Measuring Nursing Home Quality of Life: Validated Measures Are Poorly Correlated With Proxies From MDS and Quality of Life Deficiency Citationsl

Objectives: Ensuring quality of life (QoL) is an important goal of person-centered nursing home care. The provision of person-centered care relies on information captured in the Minimum Data Set 3.0 (MDS). It is unclear to what extent MDS items or QoL-related facility deficiency citations correlate with validated measures of nursing home residents' QoL. This study evaluated correlation among MDS items, facility deficiency citations, and residents’ QoL from 2 states that currently collect these measures. Design: Cross-sectional study. Setting and Participants: A total of 11,487 long-stay residents in 356 facilities in Minnesota and 13,835 long-stay residents in 851 facilities in Ohio in 2015. Methods: The outcome, QoL, was measured using validated instruments (Minnesota QoL survey and Ohio Resident Satisfaction Survey). Predictor variables included scores for Preference Assessment Tool (Section F), Patient Health Questionnaire-9 (Section D) for depressive symptoms from MDS, and count of QoL-related facility deficiency citations from the Certification and Survey Provider Enhanced Reporting database. Spearman's ranked test assessed correlation between predictor and outcome variables. Mixed effects models evaluated associations of QoL summary scores with predictor variables, adjusting for resident- and facility-level characteristics, accounting for clustering at the facility level. Results: In Minnesota and Ohio, predictor variables (Section F and D items, and facility deficiency citations) correlated significantly but poorly with QoL (coefficients ranging from 0.003 to 0.3, P < .001). In the fully adjusted mixed effects model, all predictor variables, demographics, and functional status explained <21% of the total variance in QoL among residents. These findings were consistent in sensitivity analyses stratified by 1-year length of stay and by diagnosis of dementia. Conclusions and Implications: MDS items and facility deficiency citations encapsulate a significant but very small proportion of variance in residents’ QoL. This indicates the need to measure QoL directly among residents, to plan person-centered care, and to evaluate its performance in nursing home facilities.