Discharge to the skilled nursing facility: patient risk factors and perioperative outcomes after total knee arthroplasty.

Background: Patients receiving a total knee arthroplasty (TKA) who fail to meet inpatient criteria for outpatient physical therapy or an acute rehabilitation facility are increasingly being discharged to skilled nursing facilities (SNFs). However, in some facilities, postoperative care and therapy may be suboptimal. In order to recognize the circumstances predisposing patients to a SNF discharge and quantify perioperative risks, we used a nationwide TKA database to compare those who were and were not discharged to SNFs with respect to: (I) patient and hospital characteristics; (II) comorbidities; (III) lengths of stay (LOS); and (IV) inpatient complications.
Methods: The National Inpatient Sample database was queried for all individuals who received primary TKA (ICD-9-CM 81.54) between January 1 ^st , 2009 and December 31st, 2013, yielding a total of 3,218,419 patients. Discharge disposition was readily identifiable, and the SNF patients numbered 403,575 (12.5%) vs. 2,814,574 discharged to home or a non-SNF setting (87.5%). A multi-level logistic regression analysis was conducted using patient and hospital specific factors as predictor variables in order to see if differences existed between the two cohorts. A two-tailed P value was set as the threshold for statistical significance.
Results: Patients discharged to SNFs post-TKA were older (mean, 72 vs. 65 years, P<0.001), more often female [odds ratio (OR) 1.74, P<0.001], black (OR 1.246, P<0.001), from the South (OR 1.856, P<0.001), and had various comorbidities including: preexisting psychoses (OR 1.703, P<0.001), history of drug abuse (OR 1.682, P<0.001), neurological disorders (OR 1.359, P<0.001), and depression (OR 1.334, P<0.001). The mean LOS for TKA patients discharged to SNFs was 17% longer (P<0.001). Patients discharged to SNFs were more likely to endure inpatient medical complications (OR 1.3, P<0.001), specifically pulmonary congestion or edema.
Conclusions: Patients discharged to SNF had specific demographic characteristics and risk factors, increased LOS, more frequent inpatient medical complications, and greater hospital costs. Knowledge of these risk factors may be critical from the perspective of the new value-based reimbursement system for orthopaedic surgeon to intervene early and appropriately select the patients likely and capable of completing the rigorous postoperative TKA rehabilitation.
Competing Interests: Conflicts of Interest: The authors have no conflicts of interest to declare.