Stroke In China, 1986 Through 1990
Background and Purpose Incidence of stroke varies markedly in different world populations. In seven Chinese cities, the effect of a program of risk factor modification on the incidence and mortality of stroke was studied and compared with a control population. This article describes the incidence of stroke in the control populations for the years 1986 through 1990.
Methods Incidence (first-ever strokes only) for 1986 was obtained by door-to-door interview with heads of households with subsequent verification on examination by a neurologist and review of medical and/or hospital records. In subsequent years, cases were ascertained with a three-tier monitoring system: by community health workers, local medical centers, and the Beijing Neurosurgical Institute.
Results Average annual age-adjusted incidence per 100 000 was 215.6 (261.5 for males, 174.5 for females; P<.001). There was a significant drop in the total number of cases from 137 in 1986 to 106 in 1990, but the age-adjusted rate showed a significant drop for males only (322.3 to 182.5, P<.001). Marked differences in average annual age-adjusted rates existed among the seven cities, from 486.4 for Harbin to 80.9 for Shanghai. This difference in rate among cities was found for both sexes but was more pronounced in males.
Conclusions The stroke incidence rates in China, like those in Japan, are among the higher ones in the world. In recent years, there has been an apparent decline in stroke incidence. Marked differences in rates were found between males and females with decline in incidence occurring almost exclusively in males. There were also marked differences in stroke incidence among the cities studied. These differences may result in part from differences in diet, alcohol and cigarette consumption, or prevalence of hypertension.
The epidemiology of cerebrovascular disease has been studied extensively in North America and in Europe1 2 but to a lesser extent in Asiatic countries. There have, however, been six published studies concerning the epidemiology of stroke in various parts of the People’s Republic of China, two in the Chinese language journals,3 4 and four in the English language.5 6 7 8 The present report is the first from a large stroke prevention project organized by the Ministry of Public Health of the People’s Republic of China. The specific purposes of the studies were to (1) investigate the relationship between risk factors and stroke in Chinese populations and (2) determine the effectiveness of a public health program aimed at reducing these risk factors. The first projects in this group of studies were to determine and compare in sample populations not receiving the public health program, in each of seven cities from different areas of China, the incidence and mortality of stroke over the years 1986 through 1990. The present article presents results from that portion of the study relating to stroke incidence.
Subjects and Methods
Early in 1986, awards from the Chinese Ministry of Public Health for the “7-5 China Stroke Prevention Project” were made to seven clinical centers, two coordinating centers, and a laboratory standardized center. The Ministry of Public Health was responsible for supervision of the project implementation. The seven urban clinical centers were Beijing, Shanghai, Changsha, Harbin, Yinchuan, Changchun, and Zhengzhou. The two coordinating centers, Beijing Neurosurgical Institute and Shanghai Medical University Institute, were responsible for the design and completion of the program. A steering group composed of the principal investigators of the seven collaborating centers was responsible for the scientific leadership of the trial.
In each of the seven cities, two sample populations were identified, one in which measures for intervention to reduce risk factors were carried out (subsequently referred to as community A) and a second or control population in whom no such measures were carried out (referred to as community B). In each city, each such population consisted of approximately 10 000 individuals. They were defined as all individuals living in a defined geographical area adjacent to the hospital that would be carrying out the studies. In each city, there were two such hospitals: that serving community A and that serving community B. Centers were chosen on the basis of having participated in previous epidemiological studies.
From these sample populations in each city, two cohorts of 2500 subjects each (one from community A and one from community B) were further identified for the studies of the relationship of stroke to risk factors. These latter studies will form the basis for subsequent reports. The present report provides data on the incidence of first-ever stroke from the entire community B (control) population of the seven cities, which in the year 1986 was 57 914 persons.
The defining criteria of stroke of the World Health Organization (WHO) International Classification of Diseases, 9th revision, codes 430 through 438, were used: “rapidly developing clinical symptoms and/or signs of focal, and at times global, loss of cerebral functions, with symptoms lasting more than 24 hours or lasting to death, with no apparent cause other than of vascular origin.”9 CT scan was obtained in 30% to 60% of the individuals, depending on the site of study.
For demographic data, a census of the age-sex register was obtained from local police stations or residential committees once a year from 1986 through 1990.
