Background and Purpose This study compares stroke incidence, 28-day case fatality, and hospital management for Maori, Pacific Islands people (“Pacific people”), and others (mostly Europeans) living in Auckland, New Zealand.
Methods Data come from the Auckland Stroke Study, a population-based study that registered all stroke events occurring among Auckland residents aged 15 years or more during a 1-year period ending February 29, 1992.
Results During the study year, 1803 stroke events were registered, including 82 (4.5%) in Maori, 113 (6.3%) in Pacific people, 1572 (87.2%) in Europeans, and 36 (2.0%) in others of Indian or Chinese origin. The mean±SD age of stroke patients was 55.0±16.0 years in Maori, 59.7±14.9 years in Pacific people, and 73.3±12.1 years in Europeans. Maori and Pacific people have significantly higher estimated relative risks of stroke compared with Europeans (OR, 1.34; 95% confidence interval [CI], 1.05 to 1.67 in Maori; and OR, 1.63; 95% CI, 1.33 to 1.98 in Pacific people). Maori and Pacific people also have higher estimated relative risks of death within 28 days of stroke compared with Europeans, especially men.
Conclusions This study indicates that there are important differences in stroke incidence rates and case fatality among the major ethnic groups in Auckland. The reasons for the higher incidence rates in Maori and Pacific people may be related to levels of risk factors, but this requires further investigation. Ongoing monitoring of stroke incidence and outcome should include separate reporting by ethnicity.
Stroke is primarily a condition that affects the elderly population; overall, 1 in 12 of all deaths in men and 1 in 8 of all deaths in women are ascribed to stroke. Recent decades have witnessed a decline in death rates from stroke in both men and women and in all age groups, including the oldest.1 Because of the comparatively young age of the Maori and Pacific Islands populations, it is often assumed that stroke is not an important condition in these ethnic groups. For example, in New Zealand, there were 2788 deaths attributed to stroke in 1993, but only 88 in the Maori population.2 The number of deaths among Pacific Islands people (“Pacific people”) is not routinely reported. Although age-standardized stroke death rates are higher in Maori than non-Maori, both Maori men and Maori women have experienced a favorable improvement since the 1970s.2
Information about incidence and case fatality is required to understand reasons for the decline in stroke mortality, as well as for planning hospital and health services for patients with stroke. This article reports the results of the Auckland Stroke Study, which documented all new acute stroke events occurring in Auckland residents over a 12-month period and examines ethnic differences in stroke incidence and case fatality. Auckland is the largest Polynesian city in the world and therefore provides an opportunity to investigate differences in stroke incidence and outcome in Maori and Pacific Islands peoples. Ethnic differences in total mortality and coronary heart disease have been reported,3 4 but this is the first study to report ethnic differences in stroke incidence and outcome.
Subjects and Methods
All residents of Auckland who experienced an acute stroke during the year ending February 29, 1992, were included in the study. Stroke was defined as “rapidly developing signs of focal (or global) disturbance of cerebral function, leading to death or lasting longer than 24 hours, with no apparent cause other than vascular.”5 This definition includes spontaneous subarachnoid hemorrhage but excludes subdural and extradural hematomas and transient ischemic attacks. A first-ever event refers to stroke in persons who have not previously experienced a stroke. A recurrent event in the study year was considered to have occurred if more than 28 days had elapsed from the initial episode after careful review of medical records and application of the World Health Organization criteria.
The Auckland region, with a total population of 945 000 (1991 census), comprises a quarter of New Zealand’s total population. Maori and Pacific people comprise 22% of the total Auckland population (11% Maori and 11% Pacific people), more than the national average. Ethnicity was classified by self-identification based on cultural affiliation to match the definitions used in the 1991 census. The Maori ethnic group includes those who stated New Zealand Maori as either their sole ethnic group or as one of several ethnic groups. The group from the Pacific Islands includes those who identified themselves as Samoan, Niuean, Tongan, Tokelauan, Fijian, or Cook Islander. Persons of Indian or Chinese origin were included with Europeans in all analyses because there were too few (n=36) to calculate meaningful rates, their age distribution and mean age closely matched that of the Europeans, and their inclusion enabled the application of the rates to the total population.
The methods used to identify stroke events have been described in detail elsewhere.6 7 Briefly, hospital events were identified through daily searches of hospital admission lists of all public hospitals. A wide range of conditions were investigated to ensure that all strokes that met the criteria were included. All death certificates and autopsy reports were checked on a monthly basis, and any mention of stroke as a cause of death was followed up with the person who had completed the death certificate to determine whether the death had been preceded by an acute event.
