Enhancing the understanding and application of burden of disease methods

Wyper, Grant Mark Andrew (2022) Enhancing the understanding and application of burden of disease methods. PhD thesis, University of Glasgow.

Abstract

Burden of disease studies can be used to summarise the combined impact that causes of morbidity and mortality have in a defined population in a manner that is consistent and comparable. They have become an increasingly popular means to use in knowledge exchange activities for informing national, and local, policy-making decisions. The Global Burden of Disease (GBD) study provide estimates on the burden of disease for each cause of disease and injury for individual countries, and sub-national areas for selected countries, which are adopted by many researchers and national public health institutes. Alternative national burden of disease studies exist, independent of the GBD study, which utilise country-specific data sources that are representative of the state of health in their country. The co-existence of these studies can potentially be problematic for knowledge exchange activities, particularly when estimates differ and neither party are able to explain why the differences exist. This can often lead to speculation on why the differences occur, and the monopolisation of burden of disease research by the GBD study can often be problematic in ensuring a balanced narrative emerges. Many users often just seek to appraise differences in the input data as they often lack the experience, or will, to understand methodological processes as they are often complex and exhaustive. However, these differences in methods could plausibly play a key contribution in explaining differences in results. Two such methodological choices are in relation to the choice of severity distribution in assessing the non-fatal burden of disease, and the choice of standard population that is used when estimating standardised rates. These issues are tackled in the first part of this thesis. The impact of these methodological choices were appraised using approaches from the Scottish Burden of Disease study, relative to those of the GBD study. The first aim of this thesis is to establish whether using standard GBD study severity distributions result in major differences, compared to using severity distributions that better reflect the epidemiological situation in Scotland using the example of individual cancer types. The second aim of this thesis is to determine whether using different standard populations to standardise rates leads to major differences in how causes of disease and injury are ranked using disability-adjusted life years (DALYs). The research presented in this thesis gives insights into how impactful seemingly slight methodological choices can result in major differences. The second subset of research questions relate to establishing methods for COVID-19 burden of disease assessment, and monitoring the burden of COVID-19 using DALYs. The aims of this work presented in this thesis were to: (i) Assess which European countries were likely to be most vulnerable to severe outcomes from COVID-19; (ii) Develop an international consensus method for estimating COVID-19 DALYs; (iii) Estimate COVID-19 DALYs in Scotland during 2020 and contextualise the result compared to the leading pre-pandemic causes of disease; (iv) Estimate the extent of inequality in COVID-19 DALYs in Scotland during 2020; (v) Provide an alternative way of assessing the impact of all-cause inequalities by comparing inequality-attributable DALYs to the impact of COVID-19; (vi) Monitor changes in the fatal COVID-19 burden of disease in 2021, in the context of vaccine availability, compared to 2020. The research questions posed in this thesis are investigated in the included seven firstauthor papers. The insights from each of the papers are synthesised in an explanatory essay. This essay attempts to show the impact of each of the individual papers, and how they form a cohesive body of work that enhances the understanding and application of methods for use in burden of disease assessment. The explanatory essay presents a two-tier approach. The first part considers the impact of: (i) severity distributions; and, (ii) standard populations in age-standardised rate calculations, in burden of disease assessment. The essay discusses the importance of the findings from the included papers. It also highlights their importance for other stakeholders in international burden of disease research networks’, given the impact of severity distributions on resulting estimates is now more widely understood. The second part of the explanatory essay focuses on developing consensus methods to estimate COVID-19 DALYs, and then focuses on their application in monitoring the overall, and inequalities in the, COVID-19 burden of disease in Scotland. The need for a consensus method is justified on the basis of ensuring that the comparative properties of DALYs remain. This work has synergies with the first part of the explanatory essay, as it is important that assessments are reflective of the best available country-specific data inputs. The significance of this work is discussed through its uptake within the international research community. The rationale for undertaking burden of disease assessments it to generate comparable estimates for where risk factors, causes of death, disease and injury are causing the largest public health losses. In doing this, it provides compelling evidence to inform resource allocation discussions for tackling population health needs, and the implications for which these needs will have in ensuring care services and workforces are proportionate to the challenge. Through presenting developments and the journey through these seven published works in included in this thesis, I aim to have demonstrated the importance of the work in enhancing the understanding and application of burden of disease assessment methods. It is vital that the methodological context, and any uncertainties for which estimates are produced in, is well understood.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing
• Supervisor's Name: McIntosh, Prof. Emma
• Date of Award: 2022
• Depositing User: Theses Team
• Unique ID: glathesis:2022-83124
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 12 Sep 2022 10:56
• Last Modified: 14 Sep 2022 09:12
• Thesis DOI: 10.5525/gla.thesis.83124
• URI: https://theses.gla.ac.uk/id/eprint/83124

Related URLs

• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI