Objective: To assess the association and the predictive ability of the Timed Up and Go test (TUG) on the occurrence of falls among people aged 65 and older.
Methods: A systematic English Medline literature search was conducted on November 30, 2009 with no limit of date using the following Medical Subject Heading (MeSH) terms "Aged OR aged, 80 and over" AND "Accidental falls" combined with the terms "Timed Up and Go" OR "Get Up and Go". The search also included the Cochrane library and the reference lists of the retrieved articles.
Results: Of the 92 selected studies, 11 met the selection criteria and were included in the final analysis. Fall rate ranged from 7.5 to 60.0% in the selected studies. The cut-off time separating non-fallers and fallers varied from 10 to 32.6 seconds. All retrospective studies showed a significant positive association between the time taken to perform the TUG and a history of falls with the highest odds ratio (OR) calculated at 42.3 [5.1 - 346.9]. In contrast, only one prospective study found a significant association with the occurrence of future falls. This association with incident falls was lower than in retrospective studies.
Conclusions: Although retrospective studies found that the TUG time performance is associated with a past history of falls, its predictive ability for future falls remains limited. In addition, standardization of testing conditions combined with a control of the significant potential confounders (age, female gender and comorbidities) would provide better information about the TUG predictive value for future falls in older adults.