Cochrane Systematic Reviews of Occupational Health ... · PDF fileCochrane Systematic Reviews...

Cochrane Systematic Reviews of Occupational Health Interventionso sJos VerbeekKnowledge Transfer Team, Fi i h I tit t f O ti l H lth K iFinnish Institute of Occupational Health, KuopioCoronel Institute of Occupational Health, Amsterdam

In this workshop..

• you will learn• ….you will learn….

– what the Cochrane Collaboration is

– how a Cochrane Systematic Review is produced

– what a Cochrane Systematic Review is• what framework is used• what the methods used are

– what the problems are in implementing the results of what the problems are in implementing the results of Cochrane Systematic Reviews

Presentation EASOM Summerschool / august 2008 2

Cochrane Collaboration

• Mission• Mission– to provide up to date evidence on the effects of health care

• Productsf h l h– systematic reviews of healthcare interventions

– in Cochrane Library: www.thecochranelibrary.org– according to quality handbook

• Name– after Archie Cochrane, British epidemiologist

• Organisationg– network of 50 review groups that produce systematic reviews

supported by fields and centres– involves 10.000 persons on voluntary basis and small

professional staff• Web address

– www.cochrane.org


Cochrane Occupational Health Field

• FIOH started Occupational Health Field in the Cochrane • FIOH started Occupational Health Field in the Cochrane Collaboration

– funding from FIOH, Ministry of Health, FinOHTA, Dutch f l ff lMinistry of Social Affairs, NIOSH Australia

• Cochrane Occupational Health Field (COHF)o a O upa o a a d ( O )– coordinator, information specialist, researchers

• jos verbeek, merja jauhiainen, jani ruotsalainen, marika lehtola, irja laamanen, riitta sauni, iris pasternack, maritta lehtola, irja laamanen, riitta sauni, iris pasternack, maritta kinunen, juha liira

– over 50 authors involved from all over the worldover 50 authors involved from all over the world

– officially registered with CC in May 2004


Occupational Health Interventions

Environmental interventions Behavioural Interventions Clinical Interventions

Interventions: Interventions: Interventions:Interventions:Work-site visits. Organisational changes. Technical changes. Banning

Interventions:Health promotion.Education. Info material. Reward-punishment. T h i l h

Interventions:TreatmentCounselling Vaccination

Worker Health Behaviour

Technical changes. Banning. CBT

Behaviour

Risk factor at work Disease, Disability, Injury


Verbeek, Scand J WEH 2004

Tasks Occupational Health Field

• To maintain database of occupational health intervention • To maintain database of occupational health intervention studies

– gather trials by hand searching journals• database at www.cohf.fi

• To stimulate the conducting of Cochrane Occupational Health Reviews

– We……• organise funding• write ourselves• write ourselves• organise reviewers• gather topics etc.

Reviews are located with Cochrane Review Groups– Reviews are located with Cochrane Review Groups• formally not with a field

– Field has no formal power


• mediate between reviewers and Cochrane review groups

Cochrane Review Production Process

• Title registration with a review group• Title registration with a review group– avoids duplication

• Protocol – background, search, methods

peer reviewed– peer-reviewed– publication in Cochrane Library

• Review– peer-reviewed (editors and external experts)

according to Cochrane Handbook– according to Cochrane Handbook– published in Cochrane Library– possibly published as a journal article


Cochrane Review Production Process

• Time line• Time line– protocol: six months– review

• search selection: three months• data extraction: three months• data analysis: three months• writing: three months

– peer-review: six months– time to publication: two yearstime to publication: two years

• Updating– every two years


Contents of a Cochrane Review

• Title• Title

• Inclusion criteria

• Search

• Data Analysis– comparisons

outcome data– outcome data– pooling

• forest plot

• Conclusions– implications for practice


– implications for research

Titles of Cochrane Reviews

• Titles for Cochrane reviews have a set format• Titles for Cochrane reviews have a set format– P I C O– intervention for health problem– intervention for health problem A versus intervention B – intervention A versus intervention B for health problem

in participant group/locationp p g p/


Title

• Formulate a title according to the PICO acronym• Formulate a title according to the PICO acronym


Title

• For occupational health reviews• For occupational health reviews– 'intervention' for the prevention and treatment of

occupational outcome• noise induced hearing loss, back pain, disability as a result of

back pain, work-related stress, occupational injuries

k di d i i f i l – work-directed interventions for occupational outcome

• workplace adjustments, work reorganisation

– intervention' for 'health problem' in workers exposed to workplace health hazard

• hearing loss prevention programme • workplace smoking cessation programmes (?)


Inclusion criteria

• S PICO• S PICO– study-design – participants – intervention– control– outcomeoutcome

• To be used in study/article selection– be as concrete as possible!


