supine position. Consent was gained by sending the consent

forms to the school the week prior to the examination, and parents were given an opportunity

to withdraw their child’s participation should they wish (Appendix 1). Children were free to

refuse assent at the time of examination with no consequences. Participants were given oral

hygiene education with a ratio of one instructor to five children and they were given a

toothbrush prior to the placement of the sealants, or following the examination for those in the

control group. This ensured that the teeth were clean prior to examination and placement of

GIC sealants. Children were also sent home with an information sheet (Appendix 2) after

participating to inform the parents of the need for additional dental treatment where

appropriate. Ethics approval was provided by the National Committee for Health Research,

Ministry of Health, Cambodia.

2.3.1 Selection of teeth for FS

The occlusal surfaces of FPM were checked visually to detect the presence of cavitated carious

lesions. Sealants were only applied to partially or fully erupted, non-cavitated FPM including

those with staining or early carious lesions (but no underlying enamel shadow). Sealants were

placed by a mix of Cambodian Dental students, foreign dental students and dental therapy

students, as well as Cambodian and Foreign dentists.

2.3.2 Pilot A – one year evaluation of the original cohort

The first pilot involved a representative sample of 433 children selected using randomised

cluster selection at class level from the population eligible to participate in the SEAL Cambodia

project who made up the ‘intervention’ group (Group A). The control group consisted of 249

children from schools that were not included in the project (Group B). The clinical procedure

for placement of FS was as follows, with up to two teeth (opposing arches) treated

simultaneously (1) the dentine conditioner was applied to a clean tooth and then removed with

a wet cotton pellet; (2) the tooth was dried using cotton pellets; (3) hand-mixed Fuji VII® was

55

placed onto the occlusal surface of the tooth and pressed into the fissures using the operators’

finger (no defined duration) which was lubricated with cocoa butter or Vaseline (Unilever,

USA) and no subsequent occlusal adjustment.

2.3.2.1 Clinical examination at baseline and one year follow-up

Six examiners were calibrated prior to undertaking examinations (ICC >0.85). Teeth were

scored using the methods described in the ‘WHO basic survey methods’ book. Two indices

were used: the DMF index at surface level (WHO 2013); in addition, each tooth was given a

score using the PUFA Index (Monse et al. 2010). The clinical documentation used is included

in Appendix 3 and 4. Participants were divided into terciles based on caries experience as

consistent with the significant caries index (Bratthall 2000) to define those in ‘high’, ‘very

high’, and ‘extreme’ caries groups.

Evaluation of FS was based on the proportion of material that was remaining. FS that covered

more than 2/3 of the fissure system was recorded as ‘fully retained’. If FS material was present

but covered less than 2/3 it was recorded as ‘partially retained’. If the FS was not detectable

visually it was recorded as ‘lost’.

2.3.3 Pilot B - pilot study using a revised protocol

The second part involved two convenience samples recruited based on the routine schedule of

schools to be part of the sealant program as dictated by the school health department, MOEYS.

Group D involved 101 children from a school which received fissure sealants the week after

Group C according to the original protocol used for Group A; Group C involved a cluster of

115 children who attended the schools that were scheduled for routine sealant placement the

following week.

The original protocol was modified for participants in Group C. There were four changes; the

first change was to cool the GIC liquid material by placing it in a container of ice between uses.

The second change was placing one sealant at a time. The next change was the exact timing of

mix and finger press: a timer was placed next to each operator and set for 20 seconds for the

mix and one minute for the finger press. The last change was the occlusal check and adjustment

after placement. Articulating paper was used to detect any ‘high spots’ of sealants after

placement and a large spoon excavator used to reduce any ‘high spots’.

56

Two materials were trialled in Group C using a split-mouth design alternating the side the

sealants were placed based on the day that the children presented. There were four days when

sealants were placed; for day 1 and day 2 Fuji VII® was placed on the left side while Fuji IX®

was placed on the right side. For days 3 and 4, Fuji VII® was placed on the right side and Fuji

IX® on the left. Those in Group D had FS placed using the original protocol as described in

Section 2.3.2.

2.3.2.1 Clinical examination

At baseline, the standard SEAL Cambodia form was used to record data on the type of material

placed (Appendix 5). A different form was used at follow-up at one week, one month, six

months and one year and the FS retention was rated based on presence of the GIC material as

described in Section 2.3.2.1 (Appendix 4). No measure for carious lesions was used for Pilot

B follow-up.

2.3.4 Data collection and analysis

Data were entered into SPSS Ver. 20 (IBM, NY, USA) and analysed to produce descriptive

statistics on baseline caries experience and caries incidence (Pilot A) as well as sealant

retention (Pilot A and Pilot B). Descriptive analyses were performed as well as bivariate

analyses to examine differences in caries experience and sealant retention by group

membership. Baseline caries groups were defined using the Significant Caries Index (Bratthall,

2000) where by the group was divided by terciles based on dmft. Differences in means among

groups were compared using the Kruskal-Wallis test and differences in proportions among

groups were compared using the chi-squared test. Differences in means among related samples

were examined using the Wilcoxon signed-rank test.

57

2.4 Results

2.4.1 Pilot A

Data on the sociodemographic characteristics of participants by attrition at 1-year are presented

in Table 2.1. The mean age of participants was 8.1 (SD 1.1) years-of-age and there was no

significant difference in gender or age by loss to follow-up.

Table 2.1 – Pilot A – Attrition analysis for the 1-year follow-up by gender and age

(brackets contain row percentages unless otherwise indicated).

Baseline

N (column %)

One year follow-up

N (row %)

Lost to follow-up

N (row %)

Group membership

Control 251 (36.7) 163 (64.9) 88 (35.1)

Intervention 433 (63.3) 262 (60.5) 171 (39.5)

Age group

<6 years 18 (2.6) 12 (66.7) 6 (33.3)

7 years 149 (21.8) 92 (61.7) 57 (38.3)

8 years 282 (41.2) 183 (64.9) 99 (35.1)

>9 years 235 (34.4) 138 (58.7) 97 (41.3)

Gender

Male 358 (52.3) 215 (60.1) 143 (39.9)

Female 326 (47.7) 210 (64.4) 116 (35.6)

Total 684 (100.0) 425 (62.1) 259 (37.9)

Data on caries experience by membership to the control group or intervention group is

presented in Table 2.2. There was a statistically significant difference in caries experience by

gender whereby males had a more severe dmft than females. There was also a statistically

significant difference in caries experience by membership of the control or intervention group

with those in the intervention group having a higher caries experience. There was no difference

in caries experience by loss to follow-up.

58

Table 2.2 – Pilot A - Caries experience at baseline by group membership, gender and age.

dmft

Mean (SD)

DMFT

Mean (SD)

Any caries

N (Row %)

Any PUFA

N (Row %)

Treatment group

Control 7.5 (4.1)a 2.2 (2.1) 232 (93.2) 169 (67.9)a

Intervention 8.3 (3.8) 2.3 (2.2) 412 (94.2) 336 (77.6)

Follow-up

Lost 7.9 (4.0) 2.4 (2.4) 247 (93.6) 198 (75.0)

Followed-up 8.1 (3.9) 2.1 (2.0) 397 (94.7) 307 (73.3)

Sex

Male 8.4 (4.0)a 2.2 (2.1) 334 (94.6) 269 (76.2)

Female 7.6 (3.9) 2.2 (2.1) 306 (94.2) 233 (71.7)

Age group <6 y 8.2 (4.9)a 1.9 (2.7)a 16 (94.1)a 10 (58.8)a

7 y 8.9 (3.7) 2.1 (2.1) 143 (96.0) 114 (76.5) 8 y 8.7 (3.7) 2.1 (1.9) 268 (97.1) 220 (79.7) >9 y 6.6 (3.9) 2.6 (2.4) 212 (90.2) 156 (66.4)

Overall 8.0 (3.9) 2.2 (2.2) 640 (94.4) 502 (74.0)

aP-value <0.01; Kruskal-Wallis for comparison of means among groups or χ2 test for comparison of proportions

among groups within the same column.

Data on caries incidence and sealant retention by group membership are presented in Table

2.3; participants who were lost due to follow-up were not included in this analysis. There were

no statistically significant differences in the proportion or number of teeth eligible to be sealed

by group membership. Also, there was no statistically significant difference in caries incidence

by group membership. Despite this it does appear that there may have been a 10% preventive

benefit for those in the intervention group. The retention rate for sealants was low with only

approximately one in four sealants retained (including partially retained).

59

Table 2.3 – Pilot A - Caries incidence and sealant retention by group membership

Tooth 16

n (row %)

Tooth 26

n (row %)

Tooth 36

n (row %)

Tooth 46

n (row %)

Mean combineda

Mean (SD)

Teeth eligible for

sealants

Control 128 (84.2) 123 (82.0) 103 (67.3) 112 (73.2) 3.0 (1.3)

Intervention 203 (79.6) 201 (79.4) 172 (67.5) 179 (71.0) 2.9 (1.2)

Total 331 (81.3) 324 (80.4) 275 (67.4) 291 (71.9) 2.9 (1.2)

Caries

incidencea

Control 14 (11.0) 10 (8.1) 31 (18.1) 26 (23.4) 1.0 (1.2)

Intervention 21 (10.4) 16 (8.0) 19 (18.6) 32 (18.1) 0.9 (1.1)

Total 35 (10.6) 26 (8.0) 50 (18.3) 58 (20.1) 0.9 (1.1)

Sealant

retentionb

Absent 149 (73.4) 152 (75.6) 124 (72.1) 128 (72.3) 3.0 (1.1)

Partially

present

47 (23.2) 45 (22.4) 40 (23.3) 43 (24.0) 0.8 (1.0)

Completely

present

4 (2.0) 1 (0.5) 6 (3.5) 5 (2.8) 0.1 (0.4)

aProportion of caries incidence is calculated by considering the number within ‘eligible’ teeth.

bIncludes intervention group only

Data on caries incidence by group membership, previous caries experience and age group are

presented in Table 2.4. There was a statistically significant difference in caries incidence in the

permanent dentition by baseline caries experience. Participants in the control group had a

significantly lower caries experience at baseline. There was a statistically significant difference

in caries incidence by baseline caries experience. Children in the least severe caries groups

realised between 30-50% lower caries experience depending on the category applied.

Age did not appear to be associated with the caries preventive effect.

When individual teeth were taken into account, the caries increment was dependant on the

baseline caries category. For example, among the children who had the lower right first

permanent molar eligible for placement, a positive preventive effect was seen for those children

in the high or very high baseline caries categories but no preventive effect was seen for that

tooth in children that belonged to the extreme caries category.

60

Table 2.4 – Pilot A - Caries incidence in the intervention group by baseline caries experience and agea

Tooth 16

n = 41

Tooth 26

n = 30

Tooth 36

n = 109

Tooth 46

n = 93 Mean number

INT

n (row %) CON

n (row %)

INT n (row %)

CON n (row %)

INT n (row %)

CON n (row %)

INT n (row %)

CON n (row %)

INT Mean (SD)

CON Mean (SD)

Caries group

(deciduous)

High dmft

dmft <6

7 (53.8) 6 (46.2) 5 (55.6) 4 (44.4) 18 (48.6) 19 (51.4) 11 (44.0) 14 (56.0) 0.7 (0.9) 1.0 (1.2)c

Very high dmft

dmft 6-to 9

2 (28.6) 5 (7.4) 2 (28.6) 5 (71.4) 5 (27.8) 13 (72.2) 5 (26.3) 14 (73.7) 1.1 (1.2) 1.2 (1.3)

Extreme dmft

dmft >9

8 (38.1) 13 (61.9) 5 (35.7) 9 (64.3) 18 (33.3) 36 (66.7) 18 (36.7) 31 (63.3) 0.9 (1.1) 0.9 (1.0)

Caries group

(permanent)a

High DMFT

DMFT <1

2 (50.0) 2 (50.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0.3 (0.7) 0.6 (1.1)

Very high DMFT

DMFT = 1

15 (40.5) 22 (59.5) 0 (0.0) 2 (100.0) 4 (30.8) 9 (69.2) 2 (20.0) 8 (80.0) 0.7 (0.9) 0.4 (0.8)

Extreme DMFT

DMFT >2

15 (40.5) 22 (59.5) 12 (42.9) 16 (57.1) 37 (38.5) 59 (61.5) 32 (38.6) 51 (61.4) 1.2 (1.2) 1.4 (1.2)

Pulpally involved

No pufa 14 (38.9) 22 (61.1) 0 (0.0) 1 (100.0)b 9 (52.9) 8 (47.1) 5 (41.7) 7 (58.3)b 1.0 (1.1)c 1.1 (1.2)

Any pufa 3 (60.0) 2 (40.0) 12 (41.4) 17 (58.6) 32 (34.8) 60 (65.2) 29 (35.8) 52 (64.2) 0.5 (0.8) 0.9 (1.1)

Age

<6 y 0 (0.0) 1 (100.0) 0 (0.0) 1 (100.0) 0 (0.0) 2 (100.0) 0 (0.0) 1 (100.0) 0.6 (1.1) 0.0 (0..0)

7 y 3 (33.3) 6 (66.7) 2 (33.3) 4 (66.7) 7 (35.0) 13 (65.0) 9 (50.0) 9 (50.0) 1.0 (0.7) 1.0 (1.3)

8 y 6 (42.9) 8 (57.1) 2 (20.0) 8 (80.0) 14 (35.0) 26 (65.0) 11 (32.4) 23 (67.6) 0.9 (1.3) 1.0 (1.3)

>9 y 8 (47.1) 9 (52.9) 8 (61.5) 5 (38.5) 20 (42.6) 27 (57.4) 14 (35.0) 26 (65.0) 1.3 (1.1) 1.1 (1.3)

aINT stands for “intervention group” and CON stands for “control group”; percentages are calculated based on the number of eligible teeth bP-value <0.001; χ2 test or Kruskal-Wallis test for

differences among groups within the same column cP-value <0.01; χ2 and Kruskal-Wallis test as appropriate for differences among groups within the same column

61

Data on sealant retention by caries experience and age are presented in Table 2.5. Children

with one or more pulpally involved teeth at baseline had a statistically significant greater

chance of losing the fissure protection on tooth 46. Children in the lower age groups had a

statistically significant greater chance of retaining the FS on teeth 36 and 46. Sealant retention

was statistically different among age-groups and those in the younger age group were more

likely to retain the FS material. There is some evidence that sealant retention is related to

baseline caries experience and age.

62

Table 2.5 – Pilot A - Number of retained sealants by caries experience and gender (brackets contain row percentages unless otherwise indicated)

Tooth 16

(N = 206)

Tooth 26

(N = 206)

Tooth 36

(N=178)

Tooth 46

(N=185)

Lost

N (%)

Retained

N (%)

Lost

N (%)

Retained

N (%)

Lost

N (%)

Retained

N (%)

Lost

N (%)

Retained

N (%)

All FPM Mean (SD)

Baseline caries

group (primary)

High dmft

dmft <6

47 (71.2) 19 (28.8) 50 (73.5) 18 (26.5) 43 (72.9)

16 (27.1) 43 (71.7) 17 (28.3) 0.6 (1.0)

Very High dmft

dmft 6-to 9

24 (68.6) 11 (31.4) 27 (79.4) 7 (20.6) 21 (70.0) 9 (30.0) 17 (60.7) 11 (39.3) 0.6 (0.9)

Extreme dmft

dmft >9

83 (79.0) 22 (21.0) 82 (78.8) 22 (21.2) 67 (75.3) 22 (24.7) 76 (78.4) 21 (21.6) 0.7 (1.1)

Baseline caries

group (permanent)

High

DMFT <1

41 (82.0) 9 (18.0) 41 (83.7) 8 (16.3) 40 (72.7) 15 (27.3) 40 (74.1) 14 (25.9) 0.6 (1.0)

V. High

DMFT = 1

39 (73.6) 14 (26.4) 45 (83.3) 9 (16.7) 41 (77.4) 12 (22.6) 38 (69.1) 17 (30.9) 0.6 (1.0)

Extreme

DMFT >2

74 (71.8) 29 (28.2) 73 (70.9) 30 (29.1) 50 (71.4) 20 (28.6) 58 (76.3) 18 (23.7) 0.7 (1.0)

Pulpally involved

No pufa 36 (75.0) 12 (25.0) 33 (73.3) 12 (26.7) 29 (67.4) 14 (32.6) 28 (65.1) 15 (34.9) 0.7 (1.0)

Any pufa 118 (74.7) 40 (25.3) 126 (78.3) 35 (21.7) 102 (75.6) 33 (24.4) 108 (76.1) 34 (23.9) 0.6 (1.0)

Age

<6 y 2 (66.7) 3 (33.3) 2 (66.7) 1 (33.3) 2 (50.0) 2 (50.0) 1 (25.0) 3 (75.0) 0.9 (1.5)a

7 y 33 (75.0) 11 (25.0) 33 (71.7) 13 (28.3) 28 (68.3) 13 (31.7) 31 (66.0) 16 (34.0) 0.9 (1.2)

8 y 68 (74.7) 23 (25.3) 65 (74.7) 22 (25.3) 56 (70.0) 24 (30.0) 61 (75.3) 20 (24.7) 0.7 (0.9)

>9 y 51 (75.0) 17 (25.0) 59 (84.3) 11 (15.7) 45 (84.9) 8 (15.1) 43 (71.1) 10 (18.9) 0.4 (0.9)

aP-value <0.05; χ2 test or Kruskal-Wallis test as appropriate for examining differences among groups within the same column

63

2.4.2 Pilot B

The attrition data is presented in Figure 2.1 and 58 children were excluded from the analysis

as at least one cavitated first permanent molar was present. During the one month follow-up,

there was no dropout in the control group, but there was an 8% dropout rate (N=9) in the

intervention group. This rate increased so that the incremental drop out at the the six month

follow-up for the intervention group was 36.5% (N=32). At one year follow-up, there was a

14% (N=6) dropout rate in the control group, while there was 19.1% (N=22) in the intervention

group.

Figure 2.1 – Selection and follow-up of children in Pilot B

64

Data on the sociodemographic characteristics of those in Pilot B are presented in Table 2.6.

After excluding children who did not meet the inclusion criteria, there were differences in

gender by group membership. There was a lower number of female children (37.2%) compared

to male children (62.8%) in the control group. The mean age of children in the control group

was 9.0y while the mean age in the intervention group was 9.2y, which was not considered

significant.

Table 2.6 – Pilot B – Gender and group membership after cases with caries FPM at

baseline were removed

Male

N (row %)

Female

N (row %)

Total

N (column %)

Control 27 (62.8) 16 (37.2) 43 (28.9)

Intervention 54 (50.9) 52 (49.1) 106 (71.1)

Total 81 (54.4) 68 (45.6) 149 (100.0)

At the one week and one-year follow up, there was no significant difference in the number of

sound sealants that were completely retained by material type (data presented in Table 2.7). At

the one month and six-month follow-up Fuji VII® material had a statistically significant greater

chance of being fully retained. At one month Fuji IX® material was three times more likely to

be completely lost.

65

Table 2.7 – Pilot B - Proportion of sealant retention by material type in the intervention

group (numbers are percentages)a

One week One month Six months One year

VII IX P-value VII IX P-value VII IX P-value VII IX P-value

Sound 84.0 79.7 0.317 76.5 64.2 0.005 8.9 16.1 0.041 3.7 2.7 1.000

High 5.2 0.5 0.013 0.0 0.0 n/a - - - - - -

Partially

lost

9.0 9.4 0.858 16.0 13.7 0.587 48.2 24.4 <0.001 33.0 31.4 0.014

Completely

lost

1.9 10.3 0.003 7.6 22.2 <0.001 42.9 59.5 0.006 63.3 77.1 0.016

aP-value calculated using Wilcoxon Signed Rank Test; ‘VII’ refers to Fuji VII material and ‘IX’ refers to Fuji IX

material.

At one month, there was a statistically significant difference in sealant retention by group

membership, but at one year that difference was no longer significant (data presented in Table

2.8). Within the intervention group a more than 10% increase in the rate of retention at one

month was observed, but the differences at one year were not significant.

66

Table 2.8 – Pilot B - Proportion of Fuji VII® FS retained at one month and one year by

group membershipa

One month One year

Group D

(%)

Group C

(%) P-valueb

Group D

(%)

Group C

(%) P-valueb

Sound 63.9 76.5 0.019 4.7 3.7 0.331

Partially lost 23.9 16.0 <0.001 31.8 33.0 0.892

Completely lost 9.9 7.6 0.072 63.5 63.3 0.845

aGroup D were those in Pilot B that had FS placed using the original protocol and Group C were those in Pilot B

who had FS placed using a modified protocol. bP-value calculated using Kruskal-Wallis statistic for differences among groups within the same row.

67

2.5 Discussion

The SEAL Cambodia project represents the first time that Fuji VII® FS has been applied to

prevent dental caries at large scale in Cambodia and the first time that it has been evaluated

among a population with such a severe caries experience. The present study examined two pilot

protocols, each looking at a different clinical procedure to: achieve caries prevention in the

case of Pilot A; and improve retention in the case of Pilot B. The later pilot (Pilot B) was put

in place as a response to the unfavourable findings in the first pilot and changes were based on

a critical review of the initial findings. Pilot A demonstrated that there was poor FS retention,

a smaller than expected caries preventive effect, and that statistically significant clinical

benefits were only realised in the less severe caries groups. Pilot B of the study demonstrated

that modifications in the original protocol could improve retention; it could be postulated that

this is likely associated with a better preventive increment but would need to be the subject of

future research to establish evidence.

Before considering the findings in more detail, it is appropriate to first consider the strengths

and limitations of the study. Regarding Pilot B, there were significant vulnerabilities due to the

small sample size and limited follow-up at 1-year. In addition, requiring four non-cavitated

FPM for the split mouth design meant that the sample was biased towards the less severe

spectrum of caries in a potentially eligible population where one in four children that are 6

years-of-age have at least one cavitated first permanent molar. Those factors also meant that

the children in Pilot B were not comparable to the children in Pilot A who had a more severe

caries experience.

Regarding Pilot A, the strengths were that the study was sufficiently powered to examine

clinically significant caries preventive effect; however, the sample did not reach the ideal target

and older children were included into the sampling frame. This is because in Cambodia it is

common for children to repeat grades or start school late. Therefore, it is normal to have a wide

variety of ages at each grade level. The follow-up rate for Pilot A did not reach the desired 70%

and so the sample could be vulnerable to follow-up bias; however, no differences were seen at

1-year follow-up by gender, age, or baseline caries experience. In addition, the age profile of

the children in this cohort was similar to that reported among the wider population of children

participating in the SEAL Cambodia project (One-2-One Cambodia 2016). Although the cohort

was representative of the wider SEAL project, those in the control group had a significantly

68

lower caries experience at baseline and this could have masked some of the preventive benefits

of the intervention.

