Household enrollment in government child welfare programs may be associated with reductions in young women’s engagement in sexual behaviors that could put them at risk for HIV, according to a longitudinal study conducted in two South African provinces.1 Researchers compared the behaviors of adolescents whose families received benefits from either of two cash transfer programs—one available to low-income families with children and the other to foster families—with those of adolescents who were socially and demographically similar but whose families did not receive a cash transfer. Compared with those in the comparison group, female adolescents in the cash transfer programs had reduced odds of engaging in transactional sex (odds ratio, 0.4) and age-disparate sex (0.3) in the year between baseline and follow-up. Such reductions were not found among male participants.

The data come from a prospective, observational study conducted in two rural and two urban health districts in Mpumalanga and Western Cape, South Africa. The investigators recruited one male or female adolescent aged 10–17 at random from every household in randomly selected census enumeration areas within the study districts until the desired sample size was reached. Participants completed private, face-to-face interviews in 2009–2010 and again one year later. The interviews collected demographic and family information, as well as data on whether participants had had transactional sex (sex in exchange for food, shelter, school fees, transport or money), age-disparate sex (sex with a partner more than five years older than the participant), sex without a condom, sex with multiple partners and sex after drinking alcohol or taking drugs.

Because differences in family characteristics and circumstances could account for variation in whether families had access to cash transfer programs, the researchers used propensity score matching to account for 23 possible covariates of inclusion in the program, including participant age, sex and location; caregiver characteristics and parenting practices; and household unemployment and assets. The researchers constructed a control group made up of youth whose households were not participating in a cash transfer program at both baseline and follow-up and who matched the program participants on as many of these variables as possible; participants who could not be matched or were younger than 12 at follow-up were included in descriptive analyses but excluded from regression analyses.

The full sample consisted of 1,926 females and 1,475 males, two-thirds of whom were in a cash transfer program at both baseline and follow-up. Among those whose households received cash transfers, the vast majority (99%) were in the child support program. At baseline, participants in the cash transfer and control groups had a mean age of 13 and 14, respectively. About one-third lived in informal housing, and at least two-thirds lived with one or both biological parents.

Transactional sex in the previous year was reported by 1% of both males and females in the cash transfer program and by 3% of males and 5% of females whose households did not receive a grant. About 1% of grant recipients of each sex and 3–4% of nonrecipients reported age-disparate sex, and 7% and 13–15%, respectively, reported unprotected sex. Drug or alcohol use before sex in the past year was reported by 1% of male and female grant recipients and by 3% of nonrecipient males and 1% of nonrecipient females. Finally, 8% of males and 4% of females in cash transfer programs said they had had sex with multiple partners in the past year, compared with 19% of males and 10% of females not in the programs. All types of risky sexual behaviors were associated with each other, and reports of these behaviors increased with age.

A logistic regression analysis of the propensity score–matched sample revealed that among females, household receipt of cash transfer payments at both baseline and follow-up was associated with reduced odds of having had transactional sex or age-disparate sex in the past year (odds ratios, 0.5 and 0.3, respectively) and of having ever engaged in either behavior (0.5 and 0.4, respectively). Among males, cash transfers were associated only with reduced odds of having had multiple partners in the past year (0.7). Findings were similar in multivariate models that also controlled for baseline covariates.

The researchers emphasize that the positive relationship between the receipt of cash transfers and young women’s sexual risk behaviors supports "the use of cash transfers as part of a combination prevention approach that targets structural, behavioural, and biomedical drivers of infection," and they suggest that the household financial stability promoted by the transfers may help to reduce adolescent females’ recourse to deprivation-driven sex. Because the study revealed little evidence of a relationship between household receipt of cash transfers and males’ sexual risk behaviors, the investigators suggest that new strategies for reducing young men’s HIV risk are needed.—H. Ball

REFERENCE

1. Cluver L et al., Child-focused state cash transfers and adolescent risk of HIV infection in South Africa: a propensity-score-matched case-control study, Lancet Global Health, 1(6):e362–e370.