French cross-cultural adaptation and validation of the Quality of Life-Alzheimer's Disease scale in Nursing Homes (QOL-AD NH)

Abstract

Background: No specific scale to measure Quality of Life in Alzheimer's Disease in Nursing Homes (QoL-AD NH) exists in French. We aimed to translate and culturally adapt the QoL-AD NH participant scale into a French version and evaluate its psychometric properties with residents in French nursing homes (EHPAD). Methods: First, the QoL-AD NH was cross-culturally adapted into French according to guidelines. Secondly, a convenience group of residents with mild to moderate dementia answered the Folstein’s test and the QoL-AD NH. They also answered the Dementia Quality of Life and the Geriatric Depression Scale to test convergent and divergent validity. Known-group validity was tested with a comparison group of residents without dementia. Exploratory Structural Equation Modeling (ESEM) was used after Exploratory Factor Analysis (EFA) to identify factors and measure invariance across age and mental state groups. Reliability (internal consistency, McDonald’s omega and test–retest) were also measured. Results: Following successful adaptation of the QoL-AD NH, 174 residents (mean age 86.6) from 7 nursing homes with mild to moderate dementia participated in the validation study. We retained a 3-factor model of the scale after ESEM identifying: “Intra & interpersonal environment-related QoL”, “Self-functioning-related QoL” and “Perceived current health-related QoL” that were invariant across age and mental state groups. The QoL-AD NH had acceptable convergent (ρ range 0.24–0.53) and divergent validity (ρ range − 0.43 to − 0.57) and good known-group validity with 33 residents without dementia (t(205) = 2.70, p =.007). For reliability, the results revealed very good and adequate internal consistency (α = 0.86 for total scale and ≥ 0.71 for subscales). All total omega values exceeded the threshold 0.70. The hierarchical omega was 0.50, supporting the multidimensionality of the scale. Hierarchical omega subscale values exceeded the minimal level 0.50 except for the third factor, although reliable, would deserve more items. Test–retest was good with ICC (3,1) = 0.76. Conclusions: The QoL-AD NH French participant version has globally good reliability and validity for evaluating residents' quality of life. However, further studies must rework and confirm the factor structure, test sensitivity to change and responsiveness.