Jane Lyons is the research lead for the ADR Wales Social Justice research theme. In this blog, Jane discusses the findings of the team’s most recent analysis that measures trajectories in chronic disease accrual and mortality across the lifespan in Wales, UK, with the focus on differences in outcomes by area-level deprivation.

Previous literature has shown that living in poverty and more deprived circumstances are determinants of poorer health. In order to advance health equity and improve the health of the most disadvantaged, we must identify and understand differences in health outcomes over time between different groups of people to provide evidence that quantifies disparities in our society.

The aim

Our research aimed to measure chronic disease accrual and examine the differences in time to a single chronic disease diagnosis, multiple chronic disease diagnoses, and death between socioeconomic groups in Wales.

Previous, similar chronic disease and multimorbidity studies have focused on disease-specific outcomes. By comparison, our research focused on the differences in outcomes experienced by different groups of people, to help highlight disparities and provide evidence to promote change to support a fairer and healthier society.

The approach

In our study, we utilised the Secure Anonymised Information Linkage (SAIL) Databank to link:

  • the Welsh Demographic Service Dataset
  • Welsh Longitudinal General Practice
  • the Patient Episode Database for Wales
  • the Annual District Death Extract from the Office for National Statistics.

This created a population-wide e-cohort of all Welsh residents living in Wales on 1 January 2000 with follow-up until 31 December 2019.

We incorporated a 5-year clearance period to exclude anyone who had a prior chronic disease diagnosis, from a list of 132 conditions, to create a chronic disease-free starting health state which previous similar research had not included.

This study utilised multi-state models to model the trajectories of chronic disease accrual accounting for death as a competing risk. We applied a 4-state model to follow the population from the chronic disease-free state with possible transitions to single chronic disease (morbidity), two or more chronic diseases (multimorbidity), or death. All models were adjusted for age, sex, and area-level deprivation. Models were estimated for 10, 20, 30, 40, 50, 60, and 70 years of age, both for males and females, and for all deprivation quintiles to simulate the effect of population distribution change and to allow direct comparison of trajectories of disease accrual and death between the most and least deprived individuals in Wales, UK.

Our outcome measure was Restricted Mean Survival Time (RMST) which represents the average time free from an event (development of disease or death), and is measured as the average time, in years, from entry to one health state and transition to a subsequent health state. Compared to more traditional model outputs such as odds ratios and hazard ratios, RMST has allowed us to quantify evidence of the difference in time, recorded in years, to health outcomes between different groups of people. We hope the use of RMST will help reach a wider audience to improve inclusivity of understanding scientific findings and enhance the impact of our results.

Findings

In our study, we included 965,905 individuals aged 5–100+ with an average follow-up time of 13.2 years, equating to 12.7 million person-years included in total.

Overall, results showed accelerated rates of transition from being chronic disease-free to developing chronic diseases and death across both sexes and for all age groups in individuals living in more deprived areas of Wales, demonstrating faster accrual of chronic disease and death for individuals living in worse socioeconomic circumstances.

What’s next?

Moving forward, we plan to include ethnicity in our models to examine the difference in disease pathways and health disparities between ethnic groups. We will also include individual-level metrics of deprivation.

Additionally, this multi-state model approach can be transferred into different areas of our inequalities research. For example, we plan to use multi-state models to compare short and long-term health and social outcomes in different groups of injured people in Wales. We also plan to compare outcomes of the Welsh population internationally to help improve our patient care following an injury.

Read the full journal article.