More equal but heavier: A longitudinal analysis of income-related obesity inequalities in an adult Swedish cohort

Presenter: Åsa Ljungvall, Lund University

Abstract

Over the last decade obesity has become an increasing health concern in most countries. In Sweden obesity increased by 67 % between 1990 and 2000, reaching a prevalence of 9.2 %, and in the UK and US 20-25 % of the population is classified as obese. Obesity is a risk factor for several diseases, such as different cardiovascular diseases, diabetes and some cancers. From a societal point of view it is important to learn about the underlying mechanisms to the obesity phenomenon. Using longitudinal data over a 17 year period for a Swedish cohort aged 20-68 in 1980/81, this study analyses income-related inequalities in obesity. Following this cohort and using concentration index and decomposition techniques we answer the following questions: 1) Does obesity inequality favour the rich? 2) What factors explain this inequality at different points in time? 3) How can the inequality pattern over time be explained? As income measure we use individual mean income over the 17 years. Thereby individuals do not re-rank over time and we can focus on the whether changes in obesity is related to income group belonging. We find that among females, inequality in obesity favour the rich, but the inequality declines over time (from -0.29 in 1980/81 to -0.16 in 1996/97). We find that income is a key driving force behind obesity inequality in all periods, and explains 80 to 90 % of obesity inequality. Income also explains the majority of the declined obesity inequality over time, mainly driven by a rather large decrease in income elasticity of obesity. The reduced obesity inequality, as well as the decreased income elasticity of obesity, may be tempting to infer as a positive development, perhaps a success of Swedish policy. We argue, however, that one should be careful with interpreting relative inequality measures. All else the same, decreased inequality is a reasonable aim. However, in the case of Swedish aging women, simultaneously to the reduced inequality, prevalence of obesity has increased, and it can be questioned whether any improvement has taken place. The reason to the reduced inequality is that individuals in all income groups have become more obese. Since worse-off people were obese to a larger extent at the outset, a (roughly) equal increase in obesity in absolute terms means that obesity has increased more in higher income groups in relative terms; the dispersion of obesity has become more equal. Thus, we conclude that the reduced inequality is not due to any health policy success, but rather the opposite. Since the income elasticity of obesity is a key contributor to the observed inequality, policies directed towards this factor might be the most effective. Similar trends as for women are found for men, although less pronounced. This should be taken into account when evaluating of obesity reducing policies since it may well be that men and women react differently to policy interventions.