Burden of Stroke in Indigenous Western Australians
A Study Using Data Linkage
Judith M. Katzenellenbogen, PhD; Theo Vos, PhD; Peter Somerford, MSc; Stephen Begg, MPH;
James B. Semmens, PhD; James P. Codde, PhD
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
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 (ⱖ15 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. (Stroke.
2011;42:1515-1521.)
Key Words: cerebrovascular accident 䡲 epidemiology 䡲 health policy 䡲 indigenous
T
the indigenous population was estimated using mortality and
hospitalization rates only,6 potentially underestimating the
burden on communities. Quantification of disease rates is
challenging, with under-identification of indigenous people in
hospital and death data and small case numbers often hampering accurate estimates.7–9 Consequently, detailed analyses
of the indigenous burden of stroke using alternative measures
are rare. However, in WA, the study of the health of the
population is facilitated by a well-established, comprehensive
health data linkage system10 that allows detailed epidemiological investigation of particular diseases.
In this article, we estimate the burden of stroke in indigenous compared with non-indigenous Western Australians
(aged 15 years or older) pertaining to a 5-year period between
July 1997 and June 2002, using the Disability Adjusted Life
Year (DALY) metric.11 This composite burden of disease
he indigenous population of Australia (Aboriginal and
Torres Strait Islander peoples) has a disproportionate
burden of disease,1,2 exemplified by an 11-year lower life
expectancy compared with other Australians,3 that reflects
entrenched historic, cultural, socioeconomic, and political
disadvantage. The morbidity profile for indigenous Australians is characterized by high rates of cardiovascular diseases
(including rheumatic heart disease), diabetes, and end-stage
renal failure,2,4 with cardiovascular diseases being the leading
cause of death.2,4,5 Western Australia (WA) is home to 15%
of indigenous Australians (2003 WA population 68 661) who
are spread across a vast area encompassing all levels of
remoteness and with substantial heterogeneity with respect to
language group and culture.
Stroke contributes ⬇3% of the indigenous health gap in
Australia.2 Until recently, the impact (burden) of stroke on
Received September 10, 2010; accepted November 30, 2010.
From the Curtin Health Innovation Research Institute (J.M.K.), Curtin University, Perth, Western Australia; School of Population Health (J.M.K.),
University of Western Australia, Perth, Western Australia; Centre for Burden of Disease and Cost Effectiveness (T.V.), School of Population Health,
University of Queensland, Queensland, Australia; Health Department of Western Australia (P.S., J.P.C.), Perth, Western Australia; Health Economics
Unit (S.B.), Funding and Resourcing Branch, Queensland Health, Queensland, Australia; Population Health Research (J.B.S.), Curtin Health Innovation
Research Institute, Curtin University, Perth, Australia.
Correspondence to Judith M. Katzenellenbogen, Curtin Health Innovation Research Institute, Curtin University, GPO Box U1987, Perth, Western
Australia. E-mail [email protected]
© 2011 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.110.601799
1515
1516
Stroke
Table 1.
Data Components, Sources, and Assumptions Used in Calculating Disability-Adjusted Life Years for Stroke
Component Data
June 2011
Data Sources
Assumptions
Calculation
Mortality register
Stroke-coded deaths capture fatal stroke
Death counts ⫻ standard
LE (age-specific)
Global Burden of Disease Study 1994
Life tables selected representing optimal
mortality profile globally
YLL
1. Stroke death counts,
by age, sex, and
indigenous status
2. Global standard LE
at each age
YLD
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
1. Nonfatal incidence
(28-d survivors)
● Unit record linked hospital and death data
● Admission proportions from PCSS to adjust
for out-of-hospital cases
● 8-y clearance identifies “first-ever”
cases
● Perth admission proportions apply to
indigenous population
2. Duration
Modelled in DisMod12: provides analytic
solution to a set of differential equations
describing the relationship between total
mortality and 3 rates: case fatality, incidence,
remission
● DisMod assumes mortality from all other
causes is independent of the disease
being modeled
● Conceptual model of DisMod is a
multistate life table
Input 1: Nonfatal incidence Input 2: Remission
Input 3: EMR ⫽ mortality rate in prevalent
stroke cases minus population mortality rate
Remission⫽zero Excess mortality
approximates case fatality rate
Disability data from PCSS. Mapped to EuroQol.
