Background
Adolescents and young adults (AYA) 15-24 years represented more than a quarter of all new HIV infections globally in 2023. Data from two rounds of Population HIV Impact Assessments (PHIA) from four countries were used to estimate HIV incidence and assess progress in reducing incidence in AYA overall and by sex.
Methods
PHIAs are nationally representative surveys measuring HIV outcomes. We analyzed data from two rounds of PHIAs in Eswatini (2016-17 & 2022), Lesotho (2016-17 & 2020), Malawi (2015-16 & 2020-21), and Zimbabwe (2015-16 & 2020). Participating household members ≥15 years received rapid antibody HIV testing and blood sample collection. Twelve-month HIV incidence was estimated via Limiting Antigen-Avidity (Lag) enzyme immunoassays and an algorithm identifying recent HIV infections (proxy for incidence). We generated national population-level estimates of new HIV infections in AYA with 95% probability bands (PB) using survey weights with jackknife variance and calculated proportions of new infections among AYA out of all new infections in adults ≥15 years. Rao-Scott tests were used to compare estimates from the two survey rounds.
Results
Across the four countries, 21,008 and 21,028 AYA participated in round 1 and 2, respectively. HIV incidence in AYA in round 1 surveys was lowest in Malawi (0.22, 95%PB 0.03-0.40) and highest in Eswatini (1.07, 95%PB 0.48-1.66); there were an estimated 20,900 new infections in AYA across the four countries. In round 2, incidence ranged from 0.25 (95%PB 0.06-0.44) in Malawi to 0.86 (95%PB 0.26-1.45) in Eswatini; with an estimated 25,600 new infections in AYA. No country showed significant change in HIV incidence across the two surveys for all AYA; incidence in females in Lesotho decreased from 1.49 (95%PB 0.58-2.39) to 0.34 (95%PB 0-0.74) (p=0.03). AYA accounted for larger proportions of all new adult infections from round 1 to 2 in Eswatini (39.9% v. 54.2%), Malawi (25.0% v. 46.9%) and Zimbabwe (28.4% v. 41.9%).
Conclusions
Despite progress for young females in Lesotho, data from two rounds of PHIA surveys in four countries shows no change in estimated HIV incidence overall in AYA, and indicate this age group accounts for a substantial proportion of all new adult infections. Larger studies are needed to confirm these results as are more effective prevention efforts for this age group.