Ambulatory Intensive Care for Medically Complex Patients at a Health Care Clinic for Individuals Experiencing Homelessness: The SUMMIT Randomized Clinical Trial

IMPORTANCE Intensive primary care interventions have been promoted to reduce hospitalization rates and improve health outcomes for medically complex patients, but evidence of their efficacy is limited. OBJECTIVE To assess the efficacy of a multidisciplinary ambulatory intensive care unit (A-ICU) intervention on health care utilization and patient-reported outcomes. DESIGN, SETTING, AND PARTICIPANTS The Streamlined Unified Meaningfully Managed Interdisciplinary Team (SUMMIT) randomized clinical trial used a wait-list control design and was conducted at a health care clinic for patients experiencing homelessness in Portland, Oregon. The first patient was enrolled in August 2016, and the last patient was enrolled in November 2019. Included patients had 1 or more hospitalizations in the prior 6 months and 2 or more chronic medical conditions, substance use disorder, or mental illness. Data analysis was performed between March and May 2021. INTERVENTION The A-ICU included a team manager, a pharmacist, a nurse, care coordinators, social workers, and physicians. Activities included comprehensive 90-minute intake, transitional care coordination, and flexible appointments, with reduced panel size. Enhanced usual care (EUC), consisting of team-based primary care with access to community health workers and mental health, addiction treatment, and pharmacy services, served as the comparator. Participants who received EUC joined the A-ICU intervention after 6 months. MAIN OUTCOMES AND MEASURES The main outcome was the difference in rates of hospitalization (primary outcome), emergency department (ED) visits, and primary care physician (PCP) visits per person over 6 months (vs the prior 6 months). Patient-reported outcomes included changes in patient activation, experience, health-related quality of life, and self-rated health at 6 months (vs baseline). We performed an intention-to-treat analysis using a linear mixed-effects model with a random intercept for each patient to examine the association between study group and outcomes. RESULTS This study randomized 159 participants (mean [SD] age, 54.9 [9.8] years) to the A-ICU SUMMIT intervention (n = 80) or to EUC (n = 79). The majority of participants were men (102 [65.8%]) and most were White (121 [76.1%]). A total of 64 participants (41.0%) reported having unstable housing at baseline. Six-month hospitalizations decreased in both the A-ICU and EUC groups, with no difference between them (mean [SE], −0.6 [0.5] vs −0.9 [0.5]; difference, 0.3 [95% CI, −1.0 to 1.5]). Emergency department use did not differ between groups (mean [SE], −2.0 [1.0] vs 0.9 [1.0] visits per person; difference, −1.1 [95% CI, −3.7 to 1.6]). Primary care physician visits increased in the A-ICU group (mean [SE], 4.2 [1.6] vs −2.0 [1.6] per person; difference, 6.1 [95% CI, 1.8 to 10.4]). Patients in the A-ICU group reported improved social functioning (mean [SE], 4.7 [2.0] vs −1.1 [2.0]; difference, 5.8 [95% CI, 0.3 to 11.2]) and self-rated health (mean [SE], 0.7 [0.3] vs −0.2 [0.3]; difference, 1.0 [95% CI, 0.1 to 1.8]) compared with patients in the EUC group. No differences in patient activation or experience were observed. CONCLUSIONS AND RELEVANCE The A-ICU intervention did not change hospital or ED utilization at 6 months but increased PCP visits and improved patient well-being. Longer-term studies are needed to evaluate whether these observed improvements lead to eventual changes in acute care utilization.