Abstract
Background Kazakhstan has a high HIV incidence among adolescents and young adults (AYA), and high HIV stigma contributing to low HIV testing uptake. We examined whether an AYA-developed digital crowdsourced intervention reduced HIV stigma compared with conventional public health materials among AYA in Almaty, Kazakhstan. Methods A total of 216 AYA (females:116/males:110) aged 16-24 years were recruited to the online study cohort and randomized 1:1 to the intervention or control arm. AYA were exposed to the crowdsourced intervention or control materials once a week for 5weeks, with equivalent exposures between arms. Outcomes included a total HIV stigma score assessed at baseline, immediately post-intervention and 2months post-intervention from January to August 2023. We conducted multilevel mixed models to compare changes over time by arm and sex. Results AYA in the intervention arm had significantly lower HIV testing stigma 2months post-intervention (adjusted mean change (AMC): -0.73 (-1.07, -0.39)) than AYA in the control arm (AMC: -0.06 (-0.42, 0.30); P =0.032). Female AYA in the intervention arm had significantly lower total HIV stigma immediately post-intervention (AMC: -4.91 (-7.25, -2.58)) and 2months post-intervention (AMC: -5.16 (-7.48, -2.84)) than females in the control arm (immediately post-intervention AMC: -0.03 (-2.63, 2.57) and 2-months post-intervention AMC: -0.07 (-2.70, 2.56); P =0.012, P =0.012). Conclusions The AYA-developed crowdsourced intervention decreased HIV testing stigma, although this effect was moderated by sex, and decreased total HIV stigma among female AYA. Crowdsourced interventions may be a promising way to engage communities to develop interventions to decrease HIV stigma.