69

2.5.1 Caries prevention in Pilot A

The caries experience in this group was so severe that even in the lower tercile of caries

experience there was a mean dmft > 5, which matches the caries profile of most other groups

that are considered to be ‘high caries’ experience. It has been reported that caries experience

among groups is likely to moderate the caries preventive effect achieved by a FS intervention

(Ahovuo-Saloranta et al. 2017). Within Pilot A, there was potentially a benefit that was not

statistically significant and that combined with the existing body of evidence around FS as a

caries preventive strategy meant that further investigation was warranted. This is the first time

that the material has been assessed among a group with such severe caries experience and it

could be that the indices used to examine caries progression were not sensitive enough to detect

the benefits rendered. This was also a first in terms of an upstream dental intervention,

implemented at large scale in Cambodia.

2.5.1.1 Alternatives to achieve caries prevention in FPM

As this was ‘new territory’ it was appropriate to examine alternative ways to achieve prevention

of carious lesions on FPM. That included consideration for the present protocol in terms of

understanding of the material science and consideration of other materials. The alternative

materials that could be considered for pit and fissure caries prevention could be Fluoride

Varnish (FV) or Silver Diammine Fluoride (SDF) which have been able to demonstrate a

preventive increment of around 40% on FPM (Liu et al. 2012). Alternatively, supervised ‘cross

brushing’ in schools was able to achieve significant reductions in plaque and a reduction in

caries experience (Frazão, 2011; Nourallah & Splieth, 2004). Finally, the other material for

consideration would have been resin materials for the placement of FS. Resin fissure protection

is often considered the gold standard, but it would involve considerable and unfeasible

additional infrastructure and support in terms of clinic set up and transport to clinics. While

SDF or FV require less investment, these interventions do not yet have a broad body of

evidence to support effectiveness for caries prevention on FPM; and cannot yet be

recommended for large scale interventions at population level. In contrast, the benefit of the

SEAL Cambodia protocol from an organisational and logistical perspective is that it is a

discrete, one-off intervention, with very little need for investment in infrastructure and with a

good evidence of potential benefit. For that reason, it was favourable and justified to consider

continuation of the SEAL Cambodia Community Project using the modified protocol.

70

2.5.2 The retention of FS material in Pilot B

Given that the community-based project was still ongoing during the testing of a modified

protocol, an indicator of success other than 1-year caries increment was desired and therefore

retention of FS material at 3 and 6-months was the outcome used to inform the decision making

of the SEAL Cambodia project. Although retention is not the primary outcome of interest, if

the fissure protection was lost early then it could help to explain why the caries prevention was

not as beneficial as expected in Pilot A. The theory behind FS is that the material will protect

the pit and fissure system during the most vulnerable period so of development; when the tooth

has newly erupted and enamel maturation is occurring and the tooth is more difficult to reach

with a toothbrush. Therefore, it could be argued that FS are not needed after that initial

protection period and this is the rationale behind the fact that both resin and GIC fissure

protection is thought to render the same preventive effect despite the fact that resin sealants are

better retained (Mickenautsch & Yengopal, 2013). In saying that, it could be that resin FS has

been easier to detect because it stays in bulk while GIC materials are thought to be ‘micro-

retained’ (Hu et al, 2016).

In the case of the cohort in Pilot A, new carious lesions were forming on teeth that lost clinically

detectable traces of Fuji VII® FS. The major question raised was whether or not sufficient

retention (micro or macro) had been achieved. Therefore, being able to increase the presence

of clinically detectable Fuji VII® material was thought to be a desirable indictor in the pathway

to caries prevention. Achieving better retention, therefore required a careful consideration of

the material properties and setting reaction. Subsequently this informed the modification to the

clinical protocol.

A key consideration for avoiding failure was to consider ways in which the setting reaction

could be controlled. The GIC setting reaction involves three phases (1) the decomposition of

glass in an acid-base reaction during the initial mixing (2) the second phase involves the

migration of ions that occurs during the working time and (3) the gelatinous phase as the

material gradual reconstructs in the inorganic fragments to becomes more rigid by creating a

matrix that increases in strength over the following 24-hours (Berg & Croll, 2015; Croll &

Nicholson, 2002). According to the manufacturer’s instructions the mixing should take 20

seconds and the initial set should take 1-minute; however, clinical observation suggested that

the setting reaction was happening much faster in the hotter, more humid weather (GC

Corporation, 2018). In addition, FS were placed for children in Group A (Pilot A) during April

71

which has an average temperature of 35 degrees Celsius in Phnom Penh. The limited working

time also meant that the time taken to place the material on more than one tooth could also

compromise the bonding to the tooth surface and so there should be only one tooth per mix. If

the material reaches the tooth during the second phase, then fewer free ions will be available

in the mix to allow for ionic exchange and adhesion to the tooth surface. The first phase and

bonding were further exploited by delivering the material to the tooth at the optimal time after

mixing and timing the mixing so that only 20 seconds was spent.

Timing was also important during the second phase of the setting. It is during the first and

second phase that the material is most vulnerable to moisture contamination (Berg & Croll

2015). During this time, contact with saliva can result in leaching of the ions from the material

and the result is a lower compressive strength. Therefore, meticulous moisture control was

achieved by holding the finger over the sealant for a full minute as counted by a timer. During

the original protocol anecdotal evidence suggested that operators were waiting until the second

phase of setting when the material was in the gelatinous phase before placing the finger because

they were worried that the material would stick to the glove when they removed the finger. For

an experienced operator, timing might not be such an issue because the work flow is more

efficient; however, SEAL Cambodia involved a multitude (over 100) of different operators

with varying level of experience and there is evidence that the dental students applied the

protocol with differing success to that of experienced Cambodian dental nurses (Sreang 2016).

The final measure to optimise FS retention was the adjustment of the occlusion. Given that Fuji

VII has a lower compressive strength to high viscosity (Fuji IX) GIC material then protection

of the material from the occlusion may have been important. Those protocol considerations

and the presence of multiple operators with varying levels of experience highlight the

importance of having a robust protocol, with strict timing and control measures in order to

ensure consistent results in turn better quality of care.

2.5.2.1 The pattern of material loss

Even with the control measures described in the previous paragraph, the retention of sealants

was lower than that reported in other studies (Chen & Liu 2013). There were four studies that

could be found which examined retention of Fuji VII material over multiple time points during

the first year of placement. One study reported retention rates around 80% at 1-year (Chen &

Liu, 2013), while another found retention rates around 40% (Kamala & Hegde, 2008). The key

differences among the protocols was that those studies with high retention rates only sealed

72

teeth which were fully erupted and clinically judged to be deep or susceptible to caries. In

contrast, those studies with the lower retention rates and the SEAL Cambodia project sealed

all FPM that were present or partially present in the mouth.

Another assumption had been made around the loss of FS was that there might have been a

primary loss of sealants during the first week due to unfavourable occlusal force. This may

have been possible for one in twenty Fuji VII® FS that were lost using the modified protocol

even after occlusal adjustment; however, it appears that the material is being lost gradually

over time in a more linear fashion. This linear pattern of loss was present in other retention

studies of Fuji VII GIC material (Chen & Liu 2013; Kamala & Hegde 2008). Such a pattern

could suggest a wear process rather than a gross fracture or a gross failure of adhesive bonding.

In saying that, the pattern of loss in Pilot A is not known and so, after taking into account the

properties of the material and application conditions, primary failure to bond cannot be ruled

out in that protocol.

What was not reported in the previous literature was differences in retention by tooth position

and sociodemographic characteristics. Among those in Group A, children who were in the 6-

year age-group, those who were in the less severe caries category and mandibular molars

appeared to have higher retention rates. Perhaps those younger children were more likely to

have non-cavitated deep pits and fissures whereas deep pits and fissures in older age groups

were already cavitated and hence not sealed/included in the study. This suggests some benefit

could be gained by shifting the target grade range from the second grade down to the first grade.

That mandibular teeth were more likely to have retained the sealant could be because the

material was systematically placed on the lower teeth before the maxillary teeth meaning that

the mandibular teeth received material at an early phase of the setting process. The other

possibility is that the lower teeth may have been more likely to have the finger press applied

because placing a finger in the mandibular arch is more ergonomically comfortable. This logic

supports the protocol for only placing only one sealant at a time and for holding the finger in

place for a full minute.

That those with higher caries rates were more likely to lose the FS could suggest something

about the level of acid in the oral environment is relevant for retention. The acid could be

coming from the sugary diet which drives the caries process, or due to the presence of

acidogenic plaque at the interface of the FS material and the tooth. Alternatively, it is more

likely that it could be due to acidic foods, and there are previous reports of high rates of erosion

73

among Cambodian children (Kubota et al. 2016). The presence of plaque or acid could

potentially create an erosive process on the GIC material. Unfortunately, the SEAL Cambodia

project did not collect data on those variables which drive the caries process (such as sugar

consumption) or record erosion and so it is not possible to estimate these influences or make

adaptions to the protocol which would counteract that.

2.5.2.2 High vs Low viscosity GIC

It was interesting that in Pilot B of the present study, Fuji IX® material was significantly more

likely to be lost at 1-week in a primary failure scenario. Despite this, at one-year the differences

were not significant. It is therefore possible to say that Fuji VII® low viscosity material

performed better in the early stages; however, it is not known whether the better performance

and one month and six months will render a better caries preventive effect. Also of note, the

modified protocol was designed to optimise the properties of Fuji VII®. It might be that Fuji

IX® would have performed better in the original protocol due to the longer working time which

may have given sufficient placement time despite the heat and humidity of the Cambodian

school-yard environment.

2.6 Conclusion

The original protocol used for the SEAL Cambodia project did not render the retention rates or

the caries preventive effects that were anticipated (Pilot A) therefore modifications to that

protocol were tested (Pilot B). The caries prevention achieved differed depending on the

severity of the baseline caries experience, whereby those with more extreme caries experience

realised a lower preventive increment. Pilot B applied a modified protocol and achieved better

retention rates at one month and six months compared to the original protocol. This informed

the practice of the SEAL Cambodia community project and further investigation is needed to

examine whether the higher retention rates using the modified protocol in the early stages of

Pilot B will render a greater caries preventive effect when implemented at scale.

74


Agouropoulos, A., Twetman, S., Pandis, N., Kavvadia, K. & Papagiannoulis, L. (2014) Caries-

preventive effectiveness of fluoride varnish as adjunct to oral health promotion and supervised

tooth-brushing in preschool children: A double-blind randomized controlled trial. Journal of

Dentistry, 42 (10), 1277–1283.

Ahovuo-Saloranta, A., Forss, H., Walsh, T., Nordblad, A., Mäkelä, M. & Worthington, H. V.

(2017) Pit and fissure sealants for preventing dental decay in permanent teeth. In Cochrane

Database of Systematic Reviews 2017(7):1-67.

Croll, T (2015). Glass Ionomer Restorative Cement Systems: An Update. Pediatric Dentistry,

37(2), 116–124

Bratthall, D. (2000) Introducing the Significant Caries Index together with a proposal for a new

global oral health goal for 12-year-olds. International Dental Journal, 50 (6), 378–384.

Chen, X. X. & Liu, X. G. (2013) Clinical comparison of Fuji VII and a resin sealant in children

at high and low risk of caries. Dental Materials Journal, 32 (3), 512–518.

Chher, T., Hak, S., Courtel, F. & Durward, C. (2009) Improving the provision of the Basic

Package of Oral Care (BPOC) in Cambodia. International Dental Journal, 59 (1), 47–52.

Chher T., Turton B., Hak S., Beltran E., Courtel F., Durward C. & Hobdell M. (2016) Dental


Survey. Journal of International Oral Health, 8 (1), 1.

Chia, L., Densie, I. & Morgan, C. (2015) An exploratory study of parental knowledge of early

childhood oral health care in Southland, New Zealand. NZ Dent J, 111 (1), 18–24.

Croll, T. P. & Nicholson, J. W. (2002) Glass ionomer cements in pediatric dentistry: review of

the literature. Pediatric Dentistry, 24 (5), 423–429.

de Amorim, R. G., Leal, S. C. & Frencken, J. E. (2012) Survival of atraumatic restorative

treatment (ART) sealants and restorations: a meta-analysis. Clinical Oral Investigations, 16

(2), 429–441.

Forss, H., Saarni, U.-M. & Seppä, L. (1994) Comparison of glass-ionomer and resin-based

fissure sealants: a 2-year clinical trial. Community Dentistry and Oral Epidemiology, 22 (1),

21–24.

Frazão, P. (2011) Effectiveness of the bucco-lingual technique within a school-based

supervised tooth-brushing program on preventing caries: a randomized controlled trial. BMC

Oral Health, 11 (1), 11.

Ganesh, M. & Tandon, S. (2007) Clinical evaluation of FUJI VII sealant material. Journal of

Clinical Pediatric Dentistry, 31 (1), 52–57.

GC Corporation (2018) Manufacturers instructions for GC Fuji VII radiopaque Glass Ionomer

Cement and Stabilisation Material [Online]

http://www.gcaustralasia.com/Upload/product/pdf/5/GC-Fuji-VII-CAPSULE_IFU-EN.pdf

[Accessed 30 August 2018].

http://www.gcaustralasia.com/Upload/product/pdf/5/GC-Fuji-VII-CAPSULE_IFU-EN.pdf

75

Holve, S. (2008) An observational study of the association of fluoride varnish applied during

well child visits and the prevention of early childhood caries in American Indian children.

Maternal and Child Health Journal, 12 (1), 64–67.

Hu, X., Zhang, W., Fan, M., Mulder, J. & Frencken, J. E. (2016) Frequency of remnants of

sealants left behind in pits and fissures of occlusal surfaces after 2 and 3 years. Clinical Oral

Investigations, 21 (1), 1–7.

Kamala, B. & Hegde, A. (2008) Fuji III vs. Fuji VII glass ionomer sealants–a clinical study.

Journal of Clinical Pediatric Dentistry, 33(1), 29–33.

Kubota, Y., Ly, S., Turton, B., Gonzalez, M., Abu Kasim, N., Chew, H., Durward, C., (2014)

The Relationship between Tooth Erosion and Lifestyle in Cambodian Adolescents. Presented

at the South East Asian Division Meeting, Kuching, Malaysia.

Lam, S. P. L., Baros, H., O’Grady, M. J., Kendall, G. E., Messer, L. B. & Slack-Smith, L. M.

(2012) Patterns of attendance of children under 12 years at school dental service in Western

Australia. The Open Dentistry Journal, 6, 69.

Liu, B. Y., Lo, E. C. M., Chu, C. H. & Lin, H. C. (2012) Randomized trial on fluorides and

sealants for fissure caries prevention. Journal of Dental Research, 91 (8), 753 – 758.

Macpherson, L. M. D., Anopa, Y., Conway, D. I. & McMahon, A. D. (2013) National

supervised tooth-brushing program and dental decay in Scotland. Journal of Dental Research,

92 (2), 109–113.

Marinho, V. C., Higgins, J., Logan, S. & Sheiham, A. (2003) Fluoride toothpastes for

preventing dental caries in children and adolescents. The Cochrane Library (1), 1-103.

Mickenautsch, S. & Yengopal, V. (2013) Validity of sealant retention as surrogate for caries

prevention – a systematic review. PLoS ONE, 8 (10), e77103.

Monse, B., Heinrich-Weltzien, R., Benzian, H., Holmgren, C. & van Palenstein Helderman,

W. (2010) PUFA–An index of clinical consequences of untreated dental caries. Community

Dentistry and Oral Epidemiology, 38 (1), 77–82.

Nourallah, A. W. & Splieth, C. H. (2004) Efficacy of Occlusal Plaque Removal in Erupting

Molars: A Comparison of an Electric Toothbrush and the Cross-Tooth-brushing Technique.

Caries Research, 38 (2), 91–94.

One-2-One Cambodia (2016) SEAL Cambodia Final Report. Phnom Penh, Cambodia.

Poulsen, S., Beiruti, N. & Sadat, N. (2001) A comparison of retention and the effect on caries

of fissure sealing with a glass-ionomer and a resin-based sealant. Community Dentistry and

Oral Epidemiology, 29 (4), 298–301.

Sreng R, Turton B, Chen P, Serey P, Durward C. (2016) SEAL Cambodia – Retention rates of

ART Sealants according to provider type. IADR General Session. Seoul, Republic of Korea.

Turton B, Durward C, Bach K, Manton D. (2014) Seal Cambodia – 60,000 children over 3

years. IADR SEA 28TH Annual meeting, Kutching, Malaysia.

76

Chapter 3

SEAL Cambodia – Evaluation of a modified protocol for

placement of Fuji VII Fissure Sealants

Presented in part as:

Turton B, Crombie F, Durward C, Manton D (2018) Seal Cambodia – a tale of two

protocols. ANZSPD, Brisbane, Australia.

Turton, B., Durward, C., Crombie, F., Manton D. (2016) SEAL Cambodia: Improved

Caries Prevention with a modified Protocol. IADR General Session. Seoul, Republic of

Korea

77

Abstract

Objective: To evaluate the caries preventive effect of hand-mixed GIC Fissure Protection (FS)

placed on FPM of 6-8 year-old children in Cambodia using two different protocols.

Methods: A school-based randomised controlled trial involved two cohorts. The first included

an intervention group (Group A) who received FS using the original protocol and a control

group (Group B) who did not receive FS. The second cohort involved an intervention group

(Group C) who received FS using a modified protocol and a matched control group (Group D)

who received FS using the original protocol. The modified protocol involved control of the

temperature of the material, the timing of mixing and placement, and adjustment of the

occlusion. Groups A, B, and C were followed-up at 1 y and Groups C and D were followed-up

at 2 y. Data analysis examined differences in caries incidence by Group.

Results: At 1 y and 2 y, 62.8%, and 68.0% follow-up rates were achieved, respectively. The

mean age was 8.1 y (SD 1.2) for the first cohort, and 6.6 y (SD 0.6) for the second cohort. The

baseline mean dmft for the first and second cohorts was 8.0 (SD 3.9) and 9.9 (SD 4.3)

respectively. There was no statistically significant difference in caries increment between

Groups in the first cohort at1 y. A preventive fraction of 89.1% at 1 y and 32.3% at 2 y was

achieved using the modified protocol in the second cohort (P < 0.05). Children with extreme

caries experience (dmft > 8) realised half the preventive fraction at 2 y compared to those with

dmft < 8 (22.3% and 45.8% respectively).

Conclusions: Children suffered from an extreme burden of dental caries. The modified FS

protocol had a significant impact on the caries preventive effect, although it was lower than

that reported in other studies. A more holistic approach is needed to reduce the burden of dental

caries in Cambodia.

78

3.0 Introduction – SEAL Cambodia evaluation of a modified

protocol for placing Fuji VII Fissure Sealants (FS)

The burden of dental caries in Cambodia is severe and the SEAL Cambodia project was

designed as a first step in reorientating dental services towards a more upstream approach. A

strong body of evidence supports FS as an intervention next in efficacy to tooth-brushing with

fluoridated toothpaste in terms of caries prevention. The rationale for the SEAL Cambodia

project was that by sealing the most susceptible permanent teeth, the First Permanent Molar

(FPM), soon after eruption, the proportion of permanent teeth affected by carious lesions could

be dramatically reduced.

Initially a protocol for GIC Fuji VII FS was developed and field-tested (Protocol found in

Appendix 6) and results presented in Chapter 2. The first FS protocol did not render the one

year caries preventive effect that was expected and a new application protocol was developed

aiming to increase retention (Tac et al. 2016). It was thought that the limited preventive effect

of the original protocol was due to lack of control of the temperature of the GIC material before

application, the moisture control in the mouth, the timing of mixing and placement, and the

lack of occlusal adjustment. Following discussions within the research team, several steps in

the protocol were revised and the new protocol was subsequently implemented for the

remainder of the SEAL Cambodia program. (Appendix 7)

While the pilot of the revised protocol appeared to be promising, further investigation was

required to validate the rationale of the study around reducing the burden of carious lesions and

their subsequent consequences in the permanent dentition. Over the course of the investigation

updated meta-analysis became available which suggested that there was insufficient evidence

for the use of GIC FS in achieving caries prevention and estimated preventive fractions

between 39% and -3% over two years (Ahovuo-Saloranta et al. 2017). Few high-quality studies

have examined the success of these interventions in a developing world context and among a

population which has such a severe caries experience as children in Cambodia. Furthermore,

none of the existing studies have child-reported measures (such as quality of life instruments)

as an outcome measure.

The aim of this study was to evaluate the caries prevention achieved among 6 to 8 year-old

Cambodian children using a modified GIC FS protocol shown to improve early retention rates.

79

Such a study will add to the quality of evidence around the benefits of GIC FS to reduce the

burden of caries among vulnerable populations.

3.1 Aim

To investigate the caries preventive effect of GIC FS in FPM of 6 to 8 year-old children in

Cambodia using a modified protocol that had previously improved retention rates.

3.2 Objectives

i. to evaluate the caries preventive effect of a modified FS placement protocol at one year

compared with the original protocol and a control group,

ii. to evaluate the caries preventive effect of FS placed using the modified protocol at two years

compared to a control group.

80

3.3 Methods

The present study had two parts; Part 1 was a comparison of the caries preventive effect of GIC

fissure sealants among the modified and original cohorts at one year, and Part 2 examined the

caries preventive effect of the modified cohort at two years against a contemporaneous control

group. Within the two parts there were four clinical groups: Group A was the original

intervention group, Group B was the original (negative) control group, Group C was the

modified protocol (intervention group) of the new cohort, and Group D was the positive control

group of the new cohort. All clinical examinations and clinical procedures were conducted in

a school-yard setting with battery-powered head-lights and children in a supine position.

Consent was gained by sending the consent forms to the school the week prior to the

examination and parents were given an opportunity to withdraw their child’s participation

should they wish (Appendix 1). Children were free to refuse assent at the time of examination

with no consequences. Participants were given oral hygiene education with a ratio of one

instructor to five children and they were given a toothbrush prior to the placement of the

sealants or following the examination for those in the control group. This ensured that the teeth

were clean prior to examination and placement of GIC sealants. Ethics approval was provided

by the National Committee for Health Research, Ministry of Health, Cambodia.

