Analysing equity in outpatient care in Stockholm county and the impact of using different data sources

Abstract

Healthcare plays an important role in tackling inequalities in health by providing care to groups in need, and protecting lower income groups from further impoverishment due to ill health. Continually evaluating equity in healthcare is important as it changes with altered needs in the population and changes within the healthcare system itself. The aim of this thesis was to increase knowledge on equity in outpatient healthcare based upon data from Stockholm County in the period 2005-2013. The conclusions made are based upon the results from four different studies.

In Study I, we investigated the non-response bias in a health survey with respect to socio-economic differences in healthcare utilisation by comparing data from a health survey with data of the general population in Stockholm County.

In Study II, we investigated how the income gradient in healthcare utilisation changed when taking need into account, by comparing socioeconomic differences using register data on healthcare consumption and sociodemographic background characteristics, and linked survey data.

In Study III, we compared socioeconomic differences in healthcare utilisation in Stockholm County before and after the introduction of a primary care reform in 2008 focusing on free choice of provider and free establishment for providers along with a change in reimbursement system from a need-weighted capitation based system to a system based primarily on fee-for-service.

In Study IV, we investigated changes in resource allocation to primary care clinics in areas with different levels of median income over a period of nine year and whether the introduction of the 2008 primary care reform changed the trends in resource allocation. We also investigated whether a change in resource allocation was associated with changes in rates of ambulatory care sensitive conditions and emergency ward visits in areas with a decrease and increase of resources, respectively.

The conclusions of this thesis is that although response rates in health surveys can be relatively low and differ between socioeconomic groups, relative differences in healthcare utilisation between different socioeconomic groups are representative of the source population and not biased, for most groups. Using health survey data to investigate equity in healthcare utilisation allows adjusting the analyses for healthcare needs. Using data where this is not possible is likely to underestimate socioeconomic differences.

The 2008 primary care reform did not particularly benefit groups with higher healthcare needs, on the contrary there seems to be a decline in vertical equity in primary care in Stockholm from 2007 to 2011; especially among women with poor health and men with poor mental health as well as men living in disadvantaged areas. In addition, resources have been shifting from areas with populations with higher healthcare needs to areas with populations with lesser healthcare needs in this period (2005-2013). Further investigation is needed in order to fully understand the health effects of shifting resources from poorer to richer areas.