Effect of a Cash Transfer Programme for Schooling on Prevalence of HIV and Herpes Simplex Type 2 in Malawi: A Cluster Randomized Trial

Replication paper
Original publication: Lancet
Original researchers: Sarah Baird, Richard S. Garfein, Craig McIntosh, Berk Ozler
Replication plan:

Smith's replication plan

Replication researchers: Lynette M. Smith

The Original Study

The study examines the impact of a cash transfer programme on the prevalence of human immunodeficiency virus (HIV) and herpes simplex virus type 2 (HSV-2). The study also assesses the programme’s impacts on syphilis prevalence, school enrolment, self-reported marriage, pregnancy and sexual behaviour. The authors designed and implemented an intervention in 176 enumeration areas in Zomba district in Malawi between January 2008 and December 2009.The study population consisted of all never-married girls aged 13 to 22 years. After completion of the baseline surveys, 88 enumeration areas were randomly assigned to the intervention group and the other 88 to the control group. In the intervention group, conditional and unconditional cash transfers were given to guardians of the girls and directly to the girls. In the conditional group, cash transfers were given after the girls had attended school regularly during the previous month. Data were collected from 3,796 individuals. The study found that both the conditional and unconditional cash transfer interventions significantly reduced the prevalence of HIV and HSV-2. 18 months after the baseline surveys, in the cohort of baseline school girls, weighted HIV prevalence was found to be 1.2 per cent in the intervention group and 3.0 per cent in the control group. Weighted HSV-2 prevalence was 0.7 per cent in the intervention group and 3.0 per cent in the control group. The study found that there is no significant difference between conditional and unconditional intervention groups, regarding HIV and HSV-2 prevalence. Overall, this study shows that cash transfers to unmarried school girls can help reduce risky sexual activities and the likelihood that young women will be infected with HIV and HSV-2. Further studies are needed to examine whether this intervention works in other settings.

The Replication

The first objective of this replication research is to conduct a pure replication of the study. That is, establish whether the published findings can be reproduced using the study’s own data and methods. The second objective is to investigate the robustness of the findings through additional analysis methodology including hierarchical models, permutation tests and bivariate analysis. Finally, we will extend the study by (1) directly evaluating the effect of the intervention on improving HIV awareness (having a HIV test, or gaining of HIV knowledge); (2) composing a wealth index of the participants at baseline and evaluating whether the wealth index will influence the effect of the intervention on HIV and HVS-2 prevalence; and (3) modeling the causal pathway implied by the study by explicitly investigating how much of the effect of the intervention on HIV and HVS-2 is mediated through reduced sexual behavior and enrollment in school.