| Summary: | Abstract Background: Falls are a major public health burden. Most falls occur during dynamic tasks such as walking which sometimes involve elements of dual tasking. Previous research has shown conflicting results regarding the validity of dual task tests in predicting falls in community-dwelling older adults. Aim: to assess the validity of ecologically valid, simple and clinically applicable dual task tests in predicting future falls in community-dwelling older adults. Methods: Mixed qualitative and quantitative research designs were used. A focus group study was conducted to identify ecologically valid dual tasks that older adults perceive as difficult and fall inducing. This was followed by a pilot study to assess differences between fallers and non-fallers on a large variety of simple and clinically applicable dual task tests, identified from the focus groups and from the literature (n = 49). A test- retest reliability study was conducted to assess the reliability of the tests identified through the pilot study in a sample of fallers and non-fallers. The difference between fallers and non-fallers on these dual task tests was assessed in a larger sample to validate the results of the pilot study. The validity of these tests to predict future falls was assessed in a six month prospective cohort study. Key fmdings and their implications: This project identified new tasks that predicted falls within the limitations of the small sample size. It was possible to predict falls using simple ecologically valid dual task tests that can be easily applied in a clinical environment and only require the use of simple equipment such as a stopwatch. Primary tasks were reliable in single task and dual task conditions. However, the proportionate difference between the two conditions was not sufficiently reliable. The secondary tasks were less reliable than the primary tasks. The proportionate difference in primary task performance was also not predictive of falls in older adults. The absolute difference in primary task performance was better at predicting falls, but only for one task. This might indicate that the proportionate difference between single and dual task performance is not useful in assessing the risk of falling. An assessment battery of the Timed Up and Go test in single and dual task conditions, avoiding a moving obstacle in single task and a triple task test might be valuable in assessing falls risk. Complex II single tasks and concern about falling, measured usmg the Falls Efficacy Scale- International, had similar predictive validity to dual task tests, which suggests that dual task tests might not have added value over simpler tasks and questionnaires in predicting falls. These finding should be viewed in the constraints of the small sample size and the multiple comparisons conducted.
|
|---|
