Incentivising safer sexual behaviour: Evidence from a lottery experiment on HIV prevention


Published 30.10.17

Lottery incentives targeting risk lovers led to a 20% reduction in HIV incidence

An estimated 1.2 million new HIV infections occurred in Africa alone in 2016, adding to the nearly 26 million people living with HIV on the continent. New infections continue to outpace antiretroviral (ARV) therapy uptake in most developing countries (UNAIDS 2016). Thus, the global need for effective HIV prevention programmes remains urgent.

HIV prevention programmes can be broadly divided into two types: biomedical interventions that aim at reducing susceptibility to HIV, and interventions aimed at reducing risky sexual behaviour. Several biomedical interventions, such as male circumcision and pre-exposure prophylaxis, have been shown to significantly reduce HIV transmission in field trials (Gray et al. 2007, Bailey et al. 2007, Auvert et al. 2005). Evidence from randomised controlled trials of behavioural interventions, however, is at best mixed (see McCoy et al. 2010, Padian et al. 2010 for reviews). No traditional programmes – offering some combination of risk-reduction counselling, condom promotion, referral and treatment for sexually transmitted infections – have documented significant effects on HIV incidence.

With a few exceptions, most behavioural programmes have a rather general focus, although targeting those with the highest risk of becoming infected (and/or spreading the disease) may be the most effective way to reduce HIV incidence. Importantly, such targeting is not specific to HIV prevention but holds more generally for a broad range of programmes aimed at reducing risky behaviour. We know relatively little, however, about how to design such types of programmes.

Lotteries as incentives for HIV prevention

We designed an HIV intervention, and assessed its impact in a randomised trial, to appeal specifically to people who are risk-lovers and, presumably, therefore most at risk of HIV infection (Björkman Nyqvist et al. 2017). Specifically, trial participants in Lesotho were offered the chance to participate in a lottery with relatively low expected payments but high prizes conditional on negative STI test results.

Introducing a gamble into an otherwise standard financial incentive programme has at least two potential advantages. First, with lotteries, the programme becomes relatively more attractive to individuals that are willing to take monetary risks. If the willingness to take monetary risks is correlated with other risky behaviours, and especially engaging in risky sex, then lottery incentives may better target those at higher risk of getting infected by HIV.

Second, there is growing evidence from psychology and behavioural economics that people tend to overestimate small percentages, and therefore prefer a small chance at a large reward to a small reward for sure (Kahneman and Tversky 1979, Kahneman 2011, Barberis 2013). If so, the perceived return from participating in a gamble (lottery) is higher than the return from an incentive programme that pays the expected return with certainty, or likewise lotteries may provide stronger incentives for behavioural change compared to a traditional CCT holding the budget constant.

Importantly, the use of lotteries in public health interventions is not unprecedented, including in the area of HIV prevention. To the best of our knowledge, however, this is the first large-scale randomised trial to assess its impact as a preventative tool.

Our study was a parallel group randomised trial. It had three separate arms: a control arm and two intervention arms (one low-value lottery arm and one high-value lottery arm). Participants randomised in the low-value lottery arm were eligible to win lottery prizes worth 500 malotis/South African rands, or approximately $50, every four months. In the high-value lottery arm, individuals were eligible to win lottery prizes of twice that amount.

Individuals in the intervention arms were awarded a lottery ticket if they tested negative for two curable STIs (syphilis and trichomoniasis) in the week before the lottery draw. Village-level lotteries were organised every four months and four lottery winners (one male and one female per lottery arm) per village were drawn.

Individuals in the intervention arms testing positive for any of the two STIs did not receive a lottery ticket. They could, however, continue as study participants and thus become eligible in subsequent rounds if they took the free treatment offered and subsequently remained STI negative. Participants randomised in the control arm were not eligible for lottery tickets, but all other study procedures were identical between the control and intervention arms. Anyone testing positive for an STI (regardless of arm) was offered counselling and free STI treatment, and individuals testing positive for HIV were referred to public health clinics offering AIDS treatment for appropriate follow-up.

Overall, the lottery incentives had a significant impact on HIV incidence, reducing the rate of new HIV infections among participants who were HIV negative at baseline. Over the two-year trial period, the HIV incidence rate was reduced by 2.5 percentage points, or 21.4%, leading to a 3.4 percentage points lower HIV prevalence rate at the end of the trial, in the pooled intervention group compared to the control group. To the best of our knowledge, this is the first HIV prevention intervention focusing on changes in sexual behaviour (as opposed to medical interventions) to have been demonstrated to lead to a significant reduction in HIV incidence, the ultimate objective of any HIV prevention intervention. We estimate that the cost of averting a primary HIV infection was $882, or about one-tenth of the costs of averting an HIV infection in the only other behavioural trial documenting significant reduction in HIV prevalence (Baird et al. 2012).

