Abstract
The HIV/AIDS Services Administration (HASA) provides housing, benefits, and case management for low-income people with HIV (PWH) in New York City (NYC). HASA implemented three differentiated service delivery (DSD) pilots: Permanent Supportive Housing (PSH), Transitional Supportive Housing (TSH), and Emergency Single Room Occupancy (ESRO), testing tiered case management by housing type. The PSH pilot aimed to maintain viral suppression outcomes with higher caseloads (50:1), while the TSH and ESRO pilots aimed to improve outcomes with lower caseloads (25:1). We merged programmatic data with the NYC HIV Surveillance Registry to compare 6-month viral load suppression (VLS) and 12-month durable viral suppression (DVS) between pilot and non-pilot clients with follow-up through August 2022. Adjusted risk ratios (aRR) were estimated using log-binomial models. Actual caseload ratios were lower than planned: PSH (36:1), TSH (17:1), and ESRO (24:1, near its 25:1 target) compared with ~ 34:1 in non-pilot arms. VLS rates were higher among pilot clients: PSH (77% vs. 72%; aRR 1.03, 95% CI 0.99-1.07), TSH (65% vs. 56%; 1.06, 1.01-1.11), and ESRO (50% vs. 46%; 1.08, 0.99-1.17). DVS showed similar trends: PSH (70% vs. 62%; 1.05, 0.98-1.11), TSH (51% vs. 41%; 1.13, 1.05-1.22), and ESRO (37% vs. 29%; 1.11, 0.88-1.39). Implemented partly during the COVID-19 pandemic, the PSH and TSH pilots met their goals of maintaining or improving VLS and DVS, though lower-than-planned caseload ratios may have influenced outcomes. The ESRO pilot may have improved VLS slightly, but wide confidence intervals due to under-enrollment limit interpretation of DVS effects.