Inequalities in Health : the Importance of Material/Structural Factors and Psychosocial Resources

Abstract: Socioeconomic inequalities in health are well-documented in most countries. Health differences have been shown to follow a gradient, where health status in average is somewhat poorer for every lower level in the social hierarchy. Notably, the welfare state Sweden is no exception. Considerable socioeconomic inequalities in health are found, and the magnitude of e.g. educational inequalities in health has even increased over recent decades. Important contributors, or “health determinants” of observed health inequalities, include material/structural factors, behavioral factors and psychosocial factors. The inequalities arise from uneven distributions of these health determinants accumulated over the life course. Whereas earlier research has focused on independent effects of different determinants, recent research has showed that health determinants interact in complex ways when contributing to health inequalities. However, the relative importance of the independent contributions of specific types of health determinants and of the shared contributions have not been assessed. The overall aim of this thesis was to examine possible explanations of inequalities in self-reported health among groups with different educational levels in a Swedish population, in particular how material/structural factors and psychosocial resources contribute to these inequalities. This thesis is based on four population-based studies. Studies I and IV used data from the Life & health study, based on crosssectional survey questionnaires, conducted in 2000, 2004 and 2008 in five counties in the central part of Sweden. Around 35,000 respondents were included each year with response rates varying from 60% to 67%. Studies II and III used data from a sub-sample of the Swedish national public health survey (HLV) from 2012, another cross-sectional survey questionnaire. The sub-sample was carried out in four counties in the central part of Sweden. The total number of respondents in the sub-sample was 26,706, with a response rate of 53%. Outcome variables were, in studies I, II and IV, self-rated health (SRH) and, in study III, psychological distress. The magnitudes of health inequalities were examined using rate ratios and rate differences. The associations between health determinants and health out-comes were examined using logistic regression, and the analysis of independent versus shared contributions of health determinants to health inequalities was conducted using structural equation modeling (SEM). Study I showed that relative educational inequalities in SRH were two-fold among men, unchanged from 2000 to 2008, while the inequalities initially were smaller among women but increased over time, from 1.7 to 2.1. This increase was mainly due to growing inequalities in the age group 25–34 years. The distributions of all observed health determinants were more unfavorable in low education groups; most prominent for lack of a financial buffer, smoking and low optimism. These educational differences were, for most health determinants, unchanged over time. Study II examined the association of adult SRH with adverse material conditions (eco-nomic stress in the family) in childhood as well as adverse psychosocial conditions (condescending treatment) in childhood. Both economic stress and condescending treatment in childhood were strongly associated with poorer adult SRH. These associations were attenuated, but still statistically significant after adjustment for economic stress and condescending treatment in adulthood and other risk factors. Study III showed, after adjustment for age, economic difficulties, employment status and social support, a moderate association between educational level and psychological dis-tress, where low and medium educational levels were related to a lower risk of psychological distress. However, current economic difficulties showed a strong, and positive, association with psychological distress. Study IV showed that the shared pathways, including both material/structural fac-tors (e.g. financial buffer and unemployment) and psychosocial resources (e.g. optimism and social participation), explained about 40% of educational inequalities in SRH for both men and women aged 25–74 years. The pathways including only the independent effects of psychosocial resources (14% in men and 20% in women) or material/structural factors (9% and 18%, respectively) explained substantial but smaller proportions of the inequalities. In conclusion, in an adult population in the central part of Sweden, prevalence of poor SRH was, among men, twice as high in the low education group compared to the high education group during the first decade of the new millennium. Among women, educational inequalities were initially smaller, but increased over time to the same level as among men. However, when using self-reported psychological distress as health outcome, no corresponding educational inequalities were found. Instead, economic difficulties were an important determinant of psychological distress. SRH in adulthood was significantly associated with economic stress and condescending treatment during childhood, also when the same conditions in adulthood were taken into account. Material/structural factors and psychosocial resources explained more than half of the educational inequalities in SRH, and the majority of this contribution was in the form of a shared effect of material/structural factors and psychosocial resources. A shared effect means that a material/structural factor and a psychosocial resource are strongly associated, and that the combination of the two has an effect on the educational inequalities in SRH. Therefore, to reduce educational inequalities in SRH, interventions need to address both material/structural conditions and psychosocial resources. This needs to be done across educational groups, using a life course perspective, but with more intensive interventions in lower education groups.