Why do women liver longer than men?

Journal of Human Stress

ISSN: 0097-840X (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/vzhs20

Why do Women Live Longer than Men? Ingrid Waldron Ph.D. To cite this article: Ingrid Waldron Ph.D. (1976) Why do Women Live Longer than Men?, Journal of Human Stress, 2:1, 2-13, DOI: 10.1080/0097840X.1976.9937484 To link to this article: http://dx.doi.org/10.1080/0097840X.1976.9937484

Published online: 09 Jul 2010.

Submit your article to this journal

Article views: 20

View related articles

Citing articles: 31 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=vzhs20 Download by: [University of Birmingham]

Date: 10 November 2015, At: 04:29

WHY DO WOMEN LIVE LONGER THAN MEN?

Downloaded by [University of Birmingham] at 04:30 10 November 2015

INGRID WALDRON, Ph.D.

In the contemporary United States. mortality is 60 percent higher-for males than Jor.females. for^ percent of the excess oj-male mortalit-y is due t o arteriosclerotic heart disease, which is more common among men in part because they smoke cigarettes more than women do. and apparently also because they more ofien develop the competitive. aggressive Coronay Prone Behavior Pattern. Men who do not develop this Behavior Pattern may have as low a risk of‘corona y heart disease as comparable women. Oophorectomy of young women may increase the risk of coronary heart disease, but administration oj.jemale hormones generally does not reduce risk. One third oj.the sex diflerential in mortality is due to men’s higher rates of suicide. jatal motor vehicle and other accidents. cirrhosis of’the liver. respiratory cancers and emphysema. Each of’ these causes of death is linked to behaviors which are encouraged or accepted more in mules than in .females: using guns, drinking alcohol. smoking, working at hazardous jobs, and seeming to be.fearless. Thus, the behaviors expected ot’males in our society make a major contribution to their elevated mortaliv. Part I The sex differential in mortality has increased strikingly over the past half century in the U.S. In 1920. the life expectancy for women was 56, only two years longer than that for men.’ By 1970, women’s life expectancy was 75, almost eight years longer than men’s.* In 1920. male death rates were no more than 30 percent higher than female death rates at any age. By 1970, male death rates exceeded female death rates by as much as 180 percent for 15-24-year-olds and 110 percent for 55-64-year-olds. Dr. Waldron is Assrriate Professor of Biology at the Unibersity o f Pennsylvania.

2 Journal of Human Stress

Among young adults the excess of mortality for males is due primarily to accident^.^ At older ages, cardiovascular-renal diseases make the largest contribution to higher mortality among men. Rising male mortality for these causes of death and for lung cancer has been a major component of the increase in the sex differential in m ~ r t a l i t y These .~ trends were due in part to the sizable increase in cigarette smoking by men during the first third of the twentieth ~ e n t u r y . ~ , ~ Another substantial component of the increase in the sex differential in mortality has been the decline in maternal mortality and uterine cancer due to improvements in medical care. These data suggest that a wide variety of cultural factors, including automobile use, cigarette smoking and health care, contributes to the contemporary sex differential in mortality. Further evidence of the importance of cultural factors is provided by international comparisons. which show that higher male death rates, although common, have not been universal. In many countries female death rates have exceeded male death rates at ages between one and forty, and in some cases at older ages as well.’ Higher mortality among females has been observed most frequently in nonindustrial countries. The sex differential in mortality also varies for different groups within the United States. For example. the excess of male mortality is lowest among married adults, it is 10 percent greater among single and widowed adults, and it is 50 percent greater among divorced adults (data from8). The excess of mortality for males who are not married is particularly large for causes like cirrhosis of the liver which are strongly influenced by behavior, and for diseases like tuberculosis in which health habits and care play an important role. Gove’ has argued that the major reasons why the sex mortality differential is higher among males who are not married is March. 1976

WALDRON


that men do not adjust as well as women do to being unmarried, and that men derive greater advantages from being married, both in care received and in psychological well-being. Genetic factors apparently also contribute to higher male mortality, although the evidence for this is not as strong as commonly has been believed. Males have higher mortality than females in many different species, and this has been cited as evidence for a genetic contribution to the higher mortality of men.l0 However, although higher male mortality is widespread among insects, other Arthropoda and fishes, l o higher female mortality appears to be just as common as higher male mortality among our closer relatives, the birds and mammals. l o - ' * Among humans, higher female mortality is also common at certainages, asdescribed above. However, it is striking that, wherever statistics are available, males have had higher mortality during the first year of life.' Males also have been found to have higher fetal mortality in most s t u d i e ~ , 'although ~ fetal mortality during late pregnancy is as high for females as for males in pairs of twins of opposite sex.14 in multiple births of triplets or more.'5 and in a few geographical places, for example, Scotland.'" Male mortality is higher for many different causes of death (see Table 1). Several authors l o . ' ' have inferred from these observations that genetically determined metabolic differences may contribute to the higher mortality of males. Another study which has been cited widely as evidence of the importance of genetic factors is Madigan'sls comparison of life expectancy for Roman Catholic Sisters and Brothers in teaching Orders. Madigan found that the differential in life expectancy between Sisters and Brothers has been almost as large as the differential between women and men in the general population, even though the Sisters and Brothers had more similar adult roles. However, the higher mortality of the Brothers cannot be attributed solely to genetic causes, since the Brothers smoked and drank more than the Sisters and probably were socialized differently as children, and each of these differences would contribute to higher male mortality (as discussed in detail below). March. 1976