3.3.1 Clinical procedures

Participants in Group A (intervention group) and Group D (positive control) received FS using

the original protocol. It was not intended that Group D would receive FS; however, an

organisation which had previously partnered with the SEAL Cambodia project provided FS

independently for the schools in Group D. That organisation did not receive training on the

modified protocol. The original protocol involved up to two teeth (opposing arches) being

treated simultaneously (1) the dentine conditioner was applied to a clean tooth and then

removed with a wet cotton pellet; (2) the tooth was dried using cotton pellets; (3) hand-mixed

Fuji VII® was placed onto the occlusal surface of the tooth and pressed into the fissures using

the operators’ finger (no defined duration) which was lubricated with cocoa butter or Vaseline

and no subsequent occlusal adjustment.

The modified protocol which was applied to Group C involved the same work flow but with

some modifications: (a) cooling of the GIC liquid (Fuji VII® Liquid) in a cup of ice prior to

mixing; (b) timing the mixing (20 seconds) and placement of the material (60 seconds) using

a digital timer according to the manufacturer’s instructions; (c) placing one sealant at a time;

81

(d) placing the finger over the sealant for one minute using a timer; and (e) testing and adjusting

the occlusion if needed. Appendix 6 and 7 provides further detail on the two protocols.

3.3.2 Clinical examination and questionnaire

The DMF index was recorded at surface level according to the WHO Basic Survey Methods

(WHO, 2013). Each tooth was also given a score using the PUFA Index (Monse et al. 2009).

The degree of retention of FS material and the presence of any cavitated carious lesions on the

occlusal surfaces of FPM were recorded. Evaluation of FS was based on the proportion of

remaining material. Sealants that covered more than 2/3 of the fissure system were recorded as

‘fully retained’. If FS material was present but covered less than 2/3 of the fissure system, it

was recorded as ‘partially retained’. If the FS was not detectable visually it was recorded as

‘lost’.

Participants in Group C and Group D were asked a set of questions at each examination

including questions on oral hygiene, and four questions from the Oral Symptoms Domain of

the Child Perceptions Questionnaire which had been previously validated for use in a

Cambodian Setting (Turton et al. 2015). The questions were about mouth pain, sores (ulcers),

bad breath and food impaction in the context of the previous three months. The response

options for each item were “Never” (score 0), “Once or twice” (score 1), “Sometimes” (score

2), “Often” (score 3), and “Every day or almost every day” (score 4).

3.3.3 Part 1 – Methods for One-year investigation

The first part of the study involved three groups; Group A was the original intervention group

(N = 433), Group B was the original control group (N =262) and Group C was the modified

protocol intervention group (N = 364). Although the control group of the modified protocol

(Group D the positive control) was recruited contemporaneously to the intervention group, they

were not examined at 1 y due to logistical limitations. All groups were selected based on

randomised cluster sampling using the list of eligible schools provided by the School Health

Department. The inclusion criteria for the modified intervention group involved targeting for

age (not older than 7 y) in contrast to the original cohort which targeted based on the school

grade level of the child. Examiners were calibrated against an experienced dental

epidemiologist (BT) and achieved a kappa score higher than 0.85.

3.3.4 Part 2 – Two year analysis

82

The second part of the study involved two groups: Group C as described in Part 1 as the

modified protocol, the comparative group was a positive control group who had sealants placed

using the original protocol (Group D). Follow-up was maximised by working with school

administrators, having more than one ‘wave’ of data collection, and working with teachers to

identify children who were part of the cohort. Examiners went through with school

administrators the original hand-written hard copy of the school roll based on the first day of

school enrolment as well as electronic versions to account for those children who might have

modified their names between base-line and follow-up visits. There were two waves of data-

collection with examiners returning to schools two weeks later to capture any additional

children who might have been absent during the first wave of data collection. In addition,

teachers and classmates who may have taught the child at baseline were asked to identify

children who were not able to be found using the previously described means.

3.3.5 Data analysis

Data were entered into SPSS Ver. 20 (IBM, NY, USA) and data were analysed to produce

descriptive statistics on baseline caries experience and caries incidence as well as FS retention.

Data was collected on oral symptoms based on the prevalence of impacts, a child was said to

have an impact in oral symptoms if they responded with score 3 or higher to any of the items

in the domain. Descriptive analyses were performed as well as bivariate analysis to examine

differences in caries experience and FS retention by group membership. Caries groups were

defined using the Significant Caries Index (Bratthall 2000) whereby the group was divided by

terciles based on caries experience in the primary dentition (dmft). Differences in means among

groups were compared using the Kruskal-Wallis test and differences in proportions among

groups were compared using chi-squared test. Differences in means among related samples

were examined using the Wilcoxon signed rank test.

83

3.4 Results

3.4.1 Part 1 – One year comparison of old and new cohorts.

Data on attrition for Part 1 are presented in Table 3.1. There was no significant difference in

follow-up by gender or baseline caries experience.

Table 3.1 – Part 1 – Attrition analysis for the One year follow-up by group membership

and caries experience.

Total

N (Column %)

Lost

N (Row %)

One year Follow-up

N (Row %)

Group A (Original

intervention)

Male 237 (54.7) 98 (41.4) 139 (58.6)

Female 196 (45.3) 73 (37.2) 123 (62.8)

Total 433 (41.3) 171 (39.5) 262 (60.5)

Group B (Original Control)

Male 116 (46.2) 42 (36.2) 74 (63.8)

Female 135 (53.8) 46 (34.1) 89 (65.9)

Total 251 (24.0) 88 (35.0) 163 (64.9)

Group C (Modified protocol)

Male 181 (49.7) 68 (37.6) 113 (62.4)

Female 183 (50.3) 63 (34.4) 120 (64.6)

Total 364 (34.7) 131 (36.0) 233 (64.0)

Baseline caries group

High (dmft 0-5) 318 (30.3) 108 (34.0) 210 (66.0)

Very High (dmft 6-9) 356 (34.0) 136 (38.2) 220 (61.8)

Extreme (dmft >10)) 374 (35.7) 146 (39.0) 228 (61.0)

Total 1048 (100.0) 390 (37.2) 658 (62.8)

84

Data on sociodemographic characteristics of participants in Part 1 are presented in Table 3.2.

Children were spread evenly across schools although there was one large school (School XXI

from Phnom Penh) which had twice as many participants than other schools. Half of the

children in the sample were 8 years-of-age or older.

Table 3.2 – Part 1 - Sociodemographic Characteristics of Participants.

Male

N (Row %)

Female

N (Row %)

Total

N (Column %)

Age group (years)a

6-7 266 (50.6) 260 (49.4) 526 (50.7)

>8 263 (51.5) 248 (48.5) 511 (49.3)

School code

School I 12 (41.4) 17 (58.6) 29 (2.8)

School II 10 (41.7) 14 (58.3) 24 (2.3)

School III 18 (42.9) 24 (57.1) 42 (4.0)

School IV 26 (54.2) 22 (45.8) 48 (4.6)

School V 32 (64.0) 18 (36.0) 50 (4.8)

School VI 11 (44.0) 14 (56.0) 25 (2.4)

School VII 20 (55.6) 16 (44.4) 36 (3.4)

School VIII 11 (55.0) 9 (45.0) 20 (1.9)

School IX 33 (42.3) 45 (57.7) 78 (7.4)

School X 30 (62.5) 18 (37.5) 48 (4.6)

School XI 11 (73.3) 4 (26.7) 15 (1.4)

School XII 17 (51.5) 16 (48.5) 33 (3.1)

School XIII 2 (33.3) 4 (66.7) 6 (0.6)

School XIV 9 (47.7) 10 (52.6) 19 (1.8)

School XV 25 (47.2) 28 (52.8) 53 (5.1)

School XVI 24 (57.1) 18 (42.9) 42 (4.0)

School XVII 27 (50.9) 26 (49.1) 53 (5.1)

School XVIII 35 (55.6) 28 (54.5) 63 (6.0)

School XIX 15 (45.5) 18 (54.5) 33 (3.1)

School XX 36 (52.9) 32 (47.1) 68 (6.5)

School XXI 72 (53.7) 62 (46.3) 134 (12.8)

School XXII 32 (41.6) 45 (58.4) 77 (7.3)

School XXIII 26 (50.0) 26 (50.0) 52 (5.0)

TOTAL 534 (51.0) 514 (49.0) 1048 (100.0)

aDate of birth information missing from 9 participants

85

There was a statistically significant difference in the severity of caries by age-group, group

membership and school attendance (data are presented in Table 3.3). Children in Group C were

younger than those in Groups A and B. Those in Group C had a lower DMFT and a higher

dmft. There was a statistically significant amount of variation in caries experience among

schools.

86

Table 3.3 – Part 1 - Clinical Characteristics of participants by group membership.

DMFT

Mean (SD) P-value

dmft

Mean (SD) P-value

Any caries

N (Row %) P-value

Sex

Male 1.8 (2.1) 0.397 9.1 (4.2) 0.010 520 (97.4) 0.318

Female 1.9 (2.0) 8.4 (4.2) 497 (96.7)

Age group

6-7 y 1.5 (1.8) < 0.001 9.7 (4.2) < 0.001 514 (97.7) 0.120

>8 y 2.3 (2.1) 7.8 (3.9) 492 (96.3)

Group membership

Group A 2.2 (2.1) < 0.001 7.5 (4.1) < 0.001 240 (95.6) 0.209

Group B 2.3 (2.2) 8.3 (3.8) 420 (97.0)

Group C 1.2 (1.6) 10.2 (4.3) 357 (98.1)

School

School I 3.0 (2.6) < 0.001 8.3 (4.0) < 0.001 28 (96.6) 0.031

School II 2.3 (1.9) 8.8 (3.6) 24 (100.0)

School III 2.3 (25) 5.7 (4.0) 38 (90.5)

School IV 3.1 (2.7) 8.8 (3.6) 48 (100.0)

School V 2.4 (2.4) 8.2 (3.8) 50 (100.0)

School VI 2.5 (2.2) 8.0 (4.1) 25 (100.0)

School VII 2.3 (2.0) 8.7 (3.2) 36 (100.0)

School VIII 3.1 (2.5) 8.7 (3.5) 20 (100.0)

School IX 2.1 (1.8) 7.2 (4.3) 73 (93.6)

School X 2.9 (2.7) 8.1 (3.4) 48 (100.0)

School XI 2.4 (2.8) 8.1 (4.3) 15 (100.0)

School XII 2.1 (1.5) 8.6 (4.3) 33 (100.0)

School XIII 5.2 (1.8) 6.2 (3.2) 5 (83.3)

School XIV 1.5 (1.4) 6.8 (4.1) 17 (89.5)

School XV 2.2 (1.80 7.8 (3.8) 49 (92.5)

School XVI 1.4 (1.6) 8.4 (3.7) 40 (95.2)

School XVII 1.4 (1.4) 8.3 (4.3) 52 (98.1)

School XVIII 1.7 (1.8) 8.5 (4.1) 59 (93.7)

School XIX 1.2 (1.6) 10.2 (4.4) 32 (97.0)

School XX 1.7 (1.6) 10.3 (4.3) 67 (98.5)

School XXI 1.2 (1.6) 10.4 (4.3) 131 (97.8)

School XXII 0.9 (1.5) 9.5 (4.2) 75 (97.4)

School XXIII 1.0 (1.3) 10.8 (4.4) 52 (100.0)

TOTAL 1.9 (2.0) 8.8 (4.2) 1017 (97.0)

87

There was no difference between Group A and Group B in the proportion of children who had

one or more new occlusal lesions on FPM (data presented in Table 3.4). A lower proportion of

FPM in children in Group C had one or more new lesions. At tooth level, there is a statistically

significant difference in the proportion of teeth with new lesions. Teeth in Groups B and C had

a respective 10.2% and 89.1% lower chance of becoming cavitated compared to those in Group

A.

Table 3.4 – Part 1 - Caries preventive increment by group membership at one year.

Prevalence of

children with one or

more new cavitated

lesions

N (%) P-value

Proportion of

FPM with new

carious

cavitation’s

Proportion (SE) P-value

Preventive

fractiona

Group A

Original intervention 99 (37.8) < 0.001 32.5 (67.5) < 0.001

Group B

Negative control 59 (36.2) 36.6 (78.3) 10.2

Group C

Modified protocol 20 (8.6) 4.0 (15.0) 89.1

Total 178 (27.1) 23.4 (60.0)

aCalculated by comparison of the proportion of FPM with new cavitated lesions by group against the proportion

of cavitated lesions in Group A.

88

Children in Group A and Group B had more than six times the possibility for developing new

lesions on FPM (Table 3.5) when compared with Group C once baseline caries experience,

school, and gender were adjusted for.

Table 3.5 – Part 1 - Logistic regression model for risk of developing new carious lesions

based on group membership.a

B(SE) Odds Ratio (95% CI) P-value

Constant

-2.37 (0.23) - < 0.001

Group A

Original intervention 1.80 (0.29) 6.04 (3.46, 10.56) < 0.001

Group B

Negative control 1.87 (0.27) 6.47 (3.84, 10.90) < 0.001

Group C

Modified Protocol 0.0b

aCox and Snell – 0.102; Nagelkerke – 0.148; McFadden – 0.092 bThis parameter is set to zero because it is redundant

89

3.4.2 Part 2 – Two year evaluation of the modified protocol.

There was no significant difference in the distribution of the sociodemographic participants by

gender or age (Table 3.6); children in the control group had a statistically significant greater

proportion of female participants, although some schools (Schools II, IV, and V) had

statistically significant greater proportion of males. The mean age overall was 6.6 (SD 0.6)

years with no significant difference by group membership; Group C, 8.9 years (SD 0.6); Group

D, 6.5 years-of age (SD 0.6).

Table 3.6 – Part 2 - Characteristics of participants by gender, age, and group membership

and school.

Male

N (Row %)

Female

N (Row %)

Total

N (Column %) P-valuea

Age group (years)

Six 123 (48.0) 133 (52.0) 256 (41.7) 0.678

Seven 156 (47.3) 174 (52.7) 330 (53.7)

Eight 11 (39.3) 17 (60.7) 28 (4.6)

Group membership

Group C

Modified protocol 181 (50.8) 175 (19.2) 356 (57.5)

0.031

Group D

Positive control 113 (43.0) 150 (57.0) 263 (42.5)

School

School I 21 (20.0) 84 (80.0) 105 (17.0) < 0.001

School II 24 (60.0) 16 (40.0) 40 (6.5)

School III 24 (44.4) 30 (55.6) 54 (8.7)

School IV 21 (65.6) 11 (34.4) 32 (5.2)

School V 15 (65.2) 8 (34.8) 23 (3.7)

School VI 4 (100) 0 (0.0) 4 (0.6)

School VII 4 (80.0) 1 (20.0) 5 (0.8)

School VIII 15 (46.9) 17 (53.1) 32 (5.2)

School IX 36 (53.7) 31 (46.3) 67 (10.8)

School X 72 (55.8) 57 (44.2) 129 (20.8)

School XI 32 (42.1) 44 (57.9) 76 (12.3)

School XII 26 (50.0) 26 (50.0) 52 (8.4)

Total 294 (47.5) 325 (52.5) 619 (100.0)

aP-value calculated using χ2 statistics for differences among groups within the same column

90

There were some differences in dmft by gender, and school attendance. Males and some schools

had greater caries experience reaching statistical significance (Data are presented in Table 3.7).

Table 3.7 – Part 2 - Clinical characteristics of participants according to sociodemographic factors.

dmft

Mean (SD) P-valuea

DMFT

Mean (SD) P-valuea

Any caries

N (%) P-valuea

Sex

Male 10.4 (4.4) 0.013 1.2 (1.6) 0.001 290 (98.6) 0.562

Female 9.4 (4.1) 1.2 (1.6) 320 (98.5)

Age group (years)

Six 10.2 (4.6) 0.326 0.8 (1.4) < 0.001 251 (98.0) 0.611

Seven 9.8 (4.1) 1.6 (1.7) 326 (98.8)

Eight 9.0 (3.6) 1.7 (2.1) 28 (100.0)

Group membership

Group C 10.2 (4.3) 0.052 1.2 (1.6) 0.344 349 (98.0) 0.186

Group D 9.5 (4.2) 1.3 (1.7) 261 (99.2)

School

School I 9.8 (4.2) 0.013 1.3 (1.7) 0.064 104 (99.0) 0.955

School II 10.4 (4.0) 1.4 (1.6) 40 (100.0)

School III 7.6 (3.8) 1.3 (1.5) 53 (98.1)

School IV 10.6 (4.9) 1.4 (1.5) 32 (100.0)

School V 9.5 (3.5) 1.9 (2.8) 23 (100.0)

School VI 8.0 (5.8) 0.8 (1.0) 4 (100.0)

School VII 11.8 (3.8) 1.8 (1.8) 5 (100.0)

School VIII 10.2 (4.7) 1.2 (1.6) 31 (96.9)

School IX 10.3 (4.4) 1.7 (1.6) 66 (98.5)

School X 10.4 (4.3) 1.1 (1.6) 126 (97.7)

School XI 9.5 (4.1) 0.9 (1.6) 74 (97.4)

School XII 10.8 (4.4) 1.0 (1.3) 52 (100.0)

Total 9.9 (4.3) 1.3 (1.6) 610 (98.5)

aThe P-values presented are for differences among groups within the same column.

91

Data on attrition for Part 2 are presented in Table 3.8. There was a statistically significant

difference in follow-up by gender within the Group D where by a greater proportion of females

were followed-up at two years. There were no gender differences in follow-up within Group C

and there were no differences in follow-up by baseline caries experience.

Table 3.8 – Part 2 - Participant attrition analysis for the 2 year follow-up of the second

cohort.

Lost

N (%)

2-year follow up

N (%)

Total

N (Column %) P-valuea

Group C - Modified Protocol

Male 62 (34.3) 119 (65.7) 181 (49.7) 0.284

Female 59 (32.2) 124 (67.8) 183 (50.3)

Total 121 (33.2) 243 (66.8) 364 (57.5)

Group D – Positive Control

Male 45 (39.8) 68 (60.2) 113 (43.0) 0.017

Female 40 (26.7) 110 (73.3) 150 (57.0)

Total 85 (32.3) 178 (67.7) 263 (42.5)

Baseline Caries group

High (dmft <8) 61 (35.3) 112 (64.7) 173 (27.6) 0.576

Very High (dmft 8-10) 70 (30.4) 160 (69.6) 230 (36.7)

Extreme(dmft >10) 75 (33.5) 149 (66.5) 224 (35.7)

Total 206 (32.9) 421 (67.1) 627 (100.0)

aP-value for examining differences among groups within the same column

92

The children in Group C had a statistically significant greater FS retention rate at two years.

Those in Group D who had FS placed using the old protocol had a 7.5% (SE 21.6) retention

rate while those in the Group C who had sealants placed using the new protocol had a 17.1%

(SE 29.4) retention rate (P < 0.001). Children in Group C had a statistically significant one

third lower chance of developing a new carious lesion on FPM (Table 3.9). Differences were

further amplified after participants were stratified by baseline caries experience. Those children

in the High caries group had twice the caries preventive fraction compared with the two more

extreme risk groups. There was no significant difference in the prevalence of new pulpally

involved lesions; however, those in Group C had a statistically significantly lower chance of

experiencing oral symptoms.

Table 3.9 – Part 2 - Proportion of children with new carious lesions on occlusal of FPM

at 2-years by group membership.

Group C

Modified protocol

N (%)

Group D

Positive control

N (%)

Preventive

fraction (%) P-valuea

>1 new lesions FPM 72 (29.6) 70 (39.3) 32.3 0.024

>1 new pulpally

involved FPM 64 (26.3) 57 (32.0) 17.8 0.122

>1 impacts in oral

symptoms 117 (48.1) 122 (68.5) 29.8 < 0.001

Caries group

High 10 (16.7) 16 (30.8) 45.8 0.062

Very High 27 (30.3) 28 (39.4) 23.1 0.150

Extreme 35 (37.2) 26 (47.3) 21.4 0.152

aP-value for differences among groups within the same rows

93

The mean number of new interproximal lesions on FPM was 0.3 (SD 0.6) with a statically

significant greater mean number of interproximal lesions (P = 0.037; Kruskal-Wallis) in the

Modified Protocol group (0.3; SD 0.7) compared with the positive control group (0.2; SD 0.5).

One in five children (N = 83) within this cohort developed new interproximal lesions over the

observation period. Overall, the mean number of new lesions in FPMs was 0.5 (SD 0.8) for

both groups; therefore, interproximal lesions account for over half of the FPM increment in the

intervention group C and a lower proportion in the Positive Control group D (Table 3.10). Once

participants were divided by baseline caries experience, the preventive fraction was larger for

those with a less extreme caries experience.

Table 3.10 – Part 2 - Proportion children with new cavitated occlusal and interproximal

carious lesions by group membership.

Group C

Modified

Protocol

N (%)

Group D

Positive Control

N (%)

Preventive

fraction

(%)a P-valueb

>1 new occlusal

cavitated lesions on FPM 53 (21.8) 63 (35.4) 38.4 0.002

>1 new interproximal

cavitated lesions on FPM 56 (23.0) 24 (15.2) - 0.029

Baseline Caries group

High (dmft <8) 6 (10.0) 15 (28.8) 65.3 0.015

Very High (dmft 8-10) 23 (25.8) 24 (33.8) 23.7 0.298

Extreme(dmft >10) 24 (25.5) 24 (43.6) 41.5 0.029

aPreventive fraction calculated by dividing the proportion of that in the positive control group by that in the

modified protocol group bP-value relates to differences in groups within the same row

94

Data on odds ratios for developing new cavitated lesions on FPM by group membership are

presented in Table 3.11. Two models were created using logistic regression, when all lesions

involving the occlusal surface were considered then those in the control group had a 1.54x

greater chance of developing a new lesion over two years. After the occlusal lesions that

involved interproximal surfaces were excluded; children in the positive control group had twice

the chance of developing new lesions that were situated on the occlusal surface.

Table 3.11 – Part 2 - Logistic regression models showing odds ratio for the chance of

developing one or more occlusal lesions on FPM at two-years.a

Model 2 – All new lesions involving the occlusal and approximal lesions.b


Constant 0.86 (0.14) - < 0.001

Group D

Positive Control -0.431 (0.208) 1.54 (1.02, 2.314) 0.038

Group C

Modified Protocol 0

Model 3 – new lesions involving only occlusal surface.c


Constant -1.227 - < 0.001

Group D

Positive Control 0.675 (0.221) 1.96 (1.27, 3.03)

0.002

Group C

Modified Protocol 0

aAdjusted for school of attendance and baseline caries experience bCox and Snell – 0.010; Nagelkerke – 0.014; McFadden – 0.008 cCox and Snell – 0.022; Nagelkerke – 0.032; McFadden – 0.019

95

3.5 Discussion

3.5.1 General findings

The present study examined the effectiveness of the original SEAL Cambodia protocol against

a modified protocol at one and two years. It represents an important contribution to the limited

body of evidence around the prevention of caries using GIC FS applied outside the dental

surgery given that the most recent meta-analysis reported insufficient evidence for this

technology (Ahovuo‐Saloranta et al. 2017). The modified protocol produced a large preventive

increment at one year and a moderate preventive increment at two years. Child participants had

an extreme experience of dental caries which resulted in a high incidence of open approximal

lesions on FPM that extended onto the occlusal surface. Before detailed examination of the

findings of the study, it is appropriate to examine the study strengths and limitations.

