Association Between Additional Primary Care Services and Hospital Readmission Rates

Research Objectives: To estimate the association between primary care services (PCS) utilization soon after hospital discharge on readmission rates.

Study Design: Data were obtained from the Ingenix Impact Research and Development database. The database includes enrollment, medical, and pharmacy claims data for 53 million members. Patients were selected into the study if they were continuously enrolled and had one or more of the following chronic conditions from January 1, 2005-September 30, 2008: diabetes, coronary artery disease, congestive hearth failure, chronic obstructive pulmonary disease, asthma, depression, or chronic renal failure. Patients were also required to be between the ages of 18 and 64 and have claims data that were linkable to zip code information about socioeconomic factors. The Symmetry Suite of groupers was used to find patients with the conditions of interest and to estimate their overall health status via the episode risk group (ERG) score. The ERG score predicts a patient's future need for health care services and associated costs. PCS use was defined as one or more services provided by professionals in an outpatient or office visit settings after discharge but prior to readmission. Logistic regression analyses were conducted to estimate the relationship between additional PCS use and the likelihood of a readmission within either 7 days or 30 days of discharge, while controlling for patient demographics (age, gender, location, inferred income and race) and health status (obesity and ERG score).

Population Studied: After applying the above criteria, 409,717 patients were selected into the study.

Principle Findings: For those hospitalized, 4.7% were readmitted within 7 days and 11.3% were readmitted within 30 days of hospital discharge. Of those readmitted, only 24% had PCS use after the initial discharge but prior to readmission within 7 days. This number rose to 46% for readmissions within 30 days. The likelihood of readmission was significantly (p<0.01) lower (between 38% - 64% lower, respectively for 7 and 30 days) if there was PCS use between discharge and readmission.

Conclusions: Utilization of primacy care services soon after discharge was associated with significantly lower readmission rates.

Implications for Policy: Policy-makers are interest in reducing readmission rates as a way to improve the quality of care and reduce costs. Increasing the use of relatively inexpensive primary care services soon after hospitalization discharge will help reduce readmission rates and may help accomplish this goal.