A widely implemented HIV/AIDS programme inadvertently decreased HIV testing by worsening misinformation and HIV-related stigmatising attitudes, but a simple follow-up intervention offset the programme’s adverse effects.
Nyah Phillips lent us her superb writing talents by drafting this article. We greatly appreciate her support.
Implementing widespread HIV testing in areas of high prevalence is crucial to combatting the HIV/AIDS pandemic. People with HIV infection are often asymptomatic for several years. Thus, HIV testing is important for early detection and the subsequent initiation of antiretroviral therapy (ART), which leads to lower plasma viral loads and improves health outcomes for the infected individual (Ford et al. 2018). Rapid initiation of ART also lowers the risk of transmitting HIV to sexual partners (Rodger et al. 2019). Despite the significant progress made in expanding HIV testing around the world, we still remain far from achieving testing targets set by public health officials.
Prior research has shown that the provision of general HIV/AIDS information can affect knowledge, health behaviours, and demand for health goods (Duflo et al. 2015, Dupas 2011, Godlonton et al. 2016, Ciancio et al. 2020, Kim et al. 2017, Chong et al. 2013, Banerjee et al. 2020). Other studies have found that interventions can reduce HIV-related stigmatising attitudes (Hoffman et al. 2016, Lubega et al. 2019). This suggests that HIV/AIDS programmes aimed at disseminating public health information and addressing concerns about disease-related stigma can play a central role in increasing HIV testing.
New evidence from Mozambique
In our research (Yang et al. 2023), we study Força a Comunidade e Crianças (FCC, “Strengthening Communities and Children”), a PEPFAR-funded programme that aims to raise HIV testing rates by improving knowledge about HIV/AIDS and reducing HIV-related stigmatising attitudes. We then test whether the FCC programme is complementary with or substitutable for more targeted interventions (“minitreatments”) that focus directly on information and stigma.
We do this by implementing a randomised controlled trial with 3,700 households across 76 communities in Mozambique. First, we randomised half of the communities to treatment (receiving the FCC programme) and half to the control group. Then, we randomised a subset of households in each treated community to a strong encouragement to participate in the programme (“FCC-enrolled” households). These households received home visits by FCC community workers and were assessed for inclusion in various components of the programme. Lastly, following the conclusion of the endline survey, households were randomly assigned to one of five minitreatments aimed at further encouraging HIV testing or a minitreatment control group:
- Anti-stigma: The respondent is provided with individual-specific information aimed at reducing concerns about HIV-related stigma in the community. Those who overestimated the fraction of residents in their community with stigmatising attitudes toward people with HIV were told the true (lower) value collected from the baseline survey.
- HIV/AIDS Information: The respondent watches a short video providing factual information about HIV/AIDS.
- Antiretroviral Therapy (ART) Information: The respondent watches a short video providing factual information about ART.
- Both HIV/AIDS and ART Information: The respondent receives both treatments listed in #2 and #3. They are shown both the HIV/AIDS and ART Information videos.
- High Incentive for HIV Testing: The respondent’s household is given HIV testing coupon(s) with a financial incentive of 100MZN (PPP $4.85), instead of the 50MZN (PPP $2.42) coupons offered to all other households.
- Control: None of the above minitreatments.
Our primary outcome of interest is whether anyone in the household received an HIV test in the 14 days following the endline survey, as measured by the redemption of encouragement coupons distributed during the minitreatments.
The FCC programme leads to a reduction in HIV testing, but exposure to minitreatments appears to minimise this effect
Contrary to our expectations that the FCC programme would have a positive impact on HIV testing, we find that it actually leads to lower rates of HIV testing. FCC-enrolled households not receiving any minitreatment had a 10.5 percentage point reduction in HIV testing, relative to the testing rate of 26.3% in control communities.
Estimating the average effects of minitreatments in the full sample, we find that the high incentive coupon (minitreatment #5) increased HIV testing by 7.24 percentage points. There is no average effect of the other minitreatments on the full sample. However, the anti-stigma minitreatment (#1) had a negative effect on HIV testing in control communities. It is likely that individuals in control communities did not have HIV-related stigma at the top of mind, so raising the topic of stigma may have increased its salience.
We also find that counteracting concerns about HIV-related stigma (#1), providing HIV-related information (#2), providing ART-related information (#3), and providing higher financial incentives (#5) all improve the negative impact of the FCC treatment on HIV testing. We do not find any effect of the combined HIV and ART information (#4) on the impact of the FCC treatment. It is possible that providing respondents with too much information reduces the effectiveness of all information, causing lapses in the respondents’ attention or retention.
What do secondary outcomes tell us about the FCC programme?
