Background:

Preoperative risk stratification and optimization of preoperative care may be helpful in reducing readmission rates after primary total joint arthroplasty. Assessment of the predictive value of individual modifiable risk factors without a tool to assess cumulative risk may not provide proper risk stratification of patients with regard to potential readmissions. As part of a Perioperative Orthopaedic Surgical Home model, we developed a scoring system, the Readmission Risk Assessment Tool (RRAT), which allows for risk stratification in patients undergoing elective primary total joint arthroplasty at our institution. The purpose of this study was to analyze the relationship between the RRAT score and readmission after primary hip or knee arthroplasty.

Methods:

The RRAT, which is scored incrementally on the basis of the number and severity of modifiable comorbidities, was used to generate readmission scores for a cohort of 207 readmitted patients and two cohorts (one random and one age-matched) of 234 non-readmitted patients each. Regression analysis was performed to assess the strength of association of individual risk factors and the RRAT score with readmissions. We also calculated the odds and odds ratio (OR) at each RRAT score level to identify patients with relatively higher risk of readmission.

Results:

There were 207 (2.08%) readmissions among 9930 patients over a six-year period (2008 through 2013). Surgical site infection was the most common cause of readmission (ninety-three cases, 45%). The median RRAT scores were 3 (IQR [interquartile range], 1 to 4) and 1 (IQR, 0 to 2) for readmitted and non-readmitted groups, respectively. An RRAT score of ≥3 was significantly associated with higher odds of readmission.

Conclusions:

Population health management, cost-effective care, and optimization of outcomes to maximize value are the new maxims for health-care delivery in the United States. We found that the RRAT score had a significant association with readmission after joint arthroplasty and could potentially be a clinically useful tool for risk mitigation.