Baseline data on stroke incidence in the community B population for the year 1986 and for the first half of 1987 were obtained as follows. In the initial interview in 1987, the head of every household was asked if anyone in the family had had a stroke in the previous year and up to the time of the interview and, if so, whether the individual had died of that presumed stroke. In Chinese culture, it is uncommon for any individual, particularly an elderly one, not to live with family. This fact tends to minimize missed cases (eg, a fatal stroke in someone living alone).
Upon obtaining history of stroke, the interviewer notified the local medical center, and a senior neurologist and/or neurosurgeon member of the study team was sent to examine living patients either at home or in the hospital. If a diagnosis of completed stroke was confirmed by the examination and/or hospital records, the stroke was coded as defined above. In case of death, determination was based on hospital and medical records.
For the second half of 1987 and in subsequent years, including all of 1990, strokes were detected by a three-tier monitoring and register system as follows: (1) In the study sites, all deaths (and some other serious illnesses including those related to cerebrovascular disease) were immediately reported by the patient or family to the community health worker (usually retired physicians or nurses) who in turn notified the local medical center within 3 days. (2) The local medical center then sent the study-team neurologist or neurosurgeon to the home or hospital as soon as possible, usually 1 to 3 days after receiving the notification from community health workers, to verify the diagnosis by examining the patient, hospital records, or (in case of death) medical records and to apply the correct WHO diagnostic code. (3) Demographic data and forms containing diagnostic codes were sent to Beijing yearly. The principal investigator of Beijing Neurosurgical Institute was responsible for the examination and summary of data.
To summarize, stroke diagnosis was verified by (1) decisions of neurologists/neurosurgeons on the basis of patient examination, (2) hospital records and hospital discharge certificates, and (3) in the case of patients who had died at home, reports from an eyewitness of death (usually the decedent’s relatives). To ensure uniformity of data collection and recording in the seven cities (interpopulation quality control), a meeting involving all collaborating centers was held once a year to monitor the progress of the project and to design and select several quality control indicators for examination and evaluation of all aspects of the study, eg, the organization of the monitoring system, person-times of the field visit by surveyed personnel, and number of patient visits. For the intrapopulation quality control, the coordinating center (the Department of Neuroepidemiology of the Beijing Neurosurgical Institute) sent personnel to each community every year to determine that data forms were being filled out correctly, that case diagnoses had been confirmed by senior neurologists, and the presence of any missing reports.
In community B, the average annual age-adjusted incidence of first-ever stroke per 100 000 in the seven Chinese cities for the period of the 5 years (1986 through 1990) was 215.6.
For the entire period, there was a significantly higher mean rate for males than for females (261.5 versus 174.5, respectively; P<.001). As expected, the age-specific rate of stroke rose with increasing age for both males and females, as shown in Table 1⇓. The absence of any stroke cases below the age of 35 years is possibly due to the lack of recognition by the Chinese community that stroke can occur in the young. Because stroke below age 35 is extremely uncommon, however, the number of cases unreported would be expected to have a very small effect on overall age-adjusted incidence.
Stroke incidence for the years 1986 through 1990 is shown in Table 2⇓. There was a significant drop in the total annual number of strokes from 137 in 1986 to 106 in 1990. Age-adjusted incidence rates for these 2 years confirmed the decline during this period: from 247.1 to 170.3. This decline was completely due to a drop in the incident cases in males from 85 to 54. The drop in the age-adjusted incidence in males was from 322.3 to 182.5 per 100 000 (P<.001). In females, there was no change in the number of cases (52), (Figure⇓) and no significant change in age-adjusted stroke incidence between these years (180.1 to 159.7) (P>.5).
There was a marked difference in mean annual age-adjusted incidence rates for the seven cities as seen in Table 3⇓, from a high of 486.4 in the city of Harbin to a low of 80.9 in Shanghai. This difference in incidence between different cities was found in both sexes but was more pronounced in males. For example, in males the Harbin/Shanghai ratio of age-adjusted incidence was approximately 7.4 to 1.0, whereas in females it was 4.9 to 1.0.
In each city, the mean stroke incidence was higher in males than females, but in only two was this difference significant (Changchun and Harbin).
CT scans were performed on varying percentages of stroke patients during these years. The numbers of scans and percentages are shown in Table 4⇓.
