Home
Scholarly Works
Validity of Self-Reports in Dementia Research
Journal article

Validity of Self-Reports in Dementia Research

Abstract

Background: Many clinicians and researchers use self-reports from patients. However, these self-reports may not be valid if patients are seriously cognitively impaired. We conducted the following study to determine the mental status level where self-reported data from the Geriatric Depression Scale (GDS) become invalid. Methods: This study was based on data from 1,465 individuals referred to a memory clinic for memory and/or behavior problems. We examined administration and psychometric properties of the GDS in a cross-sectional and longitudinal fashion. We used the standardized version of the Mini-Mental State Examination (MMSE) as a marker for cognitive impairment. Results: Compared to participants with an MMSE score of 29-30, those with scores of 18 or less answered fewer GDS questions (p < .01), and displayed more positive response set bias (p < .01). Comparisons of correlation patterns between GDS scores at time one and at time two for the same respondents indicated that responses of participants with lower MMSE scores could be invalid (p < .01). Discussion: Cognitive impairment may render self-reports invalid once MMSE scores decline below 20. Clinicians and researchers should be aware of this potential pitfall in using and interpreting self-reports.

Authors

Bedard M; Squire L; Minthorn-Biggs M-B; Molloy DW; Dubois S; O'donnell M; Lever JA

Journal

Clinical Gerontologist, Vol. 26, No. 3-4, pp. 155–163

Publisher

Taylor & Francis

Publication Date

June 5, 2003

DOI

10.1300/j018v26n03_13

ISSN

0731-7115

Contact the Experts team