To locate eligible people who neither died nor were admitted to a public hospital, a 25% random sample of all general practitioners in Auckland were asked to refer patients in their practices suspected of having had a stroke but who were cared for either at home or in a rest home or private hospital. Stroke events in patients enrolled with these selected practitioners were included in the study regardless of their source of referral. Those patients not so enrolled were excluded unless they were admitted to the hospital or died of the stroke; in the calculation of rates, these patients were weighted by a factor of 4 to represent the total nonfatal community-treated population. Trained nurse interviewers interviewed all patients (or in the case of fatal or speech-impaired patients, a close relative or caregiver) once it was established that the clinical and residential criteria were met and as soon after the event as possible, or in the case of death, after 6 weeks had elapsed.
Incidence rates refer to a first-ever-in-a-lifetime event, and event rates include both first-ever and recurrent strokes. Because total event rates are more useful for estimating need for services, and there was no difference in the proportion of recurrent strokes among the major ethnic groups, rates reported here include recurrent events. Age-standardized rates were calculated by the direct method using the Segi world population for the age range of 15 years and over using four age groups: 15 to 64 years, 65 to 74 years, 75 to 84 years, and 85 years and over. Rates are reported with 95% confidence intervals (CIs) and take into account the sampling fraction used for the identification of nonfatal nonhospitalized stroke.7
A logistic regression model was developed to estimate the relative risk (odds ratios [ORs]) in stroke incidence by ethnic group. The age group 85 years and over was omitted from the logistic model because of the very small number of stroke cases occurring in the few non-Europeans in the population over 85 years of age. Mantel-Haenszel relative risks, adjusting for age and sex, were determined for 28-day case fatality.
During the 12-month study period, a total of 1803 stroke events occurred among 1769 patients, 69% of which were first-ever strokes. Of all the events registered, 82 (4.5%) occurred in Maori, 113 (6.3%) in Pacific people, 1572 (87.2%) in Europeans, and 36 (2.0%) in persons of Indian or Chinese origin. Of the 71 Pacific people who provided additional information as to their country of birth, 35 were Samoan, 18 were from the Cook Islands, 13 were Tongan, 3 were Fijian, and 2 were Niuean; because the numbers of stroke events in these populations were too small and the denominators too unreliable to make meaningful comparisons, Pacific people were analyzed as one group.
In Maori and Pacific people, 70.7% and 57.5%, respectively, of all stroke events occurred in those under the age of 65 years compared with only 19.3% in Europeans. The mean±SD age of stroke patients varied according to ethnic group: 55.0±16.0 years in Maori and 59.7±14.9 years in Pacific people compared with 73.3±12.1 years in Europeans. The proportion of Maori and Pacific people who experienced a subarachnoid or intracerebral hemorrhage (23.2% and 25.7%, respectively) was greater than that in Europeans (9.8%), although when adjusted for differences in the age distribution, these are not significantly different, except for Pacific people under 65 years of age (OR, 1.45; 95% CI, 1.00 to 2.12).
Age-specific event rates rise steeply with increasing age in all age groups in women and to age 84 years in men (Table 1⇓). The numbers of stroke events in the age group 85 years and over are very small, especially in Maori and Pacific people. Age-standardized event rates are highest in Pacific men. The rates appear higher in both Pacific women and Maori women when compared with European women, but the CIs are wide and overlap between these groups (Table 1⇓). There are no differences in rates between Maori men and women, but the rates in both Pacific and European men appear higher than in women (Figure⇓).
After adjustment for age and sex, Maori and Pacific people have significantly higher estimated relative risks of stroke (OR, 1.34; 95% CI, 1.05 to 1.67 in Maori; and OR, 1.63; 95% CI, 1.33 to 1.98 in Pacific people) than Europeans. Similarly, the estimated relative risk of stroke in women is lower than that in men (OR, 0.73; 95% CI, 0.66 to 0.81).
Case fatality increases with age only in European men (Table 2⇓). Crude case fatality is higher in Maori men (33%) than in Pacific (19%) or European (20%) men and higher in Pacific women (43%) than in European (21%) or Maori (18%) women. The crude data are heavily influenced by the age distribution of cases and the small numbers of Maori and stroke deaths in Pacific people. After adjustment for age and sex, the risk of death within 28 days of stroke is higher in Maori (OR, 1.37; 95% CI, 0.90 to 2.09) and Pacific people (OR, 1.61; 95% CI, 1.15 to 2.27) than in Europeans.
Overall, 27.5% of all stroke events were managed entirely outside the hospital, either at home or in a rest home or private hospital, and there were differences between the ethnic groups with respect to admission to public hospitals: 89.0% of Maori and 78.8% of Pacific people were admitted to the hospital compared with 71.1% of Europeans. However, after adjustment for age differences between the groups, Europeans were 16% less likely to be admitted than Maori (OR, 0.84; 95% CI, 0.75 to 0.95) and 13% less likely to be admitted than Pacific people for those younger than 65 years of age (OR, 0.87; 95% CI, 0.76 to 0.99). Once in the hospital, the mean stay was greater for Maori (43 days) than for Pacific people (22 days) and Europeans (28 days). However, the median length of stay for Maori (13 days) was no different from that for Pacific people (13 days) or Europeans (15 days).
