Fragmented Ambulatory Care and Subsequent Healthcare Utilization Among Medicare Beneficiaries

OBJECTIVES: We sought to determine the associations between fragmented ambulatory care and subsequent emergency department (ED) visits and hospital admissions, while considering possible interactions between fragmentation and number of chronic conditions. STUDY DESIGN: We conducted a cohort study over 3 years among 117,977 fee-for-service Medicare beneficiaries who were attributed to primary care physicians in a 7-county region of New York and had 4 or more ambulatory visits in the baseline year. METHODS: We calculated fragmentation scores using a modified Bice-Boxerman Index and, because scores were skewed, divided them into quintiles. We used Cox regression models to determine associations between fragmentation and ED visits and, separately, hospital admissions, stratifying by number of chronic conditions and adjusting for age, gender, number of ambulatory visits, and case mix. RESULTS: Among those with 1 to 2 or 3 to 4 chronic conditions, having the most (vs the least) fragmented care significantly increased the hazard of an ED visit and, separately, increased the hazard of an admission (adjusted P < .05 for each comparison). Among those with 5 or more chronic conditions, having the most fragmented care significantly increased the hazard of an ED visit but decreased the hazard of an admission (adjusted P < .05 for each comparison). Among those with 0 chronic conditions, having fragmented care was not associated with either outcome. CONCLUSIONS: The relationship between fragmented ambulatory care and subsequent utilization varies with the number of chronic conditions. Beneficiaries with a moderate burden of chronic conditions (1-2 or 3-4) appear to be at highest risk of excess ED visits and admissions due to fragmented care.