The mean age-adjusted incidence of stroke in the present report for the years 1986 through 1990 in the People’s Republic of China, 215.6, is almost identical to the rate of 219 reported in 1985 by Li et al5 in a study carried out in six urban centers, four of which were used in the present study. However, the survey of Xue et al6 in the People’s Republic of China, which sampled an urban-rural population of almost 6 million, reported an incidence of 115.8 per 100 000, age-adjusted to a world standard. The reason for the lower incidence rate found in this latter study is unclear but may be partly accounted for by the inclusion of a large rural population. In the “big-cities” segment of the populations of the study of Xue et al, the age-adjusted incidence was somewhat higher: 146. Chen et al7 in 1992 reported an age-adjusted incidence of 216 of stroke in Beijing, close to the figure of 192 for that city found in the present investigation. The study of Hu et al10 of a Taiwanese population was carried out over a period very similar to that of the present study. The annual age-adjusted stroke incidence was found to be 330, a figure considerably higher than that reported here. The Hu study excluded all individuals younger than age 35 years, but results can probably be compared with the present study, in which no strokes below that age were recorded.
It is of interest to compare the figures of the present study to those of Japan, where stroke epidemiology has been extensively studied. Reported stroke incidence in Japan is much higher than the rate in China.11 12 Komachi et al,12 for example, in a large collaborative study found an annual average incidence of 394 per 100 000 for men and 252 per 100 000 for women. (It is not clear, however, if this was for first stroke.) It is of relevance to the present study that the study of Komachi et al also noted marked variation in these rates in different regions of Japan.
The stroke incidence rates for both Japan and China are among the highest of national or geographic population groups.1 The possible reasons for these high stroke incidences have been the subject of extensive speculation. One possibility is ethnic susceptibility, but the finding that stroke incidence in Asian populations residing in the United States approximates that of whites diminishes the importance of that factor.13
In the present study, there were striking differences in stroke incidence between the different Chinese cities despite identical case-finding methods. In both the studies of Li et al5 and the present one, age-adjusted stroke incidence has been found high in Harbin and low in Shanghai. Harbin is a northern city, Shanghai southern, and it is of interest that in the study of Xue et al a definite gradient of stroke incidence between north and south was found. The northern areas of the world—Japan,12 Sweden,14 and Finland15 —have in general reported higher incidence rates of stroke than the southern areas, eg, Umbria16 and Benghazi.17 Moreover, in both Japan12 and Finland,15 regional differences of stroke incidence have been reported.
The very high incidence of stroke in the northern city of Harbin undoubtedly reflects a complex of risk factors not solely related to latitude or climate. In the present study, for example, the cities with the second and third highest rates, Changsha and Zhengzhou, are on approximately the same latitude as Shanghai, which has the lowest rate. Rochester, Minn, a northern-latitude city, recently was reported with one of the lower incidence rates of the world.18
Many possible reasons for the extremely high stroke rates found in Harbin, Changsha, and Zhengzhou and the low rate in Shanghai remain to be investigated. There may be an increased amount of hypertension in the cities with high incidence, which would markedly elevate rates of cerebral hemorrhage and, to a lesser extent, cerebral infarction. Much higher rates of hypertension in the northern and northeastern provinces of China compared with those in southern provinces have been documented.19 Excess alcohol intake20 21 and smoking22 23 are also known to be associated with increased incidence of stroke. It is possible that one or both of these risk factors are more prevalent in the cities with high stroke incidence and less prevalent in Shanghai. In Japan, it has been speculated that various dietary habits are one factor explaining regional variation in stroke incidence.12 In the populations reported here, marked differences in dietary habits may exist (eg, between Harbin and Shanghai).
The increased stroke incidence in males may have been to some degree due to the increased percentage of male stroke patients undergoing CT. The sex difference does decline somewhat in later years when the percentage of female patients studied with CT scans rises (Table 4⇑).
The increased stroke incidence in males versus females found in this Chinese study, however, is similar to that found in varying degrees in almost all populations from other parts of the world.1 The sex differential was most striking in the present study in the two northernmost cities of China. There was, for example, little difference in incidence in the sexes found in the city of Shanghai. In some other studies, there are suggestions that more marked differences in sex incidence of stroke occur in northern areas. In the eastern areas of Finland, for example, the annual age-adjusted incidence was found to be 352 for males and 204 for females15 and in Rochester, Minn, a study for the years 1980 through 1984 found the male rate to be 168 and the female 110.18 Cerebral hemorrhage is more prevalent in men than women, and a hypothesis for the predominance of strokes in males in northern areas would be that cerebral hemorrhage was more frequent. The excess incidence of stroke in males has not, however, been found in all populations.
