Financial Incentives and Deposit Contracts to Promote HIV Retesting in Uganda: a randomized trial

Background: Frequent retesting for HIV among persons at increased risk of HIV infection is critical to early HIV diagnosis of persons and delivery of combination HIV prevention services. There are few evidence-based interventions for promoting frequent retesting for HIV. We sought to determine the effectiveness of financial incentives and deposit contracts in promoting quarterly HIV retesting among adults at increased risk of HIV. Methods and Findings: In peri-urban Ugandan communities from October-December 2018, we randomized HIV-negative adults with self-reported risk to one of three strategies to promote HIV retesting: (1) no incentive; (2) cash incentives (US$7) for retesting at 3 and 6 months (total $14); or (3) deposit contracts: participants could voluntarily deposit $6 at baseline and at 3 months that would be returned with interest (total US$7) upon retesting at 3 and 6 months (total $14) or lost if participants failed to retest. The primary outcome was retesting for HIV at both 3 and 6 months. Of 1,482 persons screened for study eligibility following community-based recruitment, 524 participants were randomized to either no incentive (N=180), incentives (N=172), or deposit contracts (N=172): median age was 25 years (IQR: 22-30), 44% were women, and median weekly income was US$13.60 (IQR: $8.16-21.76). Among participants randomized to deposit contracts, 24/172 (14%) made a baseline deposit, and 2/172 (1%) made a 3-month deposit. In intent-to-treat analyses, HIV retesting at both 3 and 6 months was significantly higher in the incentive arm (89/172 [52%]) than either the control arm (33/180 [18%], odds ratio, OR 4.8, 95% CI: 3.0-7.7, p<0.001) or the deposit contract arm (28/172 [16%], OR 5.5, 95% CI: 3.3-9.1, p<0.001). Among those in the deposit contract arm who made a baseline deposit, 20/24 (83%) retested at 3 months; 11/24 (46%) retested at both 3 and 6 months. Study limitations include measurement of retesting at the clinic where baseline enrollment occurred, only offering clinic-based (rather than community-based) HIV retesting and lack of measurement of retesting after completion of the trial to evaluate sustained retesting behavior. Conclusions: Offering financial incentives to high-risk adults in Uganda resulted in significantly higher HIV retesting. Deposit contracts had low uptake and overall did not increase retesting. As part of efforts to increase early diagnosis of HIV among high-risk populations, strategic use of incentives to promote retesting should receive greater consideration by HIV programs.