Making risk-lovers behave more like risk-averse individuals

Despite the lack of a certainty equivalent intervention arm, our study provides suggestive evidence that the lottery design played an important role in achieving the effects we document. Using baseline data from the control group, we show that risk-lovers – with the degree of risk aversion measured using hypothetical lotteries – were more than twice as likely to become infected with HIV over the trial period, thus providing support for the assumption that risk-loving behaviour in hypothetical monetary gambles translates to objectively measured risk-loving sexual behaviour. In the intervention groups, on the other hand, where participants were offered lottery incentives to promote safer sexual behaviour, we find that the risk-loving individuals were equally as likely as risk-averse individuals to become infected with HIV during the two-year trial period. That is, the lottery programme lowered HIV incidence rates in the treatment relative to the control group by making individuals with greater preference for risk at baseline behave similarly to risk-averse individuals in terms of their sexual behaviour.

The practical advantages of lottery incentives

Our findings provide suggestive evidence that lotteries can be used to target groups at higher risk of becoming infected with HIV. In addition to targeting the individuals most at risk, the use of lotteries also has practical advantages that are important if we consider scaling-up such programmes. First, the administrative costs of a lottery programme are likely to be lower compared to a traditional CCT programme, as only winners need to be paid. Second, we could think of a lottery system whereby only a fraction of the study participants are tested, reducing the cost of testing, which is an important fraction of the total cost. While the research protocol for the Lesotho study required that all project participants were offered testing, the incentive for behavioural change will, under certain conditions, remain the same if only lottery winners are tested or if STI screening is also subject to a lottery. Finally, and importantly, while we study the impact of lottery incentives for a specific outcome here, our results open up avenues to explore whether lotteries can be successfully used in many other types of programmes and domains to enhance the demand for less risky behaviour.

Editors’ note: An earlier version of this column first appeared on VoxEU.

Photo credit: Stefan Krasowski/flickr.


Auvert B, D Taljaard, E Lagarde, J Sobngwi-Tambekou, R Sitta, A Puren (2005), "Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial", PLoS Medicine.

Baird, S, R S Garfein, C McIntosh and B Ozler (2012), "Effect of a cash transfer programmeme for schooling on prevalence of HIV and herpes simplex type 2 in Malawi: a cluster randomized trial", The Lancet, 379: 1320-1329.

Bailey R, S Moses, C B Parker, K Agot, I Maclean, J N Krieger, C F M Williams, R T Campbell, J O Ndinya-Achola (2007), "Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial", The Lancet, 369(9562): 643–656, 24.

Barberis, N C (2013), "Thirty years of prospect theory in economics: A review and assessment", Journal of Economic Perspectives, 27(1): 173–196.

Björkman Nyqvist, M, L Corno, D de Walque and J Svensson (2017), “Incentivizing safer sexual behavior: Evidence from a randomized controlled trial on HIV prevention”, American Economic Journal: Applied Economics, forthcoming.

Gray R, G Kigozi, D Serwadda, F Makumbi, S Watya, F Nalugoda, N Kiwanuka, L H Moulton, M A Chaudhary, M Z Chen, N K Sewankambo, F Wabwire-Mangen, M C Bacon, C F M Williams, P Opendi, S J Reynolds, O Laeyendecker, T C Quinn and M J Wawer (2007), "Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial", The Lancet, 369(9562): 657–666.

Kahneman, D (2011), Thinking, fast and slow, New York: Farrar, Straus, and Giroux.

Kahneman, D and A Tversky (1979), "Prospect theory: An analysis of decision under risk", Econometrica, 47(2): 63–91.

McCoy S I, K Abigail and N S Padian (2010), "Behavior change interventions to prevent HIV infection among women living in low and middle income countries: a systematic review", AIDS and Behavior, 14(3): 469-82.

Padian, N S, S I McCoy, J E Balkus and J N Wasserheit (2010), "Weighing the gold in the gold standard: Challenges in HIV prevention research", AIDS, 24: 621–635.

UNAIDS (2016), "Fact sheet 2016", Geneva, available at