Adjusted for prestroke function. Regression
equation to predict age-specific DW
Disability profile of Aboriginal stroke cases
similar to cases in PCSS
3. DW not reported in
detail in this article
YLD ⫽ Incidence ⫻
duration ⫻ DW
DW at 4 mo applied to
YLD calculation for year
1. DW at 1 y applied to
remaining duration
DW indicates disability weight; EMR, excess mortality rate; LE, life expectancy; PCSS, Perth Community Stroke Study; YLD, years lived with disability; YLL, years
of life lost.
index combines fatal and nonfatal burden in a single measure
being calculated from the sum of years of life lost (YLL), the
mortality component, and years lived with disability (YLD),
the morbidity component.11
Materials and Methods
The key source of empirical data was the WA Data Linkage System,
allowing the linkage of administrative health data from ⬇7 core
datasets.10 Data from a 5-year period (July 1997–June 2002) were
aggregated to increase case numbers. To address under-identification
of indigenous cases in administrative records, any stroke case
identified as indigenous on any hospital admission between 1988 and
2002 or on the death record was coded as indigenous.
Table 1 summarizes the components of the DALY calculated here,
including associated data sources and assumptions. For the fatal
component, stroke (International Classification of Diseases version
10 codes I60 –I69) deaths were extracted from the WA mortality
database and age-specific deaths were converted to years of life lost
by multiplying these by the standard life expectancies at age of death
based on the Global Burden of Disease standard.11 For nonfatal
burden, YLD were calculated as the product of age-specific and
sex-specific nonfatal incidence counts, duration of illness, and
stroke-specific disability weights (DW). Because of the high initial
fatality after stroke,13 nonfatal incident stroke was defined as patients
with first-ever stroke cases surviving at least 28 days. Each YLD
element was estimated differently, as described.
Incidence
Incidence was estimated from linked hospital and mortality data.
First, all admissions to WA private and public hospitals coded to
acute stroke (International Classification of Diseases version 10
codes I60, I61, I63, I64, and corresponding International Classification of Diseases version 9 codes) between 1988 and 2002 were
extracted from the Hospital Morbidity Data Collection. These codes
conform to guidelines for monitoring the incidence of stroke in
Australia.14 Within-hospital data linkage allowed identification of
first-ever hospital admissions for acute stroke pertaining to the 1997
to 2002 study period, using an 8-year clearance period to exclude
cases of previous stroke admissions, thus minimizing inclusion of
existing/prevalent cases. Second, individual-level mortality records
were merged with hospital records to ascertain survival status,
identifying those incident cases who survived to 28 days. Next, data
from the 1995 to 1996 Perth Community Stroke Study15 were used
to adjust incident counts for out-of-hospital strokes (assuming
admission proportions were 91% [age 15–74 years] and 88% [age 75
years and older]) and to estimate the proportion of out-of-hospital
28-day survivors. In- and out-of-hospital estimates were summed to
provide population-based indigenous and non-indigenous total and
nonfatal incidence by age and sex.
