(Stroke. 1995;26:1774-1780.)
© 1995 American Heart Association, Inc.
Articles |
Changes in the Sex Ratio of Stroke Mortality in the Period of 1955 Through 1990
From the Department of Epidemiology, School of Public Health, University of Leuven, Leuven, Belgium.
Correspondence to Professor H. Kesteloot, Department of Epidemiology, Capucijnenvoer 33, B-3000 Leuven, Belgium.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose Stroke mortality worldwide has decreased in men and women in most industrialized countries, except in eastern European countries. The purpose of this study was to compare the sex ratio of stroke mortality between populations and over time. This approach may help to determine the factors influencing this ratio.
Methods The sex ratios (men to women) of stroke mortality between ages 55 to 64, 65 to 74, and 75 to 84 years from 27 populations between 1955 and the latest available year were analyzed using World Health Organization data. The relationship between log stroke mortality and age and the relationships between alcohol, animal fat, cigarette consumption, and urinary cation excretion and the sex ratio of stroke mortality were also analyzed.
Results The mean sex ratio of stroke mortality increased 50%, 34%, and 15% in the three age classes, respectively, over 35 years. Highly significant relationships of log stroke mortality with age exist, which vary between men and women and among countries. In general, stroke mortality changed in the same direction in both sexes but decreased earlier and more rapidly in women than in men. Alcohol consumption and urinary sodium excretion correlated positively and significantly with the sex ratio. The time trends of the sex ratio also correlated positively and significantly with the time trends of cigarette consumption. No relationship with animal fat consumption was found.
Conclusions The sex ratio of stroke mortality is increasing with time and decreasing with age. Differences in lifestyle among countries and over the last three decades may contribute partially to these differences in sex ratio.
Key Words: gender • sex ratio • hypertension • mortality • risk factors
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Between 1950 and 1990, stroke mortality has decreased markedly and almost linearly in nearly all industrialized countries, with the exception of the eastern European countries. The decrease occurred in men and in women.1 2 3 4 5 While the sex ratio of stroke mortality has been changing during the last 40 years, in any given population men and women live essentially under similar environmental conditions (eg, environmental pollution, climate, regional economic status, supply of food, and medical care). Therefore, these factors are unlikely to account for the changed sex ratio. The purpose of this study was to compare the sex difference of stroke mortality among countries over time and to ascertain whether any difference could be attributed to a difference between men and women in level of exposure to some of the risk factors that affect stroke mortality.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The data on stroke mortality in men and women in 5-year age classes from 35 to 84 years of age used in this study were provided by the World Health Organization (WHO) on a special nine-track magnetic tape (6250 bytes/in). Twenty-five countries, most industrialized, with adequate mortality data from 1955 to the latest available year (LAY) were included. Data from England, Wales, Northern Ireland, and Scotland of the United Kingdom were analyzed separately. For Hong Kong, we had data only from 1960 and for Greece only from 1956. The sex ratio of stroke mortality rates before 55 years of age were not considered because most stroke events occurred in older age classes. The mean sex ratio of stroke mortality in each of the periods (1954 to 1956, 1964 to 1966, 1974 to 1976, 1984 to 1986) and the latest available three years (either 1986-1988 or through 1989-1991) were determined as well as the percentage change of the sex ratio of stroke mortality between 1955 and the LAY. The study did not differentiate between types of stroke (ie, ischemic and hemorrhagic were combined).
The data from nine selected countries (Belgium, Cuba, Denmark, France, Hungary, Japan, the Netherlands, Sweden, and the United States), with different time trends of stroke mortality and from different geographic areas, are presented graphically to exemplify the time trends of stroke mortality (65 to 74 years) and of the sex ratio of stroke mortality for two age classes (55 to 64 and 65 to 74 years) in the period from 1950 to the LAY.
