Background. Care fragmentation is a mounting challenge for Veterans, many of whom face multiple chronic conditions requiring care from different providers and clinics. This issue is likely to intensify in light of a policy environment that increasingly encourages Veterans’ dual use of VA and community care. When patient care is dispersed across providers, clinics, and health systems, it increases risks of information loss, medication interactions, and unwieldy treatment regimens, leading to health deterioration. Veterans in poor health and those with functional limitations or resource constraints may be especially vulnerable, yet little is known about fragmentation patterns and care coordination needs in these high-risk patients. Project Objectives. The objective of this study is to advance understanding of care fragmentation and care coordination’s potential among high-risk Veterans with multiple chronic conditions. Project Methods. To achieve the study goals, we will first describe care fragmentation prevalence and variation in a national cohort of high-risk patients (Aim 1). We will then examine the relationships between care fragmentation and outcomes, including acute care utilization (Aim 2.1) and patient-reported care experience (Aim 3.1). Finally we will evaluate whether randomization to comprehensive care coordination mitigates the effects of fragmentation (Aims 2.2 and 3.2). Aim 1. We will use 2014-2015 VA, VA purchased care (including Veterans Access, Choice, and Accountability Act), and Medicare data from a national sample of high-risk patients (Care Assessment Needs hospitalization risk score ≥ 90th percentile). In Aim 1.1, we will describe primary care, outpatient care, and health system fragmentation, using validated and adapted measures of care fragmentation and discontinuity. Using a theoretical approach to compare measures, we will select a primary fragmentation measure within each domain for use in subsequent aims. In Aim 1.2, we will use these select measures to examine variation in care fragmentation by patient characteristics (e.g., age, sex, chronic conditions, housing instability, distance from VA, and enrollment in care coordination programs). Aim 2. Using select fragmentation measures identified in Aim 1.1, we will evaluate the relationship between care fragmentation and acute care utilization among high-risk Veterans. In Aim 2.1, we will determine whether care fragmentation is independently associated with higher rates of acute care utilization (including emergency department visits and hospitalizations). In Aim 2.2, we will take advantage of an ongoing randomized evaluation of an Office of Primary Care PACT-Intensive Management (PIM) Demonstration Program to test whether comprehensive care coordination influences the relationship between fragmentation and acute care outcomes. Aim 3. In Aim 3, we will analyze findings from an operations-supported survey administered to high-risk patients in the PIM De...