Member Research and Reports

Member Research and Reports

Double Disadvantage: Income Inequality, Spatial Polarization and Mortality Rates in Taiwan

Unequal societies have been hypothesized to pose a threat to population health. Previous studies have suggested that unequal income distribution and economic spatial polarization are respectively associated with various health outcomes. However, no studies have examined the joint effect of income inequality and spatial polarization on population health. Dr. Chun-Tung Kuo, a postdoctoral fellow at  the department of social and behavioral Sciences in Harvard T.H. Chan School of Public Health, and his advisor Dr. Duan-Rung Chen, a professor and the director of the Institute of Health Behaviors and Community Sciences at College of Public Health in National Taiwan University, collected publicly available data at both city/county- and township-level in Taiwan to answer this question. This study has been published in the Journal of Public Health.

Previous review work has suggested the adverse effects of income inequality on health can vary across different area sizes. Income inequality measured in larger geographical scales appears to have a stronger association with health outcomes due to it reflects the scale of social class differences. In contrast, a growing body of studies indicated that the economic polarization in small areas can be an important social determinant of health. Economic polarization is defined as a rising concentration of neighborhood affluence and poverty and can be represented by the Index of Concentration at the Extremes (ICE), which assesses the concentration of people with low and high economic resources in a certain area. The ICE can range from −1 (most deprived) to 1 (most privileged). In this study, we used multilevel datasets of 352 townships nested within 20 cities in Taiwan to examine the association between township-level ICE and mortality and further examined whether city-level income inequality moderates this association.

In 2008–12, the average age-standardized mortality in Taiwan was 470.5 per 100 000 populations. As compared to the highest income-based ICE quintile, the lowest ICE quintile was associated with an excess 171.7 deaths per 100 000 people (95% CI = 116.1, 227.3) after controlling for income inequality and population size. Additionally, one unit rise in the Gini coefficient further increased 29.9 deaths (95 percent CI = 12.4, 47.5) for the lowest ICE quintile, as compared to the highest. In other words, at higher levels of ICE index, mortality rates seem similar regardless of the level of income inequality. However, townships with lower ICE index appear to suffer in the form of higher mortality rates as the city-level income inequality rises.

In conclusion, this study suggests that the income-based ICE measure is associated with age-standardized mortality rates in Taiwan. The spatial economic polarization and income inequality have a joint effect of determining the chance of mortality in Taiwan.


To our knowledge, this is the first work using multilevel models to investigate the joint effect of income inequality and spatial economic polarization on mortality rates. A small-scale measure of income distribution such as ICE provides more information with respect to how income inequality may increase mortality. Further investigations are needed to investigate whether the concentrations of extreme affluence and poverty also affects health inequities.


Kuo, C.-T., & Chen, D.-R. (2018). Double disadvantage: income inequality, spatial polarization and mortality rates in Taiwan. Journal of Public Health, 40(3), e228–e234.