New Predictive Rehab Modeling Data From Rothman
Elizabeth Hofheinz, M.P.H., M.Ed. • Fri, January 12th, 2018
Elderly female non-Caucasians…plan for the unexpected, says a new study.
Research from Rothman Institute in Philadelphia indicates that certain patients are more likely to have an unanticipated stay in an inpatient rehab or skilled nursing facility (SNF).
The study, “Who Goes to Inpatient Rehabilitation or Skilled Nursing Facilities Unexpectedly Following Total Knee Arthroplasty?” was published in the December 20, 2017 edition of The Journal of Arthroplasty.
Alexander J. Rondon M.D., M.B.A., research fellow at Rothman Institute in Philadelphia and co-author on the study, told OTW, “Recently, the trend has shifted away from sending patients to inpatient rehabilitation facilities after knee replacement due to worse outcomes and higher costs.”
“Despite these reasons, there are still patients who may benefit from these facilities. We set out to identify these patients along with the risk factors for non-routine discharge to a rehabilitation facility.”
“This was a large retrospective review of a single institution cohort of patients where the routine standard of care was to discharge to home in over 95% of cases. This factor enabled us to investigate this topic in greater depth than had been previously reported in the literature. Furthermore, we were able to report patient risk factors and create pre and postoperative predictive models for patient disposition.”
“We examined over 40 variables and found 6 significant pre-operative risk factors for a discharge disposition other than home.”
“In descending order, age 75 or greater, female, non-Caucasian race, Medicare status, history of depression, and Charlson Comorbidity Index were predictors for patients going to inpatient rehabilitation facilities.”
“We also found the presence of in-hospital complications led to a higher likelihood of being discharged to these facilities. Both our pre and postoperative predictive models had excellent performance with area under curve values of approximately 0.80.”
“Use of our novel predictive models will help orthopedic surgeons better identify patients likely to require inpatient rehabilitation following their knee replacement. Given the well-known importance of preoperative patient education, our results will help surgeons counsel patients their likely postoperative journey and discharge disposition.”
“This study identifies pre- and postoperative risk factors that predispose patients to non-routine discharges, which allow surgeons to better predict patient post-operative disposition.”