Analyses of secondary outcomes and minitreatment effects suggest that the negative impact of the FCC programme on HIV testing is driven by worsened misinformation, leading to increases in stigmatising attitudes and a large decline in HIV testing.
We examine impacts of treatment on knowledge regarding HIV/AIDS through a series of knowledge sub-indices. We find that treatment status has a significant effect on the “transmission myth index”. These are questions about whether HIV can be transmitted in certain ways, all of which are incorrect transmission methods (i.e. mosquito bites, shaking hands, kissing, sharing food, witchcraft). The impact of treatment on this subindex suggests that respondents are more likely to believe myths about HIV transmission after receiving the FCC treatment than at baseline. We also look at the programme’s impact on responses to four questions that measure the respondents’ level of HIV-related stigmatisation. We find that treatment reduces the share of non-stigmatising responses by 1.35 percentage points. In other words, the FCC treatment increases stigmatising attitudes.
Minitreatments, which provided correct HIV-related information and countered HIV-related stigma, reduced the negative treatment effects (made treatment effects more positive) of the FCC programme on HIV testing rates. These results indicate that the FCC programme was deficient in increasing recipients’ knowledge about HIV and reducing their stigmatising attitudes.
These unexpected findings raise a subsequent question: how did the FCC programme lead to increased misinformation on HIV transmission and to worsened HIV-related stigmatising attitudes? We hypothesise that providing information that HIV is transmitted through bodily fluids could lead people to believe transmission myths that involve the sharing of bodily fluids like kissing or sharing food. Moreover, FCC home visits may have increased the salience of HIV-related stigmatising attitudes in the community if it was known that home visits were targeted towards households with HIV infection. However, because we did not anticipate these negative findings, we did not collect the information needed to empirically study these mechanisms in detail.
Policy implications and next steps
Our study shows that widespread community-level programmes seeking to increase HIV testing may fail due to deficiencies in information provision and counteracting HIV-related stigma. As evidenced by our minitreatments, however, it also suggests that there are simple ways to counteract these deficiencies. More research is needed to understand the mechanisms driving these results. Motivated by this study, we are currently conducting a new study in Mozambique on ways to hire more effective public health workers to combat HIV/AIDS.
References
Yang, D, J Allen, A Mahumane, J Riddell and H Yu (2023), "Knowledge, stigma, and HIV testing: An analysis of a widespread HIV/AIDS program." Journal of Development Economics 160: 102958.
Banerjee, A, E La Ferrara, and V Orozco-Olvera (2020), "The Entertaining Way to Behavioral Change: Fighting HIV with MTV", Working Paper.
Ciancio, A, A Delavande, and H-P Kohler (2020), "Mortality Risk Information, Survival Expectations and Sexual Behaviors", Vol. 20, pp. 817–875.
Chong, A, M Gonzalez-Navarro, D Karlan, and M Valdivia (2013), "Effectiveness and Spillovers of Online Sex Education: Evidence from a Randomized Evaluation in Colombian Public Schools", NBER Working Paper Series W18776.
Duflo, E, P Dupas, and M Kremer (2015), "Education, HIV, and early fertility: Experimental evidence from Kenya", American Economic Review, 105(9): 2757–2797.
Dupas, P (2011), "Do teenagers respond to HIV risk information? Evidence from a field experiment in Kenya", American Economic Journal: Applied Economics, 3(1): 1–34.
Ford, N, C Migone, A Calmy, B Kerschberger, S Kanters, S Nsanzimana, E Mills, G Meintjes, M Vitoria, M Doherty, and Z Shubber (2018), "Benefits and risks of rapid initiation of antiretroviral therapy", AIDS, 32(1): 17–23.
Godlonton, S, A Munthali, and R Thornton (2016), "Responding to risk: Circumcision, information, and HIV prevention", Review of Economics and Statistics, 98(2): 333–349.
Hoffmann, M, S MacCarthy, A Batson, A Crawford-Roberts, J Rasanathan, A Nunn, L A Silva, I Dourado (2016), "Barriers along the care cascade of HIV-infected men in a large Urban center of Brazil", AIDS Care, 28: 57–62.
Kim, B, B Haile, and T Lee (2017), "Promotion and persistence of HIV testing and HIV/AIDS knowledge: Evidence from a randomized controlled trial in Ethiopia", Health Economics, 26(11): 1394–1411.
Lubega, P, F Nakakawa, G Narciso, C Newman, and C Kityo (2019), "HIV Status, Role Models and Attainable Goals: Experimental Evidence on Inspiring Women in Uganda", Working Paper.
Rodger, A, V Cambiano, T Bruun, P Vernazza, S Collins, O Degen, GM Corbelli, V Estrada, A Geretti, and A Beloukas (2019), "Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): Final results of a multicentre, prospective, observational study", Lancet, 393(10189): 2428–2438.