The Pearson correlation analysis was performed between male and female stroke mortality in 1969 to 1971, 1979 to 1981, and the latest available 3 years. Univariate regression analysis for log stroke mortality and age (35 to 84 years) was performed for men and women for two groups (Czechoslovakia, Hungary, and Poland, with increasing stroke mortality [CHP group]; and the United States, Canada, and Switzerland, with decreasing stroke mortality [UCS group]) as examples of the relationship between age and log stroke mortality in 1960, 1970, 1980, and 1990. The time trends of these relationships in the two groups for both sexes were also compared.
Data for animal fat and alcohol consumption were obtained from food balance sheets provided by the Food and Agricultural Organization of the United Nations (FAO)6 for 25 countries studied. The validity of FAO data for fat and alcohol consumption has been studied, and we found that the data are sufficiently correct for use in epidemiological studies.7 8 The 24-hour urinary sodium and potassium excretion data from epidemiological surveys are available for only 17 countries.9 Cigarette consumption data (per person per year) were obtained from a WHO report.10 These data were used to analyze the relationship of the sex ratio of stroke mortality with fat and alcohol consumption, with sodium and potassium excretion, and with the level of cigarette consumption between populations.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Tables 1 through 3
show the values of the male/female (M/F) stroke mortality ratio for 27
populations in three age classes and five time periods
presented as the mean of 1954-1956, 1964-1966, 1974-1976,
1984-1986, and the latest available 3 years. The mean mortality sex
ratios decreased with age in all time periods studied. However, the
mean value of the sex ratio of stroke mortality had increased 50% in
the age class 55 to 64, 34% in the age class 65 to 74, and 15% in the
age class 75 to 84 years from 1955 to the LAY, but the mean increase in
the sex ratio became less marked with age. The M/F ratio of stroke
mortality was below one in 6, 8, and 12 of 27 populations in the age
classes 55 to 64, 65 to 74, and 75 to 84 years, respectively, in 1955
but only in 2 in the oldest age class in the LAY
(
2=22.84, P<.01). A decline in the
M/F stroke mortality ratio occurred only exceptionally (Tables 1 through 3). In Japan a substantial increase occurred compared with
other Asian countries. The highest increase in the M/F stroke mortality
ratio occurred in Hungary (+1.04, +101%, for the difference and
percentage change between 1955 and LAY, respectively) in the age class
55 to 64 years.
|
|
|
The time trends of stroke mortality for men and women between 65 and 74
years of age in nine countries, selected on the basis of different
behavior, are shown graphically in Figs 1 and 2. A nearly linear decrease occurred in both sexes in
most countries except Hungary. In the latter, stroke mortality
increased from the early 1970s,1 but the increase occurred
5 years later in women than in men. Stroke mortality did not change
much in Cuba in the last two decades. The most striking decrease of
stroke mortality occurred in Japan, where it decreased about 5 years
earlier in women than in men (Figs 1 and 2). However, the sex ratios of
the M/F stroke mortality in the nine countries considered increased
steadily (Figs 3 and 4) regardless of
whether the stroke mortality decreased or increased. The sex ratio of
stroke mortality in the United States has decreased since the 1970s; it
has then leveled off in the last decade for the age class 55 to 64 and
in the last years for age class 65 to 74 years, but it still remains
higher than it was in 1955. France had the highest sex ratio in stroke
mortality, and the Netherlands has had the highest increase over the
last 40 years.
|
|
|
|
A significant correlation exists between male and female stroke mortality. The Pearson correlation coefficients (r) are .90, .95, and .96 in the age class 55 to 64 years in 1970, 1980, and 1990, respectively; .91, .95, and .98 in the age class 65 to 74 years; and .93, .97, and .97 in the age class 75 to 84 years (n=27). All probability values are less than .001.
