reports

Time to readmission among chronically ill community-resident beneficiaries: variations by geographic area and provider type

Baltimore, MD, USA: Centers for Medicare and Medicaid Services.

Schneider Institute for Health Policy, Heller School for Social Policy and Management, Brandeis University, Waltham, MA, USA

This report is focused on the potential role of care continuity in avoiding readmission for chronically ill beneficiaries dwelling in the community. Although the concept of care continuity has been of interest for decades, there is little consensus on the best ways to identify or measure it, particularly in claims data. This report focuses on the single largest provider of evaluation and management (E&M) services for each beneficiary to identify high, medium or low continuity. The sample consists of beneficiaries with a hospitalization in 2004 who were also enrolled in Medicare in 2003. For those readmitted, the average time to readmission was 64 days, and about 48 percent had a readmission within 30 days of discharge, which is consistent with previous research. In Cox proportional hazard modeling, high continuity was associated with reduced readmissions. More specifically, those with high continuity had a 6 percent lower risk of all-cause readmission, controlling for demographic, chronic illness and geographic factors. Results are similar if you consider same-cause readmission or shift from state to Dartmouth Hospital Referral Region as the geographic unit of interest. Interactions between chronic illness or illness severity and care continuity were tested as a final step. Here, there was a significant, but extremely small interaction for medium continuity and the ACG risk adjuster.

Ultimately, these results point to the potential value of care continuity when it comes to reducing readmissions. However, further work is required to understand the best way to identify care continuity, particularly in claims data. The majority provider approach used here does not account for the number of providers seen by any one beneficiary, a limitation that could be addressed in future research. That said, the findings from this study are consistent with previous research on continuity and re-admissions.