Subscribe to E-Bulletin Donate to CEGA

Conditional Cash Transfers for the Prevention of HIV and Sexually Transmitted Infections in Tanzania

Development Challenge

Around 68 percent of all people living with HIV reside in Sub-Saharan Africa.Because only two of every five people infected with HIV globally receive treatment, effective solutions are needed to curtail the spread of the epidemic.2 Many preventative interventions, such as public health information campaigns, have been unsuccessful in altering sexual behaviors that place individuals at the most risk of contracting HIV. Conditional cash transfers (CCTs) have been successful in influencing behavior, as tying financial incentives to socially desirable outcomes can increase use of program services and initiate behavioral change.  Recent research in Malawi showed that small financial incentives could increase the level of HIV testing and condom use.3 A related study in Malawi also tested the effects of rewards for maintaining HIV negative status for one year, but did not find significant effects of incentives on HIV conversion.4 The present study tested a more intensive intervention that regularly rewards individuals for avoiding a set of sexually transmitted infections (STIs), with the ultimate aim of reducing risky sexual behaviors so as to lower HIV incidence.

Context

The research took place in Tanzania, where the HIV prevalence rate for men and women aged 15 to 49 declined from 7.1 percent in 2001 to 5.6 percent in 2009.5  Women are more likely to be infected with HIV than men, and individuals in rural areas are more vulnerable to infection than those in urban areas. While 85 percent of health care facilities in Tanzania provide some support for people with HIV, few provide testing or other prevention services.6

Evaluation Strategy

This study evaluates the effectiveness of conditional cash transfers (CCTs) as a tool for the prevention of HIV and other sexually transmitted infections (STIs). The study was a randomized evaluation involving 2,399 men and women, aged 18-30, in the Kilombero and Ulanga districts of southwestern Tanzania. Participants were tested every four months over a 12-month period for four common STIs7, a proxy for risky sexual behavior. Researchers also tested for HIV, syphilis, and herpes simplex virus 2 at the baseline and at the 12-month mark.

Within the treatment group, 660 participants were assigned to a “low-value” CCT and were eligible to receive $10 after each round of testing, conditional on negative STI results.  The other 615 participants in the treatment group were assigned to a “high-value” CCT and received $20 after each testing round, conditional on negative results. The cash transfers offered a strong incentive to participate in the study, as the transfers represented a substantial proportion of the mean annual income of $250 for study participants.  All participants also received pre and post-test counseling and were enrolled in monthly group counseling sessions. Those who tested positive were offered counseling and free STI treatment for themselves and their sexual partners. Pregnant women, those who intended to migrate from the region, and those who refused to participate if assigned to the comparison group, were ineligible to participate. The 1,124 individuals assigned to the comparison group (approximately 50 percent of the total sample) did not receive a cash transfer but were eligible for all other components of the intervention, including counseling and free treatment if they tested positive.

Results and Policy Implications

Within the first two rounds of testing (months four and eight), there were no significant decreases in STI prevalence for either treatment group. However, after 12 months, individuals in the high-value CCT group had an adjusted 27 percent lower STI prevalence rate than the comparison group. Further, the high-value CCT group also experienced an STI reduction compared with the low-value CCT group. The low-value group did not demonstrate any significant decrease in STI prevalence. However, the absence of a significant effect in the low-value CCT group should be interpreted with caution, the researchers warn, as all participants were aware of the different treatment assignments and may have adjusted their behavior in response.  Finally, there was no significant difference in the impact of CCTs on STI prevalence for men versus women.

The high social and economic costs of HIV/AIDS treatment compel experts to seek effective preventative measures as a more cost-effective approach. The results indicate that cash transfers which are sufficiently large could be an effective tool for altering risky sexual behavior and preventing STIs. 

Timeline

2009-Ongoing

1UNAIDS Global Report (2010).
2UNAIDS. AIDS Epidemic Update: November 2009. Geneva, Switzerland: UNAIDS, 2009.
3Thornton, R. The Demand for and Impact of Learning HIV Status. Am Econ Rev 2008; 98:1829-63.
4Kohler, HP; Thornton, R. Conditional Cash Transfers and HIV/AIDS Prevention: Unconditionally Promising? World Bank Econ Rev 2011 doi:10.1093/wber/lhr041
5UNAIDS Global Report (2010).
6Tanzania Service Provision Assessment Survey (2006).
7Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis, and Mycoplasma genitalium.