Previous work has shown a highly significant (r>.99)
correlation between log mortality from all causes and
age11 and between log mortality from total cancer and log
age.12 The regression analysis of log mortality
(per 100 000 persons per year) of stroke with age (between 35 and 84
years of age) in 1960, 1970, 1980, and 1990 indicates that log stroke
mortality was significantly associated with age (r>.99) in
all four time periods in both sexes (Figs 5 and 6). Stroke mortality in a given country is determined by
the intercept and slope of the equation. This is true both for the UCS
group with decreasing stroke mortality and for the CHP group with
increasing stroke mortality. In the UCS group, the intercepts and
slopes of the regression equations decreased with time in men and
women. These time trends indicate that the decrease of stroke mortality
occurred in all age classes, but in the CHP group it also occurred in
both sexes: the intercept increased and the slopes decreased with time.
Whenever stroke mortality changed, it changed in the same direction but
at a different rate for women and men. As a result, the overall sex
ratio of stroke mortality increased.
|
|
The consumption of alcohol (%E, percentage of total dietary calorie
intake) in 1965, 1975, and 1985, as obtained from the FAO, correlated
positively and significantly in 19 of 27 correlations with the level of
the sex ratio of stroke mortality in 1965, 1975, 1985, and the LAYs
among 25 countries (data on alcohol consumption only for the United
Kingdom as a whole are available) in three age classes (Table 4). The strongest association between alcohol
consumption and the sex ratio of stroke mortality is in the age class
65 to 74 years, and the correlation coefficients progressively
increased with an increase of the interval between the time of alcohol
consumption and the time of the sex ratio of mortality considered
(Table 4). The trends of the sex ratio of stroke
mortality between 1970 and the LAY correlated positively with the
trends in cigarette smoking (per person per year) between 1970 to 1985
in the two age classes, but the correlation was significant only in the
age class 55 to 64 years (n=27, r=.48, P<.05).
The 24-hour urinary sodium excretion (millimoles per 24 hours) and the
urinary sodium to potassium ratio (millimole to millimole) correlated
positively with the sex ratio of stroke mortality, but the correlation
was significant only in the age class 55 to 64 years (n=17;
r=.745, P<.01, and r=.499,
P<.05, for urinary sodium excretion and the sodium to
potassium ratio, respectively). No consistent relationship was
found between the sex ratio of stroke mortality and the consumption of
animal minus fish fat (%E, reflecting saturated fat
consumption7 ).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The M/F sex ratio of stroke mortality has increased almost to the same extent in both eastern European and in other industrialized countries, notwithstanding marked differences in the trends of stroke mortality. Between 1950 and 1990, three revisions of the International Classification of Diseases have been made (ICD 7 to 9). These revisions did not alter the classification of stroke mortality to any major extent.13 The reasons for the differences in stroke mortality between the sexes and the substantial increase in the sex ratio of stroke mortality over about 35 years from 1955 to 1990 (or to LAY) are not clear. The significant correlation between male and female stroke mortality suggests that similar factors influence male and female stroke mortality but to a different extent. However, stroke mortality changed at different time periods and with different rates between men and women (Figs 1
and 2).
The increase of stroke mortality with age can be adequately expressed
by a simple equation relating log stroke mortality to age. In all
countries, substantial changes in slope and intercept occur with time
(Figs 5 and 6). A decrease in the intercept signals a lowering of the
stroke mortality. The changes in slope over the age range of 35 to 84
years suggest that the factors influencing stroke mortality are
operative at all ages after 35 years.
In recent decades worldwide, women have had a lower mortality from ischemic heart disease compared with men in every age class (Health Statistics Annual 1986-1993, WHO). In the earlier period of this study, women showed a stroke mortality similar to men in most countries. It was even higher than that in men in some populations (sex ratio less than 1).
The difference among countries and the increasing trends of the sex ratio of stroke mortality cannot be explained by genetic differences; for example, both the French and Japanese population had a high sex ratio. Moreover, the genetic composition of populations cannot change profoundly over a period of only 30 years.
Therefore, the differences in stroke mortality and its time trends between men and women are most likely due to environmental and lifestyle factors, as the effect of these factors acting on a genetic trail.
