Can we predict length of stay on orthopaedic rehabilitation?

Abstract

Introduction: Currently more than 85,000 hip and knee replacements are undertaken each year in Australia to alleviate pain and disability [1]. There were over 16,500 presentations of fractured Neck of Femur (NOF) in Australia in 2006-07. The rates of joint replacements and NOF fractures are steadily increasing due to population growth and ageing [2], placing substantial demand on postacute rehabilitation care [3-6]. Few Australian studies have examined which variables are associated with hospital length of stay in these patient populations. Aims: We conducted this retrospective study to investigate how pain at rest and on movement (documented on admission), Type of pain medication (slow release, immediate release) used, type of operation (hip/ knee replacement vs neck of femur fracture) , age (<75 vs > 75), gender and admission Functional Inpendence measure (FIM) total score impact on length of Stay (LOS). Methodology: The sample comprised 63 patients, predominantly female (54 or 85.7%), aged from 48-104 years (mean-74). LOS ranged from 2-49 days (mean=17.2). Multiple regression examined all 8 variables simukltaneiously. Contemporary methods , including lasso regression searched for key variables by attempting to shrink regression coefficients towards zero [7]. Lasso regression has been rrecently applied to pain and rehabilitation data. Results: The full model of 8 variables accounted for 69% of variation in LOS, however, lasso regression selected 3 varibales which accounted for 61.75 of variation in LOS. These variables were admission FIM, operation type and pain on movement. Increased admission FIM and knee or hip replacement were associated with shorter LOS, neck or femur fracture and increased pain on movement were associated with longer LOS. Conclusion: Given the increased demand on rehabilitation beds, predicting length of stay using the 3 factors identified in this study may help to optimize resource allocation.