Good adherence to HIV medication is crucial to extend the length and quality of life and to eliminate transmission of HIV to others. However, in sub-Saharan Africa, only 29% of people living with HIV (PLHIV) have viral suppression, the ultimate goal of treatment. One of the reasons for this gap is food insecurity, which is a pervasive barrier to care and treatment among PLHIV. Food insecurity increases the likelihood of hunger and worsens the side effects associated with HIV medication. It also increases stress and forces PLHIV to make choices between food and other goods including HIV care and treatment.

Food insecure PLHIV who were over 18 years of age and newly starting HIV treatment in Shinyanga, Tanzania were  randomly assigned to one of three groups:

• Nutrition assessment and counseling (NAC, a standard service provided to PLHIV),
• NAC supplemented with monthly food baskets, and
• NAC supplemented with monthly cash transfers.

Participants assigned to the food or cash transfer groups received monthly food baskets or a cash transfer approximately equivalent to USD 11 once a month for up to six months, conditional on attendance at scheduled HIV appointments. Monetary incentives were transferred through mobile money services or in person at the clinic if individuals did not have access to a mobile phone. Food baskets were equivalently valued to the cash transfer and were intended to supplement but not replace the household food supply. Baskets were distributed at a shop near the health clinic on two days per month and contained local foods: maize meal (12kg), groundnuts (3kg) and beans (3kg).

This randomized evaluation of 800 PLHIV found that individuals receiving food or cash were more likely to be in care at 6 months and were more likely to have high levels of treatment adherence (defined as having HIV medication available 95% of the time) compared to individuals who only received standard services. For example, high HIV treatment adherence at 6 months was 21.6 percentage points higher in the cash group compared to the standard services group. The benefits of cash and food assistance were similar at 6 months.

At 12 months, 6 months after the intervention period was complete, cash continued to be significantly associated with better adherence and retention in care. For example, after 12 months of follow-up, retention in care remained approximately 10 percentage points higher among cash recipients than individuals who only received standard services.

In summary, the study found that short-term conditional cash and food assistance improves HIV treatment adherence and reduces disengagement from care among food-insecure people living with HIV in Tanzania. These findings suggest that short-term cash and food assistance are both valuable tools to maximize the effectiveness of HIV treatment programs and could enhance strategies to end the HIV epidemic.

Inspired to implement this design in your own work? Here are some things to think about before you get started:

• Are the behavioral drivers to the problem you are trying to solve similar to the ones described in the challenge section of this project?
• Is it feasible to adapt the design to address your problem?
• Could there be structural barriers at play that might keep the design from having the desired effect?
• Finally, we encourage you to make sure you monitor, test and take steps to iterate on designs often when either adapting them to a new context or scaling up to make sure they’re effective.

Additionally, consider the following insights from the design’s researcher:

Although cash assistance was superior to food assistance for several health outcomes measures in the study, there were additional reasons why cash may be a better choice in future studies:

• When asked about their preferences, the majority of participants would have preferred to be in the cash study group. Specifically, if given the choice, 83% of cash and 44% of food recipients would have selected cash over food assistance.
• On average, cash recipients reported spending less time and money retrieving their incentives than food recipients, as food recipients had to pick up the basket and transport it home.
• Cash transfers were cheaper to implement: the direct intervention cost for the cash transfer was USD 178 per beneficiary compared to USD 198 per beneficiary for the food transfer. This included start-up and implementation costs, but excluded research-related costs.
• Cash transfers were also easier to implement, as the study used mobile money services to distribute the cash. Food baskets incurred additional costs, including storage, packaging, and security and they were difficult to distribute in a health facility setting.
• Participants did not report any harmful events associated with the incentives, but reported myriad beneficial effects on household welfare including supplementing the household food supply, paying for school fees, and investing in businesses.

Thus, in settings where a cash transfer is appropriate to alleviate food insecurity among PLHIV, it may be preferred over a food basket. One of the key issues to determine in future studies of cash transfers is the appropriate cash transfer size. In this study, the cash transfer amount was selected to prevent undue coercion to study participants and to be consistent with the Tanzania Social Action Fund (TASAF), a government-run anti-poverty program which targets “orphans, disabled, elderly, widows/widowers, and those infected or affected by HIV/AIDS,” among other vulnerable groups. TASAF provides USD 6-18 monthly to households depending on the presence and number of vulnerable children and elderly members. The transfer for this study, at USD 11, is within this range, ensuring that the amount is policy-relevant. Future studies might also try and align their cash transfers with ongoing social protection programs that include a cash transfer component.