The strengths were associated with the robust randomised cluster sampling procedure;

however, the follow-up rates fell short of the ideal. Further measures were put in place at the

two year follow-up which resulted in a 5% increase in follow-up. The children in the study

were representative of the wider SEAL project and representative of those in their respective

province based on data from the Cambodia National Oral Health Survey (CNOHS) (Chher et

al. 2016). Despite this, it appears that the rate of approximal lesions on FPM was higher in

Phnom Penh compared with Kampot and this may have limited some of the comparability

among the second cohort (Groups C and D).

This is the first study to report caries preventive fractions of GIC sealants among a group of

children with a disproportionately severe caries experience. It has been reported in other studies

that higher caries risk was associated with lower preventive increments, as observed in the

present study (Ahovuo-Saloranta et al. 2017). Children in the lower tercile of caries experience

had a comparable caries experience to those considered to be in the ‘high risk’ group of

previously published investigations.

3.5.2 Preventive benefit of the SEAL Cambodia intervention

The caries experience at baseline played an important role in explaining the overall caries

preventive effect at two years. Those in the extreme caries groups did not realise a statistically

significant reduction in the incidence of new carious lesions on FPM. The 38% preventive

fraction achieved was at the higher end of the confidence interval reported in the most recent

96

Cochrane review (Ahovuo‐Saloranta et al. 2017). It is clear that this preventive increment

would not have been possible without the use of the revised protocol that featured additional

control measures. It is also worth commenting that 20% of the children in the intervention

group developed new carious lesions on approximal surfaces during the course of observation.

Although caries prevention/control was observed among this group, this intervention did not

address the drivers of dental caries such as dietary or oral hygiene practices. That meant that

the same teeth which may have realised some benefit on the occlusal surface could have been

compromised by the cavitated approximal lesions. As such, the value of the SEAL Cambodia

intervention lies in its place as an appropriate first step of the dental community in Cambodia

to re-orientate services from ‘bottom of the cliff’ and towards services based on a foundation

of interventions to maintain oral health. That an untold number of dental students, dentists,

therapists and hygienists from Cambodia and abroad participated in delivering this intervention

was a valuable demonstration of the possibilities in terms of the role of dental professionals to

move ‘up stream’. Seal Cambodia formed a foundation upon which the dental community

could better understand the target population and form the basis upon which more

comprehensive strategy could be built; One-2-One Cambodia 2016.

3.5.3 Protocol considerations

One of the important findings of the project was the difference in preventive increment across

two different protocols, both of which were consistent with the manufacturer’s instructions and

based on the assumption of a temperate working environment. The key difference between the

protocols was the use of control measures to regulate the operators work flow and material

temperature. While GIC materials are claimed to be moisture tolerant, it appears that there is

wide variation in setting reaction depending on the temperature of the liquid. The original

protocol used material at ‘room temperature’ that is likely to be higher than 30 °C which could

have resulted in a faster setting time and the FS material being delivered to the tooth in the

‘gelatinous’ phase rather than during the initial working phase (Berg & Croll 2015). This would

have resulted in the material failing to flow into the deep pits and fissures and failing to form

an adhesive bond on the surface of the tooth. The gelatinous phase of setting is the time at

which the material is most sensitive to leaching valuable ions on contact with moisture (Berg

& Croll 2015). Therefore, the additional moisture control provided by the one-minute finger

press was essential for protecting the material as it transitions through the setting phases. This

ensured the maximum compressive strength of the material could be achieved. Finally, the

97

temperature of the material affects the volume dispensed by one drop from the GIC liquid

bottle. At 25-30 °C approximately 115-120 drops can be dispensed. In contrast, only 100-105

drops were dispensed from a bottle that is cooler than 20 °C. The temperature of the liquid had

a major impact on all steps of the SEAL protocol.

It is possible that the choice of Fuji VII® rather than Fuji IX® could have created a scenario in

which the protocol is more technique sensitive. Fuji IX® has a two minute working time

compared to a working time for Fuji VII® of one minute. Although a pilot study demonstrated

better retention of Fuji VII® at 6 months, it is possible that there could be some benefit in using

Fuji IX® in order to allow for a more forgiving protocol. In any case, the SEAL Cambodia

project shows that an adjustment and strict monitoring of a protocol, taking into account local

conditions, can make the difference between whether an intervention benefits participants or

not.

While the preventive fraction at one year was comparable to that in other studies, the two year

preventive fraction was lower than expected (40%) and the finding that sealants were being

lost gradually over time leads to the question of whether GIC was the optimal choice of

material. It could be possible that a resin sealant with a higher retention rate could have

achieved higher two year preventive results. Unfortunately, placing resin sealants would

require significant equipment investment and it is unlikely that this would have been achievable

at scale or in the school environment in Cambodia. The alternative to a school environment

would be to transport children to an equipped clinic but that would have been more problematic

in terms of consent, transport costs, supervision, safety and time lost from class. In addition,

many of the teeth that were targeted for sealants were not yet fully erupted and therefore, resin

sealants, requiring meticulous moisture control, were contraindicated.

The next question to consider is why the retention rates were lower than those reported in other

studies (Chen & Liu, 2013). The Pilot studies (Chapter 2) demonstrated that the loss of material

is gradual, rather than because of early failure, therefore, it is proposed that the material is lost

due to wear or an erosive process. Other researchers have reported the presence of retained

GIC fragments held microscopically in the base of the fissure system (Kamala & Hegde, 2008;

Chen & Liu, 2013). It is not clear if this might be the case for the present study due to the fact

that those studies were on ART Sealants placed using high viscosity Fuji IX GIC rather than

Fuji VII. In light of the early loss of GIC, one should consider whether re-examining children

and replacing or repairing lost or deficient GIC sealants after one year should be recommended.

98

This conundrum further highlights the inadequacy of GIC fissure sealants as a stand-alone and

one-off preventive strategy.

3.5.4 Self-reported oral symptoms

One unique aspect of this study is the inclusion of a self-reported oral health measure as an

outcome. Although there was no statistically significant difference in the prevalence of new

pulpally involved lesions, and a higher rate of approximal carious lesions among Group C (the

modified protocol group) there was a significant difference in oral symptoms as reported by

the oral symptoms subdomain of the Child Perceptions Questionnaire. The proportion of

children in the original validation sample with impacts in oral symptoms was similar to those

in the control group of the second cohort (Turton et al. 2015) and the intervention appeared to

realise a one third reduction in the proportion of children experiencing oral symptoms. A

reduction in oral symptoms among a population with such a severe caries experience is

valuable and warrants further investigation into the role of dental interventions in improving

the quality-of-life for young children in Cambodia.

3.5.5 The theoretical argument for SEAL as a preventive strategy

The theory behind targeting FPM in 6 to 8 year-old children was that (a) the majority of new

lesions that occur in the early permanent dentition occur on occlusal surfaces of FPMs, (b) that

there is a window of around two years when enamel is less mature and more soluble, and (c)

that diet and oral hygiene may improve across the primary school years due to a fixed school

routine and lower consumption of sugary food (Chher et al. 2016). Unfortunately, the natural

history of caries among Cambodian children did not follow the pattern described in other

sealant studies among mostly western populations which have better access to fluoride and

lower exposure to non-nutritious foods.

In Cambodian children a large number of lesions develop in permanent teeth other than FPM

and analysis of the CNOHS data shows that two in five 12-year-old children have cavitated

carious lesions on maxillary anterior teeth. In addition, improvements in oral hygiene routines

and dietary practices don’t appear to occur until the children are older and reach secondary or

high school (Oral Health Bureau, 2014). Part of this is because primary school in Cambodia is

only four hours-per-day and 180 days-per-year. Although there is a ‘healthy food’ policy from

the school health department (Ministry of Education Youth and Sport, 2015), anecdotal

evidence would suggest that most schools make further income by allowing vendors onto the

99

school grounds to sell (often non-nutritious) foods and drinks to children. This demonstrates

that the underlying theory, based on ‘at risk’ populations in a western setting do not apply well

in a Cambodian setting.

In addition to the socio-behavioural factors compromising the success of a sealant program in

6 to 8 year-old children, the present project was confounded by the uncontrolled caries process

that was underway by the time the children arrived in the school environment. One in four

participants already had cavitated lesions in FPM at baseline. This suggests the need to address

the caries process for children before they enter the school system. At the time that the SEAL

Cambodia strategy was conceived in 2010 there had only been one published report of ECC in

a small Cambodian community of garment workers’ children (Todd et al. 1994). A more

comprehensive understanding of the natural history of dental caries, particularly among the

youngest age groups, would have been helpful in informing strategy for larger scale oral health

interventions in Cambodia.

3.6 Conclusion

The modified FS protocol rendered an improved preventive fraction compared to the original

protocol at one year. However, the two year preventive increment was not as beneficial as

expected due to a large proportion of approximal lesions on FPM. Further investigation is

needed to examine how benefit could be gained by addressing the caries burden prior to the

arrival of children in the school environment.

100


Ahovuo‐Saloranta, A., Forss, H., Walsh, T., Nordblad, A., Mäkelä, M. & Worthington, H.V.

(2017) Pit and fissure sealants for preventing dental decay in permanent teeth. The Cochrane

Library.7, CD001830.

Berg, Joel H. & Theodore P. Croll (2015) Glass Ionomer Restorative Cement Systems: An

Update. Pediatric Dentistry, 37 (2), 116–124.

Bratthall, D. (2000) Introducing the Significant Caries Index Together with a Proposal for a

New Global Oral Health Goal for 12-Year-Olds. International Dental Journal 50, (6), 378–

384.

Chen, X. X. & Liu, X.G. (2013) Clinical Comparison of Fuji VII and a Resin Sealant in

Children at High and Low Risk of Caries. Dental Materials Journal, 32 (3), 512–518.

Chher, T., Turton, B.J., Hak, S., Beltran, E., Courtel, F., Durward, C. & Hobdell, M. (2016)

Dental caries experience in Cambodia: Findings from the 2011 Cambodia National Oral Health


Kamala, B. & Hegde, A. (2008) Fuji III vs. Fuji VII glass ionomer sealants–a clinical study.

Journal of Clinical Pediatric Dentistry, 33 (1), 29-33.

Ministry of Education Youth and Sport, Cambodia (2015) Policy for food in schools.

One-2-One Cambodia (2016) SEAL Cambodia final report. Phnom Penh, Cambodia.

Oral Health Bureau, Ministry of Health Cambodia (2014) Cambodia National Oral Health

Survey Final Report. Phnom Penh, Cambodia.

Tak N, Turton B, Durward C (2015) Seal Cambodia – A comparison of two protocols; IADR

SEA 29th Annual meeting, Bali, Indonesia.Presentation ID: 071

Todd, R.V., Durward, C.S., Chot, C., So, P.K. & Im, P. (1994) The dental caries experience,

oral hygiene and dietary practices of preschool children of factory workers in Phnom Penh,

Cambodia. International Journal of Paediatric Dentistry, 4(3), pp.173-178.

Turton, B.J., Thomson, W.M., Foster Page, L.A., Saub, R.B. & Razak, I.A. (2015) Validation

of an oral health–related quality of life measure for Cambodian children. Asia Pacific Journal

of Public Health, 27 (2), NP2339-NP2349.

99

Chapter 4

Socio-behavioural risk factors for Early Childhood Caries

(ECC) in Cambodian preschool children.

Published in part as:

Turton B J, Durward C S, Manton D (2015) Early Childhood Caries and Maternal

Caries Experience in a Convenience Sample of Cambodian Pre-schoolers. Pediatric

Dental Journal 25 (1), pp. 14-18.

Turton B J, Durward CS, Manton D, Bach K, Yos C (2015) Socio-behavioural risk

factors for early childhood caries (ECC) in Cambodian preschool Children

European Archives of Paediatric Dentistry. 17(2), pp. 97-105

100

Abstract

Objective: The aim of the present study was to explore the socio-behavioural risk factors

for ECC in Cambodia.

Methods: A convenience sample of 362 Primary Caregiver (PCG)-child dyads with

children ranging from birth through to six years-of-age participated in a structured

interview and intra-oral examination.

Results: 244 of 362 (65.6%) participants had Early Childhood Caries (ECC) and 50.6%

had Severe Early Childhood Caries (sECC). There were significant associations between

caries experience and tooth-brushing, dietary and nursing habits. The Odds Ratio (OR)

for sECC in children who started brushing before the age of 18 months was 0.41 (CI 0.18,

0.93), whilst for children who continued to breast-feed after the age of two years the OR

was 5.31 (CI 1.50, 18.79).

Conclusions: The most prominent risk factors for ECC in the present study were lack of

tooth-brushing and continued breast-feeding past the age of two years.

101

4.0 Chapter 4 – Cambodia Smile cross-sectional study.

Introduction

Children in Cambodia have a high burden of dental caries and the CNOHS reported that

the mean dmft for 6 year-old children was 9.0 (SD 4.9) and the prevalence of pulpally

involved primary teeth was 64.8% (Chher et al. 2016). This burden of dental caries

suggests that school age is too late for successful preventive intervention and prior to the

present study, little was understood about the natural history of ECC in the Cambodian

context. The only published report on ECC in Cambodian preschool children prior to this

body of work was from a convenience sample of the children of factory workers in Phnom

Penh in 1994. The authors reported a mean dmft of 5.7 (SD 5.2) for children with a mean

age of 4 (SD 1.4) years (Todd et al. 1994). Aside from this study, there is very little

information available about the pattern, severity and risk factors involved with the caries

experience of Cambodian pre-schoolers. Although the cause and drivers of dental caries

are well known and well documented (Sheiham & James 2015), specific investigation in

each setting is beneficial in order to better understand how known socio-behavioural

factors influence ECC in a Cambodian environment.

Dental caries in preschool children has been given a variety of labels including ‘baby-

bottle decay’, ‘nursing caries’ and most commonly ECC and severe-ECC (sECC) (Ismail

& Sohn 1999). The American Academy of Paediatric Dentistry defines sECC as ‘any

sign of smooth-surface caries in children below the age of 3 years; from ages 3 through

5, 1 or more cavitated, missing (due to caries), or filled smooth surfaces in primary

maxillary anterior teeth or a decayed, missing, or filled score of ≥ 4 (age 3), ≥ 5 (age 4),

or ≥ 6 (age 5)’(Drury et al. 1999). All definitions involve the use of the DMF/dmf index

which has some limitations in capturing the true extent of dental caries in the Cambodian

population as the carious lesion is often extensive and this is not fully captured by the

DMF/dmf index, in addition the carious lesions are largely untreated (Chu et al. 2008;

Todd et al. 1994). To address this issue the PUFA index is used to record the severity of

untreated carious lesions by recording the extent of pulpal involvement and augments the

information provided by the DMF index, and as such more accurately describes untreated

carious lesions in high caries populations (Monse et al. 2010).

The impact of ECC on the lives of children includes pain and infection, failure to thrive,

low weight, difficulty studying and absenteeism from school, and social problems

102

(Abanto et al. 2011; Gift et al. 1992). These justify an investigation of ECC in the

Cambodian environment. Such an investigation also provides some context for the

possibility that addressing and reducing caries experience in this age group may lead to

improved life-long oral health among the wider Cambodian population.

That Cambodian children have severe caries experience that is not well described

highlights the need to understand the underlying determinants of dental caries in the

Cambodian context better. Previous authors studying young children in other countries

have proposed risk models for ECC and implicated, as contributing/associated factors,

some common variables such as bottle-feeding, prolonged night-time breast-feeding, the

consumption of sugary drinks, lack of oral hygiene habits, low socioeconomic status

(SES), and high maternal caries experience (Çolak et al. 2013). Studies relating to

Cambodia report unfavourable oral hygiene practices and weaning practices (Chu et al.

2008; Bach 2015).

Two of the most common predictors of a child’s dental caries experience are untreated

caries experience of the primary caregiver (PCG) (Abiola et al. 2009; Li & Caufield 1995;

Nowjack-Raymer & Gift 1990; Steffensen 1990) and SES (Willems et al. 2005; Desai &

Alva 1998). The typical family unit in Cambodia may be different from that in other

societies and anecdotal evidence suggests that the grand-parents often play an important

role in raising children. Previous studies have not investigated the relationship between

caries experience of the PCG and preschool children they care for in Cambodia.

The ‘Cambodia Smile’ cross-sectional study was conducted to investigate the risk

indicators for ECC in the Cambodian context to inform strategies for reducing the burden

of dental caries among young children in Cambodia.

103

4.1 Aim

To describe the epidemiological aspects of ECC in a Cambodian context in order to

identify modifiable risk factors for future intervention.

4.2 Objective

To examine the natural history of caries disease patterns and to identify the socio-

behavioural factors of ECC that are risk indicators of severe caries experience among a

convenience sample of Cambodian preschool children and their Primary Care Givers.

104

4.3 Methods

4.3.1 General approach

The present study was a cross-sectional survey of a convenience sample of 362 Primary

Caregiver-child dyads. Each PCG took part in a structured interview and both the PCG

and the child had a standardized intra-oral examination. This survey was part of the

‘Cambodia Smile’ project, an initiative by a locally-based NGO One-2-One Cambodia.

This chapter reports some of the findings from the ‘Cambodia Smile’ survey.

Data collected by the NGO was collected under the jurisdiction of an organisation

conducting a needs assessment for a health program. The de-identified data set was then

analysed for the present study and ethical approval for re-analysis of the existing data set

was provided by University of Melbourne, Health Sciences Human Ethics Sub-

Committee (Ethics ID: 1545412). The investigators followed ethical principles by

obtaining informed consent (from the individuals and from community leaders) and

ensuring that participants benefited from taking part in the survey; each participant

received a toothbrush and fluoridated (1000 ppm F) toothpaste as well as oral health

education.

4.3.2 Participants

A convenience sample of PCGs and their preschool children (younger than 6 years-of-

age) was obtained through mobile dental clinics. The local village chiefs gave consent

for the survey and invited members of the village to participate. Children from three

provinces of Cambodia took part: Kampong Speu (two different sites with 95 pairs from

Kampong Speu 1, and 92 pairs from Kampong Speu 2); Preveng (77 pairs); and Phnom

Penh (98 pairs). Interviewers and examiners consisted of senior dental students and

dentists (all native Khmer speakers) who volunteered to work in a mobile dental clinic

(operated by the NGO One-2-One Cambodia). The team comprised six examiners and

interviewers at each survey site. All were trained and underwent standardization exercises

before the study began.

All PCGs with a preschool child were invited to take part and the interviewers obtained

informed consent by explaining the study verbally and then obtaining written consent.

All interviews were conducted in Khmer.

105

4.3.3 Clinical Measures

Teeth were not cleaned and were examined wet, consistent with WHO guidelines (WHO

2013). The lead investigator trained four other examiners by examining five participants

and comparing and discussing the findings. Each tooth was given two codes, one for

DMFT status and another for PUFA status (Monse et al. 2013). The data were recorded

on paper records (Appendix 8) then entered into an Excel® spread sheet (Microsoft Corp.,

WA, USA). ECC and sECC were determined according to definitions given by the

American Academy of Pediatric Dentistry (AAPD). Caries severity was further

categorised by using the Significant Caries Index (Bratthall, 2000), which identified the

most severely affected 30% of participants. For this variable all PCGs were grouped

together and children were grouped according to age.

4.3.4 The questionnaire

Parents or Caregivers were asked a range of questions relating to oral health and general

health behaviours. Each interviewer spoke Khmer as their first language and received

training on how to conduct the interviews. The interviews were structured and conducted

in Khmer and followed a written set of questions. The items that were used along with

the response options are presented in Table 4.4 & 4.5. If the PCG responded that breast-

feeding had continued past the age of 24 months (even if the feeding had ceased at the

time of the examination), they were categorised as breast-feeding over the age of two-

years.

4.3.5 Data analysis

Data were analysed using SPSS Ver. 20 (IBM, NY, USA). Descriptive statistics were

generated prior to undertaking bivariate and multivariate analysis. Differences among

proportions were compared using the chi-squared (χ2) test. Differences between means

were compared using the Mann-Whitney U-test or the Kruskal-Wallis H-test as

appropriate. Multivariate analysis was conducted by using binary logistic regression to

examine the predictors of sECC and a multinomial stepwise regression model to predict

membership to the groups of ‘ECC’, ‘sECC’, and ‘Significant Caries’. The version of

SPSS did not have an option for producing collinearity statistics for logistic models and

therefore the same variables were entered into a linear regression model to test for multi-

collinearity. All variables identified with a statistically significant association with caries

106

experience during bivariate analysis were entered into the binary logistic regression

model.

107

4.4 Results

Data on the sociodemographic characteristics of participants by sex are presented in

Table 4.1. The children were aged from 1 month to 6 years-of-age. Mean age was 42

months (SD 19). Greater than two thirds of participants had their mother as the primary

caregiver. There was an even distribution by sex and location, however, the mean age of

participants differed significantly (P < 0.005; χ2 test) by their relationship to the PCG and

by location. Children who did not have their mother as their PCG were, on average, eight

months older and those from the Phnom Penh sample were five months younger than the

mean age (P < 0.005; t-test).

Table 4.1 – Cambodia Smile cross-sectional study - Sociodemographic characteristics of

participants.a

Male

N (%)

Female

N (%)

Mean Age in

months (SD)

Total

N (column %)

Location

Kampong Speu 1 50 (28.2) 44 (24.0) 44 (18)b 95 (26.2)

Prey Veng 37 (20.9) 39 (21.3) 42 (23) 77 (21.3)

Phnom Penh 51 (28.8) 47 (25.7) 36 (17) 98 (27.1)

Kampong Speu 2 39 (22.0) 53 (29.0) 46 (18) 92 (25.4)

Primary caregiver

Mother 124 (70.1) 135 (73.8) 40 (19)b 261 (72.1)

Other primary

caregiver

53 (29.9) 48 (26.2) 48 (17)b 101 (27.9)

Education of

primary caregiver

Illiterate 40 (22.6) 35 (19.1) 42 (19) 75 (20.7)

Preschool 48 (27.1) 58 (29.0) 42 (18) 101 (27.9)

Primary school 64 (36.2) 74 (40.4) 42 (20) 139 (38.4)

Secondary school 11 (6.2) 5 (2.7) 44 (19) 17 (4.7)

University 14 (7.9) 16 (8.7) 38 (21) 30 (8.3)

Total c 177 (48.9) 183 (50.6) 42 (19) 362 (100.0)

a Sex data missing from two participants; brackets contain row percentages unless otherwise indicated. b Significantly difference among groups within the same column; P < 0.05; χ2 test or Kruskal-Wallis test

as appropriate.

