Monitoring Impacts of WASH Interventions in

Bangladesh, the Health Impact Study “Sanitation, Hygiene education, and Water supply intervention in rural

Bangladesh” (SHEWA-B)

1

SHEWA-B Project

• Program targets 30 million rural Bangladeshi people

• Budget of over 100 million dollars

• Implemented by Government of Bangladesh with technical • Implemented by Government of Bangladesh with technical

support from UNICEF.

• Targets behavior change in sanitation & hygiene

– Support for water in arsenic-affected areas

2

SHEWA-B Intervention

• Participatory, demand-driven approach at community level

• Local NGOs subcontracted by GOB

– 10,000 Community hygiene promoters (CHP)

– 500 to 550 households per CHP– 500 to 550 households per CHP

• Intervention method include:

– Household visits, court yard meetings

– Tea stall sessions, Watsan fair

– Village theatre

3

SHEWA-B Intervention

• CHPs emphasized the promotion of:

– Hand washing with soap

– Appropriate feces disposal

– Latrine coverage and usage (without project subsidy)

– Appropriate waste disposal

– Appropriate menstrual hygiene

– Access to and use of arsenic-free water

– Safe collection and storage of drinking water

4

Intervention

areas of

SHEWA-B

5

SHEWA-B

Purpose of HIS

To investigate the health related impact of the

program interventions by:

– Assessing morbidity in children under 5 years

– Measure changes of hygiene, sanitation – Measure changes of hygiene, sanitation

and water related behavior

6

Methodology of HIS

• Multiple integrated surveys conducted at baseline, interim

assessment and midline.

• Quarterly / monthly Sentinel Surveillance??

• Sampling: clusters randomly selected

– Probability proportional to size sampling (PPS)

– 50 intervention and 50 matched control clusters.

– 10 to 17 households per cluster

7

Washing both hands with soap/ash

52 51

3430

30

40

50

60

Pe

rce

nta

ge

s

Reported baseline Observed Baseline Observed interim assessment

8

25

1311

0.7 0.4 0.7

22

17

1 0.9 1

30

0

10

20

30

Before

preparing food

Before eating Before feeding

a child

After cleaning

child’s anus*

After

defecation*

Pe

rce

nta

ge

s

Washing both hands with soap/ash-Observed

Data in next slide is not matching!

36

3035

40

45

50

Pe

rce

nta

ge

sIntervention Baseline Intervention Midline

Control Baseline Control Midline Target

9

22

17

30

24

18

2723

0

5

10

15

20

25

30

35

After cleaning child’s anus* After defecation

Pe

rce

nta

ge

s

Washing hands after defecationin previous slide handwashing with soap, Intervention went up from 17 to 30 (minor

difference only), but for control areas, according to previous slides: 18 at baseline to

23 in midline

17

2921

20

12 14 14 10

50%

60%

70%

80%

90%

100%

Washed hands with ash

Washed hands with

soap

10

125 3 6

59

5263

64

0%

10%

20%

30%

40%

Baseline Midline Baseline Midline

Intervention Control

Washed hands with

water only

Did not wash hands at

all*

Washing both hands with soap -Observed

6

7

8

9

10

Pe

rce

nta

ge

s

Intervention Baseline Intervention Midline

Control Baseline Control Midline

Target

11

0.6 0.4 0.71 1

3

0.4 0.411 1

2

0

1

2

3

4

5

6

Before preparing food Before eating Before feeding a child

Pe

rce

nta

ge

s

Open defecation by wealth category

31

18

2220

20

25

30

35

Pe

rce

nta

ge

s

Intervention Baseline

Intervention Midline

12

13

6

1 .6

13

8

.5 0

13

21

0

10

53

2

0

5

10

15

Poorest Lower middle Middle Upper middle Richest

Pe

rce

nta

ge

s

Prevalence of diarrhea last 48 hours among children < 5

years of age in the intervention and control groups

(October 2007 -- June 2009)

15%

20%

Mea

n d

iarr

hea

pre

vale

nce

Intervention Control

15%

20%

Mea

n d

iarr

hea

pre

vale

nce

Intervention Control

13

0%

5%

10%

10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6

Mea

n d

iarr

hea

pre

vale

nce

Month

2007 2008 2009

0%

5%

10%

10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6

Mea

n d

iarr

hea

pre

vale

nce

Month

2007 2008 2009

Month

2007 2008 2009

Summary of findings.

• Significant Improvement after cleaning child’s anus.

– But short of the revised target for 2009.

• Progress in Handwashing after defecation but not statistically better than in the control group

– and is short of the revised target for 2009.

• At the midline fewer than 3% of persons washed their hands with soap before food related events.

• At the midline fewer than 3% of persons washed their hands with soap before food related events.

– There was very little change from baseline to midline and very short of the revised target for 2009, set at 10%.

• The progress noted in SHEWA-B areas regarding opendefecation is much sharper within the poorest quintiles (31% at baseline and 18% at midline).

• no difference in the health of children under the age of five years between SHEWA-B and intervention communities

14

Overall Summary

• Overall, the SHEWA-B intervention affected a handful of targeted

indicators.

• These included

– improvements in hand washing with soap after cleaning a child

who has defecated,

– improvements in sanitary facilities– improvements in sanitary facilities

– a reduction in the proportion of households that were drinking

arsenic contaminated drinking water,

– A higher proportion recalling hygiene messages.

In each of the areas of improvement, there is evidence that the

poor benefitted.

15

Overall Summary 2

• The confirmed changes were quite modest.

• Big difference between reported and observed practices calls into

question the validity of including hand washing questions in

surveys.

• These changes in the minority of the indicators, even when • These changes in the minority of the indicators, even when

significant difference from the baseline, were typically quite short

of the program targets.

• Changes in the intervention community have not been substantial

enough to lead to a measurable reduction in childhood diarrhoea

or respiratory disease in the intervention communities.

16

Points for discussion

• SHEWA-B behavioural and health targets not yet broadly

achieved

– Intervention design? Or

– implementation compromised?

– time?

• Revised intervention, perhaps smaller in scope, based on the

analysis of above points?

17