It is unclear whether the sex ratio of stroke mortality is different
for hemorrhagic and ischemic stroke. If such a difference by
stroke type exists, changes in the relative frequency of
ischemic and hemorrhagic stroke could influence the sex ratio
of stroke mortality. Hemorrhagic stroke is more common in Japan and
ischemic stroke in western countries. The fact that the
increase in the sex ratio of stroke mortality was greater in Japan
compared with the United States (in the age classes 55 to 64 years and
65 to 74 years) and England and Wales (in three age classes), but lower
than in other western countries, argues against this interpretation
(Tables 1 through 3).
The fact that stroke mortalities between the sexes in the different countries are highly significantly correlated, as mentioned before, shows that basically the same risk factors are operative in men and women but at a different level. The same is true for all-cause mortality.1 14 The risk factors commonly accepted for stroke are hypertension, hypercholesterolemia, congestive heart disease, atrial fibrillation, smoking, diabetes, and alcohol intake, among others. Hypertension is the major risk factor, and its decline, due to treatment and the reduction of salt intake in a population, may be responsible for the decline in stroke mortality in men and women.15 16 17 18 19 20 21 22 Smoking frequency and alcohol consumption are lower in women than in men, and this could explain part of the difference in stroke level between men and women.23 24 25 26 27 The dietary intake of total fat and saturated fat as a percentage of energy is very similar in men and women,7 but western women have a higher high-density lipoprotein cholesterol level than men. This could be ascribable to the ability of women on a diet high in saturated fat to increase their sex hormone levels, which are important determinants of the high-density lipoprotein cholesterol levels.28 This would protect women against thrombotic stroke in the same way and by the same mechanisms that protect them against ischemic heart disease.28 29 Premenopausal and postmenopausal use of estrogen replacement therapy also could have a protective effect, but this hypothesis needs confirmation. The intake level of antioxidants and of other cations beside sodium, such as potassium and magnesium, could also influence stroke mortality, but this effect is likely to be similar in men and women.
An increase in the sex ratio of stroke mortality in the presence of its decline in men and women can only be explained by a more rapid decrease in stroke mortality among women. The level of alcohol consumption correlated significantly with the sex ratio of stroke mortality, while the time trends of cigarette smoking between 1970 and 1985 correlated significantly with the trends of sex ratio of stroke mortality during the same period, confirming their importance for the understanding of the sex ratio of stroke mortality. The higher level of stroke mortality in Japan in the early 1950s can best be explained by the high level of salt intake during that period.30 31
The higher prevalence of treatment and control of hypertension in women than in men in the last decades2 18 32 33 in some populations might partially explain the increase of the sex ratio of stroke mortality. There are exceptions, however. White women in the United States had a lesser decline in stroke mortality compared with white men during the last two decades, although white women had antihypertensive treatment more often than white men.2 In a Japanese population study,19 women and men had similar levels of antihypertensive treatment in 1960, 1970, 1980, and 1990, but the sex ratio of stroke mortality linearly increased during the last 40 years.
It would be interesting to study the sex ratio of stroke mortality and its trends for ischemic and hemorrhagic stroke separately. This may be possible in the near future because of the progress in imaging techniques. Moreover, more recent studies of stroke mortality using standard definitions and classification of stroke should remove some of the uncertainties still present in this study.
In conclusion, the differences of the sex ratio of stroke mortality among populations and their changing trends may partially be attributed to different lifestyles (such as cigarette smoking and alcohol drinking) and differences in exposure to other risk factors for stroke mortality between men and women. Most of these factors remain to be defined. Differences in the level of treatment and control of hypertension can play a role also.
Received October 12, 1994; revision received July 11, 1995; accepted July 11, 1995.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
- Kesteloot H, Sasaki S, Xie J, Joossens JV. Secular trends in cerebrovascular mortality. J Hum Hypertens. 1994;8:401-407. [Medline] [Order article via Infotrieve]
-
Klag MJ, Whelton PK, Seidler AJ. Decline in US
stroke mortality, demographic trends, and antihypertensive
treatment. Stroke. 1989;20:14-21.
