Patrick Lazarevič1, Martina Brandt2. 1. Austrian Academy of Sciences, Vienna Institute of Demography, Vordere Zollamtsstraße 3, 1030, Vienna, Austria. Electronic address: Patrick.Lazarevic@oeaw.ac.at. 2. TU Dortmund, Institute for Sociology, Emil-Figge-Str. 50, 44227, Dortmund, Germany. Electronic address: martina.brandt@tu-dortmund.de.
Abstract
BACKGROUND: Self-rated health (SRH) is arguably the most widely used generic health measurement in survey research. However, SRH remains a black box for researchers. In our paper, we want to gain a better understanding of SRH by identifying its determinants, quantifying the contribution of different health domains to explain SRH, and by exploring the moderating role of gender, age groups, and the country of residence. METHOD: Using data from 61,365 participants of the fifth wave (2013) of the Survey of Health, Ageing and Retirement in Europe (SHARE) living in fifteen European countries, we explain SRH via linear regression models. The independent variables are grouped into five health domains: functioning, diseases, pain, mental health, and behavior. Via dominance analysis, we focus on their individual contribution to explaining SRH and compare these contributions across gender, three age groups, and fifteen European countries. RESULTS: Our model explains SRH rather well (R2 = .51 for females/.48 for males) with functioning contributing most to the appraisal (.20/.18). Diseases were the second most relevant health dimension (.14/.16) followed by pain (.08/.07) and mental health (.07/.06). Health behavior (.02/.01) was less relevant for health ratings. This ranking held true for almost all countries with only little variance overall. A comparison of age groups indicated that the contribution of diseases and behavior to SRH decreased over the life-course while the contribution of functioning to R2 increased. CONCLUSION: Our paper demonstrates that SRH is largely based on diverse health information with functioning and diseases being most important. However, there is still room for idiosyncrasies or even bias.
BACKGROUND: Self-rated health (SRH) is arguably the most widely used generic health measurement in survey research. However, SRH remains a black box for researchers. In our paper, we want to gain a better understanding of SRH by identifying its determinants, quantifying the contribution of different health domains to explain SRH, and by exploring the moderating role of gender, age groups, and the country of residence. METHOD: Using data from 61,365 participants of the fifth wave (2013) of the Survey of Health, Ageing and Retirement in Europe (SHARE) living in fifteen European countries, we explain SRH via linear regression models. The independent variables are grouped into five health domains: functioning, diseases, pain, mental health, and behavior. Via dominance analysis, we focus on their individual contribution to explaining SRH and compare these contributions across gender, three age groups, and fifteen European countries. RESULTS: Our model explains SRH rather well (R2 = .51 for females/.48 for males) with functioning contributing most to the appraisal (.20/.18). Diseases were the second most relevant health dimension (.14/.16) followed by pain (.08/.07) and mental health (.07/.06). Health behavior (.02/.01) was less relevant for health ratings. This ranking held true for almost all countries with only little variance overall. A comparison of age groups indicated that the contribution of diseases and behavior to SRH decreased over the life-course while the contribution of functioning to R2 increased. CONCLUSION: Our paper demonstrates that SRH is largely based on diverse health information with functioning and diseases being most important. However, there is still room for idiosyncrasies or even bias.
Keywords:
Cross-national comparison; Epidemiology; Europe; Measurement invariance; Response behavior; Self-rated health; Survey of Health, Ageing and Retirement in Europe (SHARE)