Public interventions that pay people to adopt healthy habits have become increasingly popular over the past decade. South Africa has been experimenting with cash pay-outs to prevent HIV for a number of years now, but the results of two large trials that were released last week at the International AIDS Society conference in Vancouver are not that encouraging – well, not at first glance anyway.
Both studies were based on research that has shown the longer a young girl remains in school, the less likely she is to get HIV. So both studies aimed to encourage girls to stay in school by offering cash rewards to some girls (and, in the second study, to boys too) who did so.
The first study (named HIV Prevention Trial Network 068) took place in Agincourt in Bushbuckridge in Mpumalanga, and involved 2 448 HIV negative girls aged 13 to 20. They were divided into two equal groups. The girls in one group were given R300 a month (R100 to the girls and R200 to their guardians) provided that they attended school 80% of the time, while the girls in the other (control) group got nothing. But after three years, there was no difference between the two groups in either school attendance (which was 95%) or HIV incidence (the rate of new HIV infections, which was around 1.8% per year).
The second study (Caprisa 007) involved almost 3,000 school girls and boys in Grades 9 and 10 in schools in Vulindlela, a semi-rural area in the KwaZulu-Natal Midlands. All pupils were offered life skills classes that included tips on how to prevent HIV. But pupils in the intervention arm were given cash incentives to attend these classes, pass their tests every six months and take an annual HIV test. School attendance was high in both groups and after two years there was no difference in HIV incidence rates between the intervention and the control groups.
It’s a relief that two large studies in very different parts of the country involving poor families has proven that parents did not need cash to motivate them to send their children to school. It was the right thing to do and they all did it naturally. But while the cash didn’t make a difference to school attendance, it did make a positive impact on the lives of those who got it and reduced their risk of HIV in other ways.
According to the Mpumalanga study’s researchers, “the young women who received the cash transfer reported significantly less risk behaviours. They reported having significantly fewer sex partners, less unprotected sex, and were less likely to experience intimate partner violence compared to the young women not receiving the cash”. Meanwhile, in the Caprisa 007 study there was a significant difference in the herpes incidence, with pupils who got cash having a 30% lower incidence rate.
The results from both studies are new, and researchers are still interpreting their data. But Professor Kathleen Kahn, site lead investigator for the Mpumalanga study, says that “the cash may have helped girls not have sex with people they did not want to have sex with – this may be why they were less likely to have sex and to report less unwanted sex. Avoiding unwanted sex may also explain why we see a reduction in intimate partner violence.”
In relation to Caprisa 007, herpes is far more common than HIV so it is also possible that those getting the cash also had less sex than those who didn’t get the cash. It is also possible that the girls who got cash were less likely to have sex with older men who are more likely to have a sexually transmitted infection such as herpes and HIV.
Two years ago, research involving over 3,000 young people aged 10 to 18 in the Western Cape and Mpumalanga found that the girls from homes that received R300 child support grants were far less likely to have sex with “sugar daddies”. Even though child support grants are relatively small, the little bit of additional money was enough to push the girls over the poverty line into a safer place where they did not feel compelled to make sex deals with older men in exchange for money and things like food and clothes.
“If we give child support grants or other cash transfers to families, then we can reduce HIV risk among girls in those families,” said Dr Lucie Cluver from Oxford University, the lead author. “There have been loads of stories in the media about sugar daddies and about how girls are choosing to do this… that it’s a kind of life style choice. This data really suggests the opposite because it says that even if you give a relatively tiny amount of money to the family to put them just above the level of survival, then girls are choosing not to have sugar daddies.”
Exchanging sex for rewards is probably as old as humanity, and when a young woman comes from an impoverished family, she has very little to trade but her body. Thus, small cash injections into households with vulnerable young women may protect them against having sex with older men bearing HIV. But a Malawian study offers a cautionary tale about the dangers of dishing out cash rewards indiscriminately. The 2006 study that offered adults cash to remain HIV negative backfired when it found that the men getting the money were significantly more likely to engage in risky sex shortly after getting the cash (The men were 8.5 percentage points more likely and women were 7.5 percentage points less likely to engage in risky sex.)
The results from HPTN 068 and Caprisa 007 don’t justify state cash incentives to prevent HIV. But they do show that a little bit of extra cash can protect girls against risky behaviour that may expose them to HIV – and provide yet another argument for the dire need for jobs and, failing that, a basic income grant to protect poor households.
The studies also show yet again that there is no one magic bullet to prevent HIV. We have a motley collection of armour to protect us – mainly abstaining from sex, using condoms, men getting circumcised and all those with HIV going on antiretroviral drugs (a person with an undetectable viral load is unlikely to transmit HIV). But we lack effective ways to persuade people to use this armour. Cash may help at times but changing human behaviour is too complicated to rely solely on an ATM machine for solutions.