This earlier work suggests that both cultural and genetic factors contribute to the longer life expectancy of ~ o m e n . ' ~ Therefore, -~' we have considered both cultural and genetic factors in our analysis of the specific causes of the sex differential in mortality in the contemporary U.S. Our analysis is based on the identification of the causes of death which make a large contribution to the sex differential in mortality in the United States. Table 2 lists the seven causes of death which were responsible for at least 1 percent of all deaths in the U.S. in 1967, and for which male mortality exceeded female mortality by at least 100 percent. For each of these causes of death we considered all the major factors believed to contribute to its etiology and selected for analysis those factors which appeared to be relevant to the sex difference in mortality. Behavioral factors emerge as important determinants of the sex differential for each of the causes of death listed in Table 2. The importance of these factors is obvious in the case of accidents and suicide, and also for the respiratory diseases, which largely are due to smoking,zzas well as for cirrhosis of the liver, which is related to alcohol c o n ~ u m p t i o nThese . ~ ~ causes of death with clear behavioral components are responsible for one-third of the excess of male mortality, and arteriosclerotic heart disease is responsible for an additional 40 percent of the excess deaths among males. The data presented in the next section suggest that men have higher death rates for arteriosclerotic heart disease in large part because they smoke cigarettes more and because they more often develop the aggressive, competitive Coronary Prone Behavior Pattern. ARTERIOSCLEROTIC HEART DISEASE

Death rates for arteriosclerotic heart disease, which is primarily coronary heart disease (CHD), are twice as high for men as for women (Table 2). Cigarette smoking is associated with an elevation of CHD death rates ranging from 100 percent or more among middle-aged adults to 20 percent at the oldest ages." The elevated risk of CHD among smokers is probably due in part to the correlation between smoking and other risk factors (such as the Coronary Prone Behavior Pattern discussed below), but it is almost certainly also a direct consequence of the Journal of Human Stress

3

SEX DIFFERENTIAL IN MORTALITY

TABLE 1 Sex Mortality Ratios for All Major Causes of Death, U.S., 1967" Ratio of Male to Female Death Rates


5.9 4.9 2.8 2.7 2.4 2.0 2.0 1.8 1.7 1.6 1.6 1.5 1.4 1.4 1.4 1.4 1.3

Cause of Death

Malignant neoplasm of respiratory system, not specified as secondary Other bronchopulrnonic disease (717' 4 emphysema1 Motor vehicle accidents Suicide Other accidents Cirrhosis of liver Arteriosclerotic heart disease, including coronary disease Symptoms, senility and ill-defined c'onditions Pneumonia, except of newborn Other diseases of heart Other diseases of circulatory system Malignant neoplasm of digestive organs and peritoneum, not specified as secondary All other diseases (residual) Malignant neoplasm of other and unspecified sites Birth injuries, postnatal asphyxia, and atelectasis

0.008

Certain diseases of early infancy Other diseases peculiar to early infancy, and immaturity, unqualified Nonrheumatic chronic endocarditis and other myocardial degeneration General arteriosclerosis Vascular lesions affecting central nervous system Hypertensiveheart disease Malignant neoplasm of genital organs Diabetes mellitus Malignant neoplasm of breast

1.6

All Causes

1.3 1.2 1.2 1.o 0.89 0.89

Male Female Death Ratet Death Rate (Deaths per 100,OOO population)

50.1 24.4 39.4 15.7 41.1 18.5 357.0 14.9 32.3 17.9 18.2

8.5 5.0 14.2 5.8 17.4 9.1 175.6 8.3 19.5 11.1 11.1

53.0 32.4 20.5 11.9 29.2

36.2 22.4 14.7 8.4 21.6

15.3

11.7

26.8 17.2 96.3 22.3 17.9 14.9 0.2

20.5 14.8 83.3 22.2 20.1 16.8 24.6

-

-

1081.7

657.0

The causes of death with the highest sex mortality ratios all have major behavioral components (calculated from data in '"1. 'All causes of death were included, except those responsible for less than 1 percent of the deaths (for example, homicide). tFernale death rates have been age-adjustedto the male age distribution (see Table 21.


