Reproducibility and validity of the Dutch Life Habits Questionnaire (LIFE-H 3.0) in older adults

Objective: To examine the reproducibility, the discriminant and convergent validity and feasibility of the Dutch translation of the self-administered Life Habits Questionnaire (LIFE-H). Design: Three cross-sectional community-based studies on clinimetric properties of a measurement instrument. Subjects: Older adults (n = 85) with functional limitations due to various chronic illnesses and healthy older adults (n = 40). Measurement protocol: Participants of the reproducibility study (n = 35) filled out the LIFE-H twice in a two-week time period. In the discriminant validity study (n = 120), LIFE-H scores of healthy and ill subjects were compared. In the convergent validity study (n = 63), correlations were examined between LIFE-H, the Impact on Participation and Autonomy questionnaire and the London Handicap Scale. Results: The test—retest reliability showed a satisfactory intraclass correlation coefficient for the total overall score (0.80) but not for the categories scores. The discriminant validity study showed significant differences between the healthy and ill subjects for the 10 separate categories (P < 0.01) and the total score (P < 0.001). The correlations between the LIFE-H categories and total scores and the Impact on Participation and Autonomy Questionnaire (0.80—0.82) and London Handicap Scale (0.89—0.92) were strong. Feasibility testing showed that the subjects experienced difficulties due to the long and fairly complex instructions and structure of the LIFE-H. Conclusions: The clinimetric properties of the LIFE-H were moderate to good. The validity of LIFE-H was as good as the validity of the Impact on Autonomy and Participation and the London Handicap Scale, but the latter questionnaires were shorter and much easier to administer.