Improving the spatial and temporal resolution of burden of disease measures with Bayesian models.

This paper contributes to the field by addressing the critical issue of enhancing the spatial and temporal resolution of health data. Although Bayesian methods are frequently employed to address this challenge in various disciplines, the application of Bayesian spatio-temporal models to burden of disease (BOD) studies remains limited. Our novelty lies in the exploration of two existing Bayesian models that we show to be applicable to a wide range of BOD data, including mortality and prevalence, thereby providing evidence to support the adoption of Bayesian modeling in full BOD studies in the future. We illustrate the benefits of Bayesian modeling with an Australian case study involving asthma and coronary heart disease. Our results showcase the effectiveness of Bayesian approaches in increasing the number of small areas for which results are available and improving the reliability and stability of the results compared to using data directly from surveys or administrative sources.

Corrigendum to: Evolution of non-fatal burden estimates for cardiovascular disease in Australia: a comparison of national and state-wide methodology of burden of disease.

Objective Burden of disease studies measure the impact of disease at the population level;however, methods and data sources for estimates of prevalence vary. Using a selection of cardiovascular diseases, we aimed to describe the implications of using different disease models and linked administrative data on prevalence estimation within three Australian burden of disease studies. Methods Three different methods (A = 2011 Australian Burden of Disease Study; B = 2015 Australian Burden of Disease Study; C = 2015 Western Australian Burden of Disease Study), which used linked data, were used to compare prevalence estimates of stroke, aortic aneurysm, rheumatic valvular heart disease (VHD) and non-rheumatic VHD. We applied these methods to 2015 Western Australian data, and calculated crude overall and age-specific prevalence for each condition. Results Overall, Method C produced estimates of cardiovascular prevalence that were lower than the other methods, excluding non-rheumatic VHD. Prevalence of acute and chronic stroke was up to one-third higher in Method A and B compared to Method C. Aortic aneurysms had the largest change in prevalence, with Method A producing an eight-fold higher estimate compared to Method C, but Method B was 10-20% lower. Estimates of VHD varied dramatically, with an up to six-fold change in prevalence in Method C due to substantial changes to disease models and the use of linked data. Conclusions Prevalence estimates require the best available data sources, updated disease models and constant review to inform government policy and health reform. Availability of nation-wide linked data will markedly improve future burden estimates.

Evolution of non-fatal burden estimates for cardiovascular disease in Australia: a comparison of national and state-wide methodology of burden of disease.

Objective Burden of disease studies measure the impact of disease at the population level;however, methods and data sources for estimates of prevalence vary. Using a selection of cardiovascular diseases, we aimed to describe the implications of using different disease models and linked administrative data on prevalence estimation within three Australian burden of disease studies. Methods Three different methods (A = 2011 Australian Burden of Disease Study; B = 2015 Australian Burden of Disease Study; C = 2015 Western Australian Burden of Disease Study), which used linked data, were used to compare prevalence estimates of stroke, aortic aneurysm, rheumatic valvular heart disease (VHD) and non-rheumatic VHD. We applied these methods to 2015 Western Australian data, and calculated crude overall and age-specific prevalence for each condition. Results Overall, Method C produced estimates of cardiovascular prevalence that were lower than the other methods, excluding non-rheumatic VHD. Prevalence of acute and chronic stroke was up to one-third higher in Method A and B compared to Method C. Aortic aneurysms had the largest change in prevalence, with Method A producing an eight-fold higher estimate compared to Method C, but Method B was 10-20% lower. Estimates of VHD varied dramatically, with an up to six-fold change in prevalence in Method C due to substantial changes to disease models and the use of linked data. Conclusions Prevalence estimates require the best available data sources, updated disease models and constant review to inform government policy and health reform. Availability of nation-wide linked data will markedly improve future burden estimates.

Value of a national burden-of-disease study: a comparison of estimates between the Australian Burden of Disease Study 2015 and the Global Burden of Disease Study 2017.

BACKGROUND: Estimates of burden of disease are important for monitoring population health, informing policy and service planning. Burden estimates for the same population can be reported differently by national studies [e.g. the Australian Burden of Disease Study (ABDS) and the Global Burden of Disease Study (GBDS)]. METHODS: Australian ABDS 2015 and GBDS 2017 burden estimates and methods for 2015 were compared. Years of life lost (YLL), years lived with disability (YLD) and disability-adjusted life years (DALY) measures were compared for overall burden and 'top 50' causes. Disease-category definitions (based on ICD-10), redistribution algorithms, data sources, disability weights, modelling methods and assumptions were reviewed. RESULTS: GBDS 2017 estimated higher totals than ABDS 2015 for YLL, YLD and DALY for Australia. YLL differences were mainly driven by differences in the allocation of deaths to disease categories and the redistribution of implausible causes of death. For YLD, the main drivers were data sources, severity distributions and modelling strategies. Most top-50 diseases for DALY had a similar YLL:YLD composition reported. CONCLUSIONS: Differences in the ABDS and GBDS estimates reflect the different purposes of local and international studies and differences in data and modelling strategies. The GBDS uses all available evidence and is useful for international comparisons. National studies such as the ABDS have the flexibility to meet local needs and often the advantage of access to unpublished data. It is important that all data sources, inputs and models be assessed for quality and appropriateness. As studies evolve, differences should be accounted for through increased transparency of data and methods.