ABSTRACT This project extends the analysis of a randomized controlled trial of the Camden Coalition of Healthcare Providers’ complex care management program, an intervention that aims to reduce spending and improve healthcare quality and health outcomes by targeting “superutilizers” of the healthcare system. The program focuses on patients with chronic conditions, complex needs, and frequent hospitalizations, working with them after discharge to prevent future admissions. The program is motivated by the fact that healthcare costs (which account for almost 18 percent of GDP in the US) are heavily concentrated. It targets a group that accounts for less than two percent of the Camden population but more than a quarter of total spending at Camden hospitals. Our previous randomized controlled trial found that the care coordination program failed to reduce hospital readmissions, but it could not explain why. This follow-up research will address this unanswered question by investigating new outcomes that distinguish between two very different explanations with very different implications: a failure of trial fidelity to the program or of the underlying theory behind the program. Was postdischarge care coordination (such as getting patients to primary care appointments in a timely manner) not achieved or was it insufficient to prevent readmissions? This work is a rare opportunity to conduct a postmortem on a critically important but limited RCT and chart a path forward for improving care for medically and socially complex patients. We leverage rich Medicaid data to investigate these alternative explanations for the lack of program impact on hospital readmissions, expanding the scope of inquiry beyond hospital utilization to shed light on the program’s impact on other important healthcare utilization and health behaviors, which are key elements of the program’s theory of change. The specific aims are motivated by a need to better understand the null findings on hospital readmissions, inform the current debate over their interpretation, and investigate potential impacts of the intervention outside of the hospital setting. We will use the randomized controlled design to investigate the impact of the Camden program on (1) Quantity of outpatient care, including primary care, specialist care, prescription drugs, home care, and durable medical equipment; (2) Quality of outpatient care, as proxied by measures of care fragmentation and whether care was received in a timely manner after discharge; and (3) Patient health behaviors, including medication adherence and the stability of enrollment in Medicaid. We will also conduct interviews with program staff to understand implementation challenges, contextualize findings, and generate new hypotheses.