These findings demonstrate that there are significant ethnic differences in stroke incidence, with rates being higher for both Maori and Pacific people compared with all Europeans. A major finding from the study is the high rate of stroke in Pacific men, given the low mean age of stroke in both Maori and Pacific people.
There are a number of strengths in this study and several possible explanations for the ethnic differences in rates. The Auckland Study, with a total population of almost 1 million people, is large by world standards. The numerator is likely to be more accurate than other sources of available data. Unlike studies based on mortality statistics that rely on the undertaker for determining ethnic affiliation, this study interviewed all patients or their proxies and allowed for a self-ascribed ethnic status. There is good reason to believe that virtually all eligible acute stroke events were identified, irrespective of ethnicity, because multiple comprehensive case-finding methods were used, not only for people treated in the hospital but also for those cared for in the community. The only possible source of underrepresentation could be mild strokes occurring at home in individuals who have no general practitioner and who received no community-based services. This may be more likely to occur in Pacific people, Maori, and new immigrants for whom traditional healers may provide advice and treatment.8
Another potential source of bias in the calculation of rates is the possibility that the denominator for calculating rates taken from the census information may not be as accurate for Pacific people. The numerator may include nonresidents of Auckland, which would tend to exaggerate the rates in people from the Pacific Islands. The majority of Pacific Islands stroke patients reported living in New Zealand for more than 10 years, suggesting that at least in the older age groups, these subjects came from a relatively stable population. On the other hand, the denominator may exclude usual residents who for one reason or another are not included in the census.9
This is the first study to report the incidence and case fatality of Maori and Pacific people separately. The high stroke rate in Pacific men is a cause for concern, and it would be useful to know which population groups were most affected. A comparison of the outcome at 1 month after the stroke suggests a less favorable outcome for Maori men and Pacific women. Part of the explanation may lie in the high proportion of subarachnoid and intracerebral hemorrhage events among Pacific people under 65 years compared with European stroke patients and the associated high early case fatality. Maori have a higher incidence of middle cerebral artery aneurysms than Europeans.10 Another explanation may relate to the less healthy profile of Maori and Pacific people compared with Europeans. Maori and Pacific people who are hypertensive are less likely to receive treatment than Europeans.11 Cigarette smoking, diabetes, and obesity, which are other major risk factors for stroke, are also higher in Maori.10 12 13
Most incidence studies assume a homogeneous population, and few report differences in ethnic groups. The black/white differences in the United States consistently show higher death rates and higher hospitalization rates for stroke in blacks compared with whites. Consistent with our study, the excess stroke rate in blacks in the United States occurs mainly at younger ages (around 55 years).14 It has also been suggested that blacks in the United States have more severe strokes than whites.15 16
This study indicates that there are important differences in stroke incidence rates and case fatality among the major ethnic groups in New Zealand. The reasons for the higher incidence rates in Maori and Pacific people are not clear and require further investigation. Surveillance and monitoring of stroke incidence and outcome should continue and include separate reporting by ethnicity.
We wish to acknowledge the Health Research Council of New Zealand, the National Heart Foundation, and the Grand Lodge of Freemasons of New Zealand for their support of this research. In addition, grateful acknowledgment is made to Susanna Yuen, Siew-Choo Soo, and John Anderson for their assistance with data analysis. Alistair Stewart, Chris Bullen, and Colin Tukuitonga provided useful comments on early drafts.
- Received September 3, 1996.
- Revision received January 21, 1997.
- Accepted January 21, 1997.
- Copyright © 1997 by American Heart Association
New Zealand Health Information Service. Mortality and Demographic Data 1993. Wellington, NZ: Ministry of Health; 1995.
Bonita R, Broad JB, Anderson NE, Beaglehole R. Approaches to the problems of measuring the incidence of stroke: the Auckland Stroke Study, 1991-1992. Int J Epidemiol. 1995;24:535-542.
Ngata P, Pomare E. Cultural factors in medicine taking: a Maori perspective. New Ethicals. 1992;29:43-50.
Bathgate M, Alexander D, Mitikulena A, Borman B, Roberts A, Grigg M. The Health of Pacific Islands People in New Zealand: Analysis and Monitoring Report 2. Wellington, NZ: Public Health Commission; 1994.
Howard G, Anderson R, Sorlie P, Andrews V, Backlund E, Burke GL. Ethnic differences in stroke mortality between non-Hispanic whites, Hispanic whites, and blacks: the National Longitudinal Mortality Study. Stroke. 1994;25:2120-2125.
Kuhlemeier KV, Stiens SA. Racial disparities in severity of cerebrovascular events. Stroke. 1994;25:2126-2131.