Adjusting measurement of health inequalities: Are current measures benchmarked against the impossible?

Presenter: Jing Shen, University of Newcastle upon Tyne

Abstract

This paper investigates the measurement of health inequalities in the UK. Existing studies have measured inequality using methods based on Lorenz curves. These traditional methods may not provide measures of inequality which are informative to policy makers. Health inequalities are generated by complex interrelationships between socioeconomic variables, environment and health. If health problems are ingrained in youth and these problems later cause health inequalities among adults then policy makers will have few tools at their disposal for reducing inequality. In fact policy interventions may take years to come to fruition. We suggest that models of health inequalities need to be adjusted for socioeconomic factors to provide measures over which policy makers may have more influence.

Methods: We calculate Gini coefficients and concentration indices for a range of health outcomes using standard techniques. We then adjust for socioeconomic factors using regression analysis. This allows us to compare adjusted and unadjusted health inequalities. It may be that traditional methods are producing measures of health inequalities which are too general to be informative. We also demonstrate how health inequalities can evolve over time by looking at health over the life-course.

Data: We use the National Child Development Survey (NCDS) which is a cohort dataset from the UK. The data follows a cohort of individuals born in one week in March 1958, and there are currently 7 waves available. This data provides in depth information on childhood health and allows analysis over the life-course. We supplement this data with data from the Health Survey for England (HSE) in order to investigate inequality among older cohorts.

Results: Acknowledging the contributions of traditional measures of health inequalities, the results also show some drawbacks of these traditional measures because of the fact that many of the impacts of health in childhood take time to filter through to inequalities. This means that the policy options of decision makers are reduced and that reductions in health inequalities may take many years to take effect. The adjusted measures provide a better picture for policy makers. We also demonstrate that the aging population will initially increase health inequalities before healthy survivor effects start to reduce inequality.