A Bayesian Structural Equation Model to Predict Quality of Life in European Older Adults

TL;DR

This study uses a Bayesian model to show how loneliness, frailty, and pain affect different aspects of quality of life in older Europeans.

Contribution

A novel Bayesian structural equation model is applied to identify specific predictors of quality of life dimensions in older adults.

Findings

Loneliness significantly affects control, autonomy, and pleasure dimensions of quality of life.

Frailty is the main predictor for the self-realization dimension of quality of life.

Pain, loneliness, and frailty have unequal impacts on different quality of life dimensions.

Abstract

The ultimate goal of developed societies is to age with quality of life, but this concept is broad, and few studies have addressed which variables specifically affect its dimensions. This study aims to test a model for predicting the impact of frailty, pain, and loneliness on the four dimensions of quality of life (control, autonomy, pleasure, and self-realization). Data were drawn from a sample of 61,355 Europeans from the Wave 7 of the SHARE project, aged 60 or older (M = 71.87, SD = 8.22), from which 55.9% were females. Statistical analyses included a fully Bayesian structural equation model that evidences the significant effect of loneliness, frailty, and pain on the four dimensions of quality of life. The variables have an unequal impact on the different dimensions’ loneliness was the main predictor for the dimensions control, autonomy, and pleasure, while frailty was the main predictor to self-realization. This study supports the need to address these variables to intervene on the different aspects of quality of life in old age.

Figures