Peter Somerford

Burden of Stroke in Indigenous Western Australians A Study Using Data Linkage

Background and Purpose— Despite the disproportionate burden of cardiovascular disease among indigenous Australians, information on stroke is sparse. This article documents the incidence and burden of stroke (in disability-adjusted life years) in indigenous and non-indigenous people in Western Australia (1997–2002), a state resident to 15% of indigenous Australians comprising 3.4% of the population of Western Australia. Methods— Indigenous and non-indigenous stroke incidence and excess mortality rates were estimated from linked hospital and mortality data, with adjustment for nonadmitted events. Nonfatal burden was calculated from nonfatal incidence, duration (modeled from incidence, excess mortality, and remission), and disability weights. Stroke death counts formed the basis of fatal burden. Nonfatal and fatal burden were summed to obtain disability-adjusted life years, by indigenous status. Results— The total burden was 55 099 and 2134 disability-adjusted life years in non-indigenous and indigenous Western Australians, respectively. The indigenous to non-indigenous age-standardized stroke incidence rate ratio (≥15 years) was 2.6 in males (95% CI, 2.3–3.0) and 3.0 (95% CI, 2.6–3.5) in females, with similar rate ratios of disability-adjusted life years. The burden profile differed substantially between populations, with rate ratios being highest at younger ages. Conclusions— The differential between indigenous and non-indigenous stroke burden is considerable, highlighting the need for comprehensive intersectoral interventions to reduce indigenous stroke incidence and improve outcomes. Programs to reduce risk factors and increase access to culturally appropriate stroke services are required. The results here provide the quantitative basis for policy development and monitoring of stroke outcomes.

High body mass index overtakes tobacco as the leading independent risk factor contributing to disease burden in Western Australia

Until recently, smoking was estimated as the leading independent risk factor contributing to the burden of disease in Australia. However, the prevalence of overweight and obesity is increasing in Australia, which now has one of the highest rates of overweight and obesity in the world. Concurrently, the prevalence of tobacco smoking has significantly decreased as a result of public health efforts. This letter presents results of a recent investigation into the contribution of selected risk factors to the burden of disease in Western Australia (WA) in 2006. The study was based on the comparative risk assessment (CRA) approach developed by the Global Burden of Disease study. This approach represents a systematic evaluation of the changes in population health, which would result from modifying the distribution of exposure to a risk factor or group of risk factors in the population. Central to CRA is the calculation of the Population Attributable Fraction (PAF), reflecting the proportion of current disease burden attributable to current and past exposure to a risk factor. The PAF is determined from the relative risks (RR) of disease in those exposed (compared to those not exposed) and the prevalence of the risk factor in the population. The RRs used in the calculation of PAFs in the WA study were consistent with those used in the Australian 2003 study, while estimates of the prevalence of high body mass and tobacco use were obtained from the WA Health and Wellbeing Surveillance System. The population burden was estimated in Disability Adjusted Life Years (DALYs). DALYs provide comparable information on the years of life lost to premature mortality and the equivalent ‘healthy’ years lost due to disability, thus reflecting both fatal and non-fatal burden. Mortality data for WA in 2006 were used to determine mortality burden. Disability burden was derived by applying the Australian 2003 baseline disease models, derived for the Australian 2003 study to more recent WA data derived from death and hospitalisation trends Table 1: The changing impact of high body mass and tobacco as independent risk factors for disease burden in various Australian studies. Study High BMI Tobacco Ranka % total DALY Ranka % total DALY Australia 19962 4 4.30% 1 9.70% WA 20003 4 3.90% 1 8.60% Australia 20031 3 7.50% 1 7.80% WA 2006 1 8.70% 2 6.50%

Disability burden due to stroke in Western Australia: new insights from linked data sources

Background The disability-adjusted life year index is used extensively to compare disease burden among diseases and locations, but difficulties remain in accurately estimating the nonfatal stroke burden in years lived with disability. Aims To improve stroke-related years lived with disability estimates in Western Australia for 2000, by improving the accuracy of component inputs: nonfatal (28-day survivor) incidence, disease duration and disability (severity) weights. Methods Nonfatal stroke incidence and the mortality difference between prevalent cases and the general population were estimated from linked hospital and mortality data using the Western Australian Data Linkage System. Dismod software used these inputs to model disease duration. Disability weights were estimated from population-based stroke survey data, using indirect health valuation methods and adjusting for prestroke disability. Years lived with disability were calculated from the three components. Results The annual age-standardised nonfatal incidence (n = 1985) was higher ***In males (121/100 000) than females (96/100 000). The duration varied between 35.8 (females 15–24 years) and 3.4 years (males 85+ years). The mean pre-stroke-adjusted disability weight was higher at 4-months (0.38) than at 12-months (0.31). The age-standardised rate of nonfatal burden in males (302/100 000; 95% CI 290–314) was significantly higher than that in females (250/100 000; 95% CI 240–260). The nonfatal proportion of stroke burden (males 45%; females 37%) was higher than estimated in previous studies. Conclusion This study illustrates that previous reports most likely underestimated disability burden as a contributor to the total stroke burden in Australia. Methodological refinements will contribute to burden of disease studies elsewhere.

Excess Mortality Rates for Estimating the Non-Fatal Burden of Stroke in Western Australia: A Data Linkage Study

Background: Non-fatal stroke burden measured in Years Lived with Disability (YLD) requires valid estimates of stroke case fatality to allow modelling of disease duration. In the model, case fatality can be calculated from the absolute risk of mortality in cases in excess of that in the non-diseased. Aims: Our purpose was to estimate excess mortality rates in 28-day survivors of stroke in Western Australia and to evaluate differentials in survival by stroke type, age and time since the first stroke event. Method: Excess mortality among prevalent (first-ever plus existing) survivors was estimated from linked hospital and mortality data. Changes in excess mortality over time were calculated over a 6-year period. Results: Excess mortality increased with age for both males (21 per 1,000 in the 15- to 54-year to 109 per 1,000 in the ≧85-year age group) and females (16 and 122 per 1,000 for the 15- to 54-year and ≧85-year groups, respectively). Survival by stroke sub-types differed at ages <55 years but not >55 years. During the first year excess mortality was markedly higher, after which it was relatively constant for each age group. The assumption of constant rather than changing excess mortality in 28-day survivors of stroke had minimal effect on estimates of duration. Conclusion: Measures of excess mortality in prevalent survivors have not previously been available for estimating YLD for stroke. An analysis of all stroke types combined is not likely to substantially bias estimates of non-fatal stroke burden nor is an assumption of constant excess mortality for survivors.

Effect of clearance periods on hospital stroke incidence using linked administrative data

In data linkage studies of disease incidence, prevalence pooling, the effect of including prevalent cases as first-ever cases can be accounted for by using an appropriate clearance period to check for prior hospitalisations (1–3). Provided the clearance is long enough, the first record for an individual is assumed to reflect the first-ever stroke event. To evaluate the use of different clearance periods in identifying first-ever cases of hospitalised stroke, stroke hospital admissions in Western Australia (WA) were extracted from the Hospital Morbidity Data System covering the years 1988 to 2003 (4). Data linkage using the WA Data Linkage System (5) allowed separations belonging to particular individuals to be identified and their first stroke admissions to be selected. A 12-year clearance was used to identify incident counts for acute stroke for 2000. This procedure was repeated using hospitalisations covering shorter periods to investigate prevalence pooling over different clearance periods. A total of 3875 admissions with an acute stroke code, comprising 41% of all cerebrovascular disease admissions, were investigated. Using the 12-year clearance first-ever cases as the standard, 41% of unlinked admissions with acute stroke codes in 2000 were deemed repeat stroke admissions, either due to recurrences or multiple admissions per event. The percentage repeat acute stroke admissions reduced to 16%, 7% and 2% when admissions were linked for periods of 1 year, 2 years and 5 years, respectively. A clearance of 10 years identified o1% as prevalent cases. A similar pattern was evident in stroke admission data for 2002 using a 14-year clearance as the standard. Knowledge of the relative benefit of different lengths of clearance time can guide future analyses of hospital-based stroke incidence. These results indicate that after 10 years, o1% of those identified as first-ever will be recurrences, while after 5 years, no more than 3% would be recurrent cases. Jurisdictions in which linked hospital records exist for fewer years than in WA, a clearance of 5 years could be considered adequate for certain types of study. This guideline will be useful as other Australian states introduce or extend their linked data capacity. An alternative ‘backcasting’method (6) using a retrograde survival model to obtain correction factors for over-ascertainment is a valid alternative, being particularly useful for determining trends without loss of information from the early years of observation (6). However, the method does not identify cases at the unit record level for subsequent survival analysis or identify patient samples for more detailed studies requiring the collection of further information. Where no population-based stroke registers or linked data systems exist, incidence estimates using hospital admissions as the starting point will continue to need downward adjustments of acute stroke events to account for recurrences and multiple admissions within episodes. The analysis of the ratio between first-ever to total acute admissions calculated here found about 60% of acute events to be first-ever events, and the proportion was closely similar by age and gender. The extent to which this ratio is generalisable within and beyond Australia is uncertain because of differences in discharge practices and the coding of transfers between and within hospitals. Additionally, the proportions of recurrent and incident cases may change with time, reflecting the changing impact of the secondary prevention of stroke using new medications and interventions.

The contribution of blood alcohol concentration formulae to establishing a responsible drinking level for females

The use of formulae for calculating blood alcohol concentration (BAC) supports the recent NHMRC drinking guidelines in the establishment of a daily limit of up to 2 drinks female/4 drinks male as opposed to a 3 drink female/4 drink male ratio. BAC formulae have other educational uses as well, such as information related to alcohol consumption and drink driving BAC limits. Public health professionals are still not able to answer questions about the health damage associated with certain types of drinking patterns (number of days, length of drinking session) because of the lack of applicable research.

Strategic information for hospital service planning: a linked data study to inform an urban Aboriginal Health Liaison Officer program in Western Australia

OBJECTIVES The aim of the present study was to provide descriptive planning data for a hospital-based Aboriginal Health Liaison Officer (AHLO) program, specifically quantifying episodes of care and outcomes within 28 days after discharge. METHODS A follow-up study of Aboriginal in-patient hospital episodes was undertaken using person-based linked administrative data from four South Metropolitan hospitals in Perth, Western Australia (2006-11). Outcomes included 28-day deaths, emergency department (ED) presentations and in-patient re-admissions. RESULTS There were 8041 eligible index admissions among 5113 individuals, with episode volumes increasing by 31% over the study period. Among patients 25 years and older, the highest ranking comorbidities included injury (47%), drug and alcohol disorders (41%), heart disease (40%), infection (40%), mental illness (31%) and diabetes (31%). Most events (96%) ended in a regular discharge. Within 28 days, 24% of events resulted in ED presentations and 20% resulted in hospital readmissions. Emergency readmissions (13%) were twice as likely as booked re-admissions (7%). Stratified analyses showed poorer outcomes for older people, and for emergency and tertiary hospital admissions. CONCLUSIONS Future planning must address the greater service volumes anticipated. The high prevalence of comorbidities requires intensive case management to address case complexity. These data will inform the refinement of the AHLO program to improve in-patient experiences and outcomes.