108

Some differences existed in caries experience of the children based on sociodemographic

indicators (Table 4.2). There was a statistically significant difference in caries experience

of the children by location and relationship to the PCG. Children who had someone other

than their mother as their PCG had a higher dmft score and approximately 5% greater

prevalence of ‘any caries’ experience and sECC when compared with the overall group.

The children from Prey Veng had a higher dmft (7.0 SD 6.4); however, the children from

Kampong Speu 1 had a higher prevalence of caries (+ 10%) and a higher prevalence of

sECC, cavitated carious lesions in anterior and posterior teeth (+ 5%). Two thirds of

children had cavitated carious lesions present and one half had sECC. Approximately half

of the children had cavitated carious lesions involving the posterior teeth and nearly two

thirds of children had cavitated carious lesions involving the anterior teeth.

The bivariate analysis of PCG caries experience by caries experience of the children is

presented in Table 4.3. A significant positive association existed between child and

mother caries experience. Children of mothers with pulpally involved teeth or ‘significant

caries’ had more greater caries experience (P < 0.05). There was no statistically

significant association between the caries experience of the PCG who were not the

mothers of the children in their care.

109

Table 4.2 – Cambodia Smile Cross-sectional study - Socio-demographics characteristics by caries experience of the child.

dmft

Mean (SD)

Any caries

N (row %)

Any PUFA

N (row %)

SECC

N (row %)

Significant caries

N (Row %)

Anterior Caries

N (row %)

Posterior caries

N (row %)

Sex

Male 5.4 (5.5) 117 (66.1) 45 (25.4) 90 (52.9) 60 (33.9) 107 (60.5) 97 (54.8)

Female 5.3 (5.5) 123 (67.2) 59 (32.2) 85 (49.7) 60 (32.8) 107 (58.5) 98 (53.6)

Location (Community)

Kampong Speu 1 5.6 (5.5)a 74 (77.1)a 27 (28.1)a 61 (63.5)a 33 (34.4)a 67 (69.8)a 62 (64.6)a

Prey Veng 7.0 (6.4) 54 (67.5) 35 (43.8) 40 (59.7) 40 (50.0) 51 (63.8) 49 (31.2)

Phnom Penh 3.4 (4.8) 50 (50.5) 26 (26.3) 33 (33.7) 26 (26.3) 43 (43.4) 38 (38.4)

Kampong Speu 2 4.4 (4.4) 66 (68.0) 17 (17.5) 44 (48.4) 23 (23.7) 56 (57.7) 50 (51.5)

Primary Caregiver education

Illiterate 5.6 (5.4) 56 (72.7) 23 (29.9) 41 (56.9) 26 (33.8) 49 (63.6) 44 (57.1)

Preschool 4.2 (4.6) 62 (59.6) 27 (26.0) 44 (44.4) 26 (25.0) 53 (51.0 52 (50.0)

Primary school 5.8 (5.9) 95 (66.4) 42 (29.4) 70 (51.9) 51 (35.7) 89 (62.2) 76 (53.1)

Secondary school 6.0 (5.0) 13 (76.5) 3 (17.6) 10 (58.8) 7 (41.2) 10 (58.8) 16 (51.6)

University 5.3 (6.2) 18 (58.1) 10 (32.3) 13 (44.8) 51 (35.7) 16 (51.6) 11 (64.7)

PCG relationship to child

Mother 6.1 (5.2)b 165 (61.3)a 76 (28.3) 121 (47.5)b 90 (32.8) 151 (56.1) 132 (49.1)a

Other relationship 5.0 (5.6) 79 (76.7) 29 (28.2) 57 (58.8) 32 (31.1) 66 (64.1) 67 (65.0)

Total 5.3 (5.4) 244 (65.6) 105 (28.2) 178 (50.6) 122 (32.8) 217 (58.3) 199 (53.5)

a P < 0.01; χ2 test or Kruskal-Wallis test as appropriate for comparison of groups within the same column b P < 0.05; χ2 test or Kruskal-Wallis test as appropriate for comparison of groups within the same column

110

Table 4.3 – Cambodia Smile cross-sectional study - Maternal caries experience by Child caries experience.

dmft

Mean (SD)

Any caries

N (row %)

Any PUFA

N (row %)

SECC

N (row %)

Significant caries

N (row %)

Anterior caries

N (row %)

Posterior Caries

N (row %)

Mother

Any caries 5.1 (5.6)a 152 (63.3)b 70 (29.2) 111 (48.9) 82 (34.2) 138 (57.5) 123 (51.2)b

Any PUFA 5.8 (5.8) 108 (67.5)b 55 (34.4)a 81 (54.4)b 62 (38.8)b 98 (61.2)b 89 (55.6)b

Significant Caries 6.9 (6.1)a 63 (70.8)b 31 (34.8) 52 (65.0)a 42 (47.2)a 61 (68.5)b 54 (60.7)b

Other Caregiver

Any caries 6.3 (5.2) 70 (79.5) 24 (27.3) 50 (61.0) 28 (31.8) 58 (65.9) 58 (65.9)

Any PUFA 6.5 (5.5) 55 (79.7) 19 (27.5) 39 (60.0) 23 (33.3) 48 (69.6) 45 (65.2)

Significant Caries 6.7 (5.8) 30 (81.1) 10 (27.0) 20 (60.6) 13 (35.1) 27 (73.0) 21 (56.8)

a P< 0.01; χ2 test or Kruskal-Wallis test as appropriate for differences among groups within the same column b P<0.05; χ2 test or Kruskal-Wallis test as appropriate for differences among groups within the same column

111

The responses to the questionnaire are presented in Table 4.4 and 4.5. Nearly two thirds of

children did not have their teeth cleaned and of those who did, three in five did not have any

help with brushing. Most of the mothers (95.4%) had breast-fed the child at some stage and

more than half of the mothers with children 4 to 5 years-of-age were still breast feeding. In

addition, the great majority of children (94.4%) slept with their mother at some stage, and more

than two thirds of the children aged over three years still slept with their mother at night.

112

Table 4.4 – Cambodia Smile cross-sectional study - Questions on nursing habits and oral

health behaviours by responses from the primary-caregiver.

Oral Hygiene N

(column %)

Does your child have his/her teeth

cleaned?

Yes

No

142 (38.2)

230 (61.8)

If yes how? Toothbrush

Other

125 (88.0)

17 (12.0)

Who brushes your child’s teeth?

Child by themselves

Child with help from parent

or someone else

81 (57.0)a

61 (43.0)

How often does your child usually have

his/her teeth brushed?

Less than once a day

> Once a day

34 (23.9)a

108 (76.1)

What age did tooth-brushing start?

Given in months < 18 months

>18 months

31 (8.3)a

341 (91.7)

Does your child use toothpaste?

Yes

No

106 (74.6)a

36 (25.4)

If your child doesn’t use toothpaste, why

not?

Don’t think it’s necessary

Cost too much money

Other reason

Data missing

20 (27.8)a

27 (37.5)

25 (34.7)

70

Nursing Habits

Did (or does) your child sleep with you at

night

Yes

No

351 (94.4)

21 (5.6)

If yes, age of stopping sleeping with child

Age given in

months and

grouped into

years.

>1 year

> 2 years

> 3 years

> 4 years

351 (100.0)

331 (94.3)

291 (82.9)

244 (69.5)

Did you ever breast-feed your child?

Yes

No

355 (95.4)

17 (4.6)

Still breast feeding?

Yes

No

76 (20.4)b

296 (79.6)

If no, when did breast-feeding stop

Age in months

then grouped

into years

>1 year

> 2 years

> 3 years

> 4 years

355

(100.0)b

338 (95.2)

276 (77.7)

208 (58.6)

After age 1 did your child breast-feed at

night?

Yes

No

Don’t Know

253 (71.3)b

91 (25.6)

11 (3.1)

a Proportion within those who brush their teeth b Proportion within those who breastfeed

113

The mean age for introducing solids into the infants’ diet was just older than one year

(Table 4.5). Half of the caregivers reported adding sugar to weaning foods. Most of the

children consumed sweet drinks and sweet foods more than two times-per-day.

Only one in three PCG reported that their child had ever used a baby bottle. Among those

who had used a bottle, the mean age for stopping bottle-feeding was two years. Within

the bottle-feeding group, two thirds were bottle-fed at night beyond the age of one year,

and one third added sugar to the contents of the bottle.

114

Table 4.5 – Cambodia Smile cross-sectional survey - Questions on dietary habits by responses from the

primary caregiver.a

Dietary Habits

What age did you introduce solid foods to your child?

Given in months (Mean and standard

deviation)

13 (SD 13)

What weaning foods did you give to your baby?

Asked to list and if

sugar was added.

Sugar added

No sugar added

195 (51.6)

180 (48.4)

How many times during the day does your child have:

sweet drinks, sweets, sweet snacks, fruit, vegetables,

plain (unsweetened) milk

Frequency given as

> two times per day.

select >1

Sweet drinks

Sweets

Sweet snacks

Fruit

Vegetables

Plain milk

332 (89.2)

294 (79.0)

212 (57.0)

304 (81.7)

129 (34.7)

328 (88.2)

What does your child usually drink?

Asked to list.

Select >1.

Water

Milk

Juice

Coffee

327 (87.9)

55 (14.8)

49 (13.2)

12 (3.2)

What snacks does your child usually eat?

Asked to list. Could

select >1

Packaged snacks

Rice porridge

Rice

Fruit

Sweet fried cake

Sweet bread

Ice cream

Meat

Other

185 (49.7)

143 (38.4)

177 (47.6)

20 (5.4)

26 (7.0)

24 (6.5)

3 (0.8)

7 (1.9)

5 (1.3)

Bottle-feeding Habits

Did you ever bottle-feed your child?

Yes

No

111 (29.8)

261 (70.2)

When did bottle-feeding start? Mean age given in months

8 (SD 8)

When did bottle-feeding stop (unless still bottle

feeding)?

Give age that bottle feeding stopped

(mean age given in months)

24 (SD 23)

After Age 1, did your child bottle-feed at night?

Yes

No

Missing

72 (66.1)b

37 (33.9)

2

What did you put in the bottle?

Asked to list and

whether sugar was

added

Sugar added

No Sugar added

29 (26.1)b

82 (73.9)

After age 1, how often did you child have a bottle

during the day?

Answer given in

times per day

< 3 times per day

> 3 times per day

Don’t know

33 (29.7)b

66 (59.5)

12 (10.8)

What age do you think bottle-feeding should stop? Give age that bottle feeding stopped

(mean age given in months) 24 (SD 16)

a Data are presented as Number of participants and percentage among response options of the question unless otherwise indicated. b Proportion within those that

bottle feed.

115

There was a statistically significant difference in caries experience by certain socio-behavioural

variables (Table 4.6). Higher caries experience was found in infants who did not brush, did not

have help brushing, used toothpaste, ate more than one sweet snack as their usual snack, drank

juice as their usual drink, slept with their mother at night, or breast-fed beyond the age of two.

Children who were bottle-fed or who started tooth-brushing before 18 months-of-age had a

significantly lower caries experience. There was no significant difference in caries experience

by frequency of tooth-brushing, age of introducing solids, type of weaning food, frequency of

snacking, having sugar added to the bottle, age of stopping bottle-feeding, having a bottle at

night, or frequency of bottle-feeding during the day.

Children who used toothpaste had statistically significant differences in eating behaviours

compared with those who did not; including a higher exposure to sweet snacks. The use of

toothpaste was not included in the regression analysis because of the possibility that it was

confounded by other variables; in particular, consumption of non-nutritious foods.

116

Table 4.6 – Cambodia Smile cross-sectional survey - Bivariate analysis of dental

behaviours by child caries experience.

a P <0.05; Kruskal Wallis or χ2 test as appropriate to examine differences among groups within the same column b P <0.001; Kruskal-Wallis or χ2 test as appropriate examine differences among groups within the same column

Mean dmft

(SD)

Any caries

N (%)

sECC

N (%)

Significant

caries

N (%)

Oral Hygiene

Tooth-brushing

Yes 5.1 (5.0) a 94 (66.2) 70 (51.5) 43 (30.3)

No 5.4 (5.8) 150 (65.2) 108 (50.0) 79 (34.3)

Help with brushing

Yes 3.6 (4.8)b 31 (50.8)b 23 (39.0)a 26 (32.1)

No 6.2 (4.9) 63 (77.8) 47 (61.0) 17 (27.9)

Started brushing

Before 18m 3.8 (5.6) 11 (45.8)a 9 (37.5) 6 (25.0)

After 18m 5.3 (4.9) 83 (70.3) 61 (54.5) 37 (31.4)

Use toothpaste

Yes 5.6 (4.9)a 78 (73.6)a 58 (56.9)a 31 (30.2)

No 3.6 (5.1) 16 (44.4) 12 (35.3) 11 (30.6)

Dietary habits

>1 Usual snack is

sweet

Yes 5.6 (5.5) 38 (70.4) 31 (60.8) 24 (44.4)a

No 5.2 (5.5) 206 (64.8) 147 (48.8) 98 (30.8)

Juice as usual drink

Yes 7.0 (5.6)a 38 (77.6) 31 (66.0)a 21 (42.9)

No 5.0 (5.4) 206 (63.8) 147 (48.2) 101 (31.3)

Nursing habits

Still breast-feeding

with after 2y

No 1.4 (3.1)b 16 (25.8)b 16 (24.8)b 16 (25.8)

Yes 6.1 (5.5) 228 (73.5) 162 (55.9) 106 (34.2)

Still sleeping with

mother after age 2y

No 0.5 (1.3)b 3 (15.8)b 3 (15.8)b 3 (15.8)

Yes 5.5 (5.5) 241 (68.3) 175 (52.6) 119 (33.7)

Bottle-feeding

Ever bottle-fed

Yes 4.7 (5.1) 45 (60.8) 36 (49.3) 22 (29.7)a

No 6.6 (6.5) 49 (69.0) 33 (53.2) 33 (46.5)

117

The presence of Significant Caries was influenced both positively and negatively in the binary

logistic regression model (Table 4.7); children who started tooth-brushing before the age of 18

months (compared to those who started brushing after 18 months of age) had a 41% less chance

of presenting with sECC; children who did not have help with tooth-brushing had twice the

chance of developing sECC; and children who were still breast-feeding after the age of two

years were more than five times more likely to have sECC. There was no evidence of multi-

collinearity in the tolerance values or the variance inflation factor (VIF), and the pseudo R2

statistics indicated that the model was a good fit.

Table 4.7 – Cambodia Smile cross-sectional study - Binomial logistic regression model for

Socio-behavioural habits by presence of significant caries.a

B(SE) Odds Ratio (95% CI) P-Value

Constant -1.7 (0.64) - -

Does not have help with brushing 0.75 (0.26) 2.11 (1.28, 3.50) 0.003

Started brushing before 18m -0.89 (0.42) 0.41 (0.18, 0.93) 0.033

Breast-feeding after the age of

two years 1.67 (0.65) 5.31 (1.50, 18.79) 0.010

aR2 = 0.93 (Homer & Lemenshow), 0.06 (Cox & Snell), 0.08 (Nagelkerke). Model χ2 = 465.96

118

4.5 Discussion

In the present study the socio-behavioural risk factors associated with ECC in a

convenience sample of Cambodian children were investigated. Oral hygiene habits,

breast-feeding habits, bottle-feeding habits and dietary habits were directly associated

with caries experience. Breast-feeding beyond the age of two years and the lack of early

oral hygiene habits were strong predictors of caries experience.

Before discussing the findings further, it is appropriate to consider the strengths and

weaknesses of the present study. The strengths of the study lie in the geographically

diverse sample, the high consent rate, the relatively large number of subjects, and the

broad range of topics explored in the questionnaire. The participants came from three

rural areas as well as Phnom Penh. This increased the chance of diversity within the

sample; however, since convenience sampling was used, the findings of this study may

not be generalizable to the entire Cambodian pre-school population.

Another weakness is related to the relatively large number of interviewers and examiners

involved in data collection, which could have resulted in a lack of consistency in the data.

In addition, data collection was conducted over a period of 18 months and inter-examiner

and intra-examiner reliability were not determined. In spite of this, the survey was a good

first step at quantifying caries experience of preschool children in Cambodia.

4.5.1 Caries experience within the Cambodia Smile cross-sectional

survey

This sample had a high burden of dental caries and the pattern was more severe than that

reported for other settings (Duangthip et al 2017; Khitdee 2016). The dmft of the 6 year-

old children in the present study is consistent with that reported by the CNOHS and so it

is conceivable that the caries experience of this sample could be similar to that

experienced elsewhere in Cambodia (Chher et al. 2016). The pattern of caries suggests

that there were more children with carious lesions in the anterior segment than in the

molars. This indicated that the caries process was starting on anterior teeth and then

developing on the molars in a pattern that is typical for ECC, as the primary incisors erupt

first.

119

One of the aspects of the present study that highlights the severe nature of caries

experienced by this group is the prevalence of pulpally involved teeth. Nearly one third

of children had one or more pulpally involved teeth, compared to a Brazilian study in

which only 5% of preschool children had one or more pulpally involved teeth. And yet

the prevalence of caries in the Brazil sample is similar to the present study (approximately

two thirds of children dmft > 1) (Gradella et al. 2011). This emphasises the more severe

pattern of dental caries experienced by Cambodian preschool children.

The PCG in this sample also had a high burden of dental caries. The prevalence of caries

and the number of pulpally involved teeth is consistent with the findings of the CNOHS

(Chher et al. 2016). The presence of frank open carious lesions in the mouths of the

mothers has often been suggested to indicate a more cariogenic microflora in the mother’s

mouth that is subsequently passed on to the child (Berkowitz 2003); alternatively, and

more likely, it may be that the mothers who have unrestored cavitated lesions have higher

risk behaviours for poor oral health and that these behaviours are moderating the

expression of dental caries in their children as described in other settings (Rai & Tiwari

2018; Tiwari et al. 2018; Philip et al. 2018) .

The present study supports the general consensus of the literature that maternal caries

experience is associated with, and is a good predictor of, dental caries experience in

preschool children (Rai & Tiwari 2018; Abiola et al. 2009; Li & Caufield 1995; Nowjack-

Raymer & Gift 1990; Steffensen 1990). Of note in the present study is that that the caries

experience of non-maternal caregivers was not associated with the child’s caries

experience. It could be that the reason for this lack of association is because the children

may not be left in the care of ‘other’ caregivers until after caries risk behaviours have

been established. This theory is supported by the finding that children who were left with

non-maternal caregivers were significantly older than those children were cared for by

their mothers. The severe dental caries experience in this study supports the view that

ECC is a pressing and urgent problem in Cambodia.

4.5.2 ECC and nursing habits

One of the findings that sets the results of present study apart from others is the

considerable risk ratio for dental caries associated with breast-feeding past the age of two

years. Before this finding is discussed further, it is important to observe that beliefs and

practices around infant feeding are culturally bound and some of these may be unique to

120

the Cambodian population (Desai & Alva 1998). Some of the infant feeding practices in

this population that may differ from other countries included the finding that almost all

the mothers slept with their child at night, often for several years. All these findings are

supported by previous studies of infant feeding practices in Cambodia (Desai & Alva

1998; Ismail & Sohn 1999; Willems et al. 2005).

The literature about whether there is an association between prolonged on-demand breast-

feeding at night and dental caries experience in early childhood is divided. In many

countries such practices are very uncommon, whereas in Cambodia they are almost

universal. The literature is also confusing in that breast- and bottle-feeding practices and

ECC itself are not reported in a consistent manner (Peres et al. 2017; Chaffee et al. 2014;

Thomson et al. 1996).

Reference to ‘prolonged breast-feeding’ as being ‘more than one year’ by authors is

common and few mothers in the West continue breast-feeding longer than this (Chaffee

& Cheng, 2014). In contrast, most of the children in the present study were breast-fed for

more than two years, almost all mothers slept with their infants, and nocturnal breast-

feeding was very prevalent, even amongst many older preschool children. A recent study

from Brazil was the first to examine the relationship between ECC and breast-feeding

beyond the age of two years, and their findings are consistent with the results of the

present study; in that they also reported a significantly higher prevalence of sECC in the

children who had been breast-fed past 2 years-of-age (Chaffee et al. 2014).

Although it is important to acknowledge the potential for confounding risk factors, it is

also useful to consider how breast-milk could interact with the caries process. The

mechanisms by which the mother’s breast-milk could be promoting the caries process

include both the composition of the milk and the temporal relationship (frequency of

exposure, duration of exposure, and timing of exposure) of the substrate on the teeth. The

cariogenic capacity of HBM, in particular, its limited acid buffering capacity and

relatively high lactose content has been investigated (Bowen & Lawrence 2005; Thomson

et al. 1996). The fact that HBM does not buffer acid as effectively as other milks may not

always be a problem because human breast-milk is usually coupled with saliva which has

a good buffering capacity. This leads on to one possible explanation for the relationship

between breast-feeding over the age of two years and sECC; that is, those children who

are being breast-fed at night (whilst they are sleeping with their mothers) have a

121

prolonged exposure to the carbohydrate (namely fermentable lactose) when saliva flow

is at its lowest and when there is little or no buffering capacity in the mouth. This is often

coupled with poor oral hygiene, leading to a thick, highly cariogenic biofilm.

Another mechanism by which the caries process may be potentiated by breast milk relates

to the frequency of snacking. The fact that breast-milk requires no preparation means that

children are able to breast-feed on-demand, as compared with the time required and extra

expense to prepare a bottle. Also, previous investigators have observed that feeding

practices in Cambodia tend to be directed towards ‘appeasing the child’s desires’. That is

to say, the common understanding of parents for a child crying is that they are hungry; if

they do not cry then reportedly there is no feeding initiated. Likewise, the child is not

encouraged to complete a full meal serving (whether from breast or bottle) and so they

become hungry sooner (Rasbridge & Kulig, 1995). These factors contribute to the

increasing frequency of carbohydrate exposures in the mouth.

In addition, weaning practices and time of solids introduction to the diet may also play a

role. The most common weaning food is rice porridge (“bor bor”), which often has sugar

added. White rice has a very high starch content (approximately 82%) (Snow & O’Dea,

1981), and evidence from in vitro animal and human studies show that dietary starches

are potential cariogenic substances (Scannapieco et al. 1993). Hydrolysis of cooked

starches produces significantly greater concentrations of maltose than raw starches and

subsequently a higher demineralisation potential (Brudevold et al. 1985). With

inadequate brushing and removal of food debris, the clearance time of these starches may

be extended along with the cariogenic potential.

