Functional decline in older adults: expanding methods of ascertainment.

Background: There is substantial interest in the prevention of physical disability associated with aging. However, little is known about how to screen for declining function at sufficiently early stages to prevent frank disability. This work provides an initial description and assessment of a new self-report method for measuring function in older adults which complements standard methods and may be sensitive to subtler degrees of disablement than currently employed methods. We operationalize the method and demonstrate that it identifies a meaningful number of people as potentially disabled who are not identified by standard methods; also, we evaluate the method's reliability and validity for identifying subtle functional decline relative to existing methods.

Methods: This is a cross-sectional, observational study of a volunteer population of 231 adults 59 years and older, evaluated in a multidisciplinary laboratory. Self-reported difficulty and dependency in each of 27 tasks of daily life were assessed as the measure of disability. Those without difficulty were evaluated for other functional change, by self-report, defined as either: (a) modification of method, or (b) decreased frequency of task performance, when due to health rather than change in social circumstances. Prevalence, test-retest reliability, construct, and criterion validity were evaluated.

Results: In a study population in which from 1% to 50% of participants reported difficulty with each of 27 tasks, from 2% to 33% reported modification of task performance while not having difficulty. Few participants reported decreased frequency without modification. Persons with task modifications, particularly in mobility tasks, showed intermediate levels of performance on objectively measured tests and an intermediate number of chronic diseases, compared to those reporting difficulty and those reporting neither difficulty nor modification.

Conclusions: Cross-sectional evidence supports the ability of the proposed method to identify meaningful numbers of people with functional decline who were not identified by standard self-report methods. In some cases, these were individuals with functional losses comparable to those who did report difficulty. In other cases, these may have been individuals with earlier functional loss, perhaps representing a preclinical stage of disability. Prospective evaluation will identify whether this method predicts the onset of difficulty or disability, and whether it predicts decline with sufficient precision to identify at-risk individuals in a clinical setting.