The purpose of this thesis is to evaluate unfair inequality in healthcare use in Brazil, between 1998 and 2013, allowing for multiple social dimensions of inequality. The thesis innovates methodologically by proposing the Health Care Advantage (HCA) approach, which takes into consideration multiple social dimensions of inequality using a metric that is directly comparable to traditional bivariate measures that focus on a single dimension of inequality such as income. The thesis also has three other contributions. Firstly, it provides new empirical evidence about unfair inequality in health care in a developing country, where inequalities are particularly large and important. Secondly, it provides up-to-date national evidence about equity in the use of health care by analyzing a new wave of survey data in Brazil not previously analyzed. Thirdly, it provides the first national evidence about health care equity trends in mammography and cervical screening in Brazil, during a period of substantial health care reform. The data for the analysis comes from four large, repeated cross section sample surveys, the Health Supplement of the Brazilian National Household Sample Survey for the year 1998, 2003, 2008 and the first National Health Survey, conducted in 2013, with an average sample size of 371,000 over the four waves. After controlling for age, sex and self-assessed health, unfair inequality – or “inequity” – is observed in three different forms of care: physician visits, mammography screening and cervical screening. Overall inequity is substantially larger than income-related inequity. Over time, inequity has decreased for physician visits and cervical screening in Brazil, although for mammography there is no clear trend. Decomposition analysis shows that the main component of unfair inequality in all cases is health insurance, and its relevance increased between 1998 and 2013. For mammography and cervical screening, though not for physician visits, other key components of inequality (> 5% contribution) were region, urban status, education and income. Having children in the household was an important component of inequality in 1998, but this reduced substantially over time, as did the contribution of living in rural areas. The contribution of income to overall inequity decreased during the study period for physician visits and mammography screening, yet for cervical screening it doubled between 2003 and 2013. The methods developed in this thesis can yield useful new insights into unfair inequality in health care, and may help shift research attention away from income-related inequities that are not always the largest or most important inequities from a policy perspective.
