Care seeking for fatal neonatal illness episodes: a population-based study in rural Bangladesh...

Care seeking for fatal neonatal illness episodes: a population-based

study in rural Bangladesh

Centre for International Health

Md.Hafizur Rahman Chowdhury

8th October 2007Conversation Series- Doctoral Forum


Supervisors

• Principal supervisor• Sandra Thompson

• Associate Supervisors• Kieran McCall• Kim Peter Steatfield


Background

• 40-50% of under-five deaths occur in the first four weeks (neonatal period)

• 98% of deaths occur in developing countries and most of these occur at home

• One in three child deaths occurs in South-East Asia

• Poor or delayed care seeking contributes to up to 70% of child deaths

• Community level information on cause of death and health care seeking during the fatal episode is often unavailable


Objectives

• To investigate the consultation patterns of care seeking during fatal illness in the rural Matlab sub-district of eastern Bangladesh

• To assess the differentials of consultation by– Sex of deceased– Time period at death – Service area


Methodology

• Study was approved by Ethics Committees both at ICDDR,B and Curtin University

• Neonatal deaths were identified through a population-based demographic surveillance system

• Trained staff administered a structured questionnaire on care seeking practice to mothers at home along with a verbal autopsy tool

• All data entered into Visual Fox-Pro

• Descriptive statistics used for analysis

• Stata software 9 version used for analysis

Bangladesh at a glance• 141 million people• Density: 900/km2• GNI: US$ 470

• IMR:65 /1,000 live births

• NMR:41 /1,000 LB

• MMR:320 /100,000 LB

• Home delivery: 90%

• Life exp: 62 years

220,000 population under regular demographic surveillance

• Vital registration

• Monthly home visit

• Identification number

• Two service areas

• ICDDR,B area-110,000 pop (MCH-FP services)

• Government area-110,000 pop (Government services)

• Quality assurance for data


Matlab Study Site

Data collection

Verbal Autopsy questionnaire -structured -open-ended

Data management

Routine quality assuranceMedical assistant VA review

Data entry Centre for International Health

RESULTS


Timing of death, Matlab 2003-2004

136

57 56

30

8 6 6 8

2632

0

50

100

150

Fre

quen

cy

Less

than

1 d

ay1

day

2nd

day

3rd

day

4th

day

5th

day

6th

day

7th

day

8-14

day

s

15-2

8 da

ys

Age at death (day)

Neonatal deaths by age at death

Deaths=365


37%

16%15% 9%8% 7%

Type of provider during consultation

Health care provider N(365)

Percent

No treatment 137 37.5

Medically qualified 37.2

MBBS 87 23.8

Health Centre 33 9.0

Paramedic 16 4.4

Traditional/unqualified 25.4

Quack/village doctor 29 8.0

Kabiraj/herbalist 29 8.0

Homeopath 21 5.8

Spiritual healer 11 3.0

Pharmacy (drug seller) 2 0.6


Service area

ICDDR,B(173)

Government(192)

Health Care provider % %

No treatment 28.9 45.3

Medically qualified 54.9 21.4

Traditional/unqualified 16.2 33.3

Provider type by Service Area


Provider Type by Sex of Neonate

Sex of newborn

Male(200)

Female(165)

Health care provider % %





Provider Type by Time of Neonatal Death

Age at death (days)N=365

0-7 days(n=307)

8-28 days(n=58)

Health care provider % %





Consultation by sex of the neonate

Sex of newborn N=365

Male(n=200)

Female(n=165)

No of Consultations % %

0 30.0 46.7

1 44.0 34.6

2 25.0 18.8


Consultation by timing of death

Age at death (days)N=365

0-7 days(n=307)

8-28 days(n=58)


0 41.7 15.5

1 41.0 36.2

2 17.3 48.3


Consultation by service area

Service area N=365

ICDDR,B(n=173)

Government(n=192)


0 28.9 45.3

1 48.0 33.3

2 23.1 21.4


Conclusions• 84% died in the early (0-7days) neonatal period,

with 37% in first 24 hours

• Overall, 63% of the neonates received care either from traditional/unqualified provider or no care at all

• About 22% sought more than one consultation, including 6% received three or more.

• Multiple consultations for care, as well as consultations with a medically qualified provider, were more likely among male newborns, with late neonatal deaths, and in the ICDDR,B project area


Recommendations• Increase community awareness about prompt

early care seeking

• Greater emphasis on gender equality and female education

• Skilled attendance at delivery and postnatal checkup for appropriate neonatal care

• Integrating traditional care providers into mainstream health programs may be an approach to reducing neonatal mortality in the study setting


International Postgraduate Research Scholarship (IPRS)

Curtin University Postgraduate Scholarship (CUPS)

International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B)

Centre for International HealthCurtin University of Technology

Acknowledgements


Care seeking for fatal neonatal illness episodes: a population-based study in rural Bangladesh...

Documents

Transcript of Care seeking for fatal neonatal illness episodes: a population-based study in rural Bangladesh...