Inclusion criteria: study designs

• Evaluation research in OH• Evaluation research in OH– often more difficult to perform because

• employers, employees not interested• rapid changes in working populations and firms• unpredictable changes in organisations• interventions often carried out at group level

– work accommodations

– needs evidence other than RCTs• cluster randomised trials• cohort studies• interrupted time series• case-control studies• before-after studies


Inclusion Criteria: Study Designs

• the lower the methodological quality• the lower the methodological quality…

– the lower the power to change believesp g

– the more work it is to locate studies

– the more difficult it is to agree about study designs

– the more difficult it is to draw conclusions


Inclusion Criteria:Study Designs In Cochrane ReviewsStudy Designs In Cochrane Reviews

• Randomised Controlled Trials• Randomised Controlled Trials– cluster randomised, cross-over trials

C ll d f f S di• Controlled Before After Studies

• Interrupted Time Series


Inclusion Criteria:Randomised Controlled TrialRandomised Controlled Trial

Oi X OiRandomisation Eligible

Oc OcProcessindividuals


Inclusion Criteria:Controlled Before-After StudyControlled Before After Study

Oi X OiAssignment based on Eligible

Oc Ocsome characteristicindividuals


Inclusion Criteria:Interrupted Time SeriesInterrupted Time Series

….Oi Oi Oi X Oi Oi Oi Oi…Eligible Group of

(Oc Oc Oc OcOc Oc Oc)Group of

Individuals


Inclusion Criteria

• Ergonomic shift scheduling to prevent cardiovascular • Ergonomic shift scheduling to prevent cardiovascular disease in shift workers

• Formulate Inclusion Criteria: S PICO


Inclusion Criteria

• Intervention in shift scheduling and changes in biomarkers of heart disease in hospital g g pwards. Henrik Bøggild, Hans Jeppe Jeppesen Scand J Work Environ Health 2001;27(2):87—96

• Objectives The effect of introducing regularity, few consecutive night shifts, more weekends off, and only 2 different types of shifts (day-evening or day-night) into shift , y yp ( y g y g )scheduling on biomarkers of heart disease was studied.

• Methods Ergonomic shift criteria were introduced in a quasi-experimental controlled intervention in 4 hospital wards. Six wards participated as controls. Altogether 101 nurses and nurses’ aides were followed for 6 months with measurements of cholesterol nurses and nurses aides were followed for 6 months with measurements of cholesterol and triglycerides. The intervention led to more regular schedules and more staff having 2 shifts in 2 of the intervention wards 1 year after the intervention. The schedules among the controls became less regular and less predictable. The number of consecutive night shifts remained unchanged.

• Results After 6 months the HDL cholesterol level had increased in the intervention group, and the total cholesterol and LDL cholesterol levels and the total:HDL cholesterol ratio had decreased. Regardless of the intervention, changes in regularity were associated with the triglyceride and HDL cholesterol levels and also with the total:HDL cholesterol ratio. More ergonomic changes were associated with lower LDL h l l l l l l HDL h l l i d hi h HDL h l l cholesterol levels, a lower total:HDL cholesterol ratio, and higher HDL cholesterol

levels.

• Conclusions Increased ergonomic scheduling was possible. Lipids and lipoproteins changed as predicted, both when the changes were assessed in respect to the changes


g p , g p gin schedules that resulted from the intervention and the changes that occurred regardless of the intervention. The study suggests that scheduling based on ergonomic criteria is a possible means for reducing the risk of heart disease among shift workers.

Search and study selection

• Based on S PICO• Based on S PICO

• Usually combination of search terms forstudy type– study-type

• randomised controlled trial search string– intervention– outcome– work

• Sensitive search yieldsy– 1000 to 8000 references– most sensitive is 15 million references

• Article selection– Article meeting inclusion criteria?– In duplicate because unreliable


Data Synthesis

• What to combine?• What to combine?– similar

• S PICO• follow-up time• setting

– arbitrary judgement

• Example


Data synthesis

• Make sensible comparisons:• Make sensible comparisons:

1. Ear muffs reduce noise levels with 20 dB(A)( )2. EAR plugs reduce hearing loss with 10 dB3. Legislation reduces noise levels with 5 dB(A)4. Worker training in plug use reduces hearing loss with

15 dB5. Subsidies for employers reduce noise levels with 10 5. Subsidies for employers reduce noise levels with 10

dB(A)6. Information campaign for workers reduces hearing loss

ith 10 dBwith 10 dB7. Magnesium for noise-exposed workers reduces hearing

loss with 5 dB


Data-synthesis: comparisons

• Hearing protection vs no protection• Hearing protection vs no protection– hearing loss (plugs)– noise reduction (muffs)

