Interventions to improve water quality to prevent diarrhoeal disease: a systematic review and case study in Colombia

Transcription

1 Interventions to improve water quality to prevent diarrhoeal disease: a systematic review and case study in Colombia Thomas F. Clasen London School of Hygiene & Tropical Medicine Keppel St., London WC1E 7HT, United Kingdom thomas.clasen@lshtm.ac.uk Girardot, Mayo 2006

2 Leading Causes of Deaths from Infectious Diseases 2004 World Health Report Deaths (000s) Respiratory Infections HIV/AIDS Diarrhoeal Diseases Tuberculosis Malaria Measles 611

3 Leading Causes of DALYs from Infectious Diseases 2004 World Health Report DALYs (000s) Respiratory infections HIV/AIDS Diarrhoeal diseases Malaria Tuberculosis Measles

4 Childhood Mortality by Cause Black RE, Morris SS & Bryce J (2003). Where and why are 10 million children dying every year? Lancet 361:

5 Environmental Barriers to Faecal-Oral Transmission Primary Barrier Sanitation (proper excreta disposal) Hygiene (hand washing) Secondary Barriers Water quality (treatment & safe storage) Water quantity (personal and domestic hygiene) Hygiene (especially hand washing) Proper cooking/food handing practices

6 Emergence of the Dominant Paradigm Percentage reductions in diarrhoeal morbidity rates attributed tot water supply or excreta disposal improvements (Esrey( Esrey,, 1985) Reduction Type of Intervention No. Studies Median Range All Interventions 53 22% 0-100% Improvements in Water Quality 9 16% 0-90% Improvements in Water Quantity 17 25% 0-100% Improvements in Water Quality and Availability 8 37% 0-82% Improvements in Excreta Disposal 10 22% 0-48%

7 Consolidation of the Dominant Paradigm Expected reduction in diarrhoeal disease morbidity from improvements in one or more components of water and sanitation (Esrey( Esrey,, 1991) All Studies Rigorous Studies No. Studies Reduction No. Studies Reduction Water and Sanitation 7 20% 2 30% Sanitation 11 22% 5 36% Water Quality and Quantity 22 16% 2 17% Water Quality 7 17% 4 15% Water Quantity 7 27% 5 20% Hygiene 6 33% 6 33%

8 Ceramic Filter Designs Candle Filter

9

10 Microbiological Efficacy: Charinco, Bolivia Mean Fecal Coliform Count Charinco, Bolivia 1000 Arithmetic Mean FC Count Intervention Group Control Group Baseline March April June July Sam pling Period Clasen T, Brown J, Suntura O, Collin S (2004). Safe household water treatment and storage using ceramic drip filters: a randomised controlled trial in rural Bolivia. Water Sci & Tech 50(1):

11 Health Impact: Charinco,, Bolivia Diarrhoeal Prevalence and 95%CI Charinco, Bolivia Control Intervention Percentage Prevalence (7 day recall) Baseline Week 7 Week 13 Week 19 Week 25 Surveillance Point In a six-month randomized controlled trial by First Water and Fundación Sumaj Huasi in Charinco, Bolivia, diarrhoea prevalence among filter users was 64% less than the control group (p<0.001). The reduction was 72% among children under 5 (p<0.001) (Clasen T, Brown J, Suntura O, Collin S, Cairncross S (2004). Reducing diarrhoea through the use of householdbased ceramic water filters: a randomized, controlled field trial in rural Bolivia. Am J. Trop Med. & Hyg. 70(6):

12 Health Impact: Chiñiri iri,, Bolivia 14.0% 12.0% 13.2% Control 10.0% Intervention Diarrhoea Prevalence 8.0% 6.0% 6.7% 7.1% 7.5% 5.3% 4.7% 4.0% 2.0% 3.9% 4.1% 2.1% 3.2% 3.1% 1.5% 3.8% 2.9% 0.0% Baseline Round 1 Round 2 Round 3 Round 4 Round 5 Overall Surveillance Round In a five-month randomized controlled trial of a pilot intervention by Food for the Hungry, Int l in rural Bolivia, locally-fabricated water filters using Brazilian ceramic elements were associated with a 45.3% reduction in diarrhoea compared to the control group (p=0.02) Clasen T, Brown J, Collin S (2006) Preventing diarrhoea with household ceramic water filters: an assessment of a pilot project in Bolivia. Int l J. Environ. Health Res. (in press)

13 The encouraging results from studies of improved household water management provide a sufficient impetus for re- examining the potential health impact of interventions to improve drinking water quality.... This type of analysis should ultimately help refine the dominant paradigm, and lead to more focused guidance on the potential health impact of water quality interventions. Clasen T & Cairncross S (2004). Household water management: refining the dominant paradigm. Trop. Med. Int l Health 9(2):1-5

14 Protocol for a Cochrane Review Clasen T, Roberts I, Rabie T, Cairncross S (2004). Interventions to improve water qualtity for preventing diarrhoea. Cochrane Collaboration, 2004 Issue 2

