Innovative Behavioral Economics Incentives Strategies for Health (IBIS-Health)

The success of combination HIV prevention efforts, including HIV treatment as prevention, hinges on universal, routine HIV testing with effective treatment after HIV diagnosis. The proposed study will evaluate the comparative effectiveness and sustainability of innovative incentive strategies, informed directly by behavioral economics and decision psychology, to promote HIV testing among men and HIV treatment among HIV-infected adults in rural Uganda.

[INTRODUCTION] The success of combination HIV prevention efforts, including HIV treatment as prevention, hinges on universal, routine HIV testing with linkage to care and antiretroviral treatment initiation after HIV diagnosis. The proposed study will evaluate the comparative effectiveness and sustainability of innovative incentive strategies, informed directly by behavioral economics and decision psychology, to promote HIV testing among men and HIV and treatment among HIV-infected adults in rural Uganda. [OBJECTIVES] AIM 1: Adult men living in the study communities in rural Uganda (N=3,000) will be randomized to one of three (fixed, loss aversion, and lottery) incentive approaches and different incentive amounts that encourage HIV testing. The hypothesis is that lottery and loss aversion incentives will result in significantly higher testing uptake than fixed incentives. The investigators also hypothesize that the proportion of testers in each arm who are HIV-infected (secondary outcome) will be highest with lottery-based incentives. In sub-samples of men who do and do not test, the investigators will conduct in-depth interviews to assess perceptions, attitudes and preferences related to incentives that may affect how incentives influence testing. AIM 2: Adult men and women living in the study communities (N=400) who obtained an HIV-positive result at a community health campaign will be randomized into one of two incentive approaches that encourage HIV treatment adherence. The investigators hypothesize that a financial incentive will be more effective than no incentive in promoting HIV virologic suppression (a measure of success in ART adherence and navigation of the HIV treatment cascade) as incentives capitalize on present bias by drawing attention to a salient, immediate benefit of initiating and/or maintaining treatment, and leverage loss aversion by generating implicit loss as a result of delaying the decision to initiate ART. AIM 3 - Pilot: In order to assess the feasibility of leveraging loss aversion to increase repeat HIV testing, HIV-negative adults who are at high risk of HIV acquisition and have just tested for HIV will be randomized into one of several different incentive strategies that encourage repeat HIV testing. The incentive arms will either: a) leverage loss-aversion by requesting participants to make an initial voluntary deposit that they will lose if they do not test for HIV at a later date; or b) use a standard gain-framed incentive strategy, in which participants are told they will receive an incentive for testing again for HIV at a later date. We will compare these two types of incentive strategies to a no incentive arm as well. Results from this pilot study will also be used to inform how best to implement loss aversion-based incentives in a larger trial, and provide preliminary data to guide sample size estimates for a larger trial comparing loss aversion vs. gain-framed incentive-based strategies vs. no incentive, on the outcome of repeat HIV testing. We hypothesize that loss aversion incentives will be feasible (i.e. ≥50% of eligible adults will be willing to participate), and will result in significantly higher testing uptake than either gain-framed incentives or no incentives. Aim 3 - Trial. Assess the comparative effectiveness of deposit contracts (a form of incentives that leverages loss aversion) vs. gain-framed incentives, compared to no incentives (control), to promote repeat HIV testing among high-risk HIV-uninfected adults. In our Aim 3 pilot trial, we assessed the feasibility and acceptability of deposit contracts: a loss aversion-based strategy to incentivize retesting for HIV. As deposit contracts were found to be highly acceptable and feasible in our Aim 3 pilot in August-December 2017 (>90% of participants in the deposit contract group made deposits into the study contracts), we will now proceed with a larger trial of sufficient sample size to compare the effectiveness of loss aversion and gain-framed incentive approaches vs. no incentives, on the outcome of repeat HIV testing. We hypothesize that deposit contracts (loss aversion-based incentives) will result in significantly higher HIV retesting uptake 3- and 6-months after enrollment than either gain-framed incentives or no incentives.

The incentives will increase in value when participants are found to meet the pre-specified virologic suppression criteria at 6 weeks, 3 months and 6 months. If the virologic threshold for an incentive is missed at the 6 week or 3 month time-point, the subsequent incentive for an undetectable HIV viral load will be reset to the initial incentive value.

Men are informed that if they come for HIV testing, they will receive a specific prize (gain framing). The prize will be an item worth the same amount in US dollars as loss aversion and the expected value of a lottery prize. This gain-framed incentive resembles the form that incentives usually take in most studies and serves as a comparison to the lottery-based and loss-framed incentives.

The prizes are worth approximately the same amount as the fixed incentive and expected value of a lottery prize. At the time of randomization, study staff will inform the participant that he has won a prize. Staff will ask the participant to choose a specific prize from several choices, and then provide an opportunity for the participant to see the prize. Study staff will then tell participants that they will lose the prize if they do not participate in HIV testing. In this way, the incentive is framed as a loss rather than a gain, thereby leveraging loss aversion while not requiring men in very low-income settings to experience an actual loss.

Men are entered in a lottery that offers a chance to win high-value prizes after testing for HIV at a community health campaign. Staff will emphasize that only those who come for HIV testing will be entered into the lottery and that not everyone will win a prize. Participants were informed at enrollment about the list of prizes and corresponding probabilities of winning them, in terms that are understandable to men with low numeracy (e.g., "1 in 20" rather than 5%). The probabilities of winning prizes varied between 1-5%, with higher value prizes having lower probability.

A standard gain-framed incentive arm in which participants will be offered a voucher for coming for a repeat HIV test in the future.

