Assessing older adults’ ability to live independently is necessary to provide the diagnostic and prescriptive information to make informed health-related decisions. These preliminary analyses support the reliability and validity of the DAFS-R with this sample of elders living independently in the community. The internal consistency reliability (Cronbach’s alpha) of the 10 subscales was .68 when removing the writing a check subscale, which had no variation. Although the alpha coefficient was slightly lower than traditional standards, it should be noted that the DAFS-R tests skills from a variety of areas of functioning, including communication, financial, shopping, and medication management. Consequently, we might not expect to see high inter-item agreement. Future studies with a larger sample should be undertaken to test other aspects of reliability, particularly the temporal stability of scores on this measure.
The validity of the DAFS-R was supported by the moderate correlations of the total score and most of the subscales with the concurrently administered MMSE, which screens for CI. The DAFS-R total score also discriminated between those previously identified as having normal and impaired functioning, as measured by the previously administered Rivermead. However, future studies should explore other aspects of validity, particularly the relationship between scores on the subscales related to medication and other measures of medication skills.
There were individuals in our sample whose performance was poor enough to raise concerns about aspects of their daily functioning. Because individuals with MMSE scores less than 20 were excluded from participating in the study, our findings are limited to individuals exhibiting normal to poor memory performance and mild CI. The DAFS was able to differentiate individuals with varying standard profile scores on the RBMT memory performance instrument (normal, poor, and impaired). We would expect that cognitively impaired individuals will score lower on the DAFS-R because individuals with “poor” memory scores had difficulty with the complex tasks.
A comparison of the distributions of the DAFS and DAFS-R ( and ) suggests that the DAFS-R has a more normal distribution and is also relatively challenging, as evidenced by the lack of a ceiling effect. This preliminary evidence suggests that the DAFS-R may provide more information on this population’s functional status at the upper limit, which may be especially important for clinicians wanting to detect slight decrements in activities of daily living that might signal the onset of memory impairment. In addition, we reduced the number of items from 85 to 55, a 38% reduction, and this shorter version of the DAFS would be a more attractive screening measure to administer in a clinical setting.
Although the DAFS-R has fewer items than the original DAFS, the administration of the DAFS-R might take slightly longer than the original DAFS, primarily because the pillbox exercise is somewhat time-consuming. Future studies should compare the two versions to determine if the DAFS-R is markedly longer than the DAFS, because the decision of which version of the tool to use would be informed by the relative differences in administration time.
The limitations of this exploratory pilot study included its small convenience sample of primarily nonimpaired. Future studies should be undertaken with larger, more diverse samples. Such samples would support more complex psychometric analyses, such as confirmatory factor analysis. In addition, information from respondents with a wider range of cognitive abilities would more clearly elucidate the relationship between cognitive functioning and ability to perform activities of daily living. Future studies should also address the sensitivity of the DAFS-R to clinically meaningful changes in cognitive functioning that can affect elders’ ability to carry out activities of daily living such as those measured here.
Although there are a number of self-reports of activities of daily living, there are fewer performance measures, particularly measures that assess higher level daily living skills, including proper medication management. Given the large number of older adults who regularly take multiple prescription medications, deficits in medication management skills can have serious consequences. These are areas that may be affected by subtle changes in cognitive functioning and may signal the onset of mild CI. Consequently, a measure that emphasizes these higher level daily living skills can help providers screen for initial signs of functional decline that can be followed up by more in-depth testing in specific functional areas.