There are likely to be strong cultural and traditional influences at play in relation to infant

feeding practices in Cambodia and it is recognized that changing health-related

behaviours can be difficult, especially where habits may be regarded as “normal” and

“healthy”. However, it is encouraging to note that there has been some recent success in

modifying infant feeding habits by engaging Buddhist nuns in Banteay Meanchey

province (Cambodia). The nuns in this study met with expectant mothers in a range of

settings and provided breast-feeding advice. The results showed a measurable change in

feeding habits whereby those in the intervention group had a higher rate of exclusive

breast-feeding when compared to a control group (Crookston et al. 2007).

122

It is clear that a coordinated approach would be required to bring about changes in such

long-established and culturally ingrained early childhood feeding behaviours;

furthermore, it is important to make sure that breast-feeding recommendations are clear

and consistent with WHO and the MOH recommendations in order to avoid providing

conflicting messages (Desclaux & Alfieri 2009; Marriott et al. 2010). An open cross-

discipline discussion around this issue would be helpful. This is especially important as

changes in infant feeding practices are becoming apparent with an improvement in the

economic status of the country and the use of the baby bottle with either milk formula or

improvised (often sweetened) milk replacements is becoming more popular within many

families in Cambodia (Helen Keller International 2015).

4.5.3 ECC and oral hygiene practices in Cambodia

Although modifying infant feeding habits is an attractive target for intervention, it is

important that a multifaceted coordinated approach is adopted. Another major area of

concern highlighted in the present study was the lack of good oral hygiene practices. Most

of the preschool children did not brush their teeth, and the majority of those who did,

brushed unassisted.

The effect of tooth-brushing, and the effect of the child having assistance while tooth-

brushing was seen most markedly in the binary regression model where the odds ratio for

sECC was more than halved by the early introduction of tooth-brushing. It is of note that

the study did not determine whether the brushing was associated with the use of fluoride

toothpaste, which is known to have an important effect on reducing caries increment

(Marinho 2009). It is known that many toothpastes in Cambodia either contain no fluoride

or contain low concentrations (Chher & Hak 2005). There are significant access issues to

affordable and appropriate toothpastes in Cambodia. The findings of the present study

suggest that the lack of tooth-brushing with fluoride toothpaste in this age group could

be a significant contributor to the severity of dental caries experienced.

4.5.4 Other risk indicators for ECC

It was surprising that although snacking on sweet foods and drinks was found to be

positively associated with ECC, it did not feature in the predictive model. One possible

explanation is that these variables were masked as predictors because the snacking on

sweet foods is happening concurrently with breast-feeding from a very early age or it is

123

so ubiquitous. In two previous Cambodian studies, the estimates for the prevalence of

exclusive breast-feeding at six months-of-age ranged from 15% (National Institute of

Statistics 2015)) to 60% (Senarath et al. 2010). The finding that the diets of many young

children are not optimal may not only be contributing to poor oral health but also to poor

nutritional outcomes (National Institute of Statistics 2010).

4.6 Conclusions

The results of the present study confirmed that preschool children in Cambodia are likely

to have a severe experience of dental caries which is associated with maternal caries

experience, late initiation of oral hygiene practices, and prolonged breastfeeding. In the

present study, breast-feeding continuing past the age of two years and lack of tooth-

brushing were significant predictors of dental caries experience in preschool children.

The unique socio-behavioural habits around nursing present in these children highlight

the need for a coordinated and multifaceted approach to improving the dental caries

outcomes of Cambodian pre-schoolers. Information from the present survey may help to

inform strategies to reduce the burden of dental caries in Cambodia.

124


Abanto, J., Carvalho, T. S., Mendes, F. M., Wanderley, M. T., Bönecker, M. & Raggio,

D. P. (2011) Impact of oral diseases and disorders on oral health-related quality of life of

preschool children. Community Dentistry and Oral Epidemiology, 39 (2), 105–114.

Abiola Adeniyi, A., Eyitope Ogunbodede, O., Sonny Jeboda, O. & Morenike Folayan, O.

(2009) Do maternal factors influence the dental health status of Nigerian pre-school

children? International Journal of Paediatric Dentistry, 19 (6), 448–454.

Bach K. (2014) The Oral Health of Cambodian Pre-School-Aged Children. DCD Thesis

submitted to the Faculty of Medicine, Dentistry and Health Science of The University of

Melbourne.

Berkowitz, R. J. (2003) Causes, treatment and prevention of early childhood caries: a

microbiologic perspective. Journal of the Canadian Dental Association, 69 (5), 304–307.

Bowen, W. H. & Lawrence, R. A. (2005) Comparison of the cariogenicity of cola, honey,

cow milk, human milk, and sucrose. Pediatrics, 116 (4), 921–926.

Bratthall, D. (2000) Introducing the Significant Caries Index together with a proposal for

a new global oral health goal for 12-year-olds. International Dental Journal, 50 (6), 378–

384.

Brudevold, F., Goulet, D., Tehrani, A., Attarzadeh, F., & Van Houte, J. (1985) Intraoral

demineralization and maltose clearance from wheat starch. Caries Research, 19 (2), 136–

144.

Chaffee, B. W. & Cheng, A. (2014) Global Research Trends on Early-Life Feeding

Practices and Early Childhood Caries: A Systematic Review. Journal of Oral Diseases,

2014.

Chaffee, B. W., Feldens, C. A. & Vítolo, M. R. (2014) Association of long-duration

breastfeeding and dental caries estimated with marginal structural models. Annals of

Epidemiology, 24 (6), 448–454.

Chher T. & Hak S. (2005) Fluoride content of toothpastes sold in Cambodia 2001 and

2004. Ministry of Health, Cambodia.

Chher, T., Turton, B.J., Hak, S., Beltran, E., Courtel, F., Durward, C. & Hobdell, M.,

(2016) Dental caries experience in Cambodia: Findings from the 2011 Cambodia

National Oral Health Survey. Journal of International Oral Health, 8 (1), 1.

Chu, C. H., Wong, A. W. Y., Lo, E. C. M. & Courtel, F. (2008) Oral health status and

behaviours of children in rural districts of Cambodia. International Dental Journal, 58

(1), 15–22.

Çolak, H., Dülgergil, Ç. T., Dalli, M., & Hamidi, M. M. (2013) Early childhood caries

update: A review of causes, diagnoses, and treatments. Journal of Natural Science,

Biology, and Medicine, 4 (1), 29.

Crookston, B. T., Dearden, K. A., Chan, K., Chan, T. & Stoker, D. D. (2007) Buddhist

nuns on the move: An innovative approach to improving breastfeeding practices in

Cambodia. Maternal & Child Nutrition, 3 (1), 10–24.

125

Desai, S. & Alva, S. (1998) Maternal education and child health: Is there a strong causal

relationship? Demography, 35 (1), 71–81.

Desclaux, A. & Alfieri, C. (2009) Counseling and choosing between infant-feeding

options: overall limits and local interpretations by health care providers and women living

with HIV in resource-poor countries (Burkina Faso, Cambodia, Cameroon). Social

Science & Medicine, 69 (6), 821–829.

Drury, T. F., Horowitz, A. M., Ismail, A. I., Maertens, M. P., Rozier, R. G. & Selwitz, R.

H. (1999). Diagnosing and reporting early childhood caries for research purposes: a

report of a workshop sponsored by the National Institute of Dental and Craniofacial

Research, the Health Resources and Services Administration, and the Health Care

Financing Administration. Journal of Public Health Dentistry, 59 (3), 192–197.

Duangthip, D., Gao, S.S., Lo, E.C.M. & Chu, C.H. (2017) Early childhood caries among

5- to 6-year old children in Southeast Asia. International Dental Journal, 67 (2), 98-106.

Gift, H. C., Reisine, S. T. & Larach, D. C. (1992) The social impact of dental problems

and visits. American Journal of Public Health, 82 (12), 1663–1668.

Gradella, C. M., Bernabé, E., Bönecker, M. & Oliveira, L. B. (2011) Caries prevalence

and severity, and quality of life in Brazilian 2- to 4-year-old children. Community


Helen Keller International (2014) Assessment of Promotion of Foods Consumed by

Infants and Young Children in Phnom Penh: Assessement and Research on Child Feeding

(ARCH) – Cambodia Country Report.

Ismail, A. I. & Sohn, W. (1999) A systematic review of clinical diagnostic criteria of

early childhood caries. Journal of Public Health Dentistry, 59 (3), 171–191.

Khitdee, C. (2017) The epidemiology of Early Childhood Caries. Thai Dental Public

Health Journal, 22(sup), 4-10

Li, Y. & Caufield, P. W. (1995) The fidelity of initial acquisition of mutans streptococci

by infants from their mothers. Journal of Dental Research, 74 (2), 681–685.

Marinho, V. C. C. (2009) Cochrane reviews of randomized trials of fluoride therapies for

preventing dental caries. European Archives of Paediatric Dentistry, 10 (3), 183–191.

Marriott, B. P., White, A. J., Hadden, L., Davies, J. C. & Wallingford, J. C. (2010) How

well are infant and young child World Health Organization (WHO) feeding indicators

associated with growth outcomes? An example from Cambodia. Maternal & Child

Nutrition, 6 (4), 358–373.

Ministry of Health, Cambodia (2015) Cambodia Demographic and Health Survey

Monse, B., Heinrich-Weltzien, R., Benzian, H., Holmgren, C. & van Palenstein

Helderman, W. (2010) PUFA–An index of clinical consequences of untreated dental

caries. Community Dentistry and Oral Epidemiology, 38 (1), 77–82.

Nowjack-Raymer, R. & Gift, H. C. (1990) Contributing factors to maternal and child oral

health. Journal of Public Health Dentistry, 50 (6), 370–378.

126

Peres, K. G., Nascimento, G. G., Peres, M. A., Mittinty, M. N., Demarco, F. F., Santos,

I. S., Matijasevich, A. & Barros, A. J. D. (2017) Impact of Prolonged Breastfeeding on

Dental Caries: A Population-Based Birth Cohort Study. Pediatrics, 140 (1), e20162943.

Philip, N., Suneja, B. & Walsh, L.J. (2018) Ecological approaches to dental caries

prevention: paradigm shift or shibboleth? Caries Research, 52 (1-2),153-165.

Rai, N.K. & Tiwari, T. (2018) Parental Factors influencing the Development of early

childhood caries in Developing Nations: a Systematic Review. Frontiers in public health,

6, 64.

Rasbridge, L. A. & Kulig, J. C. (1995) Infant feeding among Cambodian refugees. MCN:

The American Journal of Maternal/Child Nursing, 20 (4), 213–218.

Scannapieco, F. A., Torres, G. & Levine, M. J. (1993) Salivary α-amylase: role in dental

plaque and caries formation. Critical Reviews in Oral Biology & Medicine, 4 (3), 301–

307.

Senarath, U., Dibley, M.J. and Agho, K.E., 2010. Factors associated with nonexclusive

breastfeeding in 5 east and southeast Asian countries: a multilevel analysis. Journal of

Human Lactation, 26(3), pp.248-257.

Sheiham, A. & James, W.P.T. (2015) Diet and dental caries: the pivotal role of free sugars

reemphasized. Journal of Dental Research, 94 (10), 1341-1347.

Snow, P. & O’Dea, K. (1981) Factors affecting the rate of hydrolysis of starch in food.

The American Journal of Clinical Nutrition, 34 (12), 2721–2727.

Steffensen, J. E. (1990) Literature and concept review: issues in maternal and child oral

health. Journal of Public Health Dentistry, 50 (6), 358–369.

Thomson, M. E., Thomson, C. W. & Chandler, N. P. (1996) In vitro and intra-oral

investigations into the cariogenic potential of human milk. Caries Research, 30 (6), 434–

438.

Tiwari, T., Wilson, A.R., Mulvahill, M., Rai, N. & Albino, J. (2018) Maternal factors

associated with early childhood caries in urban Latino children. JDR Clinical &

Translational Research, 3 (1), 83-90.

Todd, R. V., Durward, C. S., Chot, C., So, P. K. & Im, P. (1994) The dental caries

experience, oral hygiene and dietary practices of preschool children of factory workers

in Phnom Penh, Cambodia. International Journal of Paediatric Dentistry, 4 (3), 173–

178.

Willems, S., Vanobbergen, J., Martens, L. & De Maeseneer, J. (2005) The Independent

Impact of Household-and Neighborhood-based Social Determinants on Early Childhood

Caries: A Cross-sectional Study of Inner-city Children. Family & Community Health, 28

(2), 168–175.

World Health Organization (2013) Oral health surveys: basic methods. World Health

Organization.

128

Chapter 5

Evaluation of a primary care based ECC intervention -

Cambodia Smile

Souen S, Turton B, Durward C. (2015) Integrating oral health and general health in a

Cambodian setting; IADR SEA 29th Annual meeting, Bali, Indonesia.

Turton B (2015) Cambodia Smile – integrating oral health and general health in a

Cambodian setting. 2015 Global childrens nutrition and oral health symposium, UCSF,

California, USA.

Sopharith S, Turton B, Durward C, Crombie F, Manton D. (2016) Cambodia Smile –

Preliminary results from an early childhood caries intervention. IADR General Session.

Seoul, Republic of Korea.

Turton B, Durward C, Soeun S, Crombie F, Manton D (2017). Cambodia Smile – 2-year

follow-up of a community based Early Childhood Caries (ECC) intervention. Congress

of the International Association of Paediatric Dentistry. Santiago, Chile.

Turton, B., Durward, C., Soeun S., Crombie, F., Manton, D. (2018) Cambodia Smile –

Caries Prevention by primary health care providers. ORCA, Copenhagen, Denmark.

129

Objectives: To test a primary care-based intervention to prevent ECC using primary

healthcare-workers in public health centres.

Methods: A Total of 262 Mother-child dyads were recruited at birth and followed for two

years (intervention group) and a contemporaneous sample of 184 participants from ages 18 to

30 months were recruited from surrounding districts (control group). Intervention group

participants received oral health information, toothbrushes with fluoride toothpaste, fluoride

varnish (NFlor protect; Ivoclar Vivadent, Schaan, Liechtenstein), and dietary advice integrated

with existing vaccinations, de-worming, and vitamin A schedule. At follow-up, intraoral

examinations were conducted to detect carious lesions including white spot lesions using the

South East Asian recommendations for epidemiological surveys of ECC.

Results: Of the intervention group 185/262 were followed-up (70.2%) and there was no

statistically significant difference in age in months between the control group (22.2; SD 2.4)

and the intervention group (21.9; SD 2.6). There was no statistically significant difference in

caries experience by sex. A significant (p < 0.001) difference existed in the number and

prevalence of carious lesions between the intervention group (2.9; SD 3.0 and 68.1%) and

control group (8.7; SD 4.9 and 94.0%). There was a significant difference in preventive

increment depending on the components of the intervention; children who had Oral Hygiene

Education (OHE) and dietary advice only during the first 6 months of life had a 25.0%

preventive increment while those who had OHE during the first 6 months of life as well as FV

application had a 62.5% preventive increment when compared to the control group.

Conclusions: The Cambodia Smile primary care-based intervention significantly reduced

caries experience. Further advocacy is required in order to upscale this intervention across

Cambodia.

130

5.0 Chapter 5 – The Cambodia Smile intervention

The oral health of Cambodian children is amongst the poorest in the world. At age 6 y, 98% of

children have carious lesions, with a mean dmft of 9.0, and a mean of 2.7 pulpally-involved

teeth (Chher et al. 2016). This severe burden of disease has previously been reported to have

severe impacts on the Oral health Related Quality of Life of Cambodian Children (Khoun et

al. 2018; Turton 2018; Turton et al. 2015). Perhaps part of the reason for those severe impacts

are that most of the carious lesions go untreated. That is to say, the ‘D’ or ‘decayed’ component

of the dmft/DMFT index accounts for most (99%) of the disease experience. The result is that

most children enter school in a state of uncontrolled caries risk with numerous untreated carious

lesions.

Dental caries is a behaviourally moderated disease initiated by the metabolism of fermentable

carbohydrates by bacteria on the surface of the tooth, the acidic by-products of which

demineralise the enamel, subsequently leading to cavitation of the tooth surface (Sheiham &

James, 2015). The key driver in the presence of those untreated carious lesions is a diet with

large amounts of sugar. The average preschool child in Phnom Penh consumes three sugary

sweetened beverages per-day (Turton et al. 2018). While specific quantification of sugar

content was not performed in that study, it is more than likely that the consumption of sugar

by young children in Cambodia exceeds that recommended by the World Health Organization

(WHO) (Moynihan and Kelly, 2014). Unfortunately, the evidence around strategies to reduce

sugar intake by preschool children are still limited though motivational interviewing appears

to be the most promising method in other settings; however, there is not yet any evidence for

the use of such techniques for improvement in oral health in countries similar to Cambodia

(Jameson 2018; Albino and Tiwari, 2015).

Acknowledging that the diet is difficult to influence, it is important to employ another? key

mediator in reducing the impact of sugar on caries progression which is the use of fluoride

(Bernabé et al. 2016). Fluoride varnish (FV) together with tooth-brushing with fluoride

toothpaste is an attractive method for delivering fluoride in communities where reticulated

water systems for the delivery of community water fluoridation are not available. All of those

interventions have been shown to be cost-effective, safe to use and effective (Marinho et al.

2003).

A recent cross-sectional survey identified the late oral hygiene practices and unfavourable

nursing habits as risk indicators for ECC in Cambodia suggesting that the high consumption of

131

sugar by Cambodian pre-schoolers is not being moderated by the use of fluoride toothpaste

(Turton et al. 2015). Furthermore, exposure to fluoride toothpaste is occurring at an older age

than desirable. Recent evidence has supported the implementation of caries preventive efforts

before one year-of-age and even before the child is born by targeting pregnant mothers

(Phantumvanit et al. 2018; Sokal-Gutierrez et al. 2016; Leong et al. 2013; Wigen & Wang

2012).

The health professionals who have the most contact with Cambodian children and mothers

during the early years of life are the midwives and nurses (primary care providers) based in

community health centres. Nine out of ten mothers receive post-natal care and three out of four

children receive their full complement of basic vaccinations (MOH, Cambodia, 2015). There

is a growing body of evidence to support dental interventions carried out by primary health

care providers who have many opportunities to integrate oral health promotion with general

health and nutrition interventions. However, at this stage, there are no models for mobilising

primary health care providers for oral health interventions in a Cambodian setting. An

affordable, accessible and appropriate intervention to address ECC in Cambodia could involve

providing oral health education to mothers and babies, together with the application of fluoride

varnish to children, delivered alongside existing health interventions by primary health care

providers. The present study seeks to examine the outcome of a pilot intervention to deliver an

oral health intervention through the primary health care structure in Cambodia.

5.1 Aims

To investigate the effectiveness of a pilot strategy for the reduction of dental caries experience

in Cambodian preschool children through an integrated primary health care model.

5.2 Objective

To investigate the effects on dental caries experience among a sample of young children in

Kampong Speu district through packaging the delivery of oral health education and topical

fluoride applications with the existing programs for de-worming, vitamin A supplementation

and vaccinations.

132

5.3 Methods

A two year interventional study based in CHC was undertaken. Mothers were invited to

participate with their new-born babies and consent was gained both verbally (explained by

primary health care provider) and in writing (Appendix 9). Participants were matched with a

control group who were recruited after two years to make a comparative evaluation of the

outcome of the program. Therefore, individuals in the control group still benefitted from

participating in the study. Ethical review was provided by the National Ethics Committee for

Health Research in Cambodia and by the University of Melbourne, Health Sciences Human

Ethics Sub-Committee (Ethics ID: 1545412). Copies of the ethical review documents are

included in Appendix 11.

5.3.1 Health Centre and Population Selection

The study sample size was calculated based on the assumption that children in the intervention

group would have a mean d1mft at least one tooth less than those in the control group. Using

the data from the previously conducted cross-sectional survey where the mean dmft of 2 year-

olds was 3 (SD 3), 95% power, and a significance level 0.05, it was determined that a sample

size of 392 (196 in the control and 196 in the intervention) would be required. However follow-

up can be difficult in a Cambodian environment and so an arbitrary 30% was added to each

group, therefore a sample of 520 mother-child dyads was required.

Participants for the intervention group were recruited from four health centres in the province

of Kampong Speu between February and July of 2015 with mother-child dyads invited into the

study until the sample size was filled. Recruitment of the control group was conducted during

October 2016 from four different health centres in the same province. None of the children had

access to fluoridated water.

5.3.3 Clinical Protocol for the intervention

The clinical protocol involved the delivery of dietary advice, oral hygiene instruction with flip

charts, and provision of toothbrush and fluoride toothpaste for both mother and baby at baseline

and for baby at subsequent visits. The package of dietary advice, oral hygiene instruction and

delivery of fluoride toothpaste with toothbrush is what is referred to in this chapter as Oral

Health Education (OHE). In addition up to 0.4 mL of 7000 ppm fluoride varnish (NFlor Protect

FV, Ivoclar-Vivadent, Schaan, Liechtenstein) was placed on tooth surfaces once teeth erupted

into the mouth. These interventions were delivered concurrently with routine visits for

133

vaccinations, vitamin A supplementation and de-worming tablets. Two additional

opportunities were provided at 15 and 21 months-of-age to supplement the routine contacts

representing six opportunities for intervention over the first two years-of-life.

5.3.4 The questionnaire for the two-year follow-up

The questionnaire included questions on oral knowledge and behaviours and diet; however,

analyses of these aspects are beyond the present scope. All data collection forms are presented

in Appendix 9. Data analyses for the present study included data on the Family Impact Scale

(FIS), maternal education and household income level. The FIS is an OHRQoL instrument

which has been validated for use in a Cambodian population previously. The FIS includes eight

questions across three domains; family activities, family emotions, and family conflict. The

response options for each item were “Never” (score 0), “Once or twice” (score 1), “Sometimes”

(score 2), “Often” (score 3), or “Every day or almost every day (score 4). A participant was

said to have an ‘impact’ in a particular item if they responded with a score 3 or above in any

item.

5.3.5 Clinical measures

Teeth were not cleaned and were examined wet according to the guidelines described for the

South East Asian Index for Early Childhood Caries (SEA-ECC). Each tooth was given two

codes, one for SEA-ECC and another for PUFA status (Monse et al. 2010). Data from the SEA-

ECC are presented as d1mft for carious lesions limited to enamel and d2mft for carious lesions

extending to dentine and cavitated. The term ‘severity of caries’ in this study refers to the

number of lesions in an individual rather than the depth of the lesions. The Greens Simplified

Plaque index was used to assess the amount of plaque accumulation observed on the primary

right central incisor. Children were examined in the supine position using hand held torches

and mouth mirrors with appropriate universal infection control measures in place.

