Impact of Diabetes Comorbidity on Health Outcomes in Patients Undergoing Medical Rehabilitation after Lower Extremity Joint Arthroplasty

Date

2013-06-03

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Introduction: Diabetes is one of the rapidly increasing chronic conditions that can impact the health and well-being of individuals. Objectives: To determine the prevalence of diabetes in patients receiving medical rehabilitation after lower extremity joint replacement and to examine the associations between diabetes as comorbidity and outcomes including functional status, likelihood of discharge to acute care, and 90-day hospital readmission. Methods: Secondary analysis of Medicare data. We selected patients who underwent a primary hip/knee joint replacement procedure during 2007-2008. We identified diabetes-related ICD-9CM codes in the Medicare Provider Analysis and Review and Inpatient Rehabilitation Facilities Patient Assessment Instrument data files, and created a three-level diabetes status: no diabetes, non-tier diabetes (controlled diabetes), and tier diabetes (uncontrolled diabetes). The effect of diabetes status on functional status gain was estimated using multivariate regression models. Discharge to acute care (yes/no) after inpatient rehabilitation was compared against discharge to community using multinomial logistic regression. Hospital readmission (yes/no) rates were estimated using Cox regression hazard models. Results: The prevalence of controlled diabetes in the knee and hip joint replacement cohorts was 21% and 17%, respectively; uncontrolled diabetes was identified in 4% and 3% of patients, respectively. The adjusted effect of diabetes status on functional status gain was minimal. The likelihood of discharge to acute care was explained by marital status and discharge functional scores, as compared to diabetes status, in both knee and hip cohorts. Using no diabetes as the reference group, the risk of hospital readmission in the hip cohort was 19% higher for those with controlled diabetes (HR=1.19, 95% CI=1.08-1.30) and 31% higher for those with uncontrolled diabetes (HR=1.31, 95% CI=1.08-1.59). In the knee cohort the risk was 22% higher for those with controlled diabetes (HR=1.22, 95% CI=1.14-1.30) and 43% higher for those with uncontrolled diabetes (HR=1.43, 95% CI=1.26-1.61). Conclusion and Implications: Our findings indicate diabetes is an important comorbid condition across the continuum of care. Strategies to better manage diabetes, both prior to elective procedures such as joint replacement, and throughout the following rehabilitation stay and recovery phases, could improve the overall efficiency and quality of care in this population.

Description

Citation