Psychometric properties of the PAT: an assessment tool for ADL performance of older people living in residential homes.

BACKGROUND

As the world population ages, the number of people with diminished performance on the Activities of Daily Living (ADL) increases. A reliable and valid measure needs to be developed to determine the effects of interventions focused at increasing self-care abilities. We developed the Performance ADL Test (PAT) for this purpose.

OBJECTIVE

The aim of this study was to investigate the reliability and validity of the PAT in older people living in residential homes.

METHODS

The PAT contains 16 test items, covering the entire range of Basic ADL and Instrumental ADL performance in elderly people. For this assessment, 40 older people (mean age of 85 ± 7.5 years) participated. All 40 subjects lived in residential institutions in the Netherlands, were able to walk at least 10 m, could understand instructions spoken in Dutch, and had no cognitive impairment. During the first test session, subjects completed the PAT, the Groningen Activity Restriction Scale (GARS), and performance-based physical fitness tests. Two weeks later, subjects were retested on the PAT.

RESULTS

Factor analysis revealed three subscales: Organization of Performance, Gross Motor Function, and Fine Motor Function. Internal consistency (Cronbach's α) of all scales and subscales ranged from 0.731 to 0.881. Test-retest reliability (intraclass correlation) ranged from 0.316 to 0.950. Paired sample t-tests revealed no significant differences between subject performance obtained during the two test periods. Pearson's correlations between the PAT and the GARS ranged from 0.490 to 0.831, and between the PAT and the fitness tests from 0.317 to 0.781.

CONCLUSION

Although the number of participants was limited (n = 40), the PAT seems to be a useful instrument for assessing ADL performance in older people living in residential homes. In general, internal consistency, test-retest reliability, and validity were satisfactory.