5.3.6 Data collection and analysis

Data were collected on stage-appropriate forms and participants were identified by a unique

identification (ID) number (Appendix 9). Midwives were given a list of children to follow-up

and one phone call a month was made between the research co-ordinator to the health centre

to encourage follow-up. Data were entered into IBM SPSS version 23 and both descriptive and

bivariate analyses were performed. No multivariate analysis was performed because no

134

differences in caries experience by sociodemographic indicators were found and sampling

procedures controlled for variations between the groups.

135

5.4 Results

There was no significant difference in age in months between the control group (22.2; SD 2.4)

and the intervention group (21.9; SD 2.6); There was no significant difference in group

membership by gender, or sociodemographic indictors such as the mother’s education or the

household income by loss to follow-up (Table 5.1).

Table 5.1 – Attrition analysis of intervention group at 1-year.

Baseline

N (row %)

Lost

N (row %)

Followed-up

N (column %) P-value

Sexa

Male 123 (47.1) 40 (32.5) 83 (67.5) 0.157

Female 138 (52.9) 36 (26.1) 102 (73.9)

Household income

<50 22 (8.4) 6 (27.3) 16 (72.7) 0.689

51-150 160 (61.1) 44 (27.5) 116 (72.5)

151-250 74 (28.2) 25 (33.8) 49 (66.2)

>250 6 (2.3) 1 (16.7) 5 (83.3)

Total 262 (100.0) 76 (29.0) 186 (71.0)

aGender data missing from one participant.

136

There were statistically significant differences between control and intervention groups in

terms of maternal education and household income. Participants in the intervention group had

a broader spread of participants across education levels and household income levels (Table

5.2).

Table 5.2 – Sociodemographic characteristics of participants at follow-up by group.

Intervention

N (row %)

Control

N (row %)

Total

N (column %) P-value

Sexa

Male 83 (48.8) 87 (51.2) 170 (46.1) 0.359

Female 102 (51.3) 97 (48.7) 199 (53.9)

Maternal Education

Illiterate 17 (100.0) 0 (0.0) 17 (4.6) <0.001

Primary 108 (37.0) 184 (63.0) 292 (78.9)

Secondary or High-

school

58 (100.0) 0 (0.0) 58 (15.7)

Tertiary 3 (100.0) 0 (0.0) 3 (0.8)

Household income

<50 16 (100.0) 0 (0.0) 16 (4.3) <0.001

51-150 116 (38.7) 184 (61.3) 300 (81.1)

151-250 49 (100.0) 0 (0.0) 49 (13.2)

>250 5 (100.0) 0 (0.0) 5 (1.4)

Total 186 (50.3) 184 (49.7) 370 (100.0)

aGender data missing from one participant.

137

Most participants received three or fewer contacts across the intervention period and those who

had four contacts had the lowest caries experience (Table 5.3). Children who had six contacts

had the most severe caries experience.

Table 5.3 – Caries severity by number of contacts among the intervention group.

Number of contacts

d3mfta

Mean (SD) P-valueb

One (N = 63) 2.9 (2.8) 0.359

Two (N = 58) 3.5 (3.6)

Three (N = 28) 2.5 (3.1)

Four (N = 16) 1.8 (2.9)

Five (N=19) 2.9 (2.3)

Six (N=2) 4.5 (0.7)

ad3mft denotes cavitated lesions in the primary dentition. bKruskal-Wallis test for examining the difference in mean d3mft by number of contacts received.

Data on caries experience are presented in Table 5.4. There were no significant differences in

caries experience according to sociodemographic characteristics.

138

Table 5.4 – Caries experience by sociodemographic characteristics.

D1mft

(SD) P-value

D2mft

(SD) P-value

Any white spot

N (row %) P-value

Any cavity

N (row %) P-value

Any pufa

N (row %) P-value

Sex

Male 6.3 (5.1) 0.111 1.4 (2.1) 0.311 139 (81.8) 0.422 69 (40.6) 0.195 14 (8.2) 0.105

Female 5.4 (4.8) 1.2 (2.0) 160 (80.4) 71 (35.7) 9 (4.5)

Maternal

Education

Illiterate 5.6 (4.8) 0.494 1.4 (1.9) 0.662 40 (80.0) 0.098 19 (38.0) 0.519 3 (6.0) 0.225

Primary 5.8 (4.7) 1.3 (2.2) 146 (85.9) 71 (41.8) 7 (4.1)

Secondary or

High-school

6.0 (5.2) 1.2 (1.9) 107 (77.0) 48 (34.5) 13 (9.4)

Tertiary 4.2 (5.8) 0.8 (1.6) 7 (63.6) 3 (27.3) 0 (0.0)

Household

income

<50 5.4 (5.1) 0.956 1.4 (2.1) 0.112 23 (85.2) 0.488 10 (37.0) 0.981 0 (0.0) 0.563

51-150 5.9 (5.1) 1.3 (2.2) 80 (79.2) 37 (36.6) 7 (6.9)

151-250 5.8 (4.6) 1.2 (1.8) 110 (84.6) 50 (38.5) 8 (6.2)

>250 5.8 (5.2) 1.3 (2.1) 87 (77.7) 44 (38.5) 8 (7.1)

Total 5.8 (5.0) 1.3 (2.0) 229 (62.1) 140 (37.9) 23 (6.2)

aP-values denote differences among groups within the same column.

139

The mean plaque scores for the control and intervention groups were statistically significantly

different, being 2.1 (SD 1.0) in the control group and 0.9 (SD 0.4) in the intervention group (p

< 0.05). There was no significant difference in frequency of ‘non-nutritious food’ (packaged

snacks, candies and sweet drinks) ingestion by group membership.

There was a statistically significant difference in caries experience by group membership and

exposure to the intervention (Table 5.5). Overall, there was a two thirds reduction in caries

severity (based on d1mft) and a two fifths reduction in prevalence for those who were in the

intervention group. Children who received OHE (with the provision of toothbrush and

toothpaste) had half the severity (d1mft) of those in the control group and those who had one

or more FV exposures had a further one-third reduction (p < 0.05). Exposure to FV was

associated with a small reduction in the prevalence of white spot lesions and there was a one

third reduction in d1mft compared to those who only received OHE. In contrast, children who

received FV had half the d2mft compared to those who received OHE only. Children who

received OHE realised a 25% reduction in the number of cavitated lesions. There was no

significant difference in the prevalence or severity of pulpally involved lesions by group

membership or level of exposure to the intervention but sample sizes were small for this

variable.

140

Table 5.5 – Clinical characteristics by exposure to the intervention.a

d1mft

(SD)

d3mft

(SD)

Any white spot

N (%)

Any cavity

N (%)

Any pufa

N (%)

Control 8.7 (4.9)a 1.6 (2.2)a 173 (94.0)a 88 (47.8)a 17 (7.6)

Intervention

Overall 2.9 (3.0) 1.0 (1.8) 127 (68.3) 53 (28.5) 9 (4.8)

OHE Only 3.3 (3.3)b 1.2 (2.1)b 78 (69.6)b 35 (31.3)b 4 (3.6)

FV & OHE 2.3 (2.5) 0.6 (1.2) 49 (66.2) 18 (24.3) 5 (6.8)

aP-value<0.001 for comparison of means or proportions between the control and intervention groups; χ2 statistic

or Kruskal-Wallis as appropriate. bP-value<0.001 for comparison of means or proportions between the OHE and FV&OHE group; χ2 statistic or

Kruskal-Wallis as appropriate.

141

There were statistically significant differences in the prevalence of impacts and mean scores

across the FIS and all FIS subscales (Table 5.6). Participants in the control group had a 2.5

times greater prevalence of impacts across the FIS; with the FIS subscale of family activities

showing the biggest differences. Families in the control group had a three times greater chance

of having their family activities interrupted by problems with their child’s teeth or mouth.

Differences in mean scores revealed a large effect size across the FIS and small to moderate

differences across the three subscales.

Table 5.6 – Differences within the Family Impact Scale and subscale impacts and mean

scores by group membership.

Intervention Control Overall

Relative

differencea

Impacts - N (%)

Family Impact Scale 29 (15.6) 93 (50.5) 122 (33.0) 69.1

Activities subscale 16 (8.6) 70 (38.0) 86 (23.2) 77.4

Emotions subscale 21 (11.3) 70 (38.0) 91 (24.6) 70.3

Conflict subscale 9 (4.8) 36 (19.6) 45 (12.2) 75.5

Mean Scores - (SD) Effect sizec

Family Impact Scale 3.7 (4.3) 8.0 (7.9) 5.8 (6.7) 0.6

Activities subscale 1.5 (1.9) 3.8 (3.9) 2.6 (3.3) 0.2

Emotions subscale 1.5 (1.6) 2.7 (2.5) 2.1 (2.2) 0.5

Conflict subscale 0.7 (1.6) 1.6 (2.2) 1.1 (2.0) 0.4

Floor and Ceiling effects

Proportion with

minimum score (0)

56 (30.1) 41 (22.5) 97 (26.4) -

Proportion with

maximum score (24)

1 (0.5) 9 (4.9) 10 (2.7) -

aP < 0.001; Kruskal-Wallis or chi squared test as appropriate. b Calculated by dividing the change in prevalence by the baseline prevalence and expressing it as a percentage. c Calculated by dividing the differences in mean scores by the standard deviation of the intervention group.

142

5.5 Discussion

The present study demonstrated that OHI and FV delivered by primary health care providers

in a rural Cambodian setting reduced the prevalence and severity of dental caries at 2 years-of-

age. Children in the intervention group had a two fifths reduction in prevalence and a two thirds

reduction in the severity of dental caries. There was evidence that the FV provided additional

benefit to the OHE alone. This is the first ECC intervention that used an OHRQoL instrument

as an outcome measure for a preventive intervention and the reduction in caries experience led

to a corresponding reduction of the impacts as measured by the FIS. Before considering the

findings of the intervention in detail, it is appropriate to consider the strengths and limitations

of the present study.

The main ethical consideration for this project was that those in the control group did not

receive interventions known to have beneficial effects. This disadvantage was partly mitigated

by the children in the control group receiving oral health interventions at the time of recruitment

and examination at two years-of-age. This compromise in recruitment meant that there were

differences in the sociodemographic characteristics between the control and intervention

groups. This may have been due to the different processes for recruiting participants in the

group. While health centres were selected randomly by the provincial health department staff

and based on the births registered by health centre, the process of finding the children was

based on the child being present within the village at the time of examination. This process was

continued until a sample with sufficient power was recruited and that lead to a bias towards the

middle spectrum of the household income scale. Despite this, no significant differences were

found in caries experience by sociodemographic characteristics.

The study was powered to examine differences in the severity of cavitated carious lesions at

group level and so there are limitations in performing sub-group analysis by number of

exposures to the intervention or the number of pulpally involved teeth. The therapies employed

had sufficient evidence to support their use and so the key focus was to examine a package of

care rather than to establish the viability of individual therapies.

5.5.1 Differences in Oral Health Related Quality of Life

In addition to the benefits at tooth-level, it appears that the Cambodia Smile intervention also

had benefits for the family unit. This study was unique in that it offered an OHRQoL measure;

the choice to use the FIS rather than the Parental-Child Perceptions Questionnaire was based

143

on data from the validation study of those instruments which suggested that for children in the

2 year age-group the FIS would be most sensitive. Any OHRQoL measure that attempts to

assess outcomes for participants who are not yet able to articulate their own opinion comes

with limitations (Varni et al. 2007). Despite this, the differences in mean scores and impacts

across the FIS scale and subscale suggests that the intervention created some benefit in terms

of OHRQoL.

5.5.2 Differences in caries experience by participation in the intervention

The differences in caries experience by exposure to the intervention were consistent with those

observed in other studies (Marinho et al. 2003). In addition, an intuitive pattern was confirmed

whereby those who received both OHE and FV realised a greater preventive fraction. In

addition, the OHE appeared to achieve the largest preventive fraction in the reduction in the

prevalence in white spot lesions. The placement of FV appeared to slow the progression of

white spot lesions to cavitation rather than greatly reduce the presence of white spot lesions.

This is consistent with the current understanding of the role of fluoride in moderating the

progression of caries (Philip et al. 2018, Bernabe et al. 2014, Fejerskov et al. 2004). Perhaps

the tooth-brushing was effective at slowing the progression of carious lesions from the pre-

clinical to clinically detectable white spot lesions, and perhaps the FV helped to slow the

progression from white spot lesion to cavitation? The limitations in sample size means that we

cannot be confident about which therapy rendered the greater preventive effect.

The results of the present study are more favourable than other FV studies reported in recent

literature and it is worth noting that the teeth were examined wet and without cleaning,

therefore the prevalence of white spot lesions may have been underestimated. Of the studies

that report FV to be ineffective, some were underpowered (Memarpour, 2015; Oliveira, 2014)

while other studies were conducted in populations with a prevalence of ECC below 40% at

around 2 years-of-age (Agouropoulos et al. 2014; Anderson et al. 2016; Divaris et al. 2013;

Jiang et al. 2014; Oliveira et al. 2014). Some researchers recruited only participants who had

already reached 2 to 3 years-of-age with no clinical signs of dental caries (Muñoz-Millán et al.

2018; Tickle et al. 2017). Other researchers have reported limited measurable benefits for FV

being applied on a conditional basis whereby clinicians did not adhere to a strict protocol for

identifying those at risk of lesion development (Fontana et al. 2018). This highlights the need

for clinicians to be clear about the selection of candidates for FV application and in the case of

the Cambodia Smile intervention, all children in the cohort were targeted at a stage of

144

development where, after taking into account the socio-behavioural aspects of the caries

disease in a Cambodian environment, maximum benefit could be gained. In addition, children

in the present context had a much higher caries risk and much lower exposure to fluoride in

other forms when compared to other research cohorts.

Despite the gross reductions in dental experience, there were still nine children in the

intervention group who developed one or more pulpally involved carious lesions. While this is

a lower proportion of children than that in the control group, the difference was not statistically

significant, and the study was underpowered in this instance. However, the presence of nine

children among the intervention group with pulpally involved teeth suggests that the Cambodia

Smile package either didn’t reach the children in the present delivery model or that the caries

imbalance present in those children exceeded that which can be moderated by the present

intervention alone. It is interesting that children with the most severe caries experience among

the intervention group were more likely to present for FV treatments. This suggests that there

might be opportunity to augment the benefits of FV by applying silver diammine fluoride to

children who present with cavitated lesions as this has been shown to be effective in other

settings (Duangthip et al. 2016).

When examining the benefits of fluoride therapies for moderating the disease experience it is

important to point out that caries is a socio-behavioural disease and if the drivers of the process

are not controlled, then caries will continue to be a burden at population level. In the present

study, there appears to be an improvement in oral hygiene as indicated by the differences in

plaque scores. However, no significant difference in the frequency of consumption of non-

nutritious foods between the two groups was noted. This suggests that the intervention did not

lead to change in dietary behaviours. There is evidence that other techniques such as

motivational interviewing might be effective in achieving behaviour change in ‘westernised’

settings (Borrelli et al. 2015; Jamieson et al, 2018). However, it is not clear if the same methods

would be effective in a Cambodian setting and more, context specific, research is needed to

examine how those risk behaviours are moderated (Rai and Tiwari, 2018). At this stage the

Cambodia Smile intervention is focused on targeting the most easily modifiable risk factors.

The previous cross-sectional study (Ch. 4) demonstrated that children who initiated oral

hygiene practices before 18 months-of-age had half the caries experience compared with those

who did not. The present study reproduces and validates those findings and suggests that an

improvement in oral hygiene is an achievable modifiable risk factor in the Cambodian setting.

145

Furthermore, risk factors were modified using the primary health care system, non-dental

professionals, and in a way that did not create additional attendance or access burdens for the

household. The main limitation in terms of the intervention was that not all chances for access

were realised and more work is needed to increase the proportion of opportunities that are

realised.

Even though this intervention does not offer a holistic approach to managing the caries process,

it does represent a tangible, population-wide approach, and it appears to be the most practical

first step at reducing the burden of ECC in Cambodia. It was evident that although six

opportunities were provided for intervention, very few of those opportunities were realised and

this highlights the need for multiple safety nets or back-ups in terms of intervention to increase

access. Although fewer than three opportunities on average were realised, that was still enough

to render a preventive effect. Given that that the caries experience is so severe and more than

99% of lesions go untreated, a population-wide role out of the Cambodia Smile intervention to

address the most easily modified risk factors would be appropriate.

5.6 Conclusions

The Cambodia Smile intervention put into practice the WHO recommendations for ECC

interventions (Phantumvanit et al. 2018) and the result was a caries reduction greater than other

previous preventive interventions in Cambodia. This study translates existing evidence around

the benefits of tooth-brushing with fluoride toothpaste and application of fluoride varnish into

a feasible intervention that was successful in a Cambodian environment. The prevalence and

severity of ECC was reduced for those children in the intervention group, however, there was

still a notable burden of caries experience following the intervention. Therefore, additional

administrative mechanisms improving access to the Cambodia Smile protocol and providing a

more holistic intervention might be able to further reduce the caries burden and the subsequent

impacts on OHRQoL. The success of this pilot study is a valuable first step in demonstrating

to key stake holders that it is possible to reduce the impacts that children and families

experience as a result of dental caries in a Cambodian setting.

146





Dentistry, 42 (10), 1277–1283.

Anderson, M., Dahllöf, G., Twetman, S., Jansson, L., Bergenlid, A.-C. & Grindefjord, M.

(2016) Effectiveness of Early Preventive Intervention with Semiannual Fluoride Varnish


Controlled Trial. Caries Research, 50 (1), 17–23.

Baker, S.R., Foster Page, L., Thomson, W.M., Broomhead, T., Bekes, K., Benson, P.E.,

Aguilar-Diaz, F., Do, L., Hirsch, C., Marshman, Z. & McGrath, C. (2018) Structural

Determinants and Children’s Oral Health: A Cross-National Study. Journal of Dental

Research, 0022034518767401.

Borrelli, B., Tooley, E. M. & Scott-Sheldon, L. A. J. (2015) Motivational Interviewing for

Parent-child Health Interventions: A Systematic Review and Meta-Analysis. Pediatric

Dentistry, 37 (3), 254–265.

Bernabé, E., Vehkalahti, M.M., Sheiham, A., Aromaa, A. & Suominen, A.L., (2014) Sugar-

sweetened beverages and dental caries in adults: a 4-year prospective study. Journal of

Dentistry, 42 (8), 952-958.

Bernabé, E., Vehkalahti, M.M., Sheiham, A., Lundqvist, A. and Suominen, A.L., 2016. The

shape of the dose-response relationship between sugars and caries in adults. Journal of Dental

Research, 95 (2), 167-172.




Chia, L., Densie, I. & Morgan, C. (2015) An exploratory study of parental knowledge of early

childhood oral health care in Southland, New Zealand. New Zealand Dental Journal, 111 (1),

18–24.

Divaris, K., Preisser, J. S. & Slade, G. D. (2013) Surface-Specific Efficacy of Fluoride Varnish

in Caries Prevention in the Primary Dentition: Results of a Community Randomized Clinical

Trial. Caries Research, 47 (1), 78–87.

Dooley, D., Moultrie, N. M., Heckman, B., Gansky, S. A., Potter, M. B. & Walsh, M. M. (2016)

Oral health prevention and toddler well-child care: routine integration in a safety net system.

Pediatrics, 137 (1), e20143532.

Fejerskov, O. (2004) Changing paradigms in concepts on dental caries: consequences for oral

health care. Caries Research, 38 (3), 182-191.

Gussy, M. G., Waters, E. G., Walsh, O. & Kilpatrick, N. M. (2006) Early childhood caries:

current evidence for aetiology and prevention. Journal of Paediatrics and Child Health, 42 (1-

2), 37–43.

147

Hooley, M., Skouteris, H., Boganin, C., Satur, J. & Kilpatrick, N. (2012) Body mass index and

dental caries in children and adolescents: a systematic review of literature published 2004 to

2011. Systematic Reviews, 1 (1), 57.

Jiang, E. M., Lo, E. C. M., Chu, C. H. & Wong, M. C. M. (2014) Prevention of early childhood


month randomized controlled trial. Journal of Dentistry, 42 (12), 1543–1550.

Jamieson, L., Smithers, L., Hedges, J., Parker, E., Mills, H., Kapellas, K., Lawrence, H.P.,

Broughton, J.R. and Ju, X., 2018. Dental Disease Outcomes Following a 2-Year Oral Health

Promotion Program for Australian Aboriginal Children and Their Families: A 2-Arm Parallel,

Single-blind, Randomised Controlled Trial. EClinicalMedicine, (1), 43-50.

Khoun, T., Malden, P. E. & Turton, B. J. (2018) Oral health-related quality of life in young

Cambodian children: a validation study with a focus on children with cleft lip and/or palate.

International Journal of Paediatric Dentistry. 28 (3), 326-334.

Fontana, M., Eckert, G.J., Keels, M.A., Jackson, R., Katz, B., Levy, B.T. & Levy, S.M. (2018)


Dental Research, 29 (1), 24-34.

Leong, P. M., Gussy, M. G., Barrow, S.-Y. L., Silva-Sanigorski, A. & Waters, E. (2013) A

systematic review of risk factors during first year of life for early childhood caries.

International Journal of Paediatric Dentistry, 23 (4), 235–250.

Macpherson, L.M.D., Ball, G.E., Brewster, L., Duane, B., Hodges, C.L., Wright, W., Gnich,

W., Rodgers, J., McCall, D.R., Turner, S. & Conway, D.I. (2010) Childsmile: the national child

oral health improvement programme in Scotland. Part 1: establishment and

development. British Dental Journal, 209 (2), 73.

Marinho, V. C., Higgins, J., Logan, S. & Sheiham, A. (2003) Fluoride toothpastes for

preventing dental caries in children and adolescents. The Cochrane Library, CD0002278.

Memarpour, M., Fakhraei, E., Dadaein, S. & Vossoughi, M., (2015) Efficacy of fluoride

varnish and casein phosphopeptide-amorphous calcium phosphate for remineralization of

primary teeth: a randomized clinical trial. Medical Principles and Practice, 24 (3), 231-237.

Monse, B., Heinrich-Weltzien, R., Benzian, H., Holmgren, C. & van Palenstein Helderman,

W. (2010) PUFA–An index of clinical consequences of untreated dental caries. Community



Acevedo, C. & Martínez-Zapata, M. J. (2018) Effectiveness of fluoride varnish in preventing


control trial. Community Dentistry and Oral Epidemiology, 46 (1), 63–69.

