A Comprehensive Geriatric Intervention Reduces Short‐ and Long‐term Mortality in Older People With Hip Fracture

To the Editor: In their article recently published in the Journal of the American Geriatrics Society,1 Vidan et al. demonstrated that an early multidisciplinary geriatric intervention reduces in-hospital mortality in elderly patients with hip fracture, without a significant effect on length of hospital stay. We would like to contribute to this topic, presenting data from a cohort of 819 patients admitted over a 3-year period (2000–02) to the Galliera Hospital (Genoa, Italy) and managed on an alternate annual basis with a traditional approach or an orthogeriatric care model. In 2001, an orthogeriatric service was experimentally introduced in the orthopedic unit. The service provided a comprehensive multidisciplinary evaluation of elderly patients with hip fracture at admission and daily intervention for medical management. A geriatrician was responsible for medical care, and an orthopedic surgeon was responsible for fracture management, operative decisions, and discharge. The rehabilitation began during the hospital stay and continued after discharge through home services or institutional facilities. The experimental phase ended after 1 year, and only recently was orthogeriatric care included among the permanent services of the hospital. Using administrative data source and medical record review, length of hospital stay and in-hospital mortality of elderly patients admitted with hip fractures in the year before (control group 1) and in the year after (control group 2) the orthogeriatric intervention were compared. Only patients aged 70 and older were considered. Telephone interview and statewide administrative database were used to ascertain 1-year mortality. The three groups were pairwise compared using analysis of variance or chi-square tests. Survival data were analyzed using the Wilcoxon-Gehan statistics in the life-tables option in SPSS, version 13.0 (SPSS Inc., Chicago, IL). The number of subjects included in the three groups (control 1, intervention, control 2) was 272, 252, and 295, respectively. The groups were similar with regard to mean age±standard deviation (84.0±6.8, 83.6±6.6, 84.5±6.7, P=.21), the proportion of men (23.4%, 18.5%, 21.6%, P=.21), and length of stay (21±13 days, 21±11 days, 19±13 days, P=.16). The rate of in-hospital mortality for the intervention group was lower than for control 1 (4.8% vs 9.9%, P=.03) and similar to control 2 (6.8%, P=.34). Long-term mortality is shown in Figure 1. The comparison of survival curves showed a significant difference over time (P=.04), and at the end of follow-up, the survival rate was approximately 10% higher in the intervention group than in the control groups (75% vs 64.7%, P=.01 (control 1), and 66.7%, P=.04 (control 2)). Survival curves of patients in the intervention group and the two control groups defined in the text; P-value represents the global comparison of the three curves. The P-values for the three pairwise comparisons are intervention group versus control 1, P=.01; intervention group versus control 2, P=.04; control 1 versus control 2, P=.55. The current data support findings from Vidan et al.1 that demonstrated a significant reduction in in-hospital mortality with an early intensive geriatric intervention in older people with hip fracture. Even if is not possible to ascertain whether the involvement of the geriatric teams and the organization of the interventions were comparable, it appears feasible to support the hypothesis that the geriatric intervention reduced incidence and produced better clinical management of major in-hospital complications that determine a reduction in early mortality. Moreover, the current study found a long-term survival benefit, whereas Vidan et al. found only a nonsignificant trend, probably because of a smaller number of subjects. It is possible that the geriatric approach may affect 1-year mortality through comprehensive assessment and global intervention in all active problems of elderly patients. Another study2 demonstrated a strong correlation between the presence of active clinical problems upon discharge and mortality. The current study failed to demonstrate a reduction in length of stay even though many patients had a long hospital stay because of waiting to move from the hospital to rehabilitation services (home and institutional). Thus, nonmedical reasons affect total length, and subanalyses would likely be needed to evaluate the influence of geriatric intervention. It is likely that there are increasing data to support combined orthogeriatric care for older people with hip fracture even if not all published reports are consistent,3 although stronger evidence derived from large randomized trials is needed in support of this model of care. Financial Disclosure: This paper had no financial support. Author Contributions: Antonella Barone: data management, interpretation of data, and preparation of manuscript. Andrea Giusti: acquisition and interpretation of data and critical review. Monica Pizzonia and Monica Razzano: acquisition of data and literature search. Ernesto Palummeri: study concept and design. Giulio Pioli: study concept and design, analysis and interpretation of data, preparation of manuscript, and critical review. Sponsor's Role: None.