Evaluation of the Caring Ahead: Preparing for End-of-Life With Dementia Questionnaire

Abstract

ObjectivesFamily caregivers of persons with dementia rarely feel prepared for end of life although preparedness predicts outcomes in bereavement. The Caring Ahead: Preparing for End-of-Life With Dementia questionnaire was developed to measure family caregiver death preparedness. The aim of this study was to evaluate questionnaire psychometrics and refine the Caring Ahead questionnaire. DesignA quantitative cross-sectional reliability study design was used to evaluate the questionnaire. Setting/ParticipantsData were collected by mail from 134 English-speaking family caregivers of persons with dementia recruited from more than 50 congregate living facilities in Canada. Thirty-two participants completed a test-retest. MethodsAnalysis of psychometrics included exploratory factor analysis, calculation of correlation with a single-global preparedness item, Cronbach alpha, intraclass correlation coefficient (ICC) over time. ResultsA 4-factor model with 20 items emerged through exploratory factor analysis with principal factors extraction and promax rotation. The revised questionnaire includes 4 factor subscales: Actions (7 items), Dementia Knowledge (5 items), Communication (4 items), and Emotions and Support Needs (4 items). Evidence was demonstrated for concurrent validity (0.44-0.55, P < .001), internal consistency (alpha > 0.7), and reliability (ICCs > 0.7). Lower levels of preparedness were reported for “knowing what the dying process with dementia may be like” and “discussing end-of-life care and preferences with health care providers.” Conclusions and ImplicationsPreliminary evidence for validity and reliability of the refined 20-item Caring Ahead questionnaire suggests the questionnaire may be useful to clinicians and researchers seeking to assess caregivers’ feelings of preparedness, identify specific areas for intervention, and evaluate the effectiveness of caregiver interventions. Additional testing is needed to evaluate predictive validity.