Background and Purpose Stroke is the second most common cause of death in Taiwan. We studied its prevalence, risk factors, and mortality in a rural Chinese population.
Methods A door-to-door survey of stroke was conducted in two Kinmen Island townships with a total population of 26 105 people. Our target population (n=5061) consisted of all the registered residents in these townships who were aged ≥50 years on August 1, 1993. All participants were given a standardized neurological examination and a questionnaire.
Results The participation rate was 77.4% (n=3915). Ninety-six cases of completed stroke were identified. Eighty-nine patients had one, 6 patients had two, and 1 patient had three episodes of stroke. The prevalence of stroke in persons aged ≥50 years was 24.5 per 1000 (95% confidence interval, 19.7 to 29.3 per 1000). Prevalence increased with age. Statistically significant risk factors associated with stroke included hypertension, diabetes mellitus, and irregular heartbeats. Of the stroke survivors, 59% were independent in activities of daily living and 71% could walk independently.
Conclusions Compared with other countries, Taiwan has a moderately high prevalence of stroke. The risk factors for stroke in this rural region of a developing country are similar to those in developed countries. Most stroke survivors here perform their daily activities independently without outside assistance.
In Taiwan, stroke is the second most common cause of death, after cancer.1 Surviving patients often experience long-term disability. Prevalence studies allow investigators to assess the social impact of stroke. A previous study found that the decline in the age-adjusted mortality rate for stroke in Taiwan was slower than declines in Japan and in the United States.2 Few community studies on stroke prevalence have been conducted in Taiwan. A previous study conducted in 19863 needs updating since substantial changes in frequency may have occurred since then. Socioeconomic, environmental, dietary, and other factors that affect stroke in Taiwan are quite different from those in other countries and regions, especially in the rural parts of Taiwan. These considerations led us to believe that a contemporary survey in Kinmen could have considerable epidemiological interest.
The aim of the present study is threefold, namely (1) to obtain the prevalence of stroke based on a community study, (2) to determine the relationship between stroke and certain defined risk factors, and (3) to describe the relationship between stroke and physical disability in this community.
Subjects and Methods
The study of prevalence and risk factors of stroke is part of the Kinmen Neurological Disorders Survey, a community-based study that also covers other disorders such as dementia, Parkinson's disease, migraine, essential tremor, and depression.
Kinmen is located approximately 154 miles (248 km) west of Taiwan and 25 miles (41 km) east of mainland China (Fig 1⇓). It consists of four townships with a total population of 45 807 in 1993. Among the aged the population is quite stable and the rate of immigration and emigration is very low, although many younger people move to Taiwan. All the residents of Kinmen are ethnic Chinese. A large proportion of the population are farmers, and there is no modern industry on the island. The four townships have similar demographic compositions. Public health clinics provide primary medical services for the population. Secondary care is provided by local hospitals, and tertiary care must be sought in Taiwan. (A more detailed description of the population, including its geographic, climatic, economic, and other characteristics, may be found in other sources.4 5 )
We chose the two townships of Kin-Hu and Kin-Cheng primarily for logistic reasons; there is no evidence that these townships differ from other townships in Kinmen. Each township in Kinmen has several local registry offices that keep records on the vital statistics of the residents. A general health survey of Kinmen was conducted in 1990, and we obtained a list of the registered residents of these two townships for that year. We updated the survey results with the help of the local registry offices. The total population of the two target townships on the prevalence day, August 1, 1993, was 26 105 (13 690 males and 12 415 females). The target population on the prevalence day, consisting of registered residents aged ≥50 years in these two townships, was 5061. The population included residents in households and in an institution. This institution for elderly people who have no relatives on the island housed 33 individuals (25 men, 8 women) or 0.65% of our target population on the prevalence day.
The study was based on a door-to-door survey. All participants were interviewed and examined by a physician from a group of neurologists. The survey period spanned 1 year and 48 days, namely, August 1, 1993, to September 17, 1994.
We used the World Health Organization criteria to define stroke as “rapidly developing clinical signs of focal (or global) disturbance of cerebral functions, lasting for more than 24 hours or leading to death, with no apparent cause other than that of vascular origin.”6 Ischemic cerebral infarction, intracerebral hemorrhage, and subarachnoid hemorrhage were included, but transient ischemic attacks were excluded.
Approximately 2 weeks before the planned evaluation date, a letter was mailed to each prospective subject, indicating the date and place for neurological evaluation. Most of the participants came to the local health clinic for evaluation. If they were unable to come for any reason, the neurologists called on them at their homes.