March. 1976


WALDRON pharmacological effects of smoking.22A rough quantitative estimate of the contribution of smoking habits to the sex differential in CHD death rates can be obtained by comparing the sex differential among nonsmokers to the sex differential for the general population (Table 3). The contribution of cigarette smoking appears to be substantial, particularly for adults under age 65. Among middle-aged adults who have never smoked regularly, the CHD mortality for men exceeds that for women by 350 percent, while for the total sample (including smokers) men’s CHD mortality exceeds women’s by 650 percent. A variety of evidence suggests that another important cause of higher rates of coronary heart disease in men may be their involvement in

I

Ratio of Male to Female Death Rates

paid jobs and in aggressive, competitive roles, in contrast to the greater orientation of women toward family and less competitive, more supportive roles. For example, several studies have found that among men the risk of coronary heart disease is higher for those who have worked many hours overtime or who have held two jobs simultaneously.*5 Also. in a projective test, men who subsequently developed CHD were more likely to locate their stones in a “socioprofessional” setting and not in a family or recreational setting.26 A more specific formulation of the proposed hypothesis can be derived from studies of the “Coronary Prone Behavior Pattern.” A person shows the Coronary Prone Behavior Pattern if he or she is work-oriented, ambitious, aggressive.

TABLE 2 Major Causes of Higher Mortality in M e n

Cause of Death

4.9 2.8 2.7 2.4 2.0 2.0

Malignant neoplasm of respiratory system, not specified as secondary Other bronchopulmonic disease (71% emphysema) Motor vehicle accidents Suicide Other accidents Cirrhosis of liver Arteriosclerotic heart disease, including coronary disease

1.6

All causes

5.9

Male Female Death Rate Death Rate’ (Deathsper 100,ooOPopulation)

50.1 24.4 39.4 15.7 41.1 78.5 357.0

8.5 5.0 14.2 5.8 17.4 9.1 175.6

1081.7

657.0

~~

This table lists all causes of death which had a sex mortality ratio of 2.0 or more and were responsible for at least 1 percent of all deaths in the U.S. in 1967. These causes of death are responsible for three-quarters of the sex differential in mortality. (Calculatedfrom data in *Female death rates have been age-adjusted using the age-specific death rates for females and the age distribution for males to calculate the death rate which would be expected for a population of females that had the same age distribution as the male population. Thus the male and female death rates are directly comparable and are not affected by the higher proportion of females at older ages.

March, 1976


5

SEX DIFFERENTIAL I N MORTALITY

TABLE 3 Comparison of Sex Mortality Ratios for Nonsmokers and for Total Population Ratio of Male to Female Death’dbtes For Those Who Never Smoked Regularly

For the Total Sample

4.5 3.3

2.1

7.5 4.4 2.4

1.6 1.4

7.3 9.4

4.0 2.2

11.7 7.3

1.3 1.7 1.6

2.2 2.5 2.0

Coronary Heart Disease Ages 45-54


55-64 65-74 Lung Cancer Ages 45-64

65-79 Emphysema Ages 45-64

65-79 All Causes of Death Ages 45-54

55-64 65- 74 (Calculated from data in published data.)

24.

Figures for lung cancer and emphysema are approximate due to incomplete

competitive, hurried, impatient and preoccupied with deadlines.” Large prospective studies have shown that men who display this Coronary Prone Behavior Pattern are twice as likely as other men to develop or die of coronary heart disease.>‘2 8 Smaller retrospective studies” ‘I also have established that women who have coronary heart disease are more likely to display the Coronary Prone Behavior Pattern than controls. Behavior pattern may make a larger contribution to the risk of coronary heart disease than does sex ppr SP. Table 4 shows the prevalence of clinical CHD for samples of men” and women I who have clear Coronary Prone Behavior Pattern (called Type A in these studies) or clear Type B (the opposite of Type A). These data 6 Journal of Human Stress

must be interpreted with caution since the samples were small and were not obtained by systematic or even strictly comparable methods. Nevertheless, it is striking that Type B men had the same low prevalence of CHD as did Type B women. This suggests that men who adopt a less competitive and rushed style of life are just as likely to avoid CHD as comparable women. At older ages, Type A men and women had the same high prevalence of CHD. The only category in which women had substantially lower rates of CHD was the younger Type A’s, but even in this age range Type A women had more CHD than Type B men. These data suggest the hypothesis that men have more coronary heart disease than women in part because the Type A or Coronary Prone

