Effects of the informed health choices secondary school intervention after 1 year: a prospective meta-analysis using individual participant data

Critical thinking about health choices is essential to avoid being misled by unreliable information and to use reliable information appropriately. The aim of this prospective meta-analysis was to synthesize the results of 1-year follow-up data from three cluster-randomized trials of an intervention designed to teach lower secondary school students to think critically about health choices. Only one other randomized trial has evaluated a school-based intervention to teach adolescents to think critically about health choices. That trial compared two teaching strategies to teach statistical reasoning. It did not assess long-term learning-retention. We conducted the trials in Kenya, Rwanda, and Uganda. The intervention included providing a 2-3-day teacher training workshop and digital resources for ten lessons. The intervention focused on nine key concepts. We did not intervene in control schools. The primary outcome was a passing score on a test (≥ 9 of 18 multiple-choice questions answered correctly). We performed random effects meta-analyses to estimate the overall intervention effects. We calculated learning retention as the test results in the intervention schools after 1 year relative to just after the intervention, adjusted for chance. Altogether, 244 schools (11,344 students) took part in the three trials. Follow-up data was collected for 8298 students (73%). The overall odds ratio for the primary outcome after 1 year was 3.6 (95% CI: 1.9-7.1; p = 0.0001) in favor of the intervention, whereas it was 5.5 (95% CI: 3.0-10.2) just after the intervention. This corresponds to 25.6% (95% CI: 21.1-30.0%) more students in the intervention schools passing the test after 1 year versus 33.3% (95% CI: 28.7-37.8%) just after the intervention. Overall, 2273 (52.6%) of 4324 students in intervention schools had a passing score after 1 year compared to 3397 (58.1%) of 5846 students just after the intervention, indicating 88.3% learning retention. One year after the intervention, we still found a positive effect on the ability of students to think critically about health choices, but 5.5% fewer students in the intervention schools had a passing score. The certainty of the evidence was also lower due to 27% of students being lost to follow-up. The protocol for this prospective meta-analysis was registered with PROSPERO May 31, 2022, ID 336580. The three randomized trials were registered in the Pan African Clinical Trial Registry February 15, 2022, PACTR202203880375077; Apr