Interviews were conducted with a series of questions on demographics, family history, personal disease history, smoking and drinking habits, diet preferences, lifestyle, and activities of daily living (ADL). Physical examination included measurements of body weight, height, blood pressure, cardiac rhythm auscultation, and neurological evaluation. Body mass index was obtained by dividing body weight (kilograms) by height (meters) squared.
We graded ADL on a 5-point scale7 : 1, no significant disability, can perform all usual activities; 2, slight disability, cannot perform some of the activities previously able to perform but can attend to everyday needs without assistance; 3, moderate disability, requires some help but can walk without assistance; 4, moderately severe disability, cannot walk without assistance or attend to bodily needs without assistance; and 5, severe disability, bedridden, incontinent, and requires constant nursing care and attention.
Findings on all possible stroke cases were discussed at consensus meetings for diagnosis. When the subject was diagnosed as a stroke case, information was obtained concerning the date of the cerebrovascular accident, the patient's age at the time of stroke, the initial symptoms and signs, and hospital admission and diagnosis. Stroke survivors received the Mini-Mental State Examination.8 The diagnosis of dementia was made by consensus according to the Diagnostic and Statistical Manual of Mental Disorders, edition 3, revised (DSM-III-R) criteria.9
The crude point prevalence ratio was defined as the cases present on prevalence day per 1000 people. The figures were expressed with 95% confidence intervals (CIs) for a single proportion. We used t tests to determine differences between means. Logistic regression and Mantel-Haenszel10 analyses were used to test the odds ratio of risk factors in stroke survivors. A value of P<.05 was adopted for statistical significance.
Descriptive and Demographic Data
From 5061 targeted subjects, a total of 3915 (77.4%), comprising 1966 men and 1949 women, participated in the study. The remaining 1146 were not included for the following reasons: 473 persons were in town but not at home on each of three house calls; 402 were living with their families away from Kinmen during the survey period; 172 refused to participate; 28 died during the survey period before evaluation was completed; and 71 had incomplete data. The participants were older (64.6±10.4 versus 61.5±9.5 years; t=9.05, df=5509, P=.0001) and had a higher proportion of women (1966 men and 1949 women versus 635 men and 511 women; χ2=9.6, df=1, P=.002) than those who did not participate.
Among the screened individuals, 96 stroke patients (52 men and 44 women) were found on prevalence day. The prevalence of stroke was 24.5 per 1000 people aged ≥50 years (95% CI, 19.7 to 29.3 per 1000). The point prevalence ratios of men and women were 26.4 (95% CI, 19.3 to 33.5) and 22.6 (95% CI, 16.0 to 29.2) per 1000, respectively. The stroke prevalences of different age groups and sexes are presented in Table 1⇓. The prevalence increased with age and was higher for men than for women in all age groups except the group aged 50 to 59 years.
Nine of the 96 stroke survivors (9.4%) were neither seen by a physician nor managed in the hospital at the time of the acute illness. Of the 96 stroke patients, 89 (92.7%) had one episode of stroke, 6 (6.3%) had two, and 1 (1.0%) had three. At the time of the first stroke, 2 patients were aged <50 years, 32 aged 50 to 59 years, 28 aged 60 to 69 years, 23 aged 70 to 79 years, and 7 aged ≥80 years. In 4 patients we could not determine the age at onset.
Thirty-four stroke patients (35.4%) received CT scan of the brain; such medical facilities are relatively unavailable to Kinmen residents. The brain CT of these 34 patients identified the following types of stroke: cerebral infarction in 88.2% (30 cases), cerebral hemorrhage in 8.8% (3 cases), and subarachnoid hemorrhage in 2.9% (1 case).
After excluding 7 patients with severe aphasia, 5 patients with impairment of consciousness, and 1 patient with previous psychiatric disease, we classified 6 of the 83 patients (7.2%) as demented according to the DSM-III-R criteria.9 Of these, 3 were men and 3 were women. Five demented patients had one episode of stroke, and 1 had two episodes (age range, 61 to 87 years; mean age, 76.5 years). Demented stroke survivors were older than nondemented subjects (mean age, 68.1±9.8 years; P<.05, t test).
The ADL of stroke survivors was graded as follows: 36 patients (37.5%) were grade 1; 21 patients (21.9%) grade 2; 11 patients (11.5%) grade 3; 18 patients (18.8%) grade 4; and 10 patients (10.4%) grade 5. Thus, more than half could take care of their daily needs independently without outside help.
