14. Increasing HIV Testing Among Adolescents: Preliminary Outcomes of A Multi-Level Intervention To Implement Universal HIV Testing In Six Urban School-Based Health Centers

Abstract

HIV testing is a critical step in facilitating access to HIV care and preventing onward transmission. While HIV incidence is increasing among adolescents, only a small proportion is tested. The CDC recommends offering HIV testing in health care settings to persons 13 years and over regardless of risk. We initiated an intervention to implement universal HIV testing to for adolescents utilizing six school-based health centers (SBHCs) in public high school campuses in the Bronx, New York. The intervention model included system-level initiatives (development of practice work flows for offering oral, rapid HIV testing with input from SBHC staff, and use of performance improvement tools led by an Implementation Coach) and staff-level initiatives (trainings, technical assistance and incentives). Three matched-pairs of SBHCs were assigned to two cohorts to receive the same intervention: Cohort 1 initiated in Fall 2016, Cohort 2 in Fall 2017. All patients seen at least once in baseline and intervention years were included for analysis. Outcomes were the offer of an annual HIV test by the clinical staff and the acceptance of testing by the adolescents. Predictors examined included the intervention year, age > 16 years, gender, and ever having been sexually active. For each outcome, we fitted a logistic mixed effects model separately for each cohort. Odds ratios for each variable were calculated, p values set at < .01. Combining both cohorts, there were 5504 unique patients in the baseline year and 5925 in the intervention year, of which the mean age was 15.9 (SD = 1.43), 58.3% were female, 49.1% Hispanic, 26.5% Black and 36.3% had ever been sexually active. The odds of an HIV test offer for Cohort 1 increased by 14% (p = 0.0043) from baseline (14.1%) to intervention (16.8%), and for Cohort 2 by 40% (p < 0.0001) from baseline (8.8%) to intervention (11.6%). Among those offered HIV testing, the odds of test acceptance increased in Cohort 1 by 81% (p < 0.0001) from baseline (34.6%) to intervention (47.8%) and in Cohort 2 by 59% (p < 0.0001) from baseline (59.0%) to intervention (67.7%). Both analyses were adjusted for age, gender, and sexual activity. Females tended to have lower odds of being offered testing within both Cohort 1 (OR = 0.75) and Cohort 2 (OR = 0.67), (p < 0.0001). Ever having been sexually active was associated with increased odds of an HIV test offer (Cohort 1 OR = 1.56, p < 0.0001; Cohort 2 OR = 1.99, p < 0.0001), as well as HIV test acceptance (Cohort 1 OR=4.45, p<0.0001; Cohort 2 OR=3.78, p<0.0001). To date, the implementation intervention increased both the offer and acceptance of HIV testing, with males more likely to be offered testing and with sexually active students more likely to be offered and to accept testing. Implementation activities will continue for one more year for each Cohort as we explore further innovations to increase universal testing and efforts to sustain these improvements.