Although there was variation from year to year, a continuous decline in stroke incidence from 1986 through 1990 was found in this study. A limitation of the study is that the method of stroke detection was retrospective for the year 1986 and the first half of 1987 and prospective in the other years. The retrospective survey was, however, exhaustive (door-to-door). Furthermore, correction for any missed cases for the years 1986 through 1987 would have raised the number of cases for those years and therefore accentuated the degree of decline in subsequent years.
It is of interest to compare this change in stroke incidence with data from other countries for these years, although these comparisons must be made with caution since there is some variation in ages of populations studied and the standards for age adjustment used. One recent study from New Zealand on stroke incidence between 1981 and 1991 found a significant drop but only in men over 75 years24 ; another study from the Netherlands documents a continuing decline in stroke incidence in that country—greater, however, in women than in men.25 Most studies, however, have not documented a continuing decline in stroke incidence. In the United States, data for recent years from the Mayo Clinic suggest a leveling off of stroke incidence,18 and data from the National Hospital Discharge Survey for the 1970s and 1980s showed no decline in hospitalizations for stroke.26 A Swedish study of stroke in the city of Göteborg reported no change in stroke incidence for these years.27 There have even been reports from some populations of some rise in the incidence of stroke in the most recent years. A Danish study, for example, documented a significant rise in stroke incidence in the period of 1972 through 1974 and 1989 through 1990, with the rise being greater in men.28
In the present study, the decline in incidence was almost exclusively in males. Reasons for this specific fall in the male rate in the 1986-1990 period can only be speculated on. Increasingly, widespread and successful treatment of hypertension in recent years may be a factor, although the importance of its role in reducing overall stroke incidence has been questioned.29 Reduction of hypertension frequency would almost unquestionably reduce incidence of cerebral hemorrhage, but data on comparative incidence of hemorrhage and infarction in our populations as confirmed by CT scan are not yet available.
Some part of the decline in stroke incidence in males may have been due to the fact that the population reported here lived in proximity to the other studied population, which was subjected to an intensive public health educational program to reduce stroke. Effects of this program could clearly have spilled over into the neighboring population, although it is difficult to explain this factor influencing the male population only.
This study was supported by funds from the Chinese Ministry of Public Health, People’s Republic of China. We thank the following collaborative centers and investigators: Dr Wang Chong-Cheng, Beijing Neurosurgical Institute; Drs Ju Zhi-Ping and Feng Er-Juan, Institute of Neurology, Shanghai Medical University; Dr Rao Ming-Li, Bethune Medical University; Drs Dai Qing-Shun and Yang Yu-Chuan, Harbin Medical University; Dr Su Fang-Zhong, Henan Medical University; Drs Lu Xue-Feng and Yang Qi-Dong, Hunan Medical University; Drs Song Jia-Ren and Kong Fan-Yuan, Ningxia Medical College; and Cheng Hang, Department of Physiology, University of Kansas Medical Center. We also wish to thank Patricia Melching for her typing assistance.
- Received March 15, 1995.
- Revision received July 6, 1995.
- Accepted August 1, 1995.
- Copyright © 1995 by American Heart Association
Terent A. Stroke morbidity. In: Whisnant JP, ed. Stroke: Populations, Cohorts, and Clinical Trials. Oxford, UK: Butterworth-Heinemann Ltd; 1993:37-58.
Kurtzke JF. Epidemiology. In: Barnett HJM, Stein BM, Stein JP, Mohr JP, Yatsu FM, eds. Stroke: Pathophysiology, Diagnosis, and Management. New York, NY: Churchill Livingstone Inc; 1986:3-18.
Wang CC, Cheng XM, Li SC, Wang WZ, Wu SP, Wang KJ, Chou SS, Zhao F, Dai QS, Song JR, Cai Y. Epidemiological survey of neurological disorders in six urban areas of People’s Republic of China [in Chinese]. Zhong-Hua Shen-Jing Wai-ke Za-Zhi. 1985;1:2-7.