Duration
Duration was calculated separately for indigenous and nonindigenous people in DisMod II, specialized computer software that
models unknown disease measures from at least 3 known estimates,
producing consistent estimates of incidence, prevalence, case fatality, remission, and duration.12 Assuming zero remission and using
the nonfatal incidence calculated, the third input comprised the
excess mortality rate (EMR), reflecting the mortality risk in stroke
patients over and above the all-cause morality risk in the general
population.16,17
Age-specific and sex-specific EMR were calculated for prevalent
cases, defined as those recorded in the hospital admission dataset
since 1988 for acute stroke or sequelae of stroke (I69), who had
survived to 28 days, and who were still alive sometime in the 5-year
study period. The mortality risk in patients was calculated from the
ratio of deaths (attributed to age at death) and person-years of
follow-up during this observation period. The all-cause mortality in
the non-indigenous population was estimated from WA mortality
statistics, whereas that for indigenous people was based on allAustralian indigenous estimates.1 Because of uncertainty of indigenous estimates at older ages, trend lines were plotted in Excel
(Microsoft) using age-group-specific EMR up to 74 years as inputs
and extrapolating to older ages. Because male and female EMR age
Katzenellenbogen et al
trends differed, a power trend line was used for males and a
polynomial trend line was used for females.
DW
Quality-adjustment of years lived with disease are dealt with by the
use of DW. These values lie between 0 (representing perfect health)
and 1 (representing death) and are numerically equal to the complement of preference-based quality-of-life weights (DW ⫽ 1 ⫺
quality-of-life weight).11 Age-group-specific DW from a previous
WA study16 based on disability data from a community stroke study
and reflecting stroke-related disability, adjusted for prestroke disability, were utilized here for both indigenous and non-indigenous
analyses. DW estimated at 4 months (mean DW, 0.38) was applied
to YLD for the first year after stroke and DW estimated at 12 months
(mean DW, 0.31) was applied to the remaining duration.16
YLD
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
YLD were discounted (deflated) at 3% per year to take into account
society’s preference for short-term benefits, as is standard in Global
Burden of Disease studies.11 Uncertainty intervals (95%) for nonfatal
incidence and EMR were calculated using Monte Carlo simulation
(bootstrap) methods in @Risk® software.18 This software allows the
uncertainty from various sources to be captured at once, providing
uncertainty intervals that can be interpreted in a similar way to
confidence intervals. Nonfatal incident counts and the observed
deaths were assumed to follow a Poisson distribution. Uncertainty
intervals for YLD were calculated in DisMod II.
Age-Standardized Rates and Rate Ratios
Summary indigenous and non-indigenous age-standardized rates and
rate ratios were estimated by the direct method for incidence, YLD,
YLL, and DALY using the World Health Organization World
Standard population,19 a population age structure commonly used
when comparing rates between populations of varying age
distributions.
This study was performed under the auspices of the epidemiology
branch of the WA Department of Health. Ethics approval for the use
of de-identified linked records was obtained through a confidentiality
process allowing the epidemiology branch to undertake research
using the WA Data Linkage System housed in the Department of
Health.
Results
Incidence
A total of 378 indigenous and 10 285 non-indigenous people
were identified through the hospital records as having had a
first-ever stroke in the 5-year study period, with 80% of each
group surviving to 28 days. After adjustment for out-ofhospital cases, 419 indigenous and 11 441 non-indigenous
cases contributed to total rates, whereas 337 indigenous and
8993 non-indigenous cases contributed to nonfatal incidence
rates.
Total and nonfatal incidence rates increased with age, with
indigenous rates being significantly higher than nonindigenous rates at all ages to 74 years (Table 2). Indigenous
age-standardized rates ⱖ15 years were 2.6-times (95% CI,
2.3–3.0) higher in males and 3.0-times (95% CI, 2.6 –3.5)
higher in females compared with their non-indigenous counterparts. Rate ratios were even higher (males, 4.6; females,
5.8) when age was restricted to 15 to 64 years. Disparities
were of a similar magnitude for nonfatal stroke incidence.
Excess Mortality Rates and Duration
EMR increased and duration decreased with age for both
indigenous and non-indigenous people (Figure 1). EMR were
Stroke Study Using Data Linkage
1517
higher in indigenous than in non-indigenous people at all
ages, with disparities decreasing with age. The average
survival of first-ever stroke was 13 years lower in indigenous
males and 7 years lower in indigenous females compared
with their non-indigenous counterparts in the 15- to 24-year
age group. In the 75-year and older age group, differences in
duration between indigenous and non-indigenous people
were small.
