Postoperative Length of Stay May Predict Foot and Ankle Patients at Risk for Readmission

Abstract

Category: Other; Ankle; Ankle Arthritis; Arthroscopy; Diabetes; Hindfoot; Lesser Toes; Midfoot/Forefoot; Trauma Introduction/Purpose: Reduction of postoperative readmissions is an important goal for the improvement of patient outcomes and satisfaction, as well as mitigating rising healthcare costs. As a result of changing federal and insurer guidelines and requirements, hospitals monitor and report 90-day all-cause readmissions. This data can be used to drive both local and national improvements in patient care. The purpose of our study is to evaluate patient and perioperative factors associated with 90-day readmissions specifically in a foot and ankle surgery population. Methods: After obtaining Institutional Review Board approval, admissions and readmissions data from October 2012 to September 2016 at a Level 1, private, academic hospital were reviewed for patients treated surgically by four foot and ankle orthopaedic surgeons. Only foot and ankle surgical procedures were included, as identified by diagnosis codes (ICD-9/ICD-10), performed on patients from 10-95 years of age. All-cause readmission rate was recorded. A medical records review was performed to collect patient demographics, admissions, readmissions, and surgical data. Data was analyzed using an analysis of variance model with a Dunn’s post-hoc test as well as Spearman rank order correlation test, with significance set at α=0.05. Results: 6,054 patients were identified, and 4,578 met study inclusion criteria; 4,404 cases did not result in readmission and 174 cases resulted in readmission within 90 days of discharge of the index inpatient/outpatient case, leading to an overall, all-cause readmission rate of 3.80%. Between non-readmission and readmission cohorts, there were no significant differences in patient age (p=0.627) or sex (p=0.208; female: male ratio in both cohorts ˜ 60% v. 40%); however, there was a significant difference in BMI (29.8 v. 31.9, respectively; p<0.001). Readmitted patients had significantly greater lengths of stay (LOS) of index procedure (3.79 v. 2.09 average days, respectively; p<0.001). Readmitted patients had a statistically higher likelihood of diabetes (25.3% vs 14.8%, p<0.002) and history of DVT/PE (23.0% vs 9.3%, p<0.0001) compared to non-readmitted patients. Conclusion: This study identified statistically significant relationships between postoperative LOS, diabetes, and thromboembolic disease history with 90-day readmission rate. Diabetes and thromboembolic history are unmodifiable risk factors that have been shown to increase readmission in other work, while initial LOS has variable relationships with readmission depending on the patient population studied. In our foot and ankle specific population, initial length of stay was nearly doubled in the readmission population. As foot and ankle surgery often leads to decreased mobility and independence, increasing LOS may be useful in predicting patients that will require more social or medical support at and after discharge.