Heterogeneity of Provider Preferences for HIV Care Coordination Program Features: Latent Class Analysis of a Discrete Choice Experiment

BackgroundThe PROMISE study was launched in 2018 to assess revisions to an HIV care coordination program (CCP) designed to address gaps in care and treatment engagement among people living with HIV in New York City (NYC). We report on the heterogeneity of provider preferences regarding a revised CCP elicited from a discrete choice experiment (DCE). MethodsFrom January to March 2020, 152 CCP providers in NYC completed a DCE with 4 program attributes: 1) help with adherence to antiretroviral therapy, 2) help with primary care appointments, 3) help with issues other than primary care, and 4) program visit location. Each attribute had 3-4 levels. Latent class analysis (LCA) was used to detect subgroups with differing attribute importance and part-worth utility patterns. Choice simulation was used to estimate providers endorsement of eight hypothetical CCPs. ResultsLCA identified three subgroups. The two larger subgroups (n = 133) endorsed more intensive attribute levels, particularly clients receiving directly observed therapy, and home visits. The remaining smaller subgroup (n = 19) endorsed clients receiving medication reminders and meeting with clients at the program. Simulation showed that intensive medical case management programs had the highest degree of endorsement (62%). ConclusionWhile our results indicate high endorsement among providers for intensive CCP features, overall, they also suggest the need for flexible service delivery options to meet the needs of the clients that these programs serve. Additional information sharing across and within agencies may be warranted to improve the fidelity with which the CCP is implemented.