Summary objectives To estimate the incremental cost-effectiveness of a large-scale urban hygiene promotion programme in terms of reducing the incidence of childhood diarrhoeal disease in Bobo-Dioulasso, Burkina Faso. methods Total and incremental costs of the
programme were estimated retrospectively from the perspectives of the provider, from the households who change their behaviour as a result of the programme and from society (the sum of the two). The programme effects were derived from an intervention study that estimated the impact on handwashing
with soap after handling child stools through a time-series method of observing 37 319 mothers. Using data from the literature, the associated reductions in childhood morbidity and mortality were estimated. The direct medical savings and indirect savings of caregiver time and lost productivity
associated with child death were estimated from interviews with households and health workers. The cost and outcome data were combined to provide an estimate of the cost per mother who starts handwashing with soap as a result of the programme and the cost per case of childhood diarrhoea averted.
results The total provider cost (including start-up and 3-year running costs) was $302 507. Core programme activities accounted for 31% of the cost, administration 40%. The total cost to the 7286 households associated with changing behaviour during the 3 years
of programme implementation was $160 125 ($7.3 per year per household). An estimated 8638 cases of diarrhoea, 864 outpatient consultations, 324 hospital referrals and 105 deaths were averted by the programme during this time. Savings to the provider from reduced treatment
costs were estimated at $10 716 and savings to the households from averted treatment cost were $9136, resulting in a total saving to society of $19 852, increasing to $393 967 if indirect savings are included. The incremental provider cost per
case of diarrhoea averted was $33.8. The incremental cost to society was $51.3 falling to $7.9 if indirect savings are included. If the programme were to be replicated elsewhere, savings in the international research input and start-up costs could reduce provider costs
to $26.9 per case of diarrhoea averted. The annual cost of the programme represents 0.001% of the national health budget for Burkina Faso. The direct annual cost of implementing the programme at the household level represents 1.3% of annual household income. conclusion Hygiene
promotion reduces the occurrence of childhood diarrhoea in Burkina Faso at less than 1% of the Ministry of Health budget and less than 2% of the household budget, and could be widely replicated at lower cost.
Health Policy Unit, London School of Hygiene and Tropical Medicine, London, UK 2:
Ministère de la Santé, Ouagadougou, Burkina Faso 3:
Disease Control and Vector Biology Unit, London School of Hygiene and Tropical Medicine, London, UK