Assessing households' values of indoor residual spraying programs to reduce malaria: A discrete choice experiment in northern Uganda

An important method for preventing malaria is the indoor residual spraying (IRS) of insecticides to kill the mosquitoes which transmit the pathogen to humans. Previous social science research of IRS programs has found high levels of acceptance among targeted households, but has simultaneously documented a number of concerns among target households. These concerns include a poor understanding of the purpose of IRS programs, damaged property, a disruption of household activities, and health-related side effects of insecticide exposure. On the other hand, the fact that one household's participation decision can yield significant benefits to neighboring households—namely through reductions in malaria exposure—illustrates a positive externality arising from IRS participation.

We report results from a discrete choice experiment (DCE) that evaluated households' preferences for IRS programs in northern Uganda, a region with some of the highest levels of malaria exposure in the world. The DCE framework provides a rigorous methodology for quantifying the perceived tradeoffs households face when deciding whether or not to participate in an IRS program. It can also provide policymakers with estimates of how tastes for different IRS programs attributes, such as the frequency with which households are visited by spray teams, vary within a population. Here we estimate households' willingness to participate in IRS in lieu of an increase in their liquid assets, and compare our findings to previously published estimates. Based on the analysis of the choice data using a mixed logit econometric model, we find that respondents placed significant value on malaria risk reduction in the DCE: The average respondent would be willing to forego a $10.79 (95% C.I. $5.75 – $21.85) increase in her assets for a permanent 1% reduction in malaria risk. Participation in previous IRS significantly increased the likelihood of participation in the experiment. However, we also find that over 28% of respondents would pay an implicit fee—through perceived nonmonetary costs—if they participated in future IRS. We could not isolate the sources of these costs in our experiment—e.g., through disruption in household activities, side effects from insecticide exposure, etc. Based on this finding and the positive externality described above, we conclude that a policy of subsidizing IRS participation could improve the efficiency of these programs. Further research is warranted to isolate and quantify relative magnitudes of the perceived nonmonetary costs of IRS programs, and to experiment with subsidies for malaria prevention.