Use of a synthetic population to model co-benefits to air quality and health from household fuel emission mitigation policies in Kenya

Energy emissions mitigation policies bring co-benefits for health and opportunities to drive sustainable development for rapidly transitioning economies in sub-Saharan Africa. Developing methods of quantifying these co-benefits in differing demographic groups is an area of interest for policymakers to support resource allocation efforts. Using synthetic populations of three municipalities in Kenya, we assessed the impact of policies to promote the use of clean cooking fuels on exposure to ambient and household air pollution and associated age- and gender-specific mortality. Exposure to household PM2.5 for a range of cooking fuel types and informal and formal housing archetypes were simulated using the building physics software, EnergyPlus. A combined household and ambient PM2.5 exposure was calculated for each individual by weighting PM2.5 concentrations using national demographic-specific time-activity estimates. Exposure-response functions were applied to quantify the burden of mortality for six associated health outcomes. To compare the health impacts of energy policy implementation, a two-stage policy was tested through medium and long-term transitions towards successively cleaner cooking fuels prioritising liquid petroleum gas and ethanol. The resulting difference in mortality consecutively declined through the two-stage policy transition with the greatest impact after the first transition and an incremental but smaller impact after the second. The overall difference in mortality burden averted per 100 000 population relative to the baseline scenario was largest in Kisumu (males: 39.23; females: 18.09), with smaller decreases in Mombasa (males: 5.71; females: 3.03) and Nairobi (males: 1.82; females: 1.08). A sensitivity analysis showed reductions in PM2.5 exposure under the policy scenarios may be overestimated in the presence of fuel stacking practices, where households rely on multiple fuels and stoves. This model provides a proof-of-concept for the use of individual-level modelling methods to estimate demographic-specific health impacts from environmental exposures and quantitatively compare health co-benefits of household fuel emission mitigation policies.