Reconsidering the income‐health relationship using distributional regression

Abstract

We reconsider the relationship between income and health taking a distributional perspective rather than one centered on conditional expectation. Using structured additive distributional regression, we find that the association between income and health is larger than generally estimated because aspects of the conditional health distribution that go beyond the expectation imply worse outcomes for those with lower incomes. Looking at German data from the Socio-Economic Panel, we find that the risk of bad health is roughly halved when doubling the net equivalent income from 15,000 to 30,000€. This is more than tenfold of the magnitude of change found when considering expected health measures. A distributional perspective thus highlights another dimension of the income-health relation-that the poor are in particular faced with greater health risk at the lower end of the health distribution. We therefore argue that when studying health outcomes, a distributional approach that considers stochastic variation among observationally equivalent individuals is warranted.

Factors Associated with the Health Literacy on Social Determinants of Health: A Focus on Socioeconomic Position and Work Environment

“…Income is the representative index that plays a definitive role in identifying socioeconomic status. The most urgent priority for low-income-status people is making a living, ahead of health management, so these people usually report low health status [ 10 , 11 ]. Moreover, though low-income individuals may try to lead healthy lifestyles, they typically lack the capacity to purchase nutritious foods or the time for exercise compared to upper-class individuals [ 12 ].…”

A distributional regression approach to income-related inequality of health in Australia

“…Most empirical studies on the effect of income on health have used ordinary least square and IV approaches, generalized linear models, and comparisons of correlation coefficients [8][9][10]. However, Silbersdorff et al [12] argue that conventional regression techniques do not provide reliable estimates of the income-health relationship. Conventional regression techniques assess the effect of variations in covariates on the expected level of the dependent variable.…”

“…Based on a variety of income measures and health variables, several studies have confirmed the existence of a significant positive relationship between income and health (e.g., [8][9][10][11]). As pointed out by Silbersdorff et al [12], the majority of studies use linear and generalized linear models, which assess the effect of variations in independent variables on the expected level of the dependent variable. With regard to the income-health relationship, this means that variations in income are solely related to the conditional mean health outcome.…”

Noncrossing structured additive multiple-output Bayesian quantile regression models

“…In order to illustrate how this directional approach can shed new light to the analysis of quantile regression models, we consider data from the Socio Economic Panel (SOEP, 2014) collected in Germany in the year of 2012, which was also analyzed by Silbersdorff et al (2018). For this exercise, we are interested in two dimensions of inequality in the population, namely health and income.…”