Uncovering the mechanisms behind socioeconomic inequalities in stroke care and outcome through innovative statistical methods for mediation analysis
Research project
Stroke is a leading cause of death and disability in the western world, affects all population groups and involves major parts of the health care system. In Sweden, research has shown that socially underprivileged patients have poorer access to stroke care at the acute stage as well as secondary prevention after stroke and are more prone to adverse outcome. That these differences exist is well established but the question remains why, and how they can be prevented.
With this interdisciplinary project we seek to answer these questions. The project is a collaboration with Riksstroke, the Swedish stroke register, and will have access to unique linked register data. By combining these high quality data with the implementation and development of advanced statistical methods related to the emerging field of causal mediation analysis we will be able to give novel insights into the causal mechanisms behind socioeconomic inequalities in health.
The project aims to: (i) investigate the causal pathways behind socioeconomic inequalities in stroke care and outcome, (ii) evaluate and implement recent developments in mediation analysis for complex causal pathways in the stroke context, (iii) develop flexible techniques for sensitivity analysis to evaluate the impact of violations of the assumptions regarding unobserved confounding upon which mediation analysis relies, for more reliable estimates of causal pathways, and (iv) implement the developed methods in publically available software and provide a guide to mediation analysis for empirical researchers.
Through improved knowledge of the causal mechanisms behind the relationship between socioeconomy and stroke care/outcome after stroke the results of the project can provide a base for political decisions concerning priorities within health care. The project will have an impact beyond the scope of stroke outcome and care through the development and implementation of state of the art statistical methods that will provide generic tools for the investigation of complex pathways.