Use of data linkage to enhance burden of disease estimates in Western Australia: the example of stroke

Research output: ThesisDoctoral Thesis

242 Downloads (Pure)


[Truncated abstract] The Disability-Adjusted Life Year index, developed by the Global Burden of Disease Study, is used extensively to compare disease burden between locations and over time. While calculation of the fatal component of this measure, Years of Life Lost, is relatively straight-forward, the non-fatal component, Years Lived with Disability, is based on parameters that are challenging to estimate. This thesis pioneers the use of the Western Australian Data Linkage System to enhance epidemiological parameters underpinning Years Lived with Disability, providing, by way of illustration, a robust quantitative profile of burden of stroke in the state of Western Australia at the turn of the 21st century. The principal methodological objective was to utilise data linkage analytic methods for the specific requirements of burden of disease estimation. The principal stroke-related objectives were: 1. To estimate the parameters underpinning the non-fatal burden of stroke (Years Lived with Disability) in Western Australia in 2000. 2. To estimate the total burden of stroke (Disability-Adjusted Life Years) in Western Australia in 2000. 3. To investigate differentials in stroke burden between different sub-populations in Western Australia. 4. To calculate projections of stroke burden for Western Australia in 2016. Years Lived with Disability from stroke were calculated for Western Australia from nonfatal stroke incidence, expected duration and disability (severity) weights. Non-fatal incidence was estimated using linked hospital and death records of first-ever hospitalised stroke 28-day survivors in 2000. This was then adjusted for out-of-hospital cases determined from the population-based Perth Community Stroke Study. iv Analysis of mortality in hospitalised 28-day survivors using linked data revealed that the excess mortality in prevalent, rather than incident cases was the main disease-specific parameter required for modelling stroke duration using DisMod II specialised software. ... Access to data linkage and population-based stroke studies in Western Australia allowed more accurate estimation of non-fatal stroke burden, with previous reports most likely underestimating disability as a contributor to total burden. Although predominantly affecting the growing aged population, stroke also affects a sizable number under the age of 65 years, the age group where differentials in stroke burden are the greatest. The findings highlight the continued need for primary prevention efforts for all ages, targeting especially younger people in disadvantaged groups. The shift to greater disability burden in the future and the needs of disadvantaged groups must be considered when planning stroke services. The multiple studies undertaken for this thesis contribute to ongoing improvement of data quality and methodological refinements underpinning estimates of Years Lived with Disability, specifically for stroke, but applicable also to other diseases. Similar linked data approaches can be used in other Australian states in the future once infrastructure is developed, thereby improving estimates of disease burden for health policy and planning in the future.
Original languageEnglish
QualificationDoctor of Philosophy
Publication statusUnpublished - 2009


Dive into the research topics of 'Use of data linkage to enhance burden of disease estimates in Western Australia: the example of stroke'. Together they form a unique fingerprint.

Cite this