15 Interventions at Source

16 Interventions at the Household

17 Post Collection Recontamination of Water Systematic review and meta- analysis of 57 studies measuring bacteria counts for source water and stored water in the home. Results: The bacteriological quality of drinking water significantly declined after collection in many settings. Conclusion: Policies that aim to improve water quality through source improvements may be compromised by post-collection contamination. Safer household water storage and treatment is recommended to prevent this, together with point-of of-use water quality monitoring. *Wright J, Gundry S, Conroy R (2004). Household drinking water in developing countries: a systematic review of microbiological contamination between source and point-of-use. Tropical Med. Int l Health 9(1):

18 Esrey Update: Fewtrell et al. Fewtrell L, Kaufmann R, Kay D, Enanoria W, Haller L, Colford J (2005). Water, sanitation, and hygiene interventions to reduce diarrhoea in developing countries: a systematic review and metaanalysis. Lancet Infect. Dis 5:

19 No. Studies in Review No. Studies Household Source Esrey 1985* Esrey 1991* Fewtrell 2005* Clasen 2005 Review *Includes observational studies

20 Effectiveness: Intervention Type (all age)

21 Effectiveness: Intervention Point (< 5s)

22 Summary of Effectiveness All ages Intervention Type (no. trials) Estimate (random) % (1-RR) 95% CI of Estimate Heterogeneity* (Chi-square) Source (6) % 0.53 to 1.01 p< Household (32) % 0.39 to 0.73 p< Filtration (6) % 0.28 to 0.49 p=0.56 Chlorination (16) % 0.52 to 0.75 p< Solar Disinf (2) % 0.63 to 0.74 p=0.73 Flocc/Disinf (7) % 0.20 to 1.16 p< Flocc/Disinf (ex Doocy) % 0.58 to 0.82 p=0.08 Impr.. Storage (1) % 0.61 to 1.03 n.a. *Note that in a test for heterogeneity, a low p-value (eg <0.10) suggests an actual underlying difference in effect between studies that is unlikely to be attributable to chance.

23 Summary of Effectiveness Under 5s Intervention Type (no. trials) Estimate (random) % (1-RR) 95% CI of Estimate Heterogeneity (Chi-square) Source (4) % 0.71 to 1.02 p=.007 Household (25) % 0.39 to 0.81 p< Filtration (5) % 0.24 to 0.53 p=0.37 Chlorination (12) % 0.67 to 0.86 p=0.004 Solar Disinfec (0) na na na na Flocc/Disinf (7) % 0.20 to 1.37 p< Flocc/Disinf (6) ( Doocy) (6) (ex ex % 0.61 to 0.84 p=0.10 Impr.. Storage (1) % 0.47 to 0.81 n.a.

24 Oxfam Pilot-- --Colombia Cartagenita Six-month (March- September 2004) randomized controlled trial of a pilot program using locally-fabricated ceramic water filter using Katadyn candles Dabeiba Following a baseline survey to confirm no differences between persons in intervention (n=415) and control (n=265) groups, filters were randomly allocated to half of the households (n=130) in each of three communities Follow-up at monthly intervals to assess water quality and diarrhoea prevalence

25 Mean Thermotolerant (Faecal) Coliform (and 95% CI) per 100ml Site Control Intervention P-value Curvarado 154 (124, 185) 50 (30, 71) < Dabeiba 220 (193, 247) 46 (27, 65) < Cartagenita 72 (61, 83) 0 < Overall 151 (135, 166) 37 (26, 48) < Filters reduced the level of faecal bacteria at all sites, with an overall reduction in arithmetic mean TTC/100ml of 75.3%. However, only in Cartagenita did the filters eliminate all TTC in product water.

26 Microbiological Performance 60.00% Oxfam-- --Colombia Control Intervention 54.51% 50.00% 47.66% Percentage of Samples 40.00% 30.00% 20.00% 24.22% 37.34% 17.58% 10.00% 7.30% 10.55% 0.00% 0.86% 0 TTC/100ml 1-10 TTC/100ml TTC/100ml 101/1000 TTC/100ml TTC Count 47.7% and 24.2% of the samples from the intervention group had no detectible TTC/100ml or conformed to WHO limits for low risk (1--10 TTC/100ml), respectively, compared to just 0.9% and 7.3% for control group samples.

27 Health Impact All ages Children < 5 years of age Site OR 95% CI P-value OR 95% CI P-value Curvarado , , Dabeida , , Cartagenita , 0.41 < , Overall , 0.63 < , Overall, prevalence of diarrhoea was 60% less among households using filters than control households (OR = 0.40, 95% CI = 0.25, 0.63, P < ). However, the microbiological performance and protective effect of the filters was not uniform throughout the study communities, and corresponded with microbial performance. Clasen T, Garcia Parra G, Boisson S, Collin S. Household-based ceramic water filters for the prevention of diarrhoea: a randomized, controlled trial of a pilot program in Colombia. Am J Trop Med Hyg. 73(4):790-95

