Association of a modified frailty index with mortality after femoral neck fracture in patients aged 60 years and older

Kushal V. Patel, Kindyle L. Brennan, Michael L. Brennan, Daniel C. Jupiter, Adam Shar, Matthew L. Davis

Research output: Contribution to journalArticle

69 Scopus citations


Background: Frailty, a multidimensional syndrome entailing loss of energy, physical ability, cognition, and health, plays a significant role in elderly morbidity and mortality. No study has examined frailty in relation to mortality after femoral neck fractures in elderly patients. Questions/purposes: We examined the association of a modified frailty index abbreviated from the Canadian Study of Health and Aging Frailty Index to 1- and 2-year mortality rates after a femoral neck fracture. Specifically we examined: (1) Is there an association of a modified frailty index with 1- and 2-year mortality rates in patients aged 60 years and older who sustain a low-energy femoral neck fracture? (2) Do the receiver operating characteristic (ROC) curves indicate that the modified frailty index can be a potential tool predictive of mortality and does a specific modified frailty index value demonstrate increased odds ratio for mortality? (3) Do any of the individual clinical deficits comprising the modified frailty index independently associate with mortality? Methods: We retrospectively reviewed 697 low-energy femoral neck fractures in patients aged 60 years and older at our Level I trauma center from 2005 to 2009. A total of 218 (31%) patients with high-energy or pathologic fracture, postoperative complication including infection or revision surgery, fracture of the contralateral hip, or missing documented mobility status were excluded. The remaining 481 patients, with a mean age of 81.2 years, were included. Mortality data were obtained from a state vital statistics department using date of birth and Social Security numbers. Statistical analysis included unequal variance t-test, Pearson correlation of age and frailty, ROC curves and area under the curve, Hosmer-Lemeshow statistics, and logistic regression models. Results: One-year mortality analysis found the mean modified frailty index was higher in patients who died (4.6 ± 1.8) than in those who lived (3.0 ± 2; p < 0.001), which was maintained in a 2-year mortality analysis (4.4 ± 1.8 versus 3.0 ± 2; p < 0.001). In ROC analysis, the area under the curve was 0.74 and 0.72 for 1- and 2-year mortality, respectively. Patients with a modified frailty index of 4 or greater had an odds ratio of 4.97 for 1-year mortality and an odds ratio of 4.01 for 2-year mortality as compared with patients with less than 4. Logistic regression models demonstrated that the clinical deficits of mobility, respiratory, renal, malignancy, thyroid, and impaired cognition were independently associated with 1- and 2-year mortality. Conclusions: Patients aged 60 years and older sustaining a femoral neck fracture, with a higher modified frailty index, had increased 1- and 2-year mortality rates, and the ROC analysis suggests that this tool may be predictive of mortality. Patients with a modified frailty index of 4 or greater have increased risk for mortality at 1 and 2 years. Clinical deficits of mobility, respiratory, renal, malignancy, thyroid, and impaired cognition also may be independently associated with mortality. The modified frailty index may be a useful tool in predicting mortality, guiding patient and family expectations and elucidating implant/surgery choices. Further prospective studies are necessary to strengthen the predictive power of the index. Level of Evidence: Level IV, prognostic study. See Instructions for Authors for a complete description of levels of evidence.

Original languageEnglish (US)
Pages (from-to)1010-1017
Number of pages8
JournalClinical orthopaedics and related research
Issue number3
StatePublished - Mar 2014


ASJC Scopus subject areas

  • Surgery
  • Orthopedics and Sports Medicine

Cite this