Comparative Effectiveness of Functional Tests in Fall Prediction After Hip Fracture

Abstract

ObjectivesTo assess the validity of 4 functional tests in predicting falls within the first year after hip fracture. DesignProspective study of functional tests shortly after hip surgery and incident falls during 12 months' follow-up. Setting and ParticipantsThe sample comprised 173 adults with acute hip fracture, aged 65 years and older (79% women, 77% community dwelling, mean age 84.2 years), who participated in a clinical trial of vitamin D or home exercise. MethodsWe assessed 4 functional tests [Timed Up and Go test (TUG), grip strength, and knee flexor and extensor strength in the nonoperated leg] by trained study physiotherapists at baseline (1-12 days after hip fracture surgery). During 12 months' follow-up, we ascertained all fall events by monthly personal phone calls, a telephone hotline, and a patient diary. Then we compared TUG and strength test performance at baseline between future single fallers, recurrent fallers, and nonfallers over the 12-month follow-up. All analyses adjusted for age, body mass index, gender, 25-hydroxyvitamin D status at baseline, days of follow-up, and treatment allocation (the original trial tested vitamin D treatment and/or a home exercise program). ResultsNinety-two of 173 (53%) participants fell and experienced 212 falls. Participants who became recurrent fallers (n = 54) had significantly longer TUG times at baseline than those who did not fall (n = 81) in the following 12 months (mean TUG for recurrent fallers = 71.6 seconds, SD = 8.2 seconds, vs mean TUG for nonfallers = 51.4 seconds, SD = 6.9 seconds; P = .02). There were no significant differences in TUG times between single fallers and nonfallers. For all 3 strength tests, there were no significant differences between single fallers, recurrent fallers, and nonfallers. Conclusions and ImplicationsIn this population of frail older adults recruited shortly after hip fracture surgery, only the TUG test discriminated between future recurrent fallers and nonfallers over a 12-month follow-up. Because of the high incidence and serious consequences of falls in older adults after a hip fracture, it is very important to identify practical and clinically related tests to predict repeated falls in the first year after a hip fracture, which is of great public health importance.