E. Lipschutz1, J. Reichert2, M.A. Horst3, L.L. Perea2, A. Muller4, A. Coots4, A. Ong4, E.F. Reilly4, A. McCall4, A. Sigal5, C.A. Butts5 1Drexel University College Of Medicine, Philadelphia, Pa, USA 2Penn Medicine Lancaster General Health, Division Of Trauma And Acute Care Surgery, Department Of Surgery, Lancaster, PA, USA 3Penn Medicine, Data & Analytics, Philadelphia, PA, USA 4Reading Hospital, Divsion Of Trauma, Acute Care Surgery, And Surgical Critical Care, Department Of Surgery, West Reading, PA, USA 5Reading Hospital, Department Of Emergency Medicine, West Reading, PA, USA
Introduction: As the geriatric population in the United States grows, and more individuals seek assisted living/skilled nursing facilities(ASNF) care, little is known regarding outcomes after ground level fall between ASNF patients versus those who live in a primary residence(PR). We hypothesize patients in ASNF will have a lower injury severity, and lower mortality.
Methods: A retrospective review(Jan 1, 2018-Dec 31, 2022) of the Pennsylvania Trauma Systems Foundation database was performed of all geriatric patients ³65 years old who presented to a trauma center after a ground level fall. 1:1 propensity(ASNF vs. PR) score matching(PSM) was performed for pre-existing baseline characteristics. A mixed effects logistic regression model (accounting for clustering in trauma centers) was constructed with propensity matched subjects to predict mortality along with key covariates.
Results: Of 48,091 database entries, 16,226(PR:n=8113; ASNF:n=8113) patients were compared post PSM. Average age was 85.8±7.7y. ISS(PR:9.08 vs ASNF:9.19;p=0.259) and mortality [PR: 3.8%(n=310) vs ASNF: 3.7%(n=299);p=0.650] were similar between matched pairs. Hospital length of stay (HLOS) [(PR:5.4d vs ASNF:5.1d;p≤0.001)] and ICU length of stay (ICULOS) [(PR:0.83d vs ASNF:0.75d; p≤0.018)] were longer for PR patients. Of those discharged alive, 28.3%(n=2,204) of PR patients were discharged home. 64.2%(n=5,006) PR patients were discharged to ASNF vs 77.9%(n=11,094) ASNF who returned to ASNF;p≤0.001). The mortality logistic regression model can be seen in Table 1.
Conclusion: PR patients had similar injury severity and mortality to ASNF patients. In these propensity matched patients optimized for PR vs ASNF and mortality, PR patients had longer HLOS and ICULOS and were more likely to be discharged to ANSF after a ground level fall. Further prospective studies are warranted to delineate factors that contribute to PR change in disposition and effects on length of stay.