Burden of Stroke
More than 60% of indigenous nonfatal stroke burden occurred in the 15- to 54-year age group, compared with 24% in
the non-indigenous population (Figure 2). Fatal burden dominated the profile, contributing 67% of the male and 53% of
the female indigenous DALY and 57% of male and 66% of
the female non-indigenous DALY. Unlike that for nonindigenous people, the mortality burden was highest and
fairly evenly spread between 35 and 74 years in indigenous
people.
Age-specific DALY rates increased with age to age 74
years in indigenous and non-indigenous males and females
(Table 3). In the 75-year and older age group, non-indigenous
rates continued to increase. Indigenous rates were substantially higher than non-indigenous rates at all ages younger
than 75 years. Indigenous age-standardized DALY rates were
⬇3-times higher for both sexes than corresponding nonindigenous rates (Table 2). When restricting the analysis to
ages 15 to 64 years, DALY rates were ⬇5-times higher in
indigenous people.
Discussion
Our results provide evidence of the substantially higher
burden of stroke and distinct epidemiological pattern in
indigenous compared with non-indigenous Western Australians. Differentials pertaining to the population aged 15 to 64
years cover the ages for which data quality was best and
inequalities were the most substantial. The elevated burden is
reflected not only in the composite DALY estimates but also
in the epidemiological components underlying DALY. As the
first detailed population-based estimates of indigenous stroke
incidence and burden in Australia (and incorporated in a
recent national burden of disease report20), these findings
highlight the need for comprehensive intersectoral interventions to reduce indigenous incidence and improve stroke
outcomes.
The disparity in stroke incidence is reflected in an agestandardized indigenous incidence rate 2.8-times that of
non-indigenous Western Australians, increasing to 5.1 in ages
15 to 64 years. These findings are substantiated by other
indigenous health statistics in Australia that reflect stroke
age-standardized hospitalization rates that are 2.1-times4
those and stroke death rates that are 1.9-times those of
non-indigenous Australians.4,6 EMR were also substantially
higher in the indigenous population, even with the extremely
high background population mortality. The high EMR and the
lower life expectancy21 resulted in much lower indigenous
duration estimates compared with that for non-indigenous
people, particularly among younger males.
1518
Stroke
June 2011
Table 2.
Total and Nonfatal Stroke Incidence Rates, by Age Group, Sex, and Indigenous Status: Western Australia 1997–2002
Total Incidence Rates Per 100 000
Males
Females
Non-Indigenous
Age Group
N
Rate
15–34
141
10
35–54
637
47
55–64
854
208
65–74
1523
553
75 or older
2514
1490
Indigenous
95% CI
(9 –12)
32
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Rate ratio
Rate
11
81
268
(210–327)
432
32
48
863
(621–1106)
484
123
(526–581)
35
1342
(900–1783)
1091
377
(1432–1548)
29
1990
(1276–2704)
3620
1374
(140–147)
49
N
146
(43–51)
(326–427)
2.6
(47–52)
226
1.0
Indigenous
95% CI
(9 –13)
37
75
236
(183–289)
44
683
(482–885)
(355–399)
40
1286
(892–1679)
(1329–1418)
28
1920
(1218–2622)
(110–115)
33
(3.9–5.4)
95% CI
(21–53)
(29–35)
341
1.0
(190–263)
4.6
Rate
21
208
113
(2.3–3)
N
(112–134)
5772
377
1.0
ASR 15–64
95% CI
(17– 47)
211
144
Rate ratio
Rate
18
(194–222)
5669
ASR† ⱖ15
N
Non-Indigenous
3.0
(31–35)
194
1.0
5.8
(295–387)
(2.6–3.5)
(162–226)
(4.9–6.9)
Nonfatal Incidence Rates Per 100 000
Males
Females
Non-Indigenous
Age Group
N
15–34
123
Rate
35–54
556
41
55–64
754
184
9
65–74
1279
465
75 or older
1859
1102
95% UI*
(7–11)
Rate ratio
ASR† 15–64
Rate ratio
N
Rate
15
28
Non-Indigenous
95% UI*
(15–44)
1.0
10
95% UI*
(8–13)
N
Rate
18
33
95% UI*
(20–48)
61
202
(149–259)
366
27
(20–35)
67
208
(163–263)
40
716
(502–951)
399
102
(91–115)
34
526
(359–718)
(433–508)
30
1155
