Age-Specific Risk Scores Do Not Improve HIV-1 Prediction Among Women in South Africa

Background:HIV-1 risk scoring tools could help target provision of prevention modalities such as pre-exposure prophylaxis. Recent research suggests that risk scores for women aged 18-45 may not predict risk well among young women aged 18-24. We evaluated the predictive performance of age-specific risk scores compared with the existing non-age-specific VOICE risk score, developed for women aged 18-45.Methods:We conducted a secondary analysis of the Evidence for Contraceptive Options and HIV Outcomes Trial to develop and internally validate HIV-1 risk scores for women aged 18-24 and 25-35 in South Africa. Candidate predictors included baseline demographic, clinical, behavioral, and contextual characteristics readily available in clinical settings. The VOICE risk score was applied to women aged 18-35. We evaluated predictive performance of each risk score by area under the receiver operating characteristic curve (AUC).Results:Predictive performance of all risk scores was moderate, with AUC (95% confidence interval) of 0.64 (0.60 to 0.67) among women aged 18-24, 0.68 (0.62 to 0.73) among those aged 25-35, and 0.61 (0.58 to 0.65) for the VOICE risk score applied to women aged 18-35; The AUC was similar in internal validation. Among women aged 18-24, HIV-1 incidence was high even at low risk scores, at 3.9 per 100 person-years (95% confidence interval: 3.2 to 4.7).Conclusions:All risk scores were moderately predictive of HIV-1 acquisition, and age-specific risk scores performed only marginally better than the VOICE non-age-specific risk score. Approaches for targeted pre-exposure prophylaxis provision to women in South Africa may require more extensive data than are currently available to improve prediction.