Life Plans of Young Adults in Rural KwaZulu-Natal: an Intervention Study

The proposed study will investigate survival expectations and health behaviours among young adults ages 18 to 25. Potential respondents will be sampled at random from a population listing from the demographic and health surveillance system of the Africa Health Research Institute (AHRI) in Somkhele, South Africa. Potential participants will be visited at home to be recruited for the study.

1. Consent. The study will be described and participants will provide written informed consent to participate in the study, including the baseline visit, a follow-up survey, and permission to link with demographic surveillance data.
2. Baseline interview (1.5 hours)

2.a) Survey module. First, a survey questionnaire will be conducted with young adults ages 18 to 25 to learn about their survival expectations, perceptions of HIV risks and treatment, future-oriented behaviours such as smoking, alcohol use, HIV risk behaviors, savings, locus of control, mental health, time and risk preference, and life satisfaction.

2.b) Video intervention. After the survey module, study participants will be randomized into treatment and control groups. Both interview participants and interviewers will be blinded to treatment assignment prior to this stage. The treatment group will be shown a 10 minute video that provides information on the changes in HIV-related mortality and life expectancy in the Demographic Surveillance Area between 2003 and today (an increase of about 18 years in the average length of life). The video includes testimonials from community members who have lived long and fulfilling lives on HIV treatment. The goal of the video is to increase measured survival expectations, hope, locus of control, and future orientation. The control group will receive a 10-minute video clip on another topic. Participants will be asked briefly for an open-ended response to the video: "what did the video make you think about?" which will be transcribed or recorded.

Details on randomization procedures: treatment assignment will be determined at the level of the household, such that if there are multiple respondents per household, they receive the same treatment. Households will be randomized to the intervention video or control ex ante. Though treatment assignment will occur ex ante, it will only be revealed - to both the field workers and the respondents - after completing the initial survey module. Ex ante randomization has two benefits. First, as a quality control, ex ante randomization limits the opportunity for violations of treatment assignment. Second, it increases power by enabling stratification on baseline variables. Because the study is nested in the AHRI population surveillance, treated and control households can be balanced on household size, wealth, and other characteristics. Treated and controls will be randomized at a 1:1 ratio, with controls further sub-randomized 1:1 to attention-placebo vs. pure control.