• Employer incentives vs no incentives– noise reduction (legislation, subsidy)

• Worker behaviour change vs no intervention• Worker behaviour change vs no intervention– hearing loss (information, training)

• Magnesium vs no interventiong– hearing loss


Data synthesis: outcomes

• Worker training to prevent injuriesWorker training to prevent injuries

• Outcome– Study 1:y

• intervention: 12 injuries / 45 workers• control: 18 injuries / 37 workers• author's conclusions:

– RR 0.46 (95% CI 0.1 to 1.9)– no significant outcome

– Study 2:• intervention: 16 injuries / 201.000 working hours• control 26 injuries / 193.000 working hours• author's conclusions:

– RR 0.59 (95% CI 0.29 to 1.12)– no significant outcome

• No quantitative analysis possible we combined studies qualitatively:

– conclusion: based on two studies with a non-significant outcome


conclusion: based on two studies with a non significant outcome we found no evidence of effectiveness

Data synthesis: outcomes

• Recalculate all outcomes on similar scale• Recalculate all outcomes on similar scale– 2000 working hours = 1 working year (US)

• Combine in meta-analysis– Pooled:

• RR 0 59 (95% CI 0 41 to 0 93)• RR 0.59 (95% CI 0.41 to 0.93)

• Review Conclusion:– the intervention reduces injuries with 40%


Data analysis: forest plot

Review: Interventions for preventing injuries in the agricultural industry (1007007)Comparison: 01 Educational intervension versus no intervention Outcome: 01 All injuries adults

Study Relative Risk (fixed) Weight Relative Risk (fixed)or sub-category log[Relative Risk] (SE) 95% CI % 95% CI Year

Pekkarinen 1994 0.0337 (0.0997) 65.18 1.03 [0.85, 1.26] 1994Rasmussen 2003 0 0488 (0 1747) 21 23 0 95 [0 68 1 34] 2003Rasmussen 2003 -0.0488 (0.1747) 21.23 0.95 [0.68, 1.34] 2003Rautiainen 2004 0.0476 (0.2183) 13.60 1.05 [0.68, 1.61] 2004

Total (95% CI) 100.00 1.02 [0.87, 1.19]Test for heterogeneity: Chi² = 0.19, df = 2 (P = 0.91), I² = 0%

Test for overall effect: Z = 0.22 (P = 0.82)

0 2 0 5 1 2 50.2 0.5 1 2 5

Favours treatment Favours control


Conclusions

• Cochrane Review Section: • Cochrane Review Section: – Implications for practice – Implications for Research

• How active should we be in implementation?

• Implementation process:– What is current practice?p– Is there a need for change?– How can we best contribute to the change?

Can we measure a change?– Can we measure a change?


Lifting Advice to prevent back pain

• Results review• Results review– 6 RCTs and 5 Cohort Studies: no positive outcome

– For preventing back pain or back pain related disability

– MMH advice not more effective than

• no advice (3 RCTs)

• very brief advice (1 RCT)

• physical exercise (1 RCT)

• back belt use (2 RCTs)

– MMH advice and assistive devices not more effective than

d l ( )• MMH advice alone (1 RCT)

• no advice (1 RCT)

Cohort studies reported the same results


– Cohort studies reported the same results

Lifting Advice to prevent back pain

• Review Implications for practice• Review Implications for practice– This review does not provide evidence that training and

advice by themselves prevent back pain– (Mandatory training should be reconsidered)

• Practice• Practice– Wide spread practice of advice (ergonomy)

– Mandatory training according to EU regulation

– EU subsidies for training– EU subsidies for training

– Strong belief among OH practitioners


Ecosh: Economic Dimension of OSH

• Implementation facilitators• Implementation facilitators– Economic evaluation, business case, incentives for uptake of

OSH-interventions

• FP-7 EU subsidized workshops

• Helsinki 23-24 1 2009• Helsinki 23 24.1.2009– productivity and OSH interventions

• Amsterdam 8-9.5.2009Amsterdam 8 9.5.2009– economic evaluation from company's perspective

• Lodz 28-29.11.2009od 8 9 009– economic incentives for OSH uptake

• WWW.ECOSH.EU


Conclusions

• Cochrane methodology can be very well applied to a • Cochrane methodology can be very well applied to a wide variety of OH topics

• More research designs than RCTs should count as evidence

• Evidence from non-randomised studies poses sometimes problems that are not surmountable

• Implementation calls for more knowledge of ent p ti e– of current practice

– powers involved in changing current practice


Cochrane Systematic Reviews of Occupational Health ... · PDF fileCochrane Systematic Reviews...

Documents

Transcript of Cochrane Systematic Reviews of Occupational Health ... · PDF fileCochrane Systematic Reviews...