(766–1610)
888
307
(283–340)
29
936
(603–1270)
(1031–1187)
23
528
(951–2243)
2644
1003
(945–1078)
20
1380
(817–2042)
304
(258–350)
(113–120)
1.0
43
Rate
(36–47)
169
116
N
126
Indigenous
(169–202)
4571
ASR† ⱖ15
Indigenous
4422
2.6
(41–45)
180
4.2
(2.2–3.0)
(147–213)
(3.5–5)
168
88
(85–90)
1.0
28
1.0
266
3.0
(26–30)
161
5.8
(225–306)
(2.6–3.5)
(132–190)
(4.7–7)
ASR indicates age-standardized rate; CI, confidence interval; UI, uncertainty interval.
*UI indicates uncertainty interval calculated in DisMod II, derived from bootstrap methods, can be interpreted similar to confidence intervals.
†ASR indicates age-standardized rate standardized to World Health Organization Standard Population.
The high levels of total stroke burden in indigenous
Western Australians concurs with the findings of an analysis
in the Northern Territory, where the Aboriginal population
had an all-cause DALY rate 2.5-times that of the nonAboriginal population, with rate ratios increasing to 4.1 in the
35- to 54-year age group.22 Cardiovascular disease was the
highest contributor to total indigenous disease burden in that
study; stroke was not reported separately.22 Outside Australia,
differentials in stroke burden also have been shown for ethnic
minorities. A 2-fold ethnic differential in stroke incidence
was reported in urban population-based studies in the United
States and United Kingdom, with greater differentials in
younger people.23–26 Differentials in stroke mortality were
somewhat lower than differentials in incidence among blacks
in the United States compared with the general population.23
Maori and Pacific Islander people experience higher incidence, disability, and mortality from stroke than other New
Zealanders.27 In North America, quantification of differentials pertaining to indigenous populations has proved unreli-
able because of difficulties in identifying indigenous cases in
routine data.28
The uncertainties in our estimates reflect the challenges of
obtaining good-quality epidemiological data for indigenous
people.7 Although our case numbers were small, the percentage of indigenous cases having consistent indigenous codes
on the linked hospital and mortality records (84%) corresponds closely with the proportion of correct codes found in
the 2000 WA validation study of indigenous coding on
hospital records.9 Consequently, our estimates are less likely
to be underestimated as a result of under-identification. A
number of assumptions also had to be made in the calculation
of indigenous YLD. First, data from an urban populationbased stroke study were used to account for out-of-hospital
fatal cases. A pilot study to evaluate this assumption found
little evidence of difference by Aboriginality and residential
location, suggesting that its use in this study is acceptable,
given the lack of more definitive data. Second, DW were
assumed to be the same for the indigenous population as the
Katzenellenbogen et al
Stroke Study Using Data Linkage
1519
Figure 1. Excess mortality rate (EMR)
and modeled duration of nonfatal stroke,
by age group, sex, and indigenous status. Western Australia 1997 to 2002.
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
rest of WA, but indigenous people generally have more
comorbidity,7 may have a different severity profile, and may
value health-related quality of life differently. However, the
DW utilized in this analysis account for prestroke comorbidity and disability. Additionally, the global approach advo-
cates using the same DW across different socioeconomic and
cultural contexts to allow valid comparisons between
populations.11
Despite these limitations and assumptions, the results
document extreme differentials by Aboriginality in stroke
Figure 2. Comparison of indigenous and non-indigenous burden (disability-adjusted life years [DALY]) disaggregated into fatal (years of
life lost [YLL]), and nonfatal (years lived with disability [YLD]) components, by age and sex.