Risk Factors and Lifestyles
The Mantel-Haenszel method was used first for univariate analysis. Only statistically significant factors were taken into further stepwise logistic regression analysis to determine their significance and odds ratio. The significant univariate variables included irregular heartbeat, meat preference, milk drinking, smoking, alcohol drinking, exercise habits, diabetes mellitus, hypertension, and heart disease. Only hypertension, irregular heartbeat, and diabetes mellitus were noted as significant variables after we performed logistic regression analysis and controlled for sex and age (Table 2⇓). Alcohol and smoking did not significantly increase the risk of stroke in the regression analysis. Body mass index and diastolic blood pressure of stroke survivors were found to be no different from those of individuals who never suffered a stroke. The stroke survivors, when examined, registered higher systolic blood pressures than those without a history of stroke (145±24 versus 139±23; P<.05). Thirty-six percent of the stroke patients ate meat less than once per week, compared with only 19% of the individuals who had not suffered a stroke (P<.001). No other diet preference was found. Thirty-six percent of the stroke patients exercised more than once per week compared with 18% of those who had not suffered a stroke (P<.05).
We compared our findings in Kinmen with those in other countries and regions, although we noted that differences in methodology, medical facilities, and the age-sex distribution must be considered in interpreting these comparisons. Comparing Kinmen with Finland,11 we found that the Kinmen study yielded a higher prevalence in persons aged ≥50 years. Compared with Japan12 and the United States,13 Kinmen appears to have a lower rate, particularly in the older group. Kinmen has approximately the same rate as mainland China.14 This is not surprising since the people on mainland China and in Kinmen share the same racial background and have very similar diet and lifestyle. In a previous study,15 the age-specific incidence of stroke in Taiwan was higher than that in the United Kingdom and the United States but similar to that in Japan and mainland China. Alter et al16 concluded that worldwide variation in age- and sex-adjusted stroke incidence rates is relatively small. Wide variations in prevalence are likely due to different mortality rates. We believe that the higher prevalence of stroke in the oldest age group in developed countries may be due to differences in the severity of stroke and/or management of the illness.
Compared with data obtained in two previous stroke prevalence studies in Taiwan—Pan's study17 of 1984 and the study of Hu et al3 of 1989—the age-specific prevalence rates of the present study were lower in the group aged 60 to 69 years and higher in the group aged ≥70 years (Fig 2⇓). Since the age-adjusted mortality rates of stroke have decreased in Taiwan,2 the increased prevalence in the older age group may be explained simply by the fact that people now live longer than they did in the early 1980s. On the other hand, the decreased prevalence in the group aged 60 to 69 years indicates that the incidence rate of stroke may be decreasing in Taiwan. However, these conclusions must be confirmed in subsequent studies.
The few available community studies suggest that 52% to 87% of surviving patients can walk without assistance and 43% to 68% can conduct ADL independently without outside assistance.18 Other community studies have reported that at least 50% of the long-term survivors of a stroke are independent in physical function as measured by ADL. Despite the high case-fatality rate of stroke, a majority of the surviving patients returned home with relatively good recovery of physical function.
Stroke is an important cause of dementia in the elderly. The prevalence rates of dementia reported in stroke cohorts are highly variable (5.8% to 26.3%).18 19 20 21 The wide range may be due to the different diagnostic criteria used, different sources of patients, variable age groups of patients, and the length of time elapsed after the stroke. The likely reason for the lower rate of dementia in our study was that we included subjects aged >50 years rather than setting the cutoff age at 60 years, as in most studies. The lower rate may also be due to the fact that we conducted a door-to-door survey. However, we cannot exclude the possibility that some patients with dementia and a history of stroke were missed in the study. Our results support the view expressed by Tatemichi et al19 that advanced age is a significant factor of dementia after stroke.
Risk factors are better determined through an incidence study. The bias in our study is caused by the stroke patients who died soon after the stroke and the stroke survivors who changed lifestyles and habits, which could not be taken into consideration. Although the ethnicity and lifestyle of Kinmen differ from those of Western countries, the risk factors of stroke in a rural region of a developing country are still similar to those in the developed countries. The two interesting findings in the stroke survivors in Kinmen are that they ate less meat than their fellow residents and exercised more. Living standards of the population in Kinmen were similar to those found in other rural regions in Taiwan. A large-scale dietary survey conducted in Taiwan from August 1986 to February 198822 found that the average meat and poultry intake per person per day was 125.5 g. People consumed more pork than other kinds of meat. Based on the Seven Countries Study,23 Taiwan is in the middle range of meat consumption per person. We believe that successful efforts in health education have influenced stroke survivors to change their lifestyle or to follow a healthier diet.
