Association for Public Policy Analysis & Management

This study in the Free State Province of South Africa randomized 650 AIDS patients receiving anti-retroviral treatment (ART) into a Control group, a treatment group receiving adherence support, and a treatment group receiving adherence support and a nutritious food supplement. Time allocation patterns reflecting labor productivity over 18 months were analyzed tackling the methodological aspects. While changes between survey rounds 1 and 3 were significant for several variables, differences in changes between control and treatment groups were generally insignificant. Second, estimated parameters from dynamic random effects models for food insecurity levels indicated that higher disability grants were likely to reduce food insecurity. Third, tackling the coexistence of under-nutrition and obesity among patients, Body Mass Index was a significant predictor of time allocation patterns. Finally, CD4 cell counts were significant predictors of time spent on productive activities. The policy implications of the findings for food and health policies for sub-Saharan Africa are discussed.