28 Summary and Conclusions Interventions to improve water quality are effective in preventing diarrhoeal disease, a major killer of children under 5 years of age Interventions at the household level are about twice as effective as those at the source Effectiveness was conditioned upon sufficient water quantity, and increased with compliance and improved sanitation In a pilot program in Colombia by Oxfam GB, the filters were associated with a 75.3% reduction in the mean faecal bacteria in drinking water and a 60% reduction in diarrhoea Variation in results by setting suggests that actual effectiveness may depend on local conditions (exposure to other sources of transmission) and programmatic support to ensure compliance

29 Interventions to Improve Water Quality for the Prevention of Diarrhoeal Disease: A Cost-Effectiveness Analysis Thomas Clasen, Damian Walker, Laurence Haller & Sandy Cairncross

30 Interventions to Improve Water Quality for the Prevention of Diarrhoeal Disease: A Cost-Effectiveness Analysis Thomas Clasen, Damian Walker, Laurence Haller & Sandy Cairncross

31 Outline Description of water quality interventions in preventing diarrhoea source and household interventions Overview of methods Estimates of cost Estimates of effectiveness and DALYs averted CE results Limitations and Qualifications

32 Overview of methods WHO Generalized CEA Interventions: 3 source and 4 household approaches to improving microbial quality of drinking water Cost data: WHO data and generic cost protocol sent to 12 programme implementers Effectiveness data: Cochrane Review Population model: WHO PopMod Summary measure: $ per DALY averted for 11 WHO epidemiological sub-regions today, results for Afr-E E and Sear-D

33 WHO epidemiological sub-regions Botswana Burundi Central African Republic Congo Côte d'ivoire Democratic Republic of Congo Eritrea Ethiopia Kenya Lesotho Afr-E Malawi Mozambique Namibia Rwanda South Africa Swaziland Uganda United Republic of Tanzania Zambia Zimbabwe Sear-D Bangladesh Bhutan Democratic People s Republic of Korea India Maldives Myanmar Nepal

34 Estimating Costs US$ per person covered per year estimated For each intervention, a mean and range of costs were computed Assumed linear cost of scaling up to 100% coverage (minimal fixed costs)

35 Sources for Cost Data Source (regional mean cost of communal tapstand, borehole and dug well based on WHO data) Household chlorination (estimates from 17 country programmes) filtration (estimates from Cambodia and Bolivia programmes) solar disinfection (estimates from 7 country programmes) flocculation / disinfection (estimates from 5 country programmes)

36 Summary of Cost Estimates $6.00 Mean annual cost per person in US$ of source and household Interventions (error bars represent range of costs) $5.00 $4.95 $4.00 $3.60 Cost $3.00 $2.61 $3.03 $2.00 $1.88 $1.00 $0.66 $0.63 $- Source-Africa Source-Asia Source-LA&C Chlorination Ceramic Filtration Solar Disinfection Flocculation- Disinfection System

37 DALYs Averted Yearly DALYs averted from implementation of water quality interventions to prevent diarrhoea at 100% coverage level 120 Yearly DALYs Averted (millions) Chlorination Ceramic Filtration Solar Disinfection Flocculation Disinfection Source 20 0 Afr-E WHO Epidemiological Sub Region Sear-D

38 CE plane for Afr-E

39 CE plane for Sear-D

40 Interpreting ICERs Assumes all the household interventions are mutually exclusive, i.e. if one is chosen another cannot be Flocculation / disinfection was (strongly) dominated by the three other household interventions (solar disinfection, chlorination and filtration) in both regions, i.e. they were all more effective and less costly Solar disinfection was (weakly) dominated by chlorination in both regions, but this assumes it is possible to deliver chlorination to less people without any change in CE

41 Interpreting CE data Following the recommendation of the CMH, WHO uses the following categories: $ per DALY averted < GNI per capita = highly CE $ per DALY averted * GNI per capita = CE $ per DALY averted > 3* GNI per capita = not CE Afr-E E GNI per capita: $369 Sear-D D GNI per capita: $276 Data suggest interventions to improve water quality are: highly CE in Afr-E E and Sear-D

42 Acknowledgements CDC (S. Luby,, R. Quick J. Crump, T. Chiller, E. Mintz) Proctor & Gamble (G. Allgood,, B. Keswick) WHO (J. Bartram, B. Gordon, L. Haller) UC Berkeley (J. Colford) University of Wales (L. Fewtrell) Colleagues at LSHTM (V. Curtis, I. Roberts, T. Rabie,, L. Smith, W. Schmidt, S. Thomas)

43 Acknowledgements Oxfam GB Colombia (G. Garcia Parra,, F. Vidal, LM Londonao,, T. Forster, A Bastable) CDC (S. Luby,, R. Quick J. Crump, T. Chiller, E. Mintz) Proctor & Gamble (G. Allgood,, B. Keswick) WHO (J. Bartram,, B. Gordon, L. Haller) Johns Hopkins University (S. Doocy) University of Bristol (S. Gundry, J. Wright) UC Berkeley (J. Colford) University of Wales (L. Fewtrell) University of North Carolina (M. Sobsey,, J. Brown) Colleagues at LSHTM (S. Cairncross,, V. Curtis, I. Roberts, T. Rabie,, L. Smith, W. Schmidt, S. Thomas)