Overall, we found that stroke prevalence in a rural Taiwan island community was in the middle range by worldwide standards. The survivors achieved high levels of function in caring for themselves. Hypertension, irregular heartbeat, and diabetes were significant risk factors. Follow-up studies in this population could provide useful information about changing risk factors and survival patterns.
This study was supported in part by National Science Council grants (NSC 85-2331-B-075-002 and NSC 85-2331-B-075-107Y).
Reprint requests to Dr Hsiu-chih Liu, Neurological Institute, Veterans General Hospital-Taipei, 11217, Taiwan. E-mail firstname.lastname@example.org.
- Received February 5, 1996.
- Revision received May 13, 1996.
- Accepted May 13, 1996.
- Copyright © 1996 by American Heart Association
Department of Health, Executive Yuan. Health Statistics, Republic of China 1994. Taipei, Taiwan: National Health Administration, Taiwan, ROC; 1994.
Su CL, Chang SF, Hung TP. Secular trends in mortality for cerebrovascular diseases in Taiwan (1959-1989) [in Chinese, with English abstract]. J Formos Med Assoc. 1992;91:S52-S61.
Hu HH, Chu FL, Chiang BN, Lan CF, Sheng WY, Lo YK, Wong WJ, Luk YO. Prevalence of stroke in Taiwan. Stroke. 1989;20:858-863.
Liu HC, Wang SJ, Fuh JL, Liu CY, Lin KP, Lin CH, Wang PN, Lin KN, Wang HC, Chen HM, Chang R, Larson EB, Wu GS, Chou P, Teng EL. The Kinmen Neurological Disorders Survey (KINDS): a study by neurologists of a Chinese population. Neuroepidemiology. In press.
Hatano S. Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ. 1976;54:541-553.
Folstein MF, Folstein SE, McHugh PR. ‘Mini-Mental State': a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:122-129.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed, revised. Washington, DC: American Psychiatric Association; 1987.
Mantel N, Haenszel W. Statistical aspects of the analysis of the data from retrospective studies of disease. JNCI. 1959;22:719-748.
Aho K, Reunanen A, Aromaa A, Knekt P, Maatala J. Prevalence of stroke in Finland. Stroke. 1986;17:681-686.
Urakami K, Igo M, Takahashi K. An epidemiologic study of cerebrovascular disease in western Japan: with special reference to transient ischemic attacks. Stroke. 1987;18:396-401.
Matsumoto N, Whisnant JP, Kurland LT, Okazaki H. Natural history of stroke in Rochester, Minnesota, 1955 through 1969: an extension of a previous study, 1945 through 1954. Stroke. 1973;4:20-29.
Li SC, Schoenberg BS, Wang CC, Cheng XM, Bolis CL, Wang KJ. Cerebrovascular disease in the People's Republic of China: epidemiologic and clinical features. Neurology. 1985;35:1708-1713.
Hu HH, Sheng WY, Chu FL, Lan CF, Chiang BN. Incidence of stroke in Taiwan. Stroke. 1992;23:1238-1241.
Pan BJ. A Survey of Cardiovascular Disease, Diabetes Mellitus and Hypertension, and Analysis of the Relative Risk Factors in People of 6 Areas in Taiwan. Taipei, Taiwan: National Taiwan University, College of Medicine; 1984. Thesis.
Wade DT, Hewer RL. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry. 1987;50:177-182.
Tatemichi TK, Desmond DW, Mayeux R, Paik M, Stern Y, Sano M, Remien RH, Williams JBW, Mohr JP, Hauser WA, Figueroa M. Dementia after stroke: baseline frequency, risks, and clinical features in a hospitalized cohort. Neurology. 1992;42:1185-1193.
Tatemichi TK, Foulkes MA, Mohr JP, Hewitt JR, Hier DB, Price TR, Wolf PA. Dementia in stroke survivors in the stroke data bank cohort: prevalence, incidence, risk factors, and computed tomographic findings. Stroke. 1990;21:858-866.
Lee NY, Chu YC, Chang CP, Shieh MJ, Kao MD. Dietary survey in Taiwan area, 1986-1988 [in Chinese with English abstract]. J Chinese Nutr Soc.. 1991;16:39-60.
Kromhout D, Keys A, Aravanis C, Buzina R, Fidanza F, Giampaoli S, Jansen A, Menotti A, Nedeljkovic S, Pekkarinen M, Simic BS, Toshima H. Food consumption patterns in the 1960s in seven countries. Am J Clin Nutr. 1989;49:889-994.