1520
Stroke
June 2011
Table 3. Disability-Adjusted Life Years as Rates Per 100 000 by Indigenous Status, Age, and Sex, and as
Age-Standardized Rates and Age-Standardized Rate Ratios
Males
Females
Non-Indigenous
Age Group
Indigenous
Non-Indigenous
Indigenous
Count
Rate
Count
Rate
Count
15–24
294
45
33
115
25–34
465
67
82
318
35–44
1226
172
252
45–54
2432
382
196
55–64
3716
907
200
3589
2541
649
235
3675
65–74
6418
2332
246
9428
5500
1902
215
6821
75⫹
10 630
6302
126
8512
18 178
6900
91
6206
Total
25 180
ASR* ⱖ15
(95% CI)
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
RR
ASR* 15 ⫺64
RR
(95% CI)
Count
Rate
279
45
45
153
720
106
34
128
1331
866
121
197
996
1746
1835
298
183
1492
1135
29 919
640
(633–648)
2027
999
573
(1909–2145)
1.0
(95% CI)
(95% CI)
Rate
3.2
1598
(567–580)
1.0
(3.0–3.4)
243
(237–248)
2.8
(2.6–3.0)
1133
(1052–1213)
1.0
(1499–1697)
4.7
194
995
(190–199)
(921–1069)
1.0
(4.3–5.0)
5.1
(4.7–5.5)
ASR indicates age-standardized rates; CI, confidence interval; RR, rate ratio.
*ASR indicates age-standardized rates standardized separately to World Health Organization World Standard Population 15 y and older
and 15– 64 y.
burden, the scale of which has been mirrored elsewhere for
other conditions.1,4,6,20,29 The number of individuals surviving
stroke and living with a disability is a major concern. The
underlying causes of these health disparities are complex,
with socioeconomic, historical, and environmental disadvantage operating through behavioral and physiological risk
factors influencing health outcomes. Although this study was
not designed to attribute causality, a range of preventive and
therapeutic strategies are warranted. It is apparent that the
multiple determinants of indigenous health in general and
stroke specifically require a committed, long-term, multipronged, and intersectoral response.30,31 Within the health
sector, a comprehensive, integrated, and culturally sensitive
primary health care program is required, providing training
and support for indigenous and non-indigenous health care
workers.30,31 Reducing highly prevalent behavioral factors
associated with stroke (for example, smoking and substance
abuse) and management of physiological risk factors,
including hypertension, diabetes, and rheumatic heart disease, have been identified as priorities.29 In WA, indigenous needs in the acute stroke care context were found to
relate to dissemination of knowledge about stroke and
services, addressing cultural needs, geographical isolation,
access to services, and improved recognition of the impact
on and support for families.31 Access to acute care and
rehabilitation, particularly stroke units outside the metropolitan areas, is considered the foundation of secondary
and tertiary prevention.31
Policies to address the indigenous disease burden have
been disappointing32 and indigenous health remains under-
funded relative to the need.32,33 The Commonwealth “Closing
the Gap” strategy34 is promising in terms of increasing
funding and focus. Estimates of the burden of stroke, as
reported here, can provide the quantitative base from which to
monitor stroke outcomes.20
Acknowledgments
The Data Linkage Branch within the Western Australia Department
of Health linked the hospitalization and mortality data using the
Western Australia Data Linkage System.
Sources of Funding
Judith Katzenellenbogen was supported by a postgraduate scholarship from the National Health and Medical Research Council
(Australia).
Disclosures
None.
References
1. Hill K, Barker B, Vos T. Excess indigenous mortality: are indigenous
Australians more severely disadvantaged than other indigenous populations? Int J Epidemiol. 2007;36:580 –589.
2. Vos T, Barker B, Begg S, Stanley L, Lopez AD. Burden of disease and
injury in Aboriginal and Torres Strait Islander Peoples: the indigenous
health gap. Int J Epidemiol. 2009;38:470 – 477.
