Research That Matters (January 17 - 20, 2008)

Purpose Each day more than 1000 Americans die in nursing homes (NHs), and approximately 500 die in residential care/assisted living (RC/AL) facilities. Annual mortality rates range from 14% in RC/AL facilities to 34% in NHs. Yet, prior to the current study, the dying experience across this range of long-term care (LTC) settings has remained largely unexamined. One reason for this gap in knowledge is the lack of empirically designed and psychometrically sound instruments to examine the dying experience in LTC. This presentation describes the development of such a measure using mixed methodology to answer two research questions: 1) What are the domains of a good death for residents who die in LTC settings? and 2) How can they be measured? The presentation also will recount the methodological and conceptual challenges of end-of-life research and the suitability of mixed methods approaches for overcoming those challenges. Methods The project used a concurrent, triangulated mixed methods strategy, such that the quantitative and qualitative components were conducted simultaneously and results compared upon completion. Quantitative Component: Participants were family (n = 439) and staff (n = 332) caregivers of 633 decedents from a stratified random sample of 117 RC/AL facilities and 31 NHs in four states (FL, MD, NC, NJ). Trained interviewers asked respondents, post-death, to rate statements describing potentially important aspects of the quality of dying using a 5-point Likert scale. We tested 36 candidate items using two levels of exploratory factor analysis with communalities set to unity and multiple criteria to determine factor structure and interpretability of two versions of a new measure, the Quality of Dying in Long-Term Care (QOD-LTC and QOD-LTC-C). Qualitative Component: We conducted this component to confirm/disconfirm the domains of a good death identified in the quantitative component. It consisted of ten homogeneous focus groups drawn from a purposive sample of LTC residents (2 groups; n = 11); family caregivers (2 groups; n = 19); paraprofessional staff (3 groups; n = 20); and licensed staff (3 groups; n = 15) from five NHs and eight RC/AL facilities that participated in the quantitative study. Groups were conducted by a trained facilitator using a semi-structured interview guide. Data were analyzed using grounded theory techniques by five analysts in order to elicit manifest and latent themes. Results The resulting 11-item QOD-LTC (alpha = .66), appropriate for surrogate respondents for all LTC decedents, consisted of three domains (i.e., personhood, closure, preparatory tasks). Five domains (i.e., purpose, closure, control, social connection, preparatory tasks) comprised the 23-item QOD-LTC-C (alpha = .85) for surrogate respondents of cognitively intact decedents. The qualitative portion of the study confirmed that these domains are appropriate for LTC and produced four additional themes. Implications The QOD-LTC is a psychometrically acceptable measure of the quality of the dying experience, developed for and tested in LTC settings, and conceptually validated by diverse focus groups . The QOD-LTC constitutes a tool with which to measure and enhance care for this vulnerable and rapidly growing population. Focus group findings also suggest areas for future research.