[Abstract/Free Full Text] - Ueshima H. Declining trend in blood pressure contributes to fall in stroke mortality in Japan. WHL. Newsletter No. 34, April 1994.
- McGovern PG, Burke GL, Sprafka JM, Xue S, Folsom AR, Blackburn H. Trends in mortality, morbidity, and risk factor levels for stroke from 1960 through 1990: the Minnesota heart survey. JAMA. 1992;268:753-759. [Abstract]
- Bonita R. Epidemiology of stroke. Lancet. 1992;339:342-344. [Medline] [Order article via Infotrieve]
- Food Balance Sheets, 1984-86 Average. Rome, Italy: Food and Agriculture Organization of the United Nations; 1991.
-
Sasaki S, Kesteloot H. Value of food and
agriculture organization data on food-balance sheets as a data
source for dietary fat intake in epidemiologic studies.
Am J Clin Nutr. 1992;56:716-723.
[Abstract/Free Full Text] - Sasaki S, Kesteloot H. Wine and non-wine alcohol: differential effect on all-cause and cause-specific mortality. Nutr Metab Cardiovasc Dis. 1994;4:177-182.
-
Sasaki S, Zhang XH, Kesteloot H. Dietary
sodium, potassium, saturated fat, alcohol and stroke mortality: an
ecological approach. Stroke. 1995;26:783-789.
[Abstract/Free Full Text] - Masironi R, Rothwell K. Smoking trends and effects worldwide. World Health Stat Q. 1988;41:228-241. [Medline] [Order article via Infotrieve]
- Kesteloot H. Nutrition and health: an epidemiological approach. Uit Verhandelingen van de Koninklijke Academie voor Geneeskunde van België. 1993;5:399-423.
- Kesteloot H, Sasaki S, Verbeke G, Joossens JV. Cancer mortality and age: relationship with dietary fat. Nutr Cancer. 1994;22:85-98. [Medline] [Order article via Infotrieve]
- Uemura K, Pisa Z. Trends in disease mortality in industrialized countries since 1950. World Health Stat Q. 1988;41:155-178. [Medline] [Order article via Infotrieve]
- Kesteloot H, Joossens JV. Worldwide mortality changes from all causes and from cardiovascular diseases between 1968 and 1988. Acta Cardiol. 1993;5:421-439.
- Kannel WB, Wolf PA, Verter J, McNamara PM. Epidemiologic assessment of the role of blood pressure in stroke. JAMA. 1970;214:301-310. [Medline] [Order article via Infotrieve]
-
Shimamoto T, Komachi Y, Inada H, Doi M, Iso H, Sato S,
Kitamura A, Iida M, Konishi M, Nakanishi N, Terao A, Naito Y, Kojima S.
Trends for coronary heart disease and stroke and their
risk factors in Japan. Circulation. 1989;79:503-515.
[Abstract/Free Full Text] - MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease, I: prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet. 1990;335:765-774. [Medline] [Order article via Infotrieve]
- Bonita R, Beaglehole R. Cerebrovascular disease: explaining stroke mortality trends. Lancet. 1993;341:1510-1511. [Medline] [Order article via Infotrieve]
- Shimamoto T, Iso H, Sankai T, Iida M, Naito Y, Sato S, Kitamura A, Kiyama M, Konishi M, Terao A, Baba S, Ozawa H, Komachi Y. Can blood pressure in the elderly be reduced? Findings from a long-term population survey in Japan. Am J Geriatr Cardiol. 1994;3:42-50.
- Joossens JV, Kesteloot H. Trends on systolic blood pressure, 24-hour sodium excretion, and stroke mortality in the elderly in Belgium. Am J Med. 1991;90(suppl 3A):5S-11S.
- Perry IJ, Beevers DG. Salt intake and stroke: a possible direct effect. J Hum Hypertens. 1992;6:23-25. [Medline] [Order article via Infotrieve]
- Xie JX, Sasaki S, Joossens JV, Kesteloot H. The relationship between urinary cations obtained from the INTERSALT study and cerebrovascular mortality. J Hum Hypertens. 1992;6:17-21. [Medline] [Order article via Infotrieve]
-
Gorelick PB. Alcohol and stroke.