3. Australian Bureau of Statistics. Experimental estimates of the Aboriginal
and Torres Strait Islander population 1991 to 2009. Canberra, Australia:
Australian Bureau of Statistics; 2004:ABS catalog number 3238.0.
4. Australian Bureau of Statistics, Australian Institute of Health and
Welfare. The health and welfare of Australia’s Aboriginal and Torres
Strait Islander peoples. Canberra, Australia: ABS and AIHW; 2005.
Katzenellenbogen et al
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
5. Zhao Y, Dempsey K. Causes of inequality in life expectancy between
indigenous and non-indigenous people in the Northern Territory,
1981–2000: a decomposition analysis. Med J Aust. 2006;184:490 – 494.
6. Thrift AG, Hayman N. Aboriginal and Torres Strait Islander peoples and
the burden of stroke. Int J Stroke. 2007;2:57–59.
7. Australia Institute of Health and Welfare. Data quality of Aboriginal and
Torres Strait Islander peoples. Canberra, Australia: AIHW; 2004:catalog
number HW179.
8. Draper GK, Somerford PJ, Pilkington AG, Thompson SC. What is the
impact of missing indigenous status on mortality estimates? An
assessment using record linkage in Western Australia. Aust N Z J Public
Health. 2009;33:325–331.
9. Young MJ. Assessing the quality of identification of Aboriginal and
Torres Strait Islander people in Western Australia Hospital Data, 2000.
Perth, Australia: Department of Health of Western Australia; 2001:Health
Information Centre Occasional Paper 13.
10. Holman CD, Bass AJ, Rouse IL, Hobbs MS. Population-based linkage of
health records in Western Australia: development of a health services
research linked database. Aust N Z J Public Health. 1999;23:453– 459.
11. Murray CL, Lopez AD, eds. The Global Burden of Disease: a comprehensive assessment of mortality and disability from diseases, injuries, and
risk factors in 1990 and projected to 2020. Cambridge, MA: Harvard
School of Public Health; 1996.
12. Barendregt JJ, Oortmarssen GJ, Vos T, Murray CJL. A generic model for
the assessment of disease epidemiology: the computational basis of
DisMod II. Population Health Metrics. 2003;1:4.
13. Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology:
a review of population-based studies of incidence, prevalence, and casefatality in the late 20th century. Lancet Neurol. 2003;2:43–53.
14. Jamrozik K, Dobson A, Hobbs MS, McElduff P, Ring I, D’Este K, Crome
M. Monitoring the incidence of cardiovascular disease in Australia.
Canberra, Australia: Australian Institute of Health and Welfare; 2001:
catalog number CVD 16.
15. Jamrozik K, Broadhurst RJ, Lai N, Hankey GJ, Burvill PW, Anderson
CS. Trends in the incidence, severity, and short-term outcome of stroke in
Perth, Western Australia. Stroke. 1999;30:2105–2111.
16. Katzenellenbogen JM, Vos T, Begg S, Somerford P, Codde JP, Semmens
JB. Disability burden due to stroke in Western Australia—new insights
from linked data sources. Int J Stroke. 2010;5:269 –277.
17. Katzenellenbogen JM, Vos T, Somerford P, Begg S J, Semmens JB,
Codde JP. Excess mortality rates for estimating the non-fatal burden of
stroke in Western Australia: a data linkage study. Cerebrovasc Dis.
2010;30:57– 64.
18. Palisade. @RISK®: advanced risk analysis for spreadsheets. New York,
NY: Palisade; 2004.
19. Ahmad OB, Boshci-Pinto C, Lopez AD, Murray CJL, Lozano R, Inone
M. Age standardisation of rates: a new WHO standard. Geneva: World
Health Organization. Available at: http://www.who.int/healthinfo/
paper31.pdf. Accessed March 15, 2008.
