Resumen de Economy of prevention: obesity and socioeconomic gradient
Introduction: The figures on prevalence of obesity have shown the existence of a health problem, which have become a global concern in public health. The growing trend followed by obesity in adults and the existence of a socioeconomic gradient in obesity, as well as the possible impact of the economic crises on health, have motivated the development of this Doctoral Thesis, whose main aims are: to evaluate the prevalence of obesity among adults and to measure the socioeconomic gradient in Spain and the Canary Islands, and to analyse the evolution of both the obesity prevalence and the inequalities related to different economic, social and/or territorial dimensions, considering the possible changes as of the arrival of the economic crisis, both in Spain and in the Canary Islands.
Methodology: The anthropometric indicator used to measure the obesity is the body mass index (BMI), which is the dependent variable. Chapter 1 studies the BMI and its socioeconomic disparities using the Spanish National Health Survey (SNHS) 2011-12 and applying an innovative methodology, which is based on the integration of the intersectionality theory and the multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA). Intersectional strata defined by combining categories of explanatory variables (gender, age, income, educational achievement and living alone) are created. Besides, two multilevel models are developed to obtain information on stratum-specific BMI averages and the degree of the clustering of BMI within strata expressed by the intra-class correlation coefficient (ICC). While the first model is a simple variance components analysis that provides a detailed mapping of the BMI disparities among the population, the second model includes the explanatory variables as a way to identify stratum-specific interactions. Chapter 2 considers the possible changes in BMI and its socioeconomic disparities from the economic crisis. The methodology is the same as in the previous chapter, but three different approaches are included: the socioeconomic, the regional and the socioeconomic+regional approaches. To study the trend followed by inequalities in the BMI, the analysis comprises the SNHS 2006-07, 2011-12 and 2016-17 and compares the stratum-specific BMI averages obtained in each survey. Besides, the ICC provides information about the impact of these different approaches on the distribution of the BMI in Spain. Finally, chapter 3 analyses the obesity prevalence, its social gradient and their evolution in recent years, especially throughout the recession period, in the Canary Islands. By doing so, a temporary period of ten years is evaluated through the Canary Islands Health Surveys 2004, 2009 and 2015. The analysis is based on linear regression models, which are divided by sex and adjusted by variables such as age, educational attainment and the island of residence. Likewise, the models also include year dummies and the corresponding interactions between time and education categories.
Results: The results for Spain suggest a moderate but meaningful clustering of individual BMI within the intersectional strata defined by gender, age, income, educational attainment and living alone (ICC=12.4%), as well as the existence of small stratum specific interaction effects (ICC=1.9%) in 2011-12. In addition, the existence of a socioeconomic gradient in BMI, which is more predominant in women and more accentuated by educational achievements, is confirmed. The socioeconomic inequalities, far from remaining unchanged, were expanded during the first years of the economic recession and were reduced in the post-crisis period. Besides, an approach that comprises socioeconomic, demographic and regional dimensions increases the capability of intersectional strata to explain the variation in the BMI in periods of economic crisis. While in Spain the BMI has increased over the whole studied period (2006-2017), although to a lesser extent in the latter year, the BMI of the Canary population and that of women increased in 2009 (+0.33 and +0.59 units, respectively), with the exception of women with medium education who reduced their BMI in 0.21 units, and decreased in 2015. Regarding obesity prevalence in the Islands, it also increased in 2009 and decreased in 2015 (from 17.95% to 19.54% and to 18.64%). Further, an inverse correlation between education and BMI is observed in this region, especially among women, as well as statistically significant differences among some islands.
Conclusions: A socioeconomic gradient is observed in Spain and in the Canary Islands. Likewise, a countercyclical behaviour is detected in the BMI of both populations, as well as in the evolution of the socioeconomic inequalities in BMI in Spain. Because intersectional strata have some influence on BMI, policymakers should design interventions that are not only targeted at the entire population, but also at those strata presenting higher BMIs. Besides, regional interventions throughout recessions and the integration of an insular perspective into public policies are important at tackling the obesity, given the influence of territory in BMI. Finally, educational attainment has been a relevant protective factor, especially for women, from the Islands, with medium education in crisis times. In summary, this research has contributed to expand the knowledge about obesity and its socioeconomic disparities in these particular populations, providing useful information for the design of new preventive actions to reduce the high obesity prevalence and the socioeconomic gradient in BMI.
