Abstract
BACKGROUND: The Risk Assessment Prediction Tool (RAPT) is a validated 6-question survey designed to predict primary total joint arthroplasty (TJA) patients' discharge disposition. It is scored from 1 to 12 with patients stratified into high-, intermediate-, and low-risk groups. Given recent advancements in rapid-discharge protocols and increasing utilization of home services, the RAPT score may require modified scoring cutoffs. METHODS: A retrospective chart review of all patients undergoing primary TJA at a single academic center over 14 months was performed. The RAPT score was implemented during the sixth month. Patients undergoing revision TJA, complex TJA, and TJA after resection of malignancy were excluded. Outcomes before and after RAPT implementation were analyzed with additional subanalysis investigating of post-RAPT data. RESULTS: A total of 1264 patients (624 Pre-RAPT and 640 Post-RAPT) were evaluated. The post-RAPT group (245 total hip arthroplasty and 395 total knee arthroplasty) experienced significant decreases in mean hospital length of stay (2.22 days pre-RAPT to 1.82 days post-RAPT, P < .001) and the proportion of patients discharged to facility (21.8% pre-RAPT to 15.2% post-RAPT, P = .002). The modified system demonstrated the highest overall predictive accuracy at 92% and was found to be predictive of hospital length of stay. CONCLUSION: Owing to the recent trends favoring in-home services over rehab facility after discharge, previously published RAPT scoring cutoffs are inaccurate for modern practice. Using mRAPT cutoffs maximizes the number of patients for whom a discharge prediction can be made, while maintaining excellent predictive accuracy.