Stroke Study Using Data Linkage
1521
20. Vos T, Barker B, Stanley L, Lopez A. The burden of disease and injury
in Aboriginal and Torres Strait Islander peoples 2003. Brisbane, Australia: Centre for Burden of Disease and Cost Effectiveness, School of
Population Health, University of Queensland; 2008.
21. Australian Bureau of Statistics. Experimental Life Tables for Aboriginal
and Torres Strait Islander Australians, 2005–2007. Canberra, Australia:
ABS; 2009:catalog number 3302.0.55.003.
22. Zhao Y, Guthridge S, Magnus A, Vos T. Burden of disease and injury in
Aboriginal and non-Aboriginal populations in the Northern Territory.
Med J Australia. 2004;180:498 –502.
23. Stansbury JP, Jia H, Williams LS, Vogel WB, Duncan PW. Ethnic
disparities in stroke: epidemiology, acute care, and postacute outcomes.
Stroke. 2005;36:374 –386.
24. Wolfe CDA, Rudd AG, Howard R, Coshall C, Stewart J, Lawrence E,
Hajat C, Hillen T. Incidence and case fatality rates of stroke subtypes in
a multiethnic population: the South London Stroke Register. J Neurol
Neurosurg Psych. 2002;72:211–216.
25. Sacco RL, Boden-Albala B, Gan R, Chen X, Kargman DE, Shea S, Paik
MC, Hauser WA. Stroke incidence among white, black, and Hispanic
residents of an urban community: the Northern Manhattan Stroke Study.
Am J Epidemiol. 1998;147:259 –268.
26. Kissela B, Schneider A, Kleindorfer D, Khoury J, Miller R, Alwell K,
Woo D, Szaflarski J, Gebel J, Moomaw C, Pancioli A, Jauch E, Shukla
R, Broderick J. Stroke in a biracial population: the excess burden of
stroke among blacks. Stroke. 2004;35:426 – 431.
27. Carter K, Anderson C, Hacket M, Feigin V, Barber PA, Broad JB, Bonita
R, Auckland Regional Community Stroke (ARCOS) Study Group.
Trends in ethnic disparities in stroke incidence in Auckland, New
Zealand, during 1981 to 2003. Stroke. 2006;37:56 – 62.
28. Gillum RF. The epidemiology of stroke in Native Americans. Stroke.
1995;26:514 –521.
29. Australia Institute of Health and Welfare. Heart, stroke and vascular
diseases–Australian facts. Canberra, Australia: Australian Institute of
Health and Welfare, National Heart Foundation of Australia. 2004:
Cardiovascular series number 22.
30. Brown A, Morrissey MJ, Sherwood JM. Uncovering the determinants of
cardiovascular disease among indigenous people. Ethnicity Health. 2006;
11:191–210.
31. National Stroke Project Steering Committee. National Stroke Unit Program: Aboriginal Stroke Project. Canberra: National Stroke Foundation
2004. Available at: http://www.cucrh.uwa.edu.au/project/files/Aboriginal_
Stroke_Report.pdf. Accessed January 15, 2008.
32. Ring IT, Brown N. indigenous health: chronically inadequate responses to
damning statistics. Med J Australia. 2002;177:629 – 631.
33. Houston S, Mooney G. Aboriginal health strategy: a case for greater
imagination and priorities. Aust N Z J Public Health. 2001;25:476 – 477.
34. Rudd K. Apology to Australia’s indigenous peoples. Canberra, Australia:
Hansard, House of Representatives, Parliament of Australia; 2008.
Burden of Stroke in Indigenous Western Australians: A Study Using Data Linkage
Judith M. Katzenellenbogen, Theo Vos, Peter Somerford, Stephen Begg, James B. Semmens
and James P. Codde
Downloaded from http://stroke.ahajournals.org/ by guest on June 17, 2017
Stroke. 2011;42:1515-1521; originally published online April 14, 2011;
doi: 10.1161/STROKEAHA.110.601799
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2011 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/42/6/1515
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/