Stroke. 1987;18:268-271.
[Abstract/Free Full Text] -
Ueshima H, Ohsaka T, Asakura S. Regional
differences in stroke mortality and alcohol consumption in
Japan. Stroke. 1986;17:19-24.
[Abstract/Free Full Text] -
Camargo CA. Moderate alcohol consumption and
stroke, the epidemiologic evidence. Stroke. 1989;20:1611-1626.
[Abstract/Free Full Text] - Shinton R, Beevers G. Meta-analysis of relation between cigarette smoking and stroke. BMJ. 1989;298:789-794.
- Wolf PA, D'Agostino RB, Kannel WB, Bonita R, Belanger AJ. Cigarette smoking as a risk factor for stroke: the Framingham study. JAMA. 1988;259:1025-1029. [Abstract]
- Kesteloot H, Sasaki S. On the relationship between nutrition, sex hormones and high-density lipoproteins. Acta Cardiol. 1993;4:353-363.
- Barrett-Connor E, Bush TL. Estrogen and coronary heart disease in women. JAMA. 1991;265:1861-1867. [Abstract]
- Ueshima H, Iida M, Shimamoto T, Konishi M, Tsujioka K, Tanigaki M, Nakanishi N, Ozawa H, Kojima S, Komachi Y. Multivariate analysis of risk factors for stroke: eight-year follow-up study of farming villages in Akita, Japan. Prev Med. 1980;9:722-740. [Medline] [Order article via Infotrieve]
- Sasaki N. High blood pressure and the salt intake of the Japanese. Jpn Heart J. 1962;3:313-324. [Medline] [Order article via Infotrieve]
- Bonita R, Beaglehole R. Increased treatment of hypertension does not explain the decline in stroke mortality in the United States, 1970-1980. Hypertension. 1989;13(suppl I):I-69-I-73.
- WHO Regional Office for Europe. Study on hypertension control monitoring at community level. EUR/ICP/NCD 226, Denmark, 1994.
This article has been cited by other articles:
 |
|
 |
|
  H. J. Fullerton, Y. W. Wu, S. Zhao, and S. C. Johnston Risk of stroke in children: Ethnic and gender disparities Neurology, July 22, 2003; 61(2): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.J. Fullerton, D.M. Chetkovich, Y.W. Wu, W.S. Smith, and S.C. Johnston Deaths from stroke in US children, 1979 to 1998 Neurology, July 9, 2002; 59(1): 34 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Fukuda, H. Yao, S. Ibayashi, T. Nakahara, H. Uchimura, M. Fujishima, and E. D. Hall Ovariectomy Exacerbates and Estrogen Replacement Attenuates Photothrombotic Focal Ischemic Brain Injury in Rats Editorial Comment Stroke, January 1, 2000; 31(1): 155 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Skarsgard, C. Van Breemen, and I. Laher Estrogen regulates myogenic tone in pressurized cerebral arteries by enhanced basal release of nitric oxide Am J Physiol Heart Circ Physiol, November 1, 1997; 273(5): H2248 - H2256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Barrett-Connor Sex Differences in Coronary Heart Disease: Why Are Women So Superior? The 1995 Ancel Keys Lecture Circulation, January 7, 1997; 95(1): 252 - 264. [Full Text]
|
|
|
|
|
|
L. B. Morgenstern, W. D. Spears, D. C. Goff, J. C. Grotta, and M. Z. Nichaman African Americans and Women Have the Highest Stroke Mortality in Texas Stroke, January 1, 1997; 28(1): 15 - 18. [Abstract] [Full Text]
|
|
|
|
|
|
M. Peltonen and K. Asplund Age-Period-Cohort Effects on Stroke Mortality in Sweden 1969-1993 and Forecasts Up to the Year 2003 Stroke, November 1, 1996; 27(11): 1981 - 1985. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||