Li ZS, Yang QD, Chen SM, Shu Q, Fu YG. Epidemiological survey of cerebrovascular disease in rural areas of China [in Chinese). Zhong-Hua Shee-Jing Wai-ke Za-Zhi. 1989;5(suppl):7-11.
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.
Shi YM, Zhou GZ, Wang GQ, Xue GB, Li GH, Yang YF. An epidemiologic investigation and analysis on stroke in Hongkou District, Shanghai, the P.R. China [in Chinese]. Zhong-Guo Sheng-Jing Jing-Shen Ji-Bing Za-Zhi. 1985;11:262-264.
World Health Organization. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death. 9th Revision. Geneva, Switzerland: World Health Organization; 1977:vol 1.
Hu HH, Sheng WY, Chu FL, Lan CF, Chiang BN. Incidence of stroke in Taiwan. Stroke. 1992;23:1237-1241.
Tanaka H, Ueda Y, Date C, Baba T, Yamashita H, Hayashi M, Shoji H, Owada K, Baba K-I, Shibuya M, Kon T, Detels R. Incidence of stroke in Shibata, Japan: 1976-1978. Stroke. 1981;12:460-466.
Komachi Y, Tanaka H, Shimamoto T, Handa K, Iida M, Isomura K, Kojima S, Matsuzaki T, Ozawa H, Takahashi H, Tsunetoshi Y. A collaborative study of stroke incidence in Japan: 1975-1979. Stroke. 1984;15:28-36.
Yatsu FM. Strokes in Asians and Pacific-Islanders, Hispanics, and native Americans. Circulation. 1991;83:1471-1472.
Terent A. Increasing incidence of stroke among Swedish women. Stroke. 1988;19:598-603.
Tuomilehto J, Sarti C, Narva EV, Salmi K, Sivenius J, Kaarsalo E, Salomaa V, Jorma T. The FINMONICA Stroke Register: community-based stroke registration and analysis of stroke incidence in Finland, 1983-1985. Am J Epidemiol. 1992;135:1259-1270.
Ricci S, Celani MG, La Rosa F, Vitali R, Duca E, Ferraguzzi R, Paolotti M, Seppoloni D, Caputo N, Chiurulla C, Scaroni R, Signorini E. SEPIVAC: a community-based study of stroke incidence in Umbria, Italy. J Neurol Neurosurg Psychiatry. 1991;54:695-698.
Ashok PP, Radhakrishnan K, Sridharan R, El-Mangoush M. Incidence and pattern of cerebrovascular diseases in Benghazi, Libya. J Neurol Neurosurg Psychiatry. 1986;49:519-523.
Broderick JP, Phillips SJ, Whisnant JP, O’Fallon WM, Bergstralh EJ. Incidence rates of stroke in the eighties: the end of the decline in stroke? Stroke. 1989;20:577-582.
Wolf PA, D’Angostino RB, Odell P, Belanger AJ, Hodges D, Kannel WB. Alcohol consumption as a risk factor for stroke: the Framingham Study. Ann Neurol. 1988;24:177. Abstract.
Gorelick PB, Rodin MB, Langenberg P, Hier DB, Costigan J. Weekly alcohol consumption, cigarette smoking and the risk of ischaemic stroke: results of a case control study at three urban medical centers in Chicago, Illinois. Neurology. 1989;3:339-343.
Shinton R, Beevers G. Meta-analysis of relation between cigarette smoking and stroke. BMJ. 1989;298:789-794.
Niessen LW, Barendregt JJ, Bonneux L, Koudstaal PJ. Stroke trends in an aging population: the technology assessment methods project team. Stroke. 1993;24:931-939.
Modan B, Wagener DK. Some epidemiological aspects of stroke: mortality/morbidity trends, age, sex, race, socioeconomic status. Stroke. 1992;23:1230-1236.
Harmsen P, Tsipogianni A, Wilhelmsen L. Stroke incidence rates were unchanged, while fatality rates declined, during 1971-1987 in Göteborg, Sweden. Stroke. 1992;23:1410-1415.
Jorgensen HS, Plesner AM, Hubbe P, Larsen K. Marked increase of stroke incidence in men between 1972-1990 in Frederiksberg, Denmark. Stroke. 1992;23:1701-1704.
Bonita R, Beaglehole R. Increased treatment of hypertension does not explain the decline in stroke mortality in the US: 1970-1980. Hypertension. 1989;13(suppl):169-173.