CASP, a measure of quality of life in old age : evaluation of its psychometric properties and association with mortality : the HAPIEE study

Recent literature has documented the importance of Control, Autonomy, Self-realisation, and Pleasure (CASP) scores, a measure of generic QOL in older age, in predicting subsequent all-cause mortality among older adults. However, there have been no large scale studies of the relationship between quality of life and mortality in the Central and Eastern Europe region and little is known about the potential mechanisms underlying this relationship. The main purpose of this study was to examine the predictive ability of CASP12v3 score for mortality in population samples from three countries from Central and Eastern Europe (CEE) and the former Soviet Union (FSU) (the Czech Republic, Russia, and Poland). Furthermore, this study aimed to explore the possible causal mechanisms through which quality of life may affect mortality risk. In order to fulfil the main purpose of the study, the psychometric properties of different CASP versions were evaluated. Various factors that influence the quality of life in early old age were also identified using the validated CASP scale. Data from the prospective population-based HAPIEE (Health, Alcohol and Psychosocial factors in Eastern Europe) study was used. At the baseline survey, between 2002-2005, 28,947 men and women aged 45-69 years were recruited from Novosibirsk (Russia), Krakow (Poland) and seven Czech towns. Quality of life was assessed using CASP12v3. The analytic sample comprised 11,476 retired individuals aged 50 and older who completed the retirement questionnaire, including the questions on quality of life (Czech Republic n=2,742; Russia n= 3,804; Poland n=4,930). The main outcome variable was all-cause mortality. Deaths in the three cohorts were ascertained using local and national death registers. The results of psychometric analyses have demonstrated that the 12-item version of CASP (CASP12v3) is a valid and reliable tool for assessing QOL in the CEE/FSU populations. Kaplan–Meier survival curves and log-rank test were used to compare the country-specific CASP12v3 tertile categories, and Cox proportional hazards regression was used to model the associations between CASP score and the risk of death after adjusting for a variety of possible covariates. Using structural equation modeling, mediation analyses were performed to quantify the direct and indirect effects of CASP on all-cause mortality through behavioral, physical health, and social network pathways. There was a significant graded relationship between CASP12v3 and all-cause mortality (P < 0.001). Compared with participants with a high QOL (tertile 1), participants with low QOL (tertile 3) had significantly a higher risk of death (age- and sex-adjusted HR: 2.60, 95% confidence interval (CI): 1.87 to 3.60 in Czech Republic; HR: 2.00, 95% CI: 1.63 to 2.45 in Russia; HR: 1.98, 95% CI: 1.61 to 2.44 in Poland. All these associations remained significant albeit somewhat attenuated after adjustment for socio-demographic variables, health behaviors and depressive symptoms (HR: 1.75, 95% CI: 1.21 to 2.51 in Czech Republic; HR: 1.70, 95% CI: 1.35 to 2.13 in Russia; HR: 1.43, 95% CI: 1.08 to 1.89 in Poland). However, when physical health variables were incorporated in the successive Cox regression model for further adjustment, the impact of low quality of life on all-cause mortality attenuated to statistical non-significance. Mediation analyses confirmed that low quality of life at baseline has a significant negative direct effect on survival. Moreover, physical health partially mediated the effect of baseline CASP12v3 score on mortality in the Czech Republic and Poland, while full mediation was indicated in the Russian data. Among Polish older adults, indirect effect of CASP12v3 on mortality risk was also mediated through frequency of contact with family. These results indicate that the inverse association between CASP12v3 score and mortality risk is mediated by self-rated physical health status and frequency of contact with family, such that the impact of CASP12v3 score beyond these risk factors was less important. Therefore, strategies for improving survival at older ages should focus on control of these variables in the mediating process, particularly declines in physical health.