Abstract: Differences In Ratings of Digital Crowdsourced Content and Conventional Public Health Materials Designed to Reduce HIV Stigma and Promote HIV Self-Testing Among Adolescents and Young Adults In Kazakhstan: A Randomized Controlled Trial (Society for Social Work and Research 31st Annual Conference)

Schedule:

Sunday, January 24, 2027

Union Station 22 (Hilton San Francisco Union Square)

* noted as presenting author

Yi Yihang Sun, MSW, Ph.D. Candidate, Columbia University, New York, NY

Sara Landers, PhD, LMSW, Postdoctoral Research Fellow, Brown University, RI

Gaukhar Mergenova, MD, MS, Project Director, Columbia University Global Health Research Center of Central Asia (GHRCCA), Almaty, Kazakhstan

Alissa Davis, PhD, Associate Professor, Columbia University, New York, NY

Background

Kazakhstan has one of the fastest-growing HIV epidemics among adolescents and young adults (AYA). Little is known about how AYA rate crowdsourced content compared with conventional public health materials across multiple timepoints. This paper examined differences in ratings of digital crowdsourced content and conventional public health materials among AYA in Kazakhstan using a randomized controlled trial (RCT).

Methods

We used data from the RCT with 216 AYA (intervention: 111; control: 105) aged 16 to 24 years in Almaty, Kazakhstan. Participants viewed assigned content at baseline, week 1, week 2, week 3, and month 1. AYA in the intervention arm viewed AYA-developed digital crowdsourced content. AYA in the control arm viewed conventional public health materials. After viewing each content, AYA responded to eight content rating items. We used multilevel mixed models and estimated marginal means to examine between-arm differences, within-arm time trends, and between-arm differences in change over time.

Results

No statistically significant between-arm differences were observed at any timepoint for content quality, interest, perceived effectiveness at reducing HIV stigma, other AYA's interest, or likelihood of sharing with a friend. At baseline, AYA in the control arm were more likely than those in the intervention arm to report willingness to post the content on social media (76% vs 51%; OR = 3.04; 95% CI, 1.09 to 8.46; P = .033) and that the content would motivate other AYA to get tested for HIV (98% vs 85%; OR = 9.81; 95% CI, 2.58 to 37.25; P < .001). At month 1, AYA in the intervention arm had higher odds of reporting that the content had changed how they view people living with HIV (24% vs 8%; OR = 3.70; 95% CI, 1.33 to 10.00; P = .012). Within the intervention arm, willingness to post on social media was higher than baseline at week 1 (OR = 5.32, P < .001), week 2 (OR = 3.03, P = .011), and month 1 (OR = 4.20, P = .002), while in the control arm we only observed a significant increase at week 1 (OR = 3.18, P = .036). Compared with the control arm, AYA in the intervention arm showed more increases from baseline in reporting that the content would motivate other AYA to get tested for HIV across all follow-up timepoints, and in reporting that the content had changed their view of people living with HIV at month 1 (OR = 0.37; 95% CI, 0.14 to 0.93; P = .035).

Conclusions

AYA in both arms rated content similarly on quality, interest, and perceived effectiveness. However, crowdsourced content was more effective at changing AYA's views of people living with HIV and at motivating HIV testing over time. These findings suggest that AYA can help create intervention content that is more meaningful and relatable to their peers. Youth-serving organizations should consider forming youth advisory boards to involve AYA in designing digital HIV prevention materials, center youth voices, and promote culturally grounded content to reduce HIV stigma and promote HIV self-testing.