Comanagement of geriatric patients with hip fractures: a retrospective, controlled, cohort study

Abstract

The objective of this 3-year retrospective, controlled, cohort study is to characterize an interdisciplinary method of managing geriatric patients with hip fracture. All patients aged 65 years or older admitted to a single academic level I trauma center during a 3-year period with an isolated hip fracture were included as participants for this study. Thirty-one geriatric patients with hip fracture were treated with historical methods of care (cohort 1). The comparison group of 115 similar patients was treated under a newly developed, institutional comanagement hip fracture protocol (cohort 2). There were no differences in age, sex distribution, or comorbidity distribution between the 2 cohorts. Patients requiring intensive care unit (ICU) admission decreased significantly from 48% in cohort 1 to 23% in cohort 2 (P = .0091). Length of ICU stay for patients requiring ICU admission also decreased significantly, from a mean of 8.1 days in cohort 1 to 1.8 days in cohort 2 (P = .024). Total hospital stay decreased significantly, from a mean of 9.9 days in cohort 1 to 7.1 days in cohort 2 (P = .021). Although no decrease in in-hospital mortality rates was noted from cohort 1 to cohort 2, a trend toward decreased 1-year mortality rates was seen after implementation of the hip fracture protocol. Hospital charges decreased significantly, from US$52 323 per patient in cohort 1 to US$38 586 in cohort 2 (P = .0183). Implementation of a comanagement protocol for care of geriatric patients with hip fracture, consisting of admission to a geriatric primary care service, standardized perioperative assessment regimens, expeditious surgical treatment, and continued primary geriatric care postoperatively, resulted in reductions in lengths of stay, ICU admissions, and hospital costs per patient. On an annualized basis, this represented a savings of over US$700 000 for our institution.

Keywords: comanagement; geriatric; hip fracture.

Figure 1.

Box and whisker plot of intensive care unit (ICU) days and hospital lengths of stay (LOS, in days) for patients with hip fracture preprotocol (“pre”, cohort 1) and after protocol implementation (“protocol”, cohort 2). Median, upper median, and lower median values are plotted with standard deviation bars. Outliers are indicated by dots. Hospital ICU stays and total LOS decreased significantly after implementation of protocol. Analysis of variance,*P = .024, †P = .021.