A loss-aversion framed incentive in which participants will be asked to voluntarily make a deposit that can be retrieved, with interest on the deposit, if they come for an HIV test in the future (i.e. for repeat testing)

Proportion of participants who receive an HIV test at a community health campaign between intervention groups

AIM 1 - TESTING TRIAL Inclusion Criteria: Male ≥18 years Resident (≥6 months) in one of 4 study communities Exclusion Criteria: Plan to move <6 months from study start AIM 2 - TREATMENT TRIAL Inclusion Criteria: ≥18 years Resident (≥6 months) in one of 4 study communities HIV positive Exclusion Criteria: Plan to move <6 months from study start AIM 3 - REPEAT TESTING PILOT Inclusion Criteria: HIV-negative by rapid HIV antibody testing at pilot trial baseline, Ages 18 - 59 years old, Attendee of high-risk site of HIV acquisition (e.g. bars, trading centers, etc.) in the region Exclusion Criteria: Intention to move away from the community in the 3 months from time of recruitment AIM 3 - REPEAT TESTING TRIAL Inclusion Criteria: HIV-negative by rapid HIV antibody testing at time of recruitment, Ages 18 - 59 years old, Reported willingness to retest for HIV in the six months following recruitment, Sexual risk behavior, defined as at least one of the following self-reported risks in the 12 months prior to recruitment: >1 sexual partner, or known HIV-infected sexual partner, or sexually transmitted infection, or paid or received compensation or gifts for sex. Exclusion Criteria: Intention to move away from the community for >=4 consecutive months during the six months following recruitment A history of testing for HIV >=3 times in the 12 months prior to recruitment

Chamie G, Schaffer EM, Ndyabakira A, Emperador DM, Kwarisiima D, Camlin CS, Havlir DV, Kahn JG, Kamya MR, Thirumurthy H. Comparative effectiveness of novel nonmonetary incentives to promote HIV testing. AIDS. 2018 Jul 17;32(11):1443-1451. doi: 10.1097/QAD.0000000000001833.

Thirumurthy H, Ndyabakira A, Marson K, Emperador D, Kamya M, Havlir D, Kwarisiima D, Chamie G. Financial incentives for achieving and maintaining viral suppression among HIV-positive adults in Uganda: a randomised controlled trial. Lancet HIV. 2019 Mar;6(3):e155-e163. doi: 10.1016/S2352-3018(18)30330-8. Epub 2019 Jan 16.

Ndyabakira A, Chamie G, Emperador D, Marson K, Kamya MR, Havlir DV, Kwarisiima D, Thirumurthy H. Men's Beliefs About the Likelihood of Serodiscordance in Couples with an HIV-Positive Partner: Survey Evidence from Rural Uganda. AIDS Behav. 2020 Mar;24(3):967-974. doi: 10.1007/s10461-019-02531-7.

Ndyabakira A, Getahun M, Byamukama A, Emperador D, Kabageni S, Marson K, Kwarisiima D, Chamie G, Thirumurthy H, Havlir D, Kamya MR, Camlin CS. Leveraging incentives to increase HIV testing uptake among men: qualitative insights from rural Uganda. BMC Public Health. 2019 Dec 30;19(1):1763. doi: 10.1186/s12889-019-8073-6.

Schaffer EM, Gonzalez JM, Wheeler SB, Kwarisiima D, Chamie G, Thirumurthy H. Promoting HIV Testing by Men: A Discrete Choice Experiment to Elicit Preferences and Predict Uptake of Community-based Testing in Uganda. Appl Health Econ Health Policy. 2020 Jun;18(3):413-432. doi: 10.1007/s40258-019-00549-5.

Chamie G, Ndyabakira A, Marson KG, Emperador DM, Kamya MR, Havlir DV, Kwarisiima D, Thirumurthy H. A pilot randomized trial of incentive strategies to promote HIV retesting in rural Uganda. PLoS One. 2020 May 29;15(5):e0233600. doi: 10.1371/journal.pone.0233600. eCollection 2020.

Kavanagh NM, Schaffer EM, Ndyabakira A, Marson K, Havlir DV, Kamya MR, Kwarisiima D, Chamie G, Thirumurthy H. Planning prompts to promote uptake of HIV services among men: a randomised trial in rural Uganda. BMJ Glob Health. 2020 Nov;5(11):e003390. doi: 10.1136/bmjgh-2020-003390.

Marson K, Ndyabakira A, Kwarisiima D, Camlin CS, Kamya MR, Havlir D, Thirumurthy H, Chamie G. HIV retesting and risk behaviors among high-risk, HIV-uninfected adults in Uganda. AIDS Care. 2021 May;33(5):675-681. doi: 10.1080/09540121.2020.1842319. Epub 2020 Nov 10.

Chamie G, Napierala S, Agot K, Thirumurthy H. HIV testing approaches to reach the first UNAIDS 95% target in sub-Saharan Africa. Lancet HIV. 2021 Apr;8(4):e225-e236. doi: 10.1016/S2352-3018(21)00023-0.

Chamie G, Kwarisiima D, Ndyabakira A, Marson K, Camlin CS, Havlir DV, Kamya MR, Thirumurthy H. Financial incentives and deposit contracts to promote HIV retesting in Uganda: A randomized trial. PLoS Med. 2021 May 4;18(5):e1003630. doi: 10.1371/journal.pmed.1003630. eCollection 2021 May.

Camlin CS, Marson K, Ndyabakira A, Getahun M, Emperador D, Byamukama A, Kwarisiima D, Thirumurthy H, Chamie G. Understanding the role of incentives for achieving and sustaining viral suppression: A qualitative sub-study of a financial incentives trial in Uganda. PLoS One. 2022 Jun 24;17(6):e0270180. doi: 10.1371/journal.pone.0270180. eCollection 2022.