Oliveira, B. H., Salazar, M., Carvalho, D. M., Falcão, A., Campos, K. & Nadanovsky, P. (2014)

Biannual Fluoride Varnish Applications and Caries Incidence in Preschoolers: A 24-month

Follow-Up Randomized Placebo-Controlled Clinical Trial. Caries Research, 48 (3), 228–236.

148

Phantumvanit, P., Makino, Y., Ogawa, H., Rugg‐Gunn, A., Moynihan, P., Petersen, P.E.,

Evans, W., Feldens, C.A., Lo, E., Khoshnevisan, M.H. & Baez, R., 2018. WHO global

consultation on public health intervention against early childhood caries. Community Dentistry

and Oral Epidemiology, 46 (3), 280-287.

Philip, N., Suneja, B. & Walsh, L.J. (2018) Ecological approaches to dental caries prevention:

paradigm shift or shibboleth? Caries Research, 52 (1-2), 153-165.

Rai, N.K. & Tiwari, T. (2018) Parental Factors influencing the Development of early childhood

caries in Developing Nations: a Systematic Review. Frontiers in Public Health, 6, 64.

Shearer, D. M., Thomson, W. M., Broadbent, J. M. & Poulton, R. (2011) Maternal oral health

predicts their children’s caries experience in adulthood. Journal of Dental Research, 90 (5),

672–677.

Sheiham, A. & James, W.P.T. (2015). Diet and dental caries: the pivotal role of free sugars

reemphasized. Journal of Dental Research, 94 (10), 1341-1347.

Sokal-Gutierrez, K., Turton, B., Husby, H. & Paz, C.L. (2016). Early childhood caries and

malnutrition: Baseline and two-year follow-up results of a community-based prevention

intervention in rural Ecuador. BMC Nutrition, 2 (1), p.73.

Tickle, M., O’Neill, C., Donaldson, M., Birch, S., Noble, S., Killough, S., Murphy, L., Greer,

M., Brodison, J., Verghis, R. & Worthington, H.V. (2017) A randomized controlled trial of

caries prevention in dental practice. Journal of Dental Research, 96 (7), 741-746.

Turton, B., Durward, C., Manton, D., Bach, K. & Yos, C. (2016) Socio-behavioural risk factors

for early childhood caries (ECC) in Cambodian preschool children: a pilot study. European

Archives of Paediatric Dentistry, 17 (2), 97-105.

Turton, B. J., Thomson, W. M., Page, L. A. F., Saub, R. B. & Razak, I. A. (2015) Validation

of an Oral Health–Related Quality of Life Measure for Cambodian Children. Asia-Pacific

Journal of Public Health, 27 (2), NP2339–NP2349.

Varni, J.W., Limbers, C.A. & Burwinkle, T.M. (2007) How young can children reliably and

validly self-report their health-related quality of life?: An analysis of 8,591 children across age

subgroups with the PedsQL™ 4.0 Generic Core Scales. Health and Quality of Life Outcomes,

5 (1), 1.

Weintraub, J. A., Ramos-Gomez, F., Jue, B., Shain, S., Hoover, C. I., Featherstone, J. D. B. &

Gansky, S. A. (2006) Fluoride Varnish Efficacy in Preventing Early Childhood Caries. Journal

of Dental Research, 85 (2), 172–176.

Wigen, T. I. & Wang, N. J. (2012) Parental influences on dental caries development in

preschool children. An overview with emphasis on recent Norwegian research. Norsk

Epidemiologi, 22 (1), 13-19.

Wong, M., Glenny, A.-M., Tsang, B. W., Lo, E., Worthington, H. V., & Marinho, V. C. (2011)

Cochrane review: Topical fluoride as a cause of dental fluorosis in children. Evidence-Based

Child Health: A Cochrane Review Journal, 6 (2), 388–439.

151

Chapter 6

Overview, limitations and recommendations of the SEAL

Cambodia and Cambodia Smile projects

152

6.0 Chapter 6 - Overview, limitations, and recommendations of the

SEAL Cambodia and Cambodia Smile projects

It is important that therapies for controlling dental caries are critically evaluated in each context

in which they are applied. There is evidence from other settings that FS and fluoride toothpaste,

and FV can render some benefit in the control of the caries process (Peterson et al. 2004;

Yengopal et al. 2009); however, adaptation of those therapies to the Cambodian context had

not yet been explored. The situation in Cambodia is challenging due to the severe burden of

dental caries, an unfavourably positioned work force, and the limited availability of affordable

fluoride toothpaste or fluoridated water. Furthermore, the dental caries disease as a complex

problem amplified by the rising social inequality, and the limited political accountability

(Baker et al. 2018) which reinforces the need for robust multidisciplinary and multisectorial

approaches. This might only be possible when there is evidence around the most appropriate

strategies for control of the caries process.

The present thesis explores the adaptation of preventive therapies for the Cambodian context.

The SEAL Cambodia intervention was successful for preventing dental caries on first

permanent molars using a modified protocol; however, the preventive increment was not as

great as was hoped. The Cambodia Smile intervention was successful at greatly reducing the

burden of dental caries for 2 year-old children among a pilot sample. Both of these interventions

had some significant limitations which present opportunities for further investigation.

6.1 Summary of findings from the ‘SEAL Cambodia’ project

The SEAL Cambodia project was the result of a collaboration among local and international

partners and the stated aim was to provide FS for 60,000 children between the ages of 6 and 8-

years as a way of reducing the caries burden in the permanent dentition. The present study was

aimed at monitoring and evaluating the efficiency of the treatment and the major strength of

the study was that problems in the pilot protocol were identified and remedied with a modified

protocol. The major limitation of the study and the intervention as a package is that it did not

address holistic causes of the caries process. If the project was aimed at managing the caries

process, then outcomes such as improved tooth-brushing habits and altered dietary patterns

would have been outcomes of interest. In contrast, the study focused on a single therapy that

was refined to prevent cavitated lesions on a single tooth surface and as such the failure to

control the wider caries process limited the benefit that the FS therapy in question was able to

153

achieve. The caries process that started, in many cases, on approximal surfaces extended to

occlusal surfaces and compromised the preventive benefit observed on FPM.

As well as limitations in protocol, there were also limitations in the sampling procedures. In

the original cohort (those in Pilot A of Chapter 2), the sampling procedure, while randomised,

produced a control group who had significantly lower caries experience at baseline and this

could have masked some of the benefits rendered within the original pilot study. In the second

cohort (those in Group C and Group D of Chapter 3), the positive control group had a similar

baseline caries experience to that of the modified protocol group but a significantly lower

subsequent incidence of proximal caries. Again, these differences could have masked some of

the benefit of the FS procedure. Despite this, the modified protocol did achieve 2-year

preventive increments which are consistent with the higher end of the confidence interval

around expected benefits from GIC FS procedures (Ahovuo-Saloranta et al. 2016).

One of the benefits that wasn’t captured in the present investigation was the organisational and

structural changes that occurred as a result of the dental community in Cambodia rallying

around an upstream intervention. That meant that 60,000 children experienced dentistry that

was not focused on surgical intervention but on prevention and that the networks dedicated to

paediatric oral heath in Cambodia became more proficient at providing logistical support for

interventions at large scale (One-2-One Cambodia 2016). That the SEAL Cambodia project

represented a first effort to deploy preventive interventions at scale in a Cambodian context is

itself a potentially significant outcome of interest and value that is not captured in the

presentation of caries preventive fractions.

The rationale for the intervention followed conventional knowledge that the most effective

setting to address oral health is in a school where it is logistically easier to deliver an

intervention and the target group is captive in a well-controlled environment (Monse et al.

2010). It also followed the evidence that, next to tooth-brushing with fluoride toothpaste, FS is

the second most effective way to prevent dental caries in children. Although the rationale

behind selecting FS as a key intervention was justified, critical analysis of the data from the

participants of the SEAL Cambodia project led to a better understanding of the natural history

of caries among the Cambodian school population. In particular, the populations amongst

whom other sealant studies have been conducted, experience the majority of their new carious

lesions on permanent teeth on the occlusal surfaces of their FPM. In contrast, one in five

participants in the SEAL Cambodia group developed frank cavitation on an approximal surface

154

that extended to the occlusal surface over the two-year observation period. In this situation FS

is not effective in protecting that tooth from the progression of caries or managing the majority

of lesions among a population. Therefore, although the Fuji VII® FS did render benefit,

managing the majority of carious lesions amongst Cambodian primary school children would

potentially require a combination of approaches. These might include supervised tooth-

brushing and complimentary secondary preventive techniques (Goldman et al. 2017; Ruff &

Niederman, 2018).

SEAL Cambodia is not the only large-scale preventive oral health intervention in Cambodia.

The two other prominent interventions are the Fit For School (FFS) intervention which

rendered a 20% preventive increment (Duijster et al. 2017) and the Bright Smiles Bright

Futures (BSBF) project which has not been evaluated in terms of caries prevention or behaviour

change. The FFS intervention features daily tooth-brushing with 1450 ppm fluoride toothpaste

and it focuses on creating a healthy environment first and foremost (Monse et al. 2010). The

overall caries preventive increment was somewhat higher than that achieved in the SEAL

Cambodia project and it had a larger impact over a wider target group within the primary school

environment. In addition, it is probable that the tooth-brushing slowed the progression of both

approximal and occlusal lesions. In contrast, the SEAL Cambodia strategy had a very specific

function in the prevention of dental caries on the occlusal surfaces of FPM. The FFS project

did not collect any self-reported data on oral symptoms and so it is not possible to say whether

the tooth-brushing resulted in an improvement in OHRQoL, although other similar studies have

reported such improvements (Clark, 2017). The mantra still stands that tooth-brushing with

fluoride toothpaste is the single most effective way to prevent dental caries. In the Cambodian

setting the preventive fraction achieved through the FFS program was lower than that achieved

in other settings, however, that a reduction was achieved reinforces the evidence that tooth-

brushing with fluoride toothpaste should be the first step in implementing any preventive

program. Further investigation could examine how FS could supplement other primary

preventive interventions such as a school-based tooth-brushing program or secondary

interventions such as simple ART restorations in a school setting. It is clear that FS are not a

stand-alone intervention; however, in this case they were a good first step in reorienting dental

services towards a more-upstream approach and an entry point for reorientation of services.

155

6.2 Summary of findings from the Cambodia Smile project

The data from the SEAL Cambodia cohort combined with the cross-sectional data on ECC

from Chapter 4 highlighted that school-age was too late to address the caries problem for

Cambodian children. That most children were exhibiting the signs of dental caries in the early

years of life, and those with an extreme caries burden (dmft > 9) did not respond favourably to

preventive therapies led to the conclusion that dental caries should be managed before children

reach school-age. The risk modelling among a convenience sample of Cambodian pre-

schoolers helped to inform the strategy for the Cambodia Smile pilot even in the presence of

clear shortcomings in the sampling frame of the cross-sectional study (Chapter 4). As a

convenience sample was involved, the findings of the study could not be extrapolated to

represent the wider population. Also, the sample size was not large enough to allow for

meaningful sub-group analysis which limited the investigation in terms of better understanding

of dietary habits across various age groups. Subsequently, the value of the dietary indicators

included in the survey were limited and failed to inform the risk profile. Holistic investigation

of caries in any setting requires understanding of the drivers of the caries process. Therefore,

further investigations are warranted to better understand the dietary patterns, particularly

around nursing habits and complimentary feeding that contribute to the caries process among

preschool children in Cambodia. In addition, investigation is needed to understand the

mediators of those behaviours and the context specific mechanisms for change (Rai and Tiwari,

2018). Despite the limitations of the cross-sectional survey undertaken, the data was a valuable

first attempt at assessing need and modelling the risk indicators among a group of preschool

Cambodian children in the absence of other information.

Although the Cambodia Smile intervention was successful in reducing caries increment among

those in the intervention, there were some limitations that would argue caution around the

interpretation of the results. In particular, the control and intervention groups were recruited

using different sampling procedures and that led to less diversity among the control group in

terms of sociodemographic characteristics. This could have played a role in either amplifying

or masking benefits of the intervention although there were no significant differences in caries

experience by sociodemographic characteristics among the sample. The second limitation is

that although the final follow-up rate was acceptable at around 70%, follow-up or attendance

at each opportunity for the preventive intervention was much lower. That meant that not all

children received FV and the sample was not powered to accurately examine differences based

156

on varying levels of the intervention. Further investigation, at a greater scale, is needed to

understand how those who realized those opportunities for intervention differed from those

who did not. Also, that some children in the intervention group still experienced a severe caries

burden, and one in twenty five children in the intervention group experienced dental abscesses,

suggests that the intervention benefited some participants more than others and the

characteristics of those participants who did not realize the benefits should be further explored.

The identification of late introduction of tooth-brushing habits in the cross-sectional survey

was consistent with the findings of the CNOHS (Chher et al. 2016) and indicated that the

preschool population in Cambodia experienced almost no exposure to fluoride. As such, the

benefits of the Cambodia Smile pilot intervention could largely be attributed to the introduction

of fluoride in a previously non-fluoridated community. The Cambodia Smile intervention was

successful at using fluoride to moderate the relationship between an unfavourable diet and

caries severity. The study replicates the well documented ability of fluoride to reduce disease

burden (Bernabé et al. 2016; Kumar & Moss 2008). The intervention was not successful at

adjusting behaviour around the key driver of the disease, the diet, as evidenced by the absence

of any significant differences in the consumption of non-nutritious foods among the control

and intervention group. The application of fluoride in the form of FV and fluoride toothpaste

is the best effort of a health profession at moderating the symptoms of a behavioural disease.

That the benefits were seen both at tooth level and in the household group, as measured by the

FIS, is sufficient justification for upscaling the Cambodia smile pilot as a first step at reducing

the caries burden at a population level.

It is worth noting that the benefits of FV observed in the present study are greater than those

reported in some more recent studies. This could be explained by the fact that the FV was

applied at a pivotal time, as informed by the cross-sectional data demonstrating that carious

lesions are forming in the first year of life. In contrast, those studies that did not observe the

benefits of FV may have been limited by the fact that the population already had a prevalence

of ECC lower than 40% (Agouropoulos et al. 2014; Anderson et al. 2016; Divaris et al. 2013;

Jiang et al. 2014; Oliveira et al. 2014); the participants were not at optimal age or caries risk to

realize benefit (Muñoz-Millán et al. 2018; Tickle et al. 2017) or the protocol for selection of

cases by the working health professional did not ensure that all those at risk received the FV

(Fontana et al. 2018). Further investigations at a larger scale would be worthwhile to build

evidence for ideal timing, target groups, and policy mechanisms that would inform those health

professionals in practice about the best protocol for selecting cases for application. The

157

protocol for FV application in Cambodia Smile in which every child receives FV application

at each opportunity appears to be suitable for rendering preventive benefits in a Cambodian

setting at dose that is appropriate. This could be confirmed by close monitoring during the

upscale phase to observe both differences in caries and presence of enamel defects.

Furthermore, given the attendance patterns among the target community, applying FV at scale

would not be possible outside of the demonstrated model in a primary health care setting and

by primary health care providers.

Although the fluoride applications were successful, even if the same preventive increments are

realized at population level, there will still be a proportion of children in Cambodia who still

develop carious lesions and for whom the intervention would not work. Therefore, further

research from a mixed-methods approach is needed to better understand the social, structural,

and behavioural aspects of the caries experience in the Cambodian context so that a more

definitive strategy can be implemented. In the meantime, it is clear that the Cambodia Smile

intervention was successful for a large proportion of the group and that is sufficient justification

for upscaling along with close monitoring.

158

6.3 Conclusions

The present series of investigations are an example of the adaptation of strategies to achieve

optimal results in a specific setting. The SEAL Cambodia project adds to the body of evidence

around the use of GIC FS for the prevention of carious lesions on FPM. Further investigation

could focus on the role of FS within a more holistic caries preventive approach in Cambodian

school children. The SEAL Cambodia project was adapted to achieve a favourable preventive

fraction of 90% at 1 year. When the 2-year preventive fraction (30%) was less favourable due

to the presence of approximal lesions, the Cambodia Smile project was a response to the

realization that school-age was too late to address the caries disease among Cambodian

children. The Cambodia Smile intervention builds evidence for the introduction of topical

fluorides at a young age and in this case, such an application (in combination with provision of

fluoride toothpaste) rendered a 66% reduction in caries severity. It is unique in that it is the

first study to demonstrate benefits of such of an intervention in terms of OHRQoL.

Further investigation is needed to better understand the social, structural and behavioural

aspects of the ECC phenomenon in Cambodia in order to better inform strategy for behaviour

change to address the diet. The Cambodia Smile pilot provides sufficient evidence to justify

up-scale and monitoring in order to reduce the disease burden among a larger proportion of

Cambodian children.

159





Dentistry, 42 (10), 1277–1283

Anderson, M., Dahllöf, G., Twetman, S., Jansson, L., Bergenlid, A.-C. & Grindefjord, M.

(2016) Effectiveness of Early Preventive Intervention with Semiannual Fluoride Varnish


Controlled Trial. Caries Research, 50 (1), 17–23.

Ahovuo-Saloranta, A., Forss, H., Hiiri, A., Nordblad, A. & Mäkelä, M. (2016) Pit and fissure

sealants versus fluoride varnishes for preventing dental decay in the permanent teeth of children

and adolescents. The Cochrane Library, CD003067.

Baker, S.R., Foster Page, L., Thomson, W.M., Broomhead, T., Bekes, K., Benson, P.E.,

Aguilar-Diaz, F., Do, L., Hirsch, C., Marshman, Z. & McGrath, C. (2018) Structural

Determinants and Children’s Oral Health: A Cross-National Study. Journal of Dental

Research, In Print:0022034518767401

Bernabé, E., Vehkalahti, M. M., Sheiham, A., Lundqvist, A. & Suominen, A.L. (2016) The

shape of the dose-response relationship between sugars and caries in adults. Journal of Dental

Research, 95 (2), 167-172.




Clark, E. (2017) Supervised tooth-brushing in Northland. (Doctoral dissertation, University of

Otago).

Duangthip, D., Chu, C.H. & Lo, E.C.M. (2016) A randomized clinical trial on arresting dentine

caries in preschool children by topical fluorides—18 month results. Journal of Dentistry, 44,

57-63.

Divaris, K., Preisser, J. S. & Slade, G. D. (2013) Surface-Specific Efficacy of Fluoride Varnish

in Caries Prevention in the Primary Dentition: Results of a Community Randomized Clinical

Trial. Caries Research, 47 (1), 78–87.

Duijster, D., Monse, B., Dimaisip-Nabuab, J., Djuharnoko, P., Heinrich-Weltzien, R., Hobdell,

M., Kromeyer-Hauschild, K., Kunthearith, Y., Mijares-Majini, M.C., Siegmund, N. &

Soukhanouvong, P. (2017) ‘Fit for school’–a school-based water, sanitation and hygiene

programme to improve child health: Results from a longitudinal study in Cambodia, Indonesia

and Lao PDR. BMC Public Health, 17 (1), 302.

Fontana, M., Eckert, G.J., Keels, M.A., Jackson, R., Katz, B., Levy, B.T. & Levy, S.M. (2018)


Dental Research, 29 (1), 24-34.

Goldman, A., Leal, S. C., Amorim, R. G. de & Frencken, J. E. (2017) Treating High-Caries

Risk Occlusal Surfaces in FPM through Sealants and Supervised Tooth-brushing: A 3-Year

Cost-Effective Analysis. Caries Research, 51 (5), 489–499.

160

Jiang, E. M., Lo, E. C. M., Chu, C. H. & Wong, M. C. M. (2014) Prevention of early childhood


month randomized controlled trial. Journal of Dentistry, 42 (12), 1543–1550.

Kumar, J. V. & Moss, M. E. (2008) Fluorides in dental public health programs. Dental Clinics

of North America, 52 (2), 387–401.

Monse, B., Naliponguit, E., Belizario, V., Benzian, H. & Helderman, W. van P. (2010)

Essential health care package for children-the “Fit for School”program in the Philippines.

International Dental Journal, 60 (2), 85–93.


Acevedo, C., & Martínez-Zapata, M. J. (2018) Effectiveness of fluoride varnish in preventing


control trial. Community Dentistry and Oral Epidemiology, 46 (1), 63–69.

Oliveira, B. H., Salazar, M., Carvalho, D. M., Falcão, A., Campos, K. & Nadanovsky, P. (2014)

Biannual Fluoride Varnish Applications and Caries Incidence in Preschoolers: A 24-month

Follow-Up Randomized Placebo-Controlled Clinical Trial. Caries Research, 48 (3), 228–236.

One-2-One Cambodia (2016) Final report for SEAL Cambodia. Phnom Penh, Cambodia.

Petersson, L., Twetman, S., Dahlgren, H., Norlund, A., Holm, A.K., Nordenram, G., Lagerlöf,

F., Söder, B., Källestål, C., Mejàre, I. & Axelsson, S., (2004) Professional fluoride varnish

treatment for caries control: a systematic review of clinical trials. Acta Odontologica

Scandinavica, 62 (3),170-176.

Rai, N.K. & Tiwari, T. (2018) Parental Factors influencing the Development of early childhood

caries in Developing Nations: a Systematic Review. Frontiers in Public Health, 16 (6) 64.

Ruff, R. R. & Niederman, R. (2018) Comparative effectiveness of school-based caries

prevention: a prospective cohort study. BMC Oral Health, 18 (1), 53.

Tickle, M., O’Neill, C., Donaldson, M., Birch, S., Noble, S., Killough, S., Murphy, L., Greer,

M., Brodison, J., Verghis, R. & Worthington, H.V. (2017) A randomized controlled trial of

caries prevention in dental practice. Journal of Dental Research, 96 (7), 741-746.

Yengopal, V., Mickenautsch, S., Bezerra, A. C. & Leal, S. C. (2009) Caries-preventive effect

of glass ionomer and resin-based fissure sealants on permanent teeth: a meta analysis. Journal

of Oral Science, 51 (3), 373–382.

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s:

Turton, Bathsheba

Title:

Early childhood caries in Cambodia

Date:

2018

Persistent Link:

http://hdl.handle.net/11343/220788

File Description:

Early childhood caries in Cambodia

Terms and Conditions:

Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the

copyright owner. The work may not be altered without permission from the copyright owner.

Readers may only download, print and save electronic copies of whole works for their own

personal non-commercial use. Any use that exceeds these limits requires permission from

the copyright owner. Attribution is essential when quoting or paraphrasing from these works.

Early Childhood Caries in Cambodia - Minerva Access

Documents

Transcript of Early Childhood Caries in Cambodia - Minerva Access