A Tool to Assess Participation in People With COPD: Validation of the Late Life Disability Instrument.

Abstract

Participation in life roles is a critical patient-centered health outcome associated with morbidity and mortality in older adults, but it is not measured routinely in people with COPD. We aimed to validate a participation measure, the Late Life Disability Instrument (LLDI), in people with COPD. To what extent does the LLDI demonstrate test-retest measurement error and reliability, internal consistency, construct and face validity, and floor or ceiling effects when applied to people with COPD? In this cross-sectional study, LLDI scores were compared with scores on measures of theoretically related constructs and between groups based on symptom severity, prognosis, and frailty. A subsample (n = 36) completed the LLDI a second time over the phone within one week. Participants and health-care professionals were asked about the relevance, comprehensiveness, and comprehensibility of the LLDI. Floor and ceiling effects were explored, and the internal consistency (Cronbach's α) of the LLDI was calculated. Ninety-six older adults with COPD participated. The frequency and limitation domains of the LLDI showed excellent test-retest reliability (two-way random effect intraclass correlation coefficient, 0.90 [standard error of measurement, 1.74 points] and 0.90 [standard error of measurement, 3.16 points], respectively). Both domains showed fair correlations with physical function, depression, and quality of life (r = 0.38-0.59). The relationship with anxiety was poor for the LLDI frequency domain (r = -0.21) and fair for LLDI limitation domain (r = -0.45). Both domains discriminated between people with different symptom severity, prognosis, and frailty (P ≤ .026). Neither domain showed floor or ceiling effects, and Cronbach's α was 0.69 and 0.91 for the LLDI frequency and limitation domains, respectively. All healthcare professionals and most participants agreed that the LLDI measures participation (79%) and that the items were relevant (81%). The LLDI shows test-retest reliability, internal consistency, and construct and face validity in people with COPD. The LLDI can be used to assess participation in this population.