March. 1976


WALDRON Behavior Pattern is more prevalent among men. In a large sample of employed adults women were slightly less Type A than men (Shekelle, personal communication). Housewives may be even less Type A than employed women,3' and about half of adult women are house wive^.^^ Aggressiveness and competitiveness are two key components of the Coronary Prone Behavior Pattern. Maccoby and Jacklin (in their review of nearly 2000 studies of sex differences in behaviorJ4) conclude that, on the average, males are more aggressive and competitive than females. Why do males develop more aggressiveness and competitiveness - more of the Coronary Prone Behavior Pattern -than females? Genetic factors make some contribution to the sex differences in aggressiveness," but the extent of

aggressiveness among males varies enormously, depending on child-rearing and cultural condit i o n ~ Sex . ~ ~differences in competitiveness are fostered by parents and schools who push boys to achieve in the occupational world and girls to seek success in the family ~ p h e r e . ~Occupa~.~~-~~ tional achievement apparently requires competitiveness, since in our society there are seldom as many jobs (particularly rewarding, high status jobs) as there are people who want and can do them.39,40In the family sphere, on the other hand, warmth and love are believed to be much more appropriate and aggressive competitiveness much less appropriate than in the business Evidence that cultural pressures and expectations do have a substantial influence on the development of the Coronary Prone Behavior Pattern comes from the observa-

TABLE 4

Relation of Clinical Coronary Heart Disease t o Age, Sex and Behavior Pattern Percent with Clinical Coronary Heart Disease (Sample Size) Type A Behavior Pattern

Type B Behavior Pattern

Premenopausal Females Men less than 50 years old

10% (82) 24% (611

3% (110) 3% (67)

Postmenopausal Females Men more than 50 years old

37% 41%

9% 6%

(43)

(22)

(22) (17)

The Type A Behavior Pattern is the competitive, aggressive, impatient Coronary Prone Behavior Pattern, and Type B is its opposite. For these samples, the prevalence of clinical coronary heart disease among Type B men is as low as the prevalence among Type B women and is substantially lower than the prevalence among Type A women. Although interpretations must be made with caution because systematic sampling methods were not used, these data suggest the hypothesis that more men than women die of coronary heart disease in part because more men develop the Type A or Coronary Prone Behavior Pattern. (Data from 3 1 and 32.)

March. 1976


7


SEX DIFFERENTIAL I N MORTALITY tion that this Behavior Pattern rarely develops in the social environment of many nonindustrial 35 societie~.~’ Thus, a variety of evidence suggests that cultural and socioeconomic pressures related to the role of men in our society push them to develop the Coronary Prone Behavior Pattern. and that this makes a major contribution to men’s higher risk of coronary heart disease. Many aspects of this hypothesis need further testing; some of this testing has been started. Although smoking and the Coronary Prlone Behavior Pattern appear to be the most important behavioral factors contributing to the sex differential in arteriosclerotic heart disease, other behavioral differences also may play a role. For example, women attend church more often than men and frequent church attenders of both sexes have a substantially lower death rate from arteriosclerotic heart disease, at least in one Protestant community ~ t u d i e d . Ex~’ trapolating the risk differential nationally leads to a prediction that men’s death rates for arteriosclerotic heart disease would be 7 percent higher, based on sex differences in church1 attendance and exclusive of related differences in smoking. Bengtsson and co-worker~’~ reach conclusions similar to ours in their study of sex differences in coronary heart disease in 50-54 year old Swedish women and men. They conclude that men’s higher rates of CHD are related to their higher rates of smoking and drinking alcohol, higher aggression and achievement scores, and greater self-reported stress. These authors believe that additional factors also icontribute to the observed sex differences in CHD. Further evidence for the substantial contribution of cultural factors to the sex differential in coronary heart disease is provided by the wide cross-cultural variation in the size of this differential. In some countries the sex differential is much smaller than in the US.” ” For example, in l%O in Greece and Hungary, arteriosclerotic heart disease mortality was only 30 percent higher for males than for females.44The age trend of the male excess also varies widely, with a peak at premenopausal ages in the U.S. and many European countries but a peak at postmenopausal ages in Japan and Colombia 8 Journal of Human Stress

(data On the other hand, men do have higher arteriosclerotic heart disease death rates in all countries studied, and this suggests that genetic factors also contribute to the sex differential. SEX HORMONES AND CHD Most previous discussions of the sex differential in coronary heart disease have focused primary attention on the hypothesis that this sex differential is a result of the physiological effects of the sex hormones. The evidence for this hypothesis is suggestive, but it is ambiguous and inconsistent. Castration of men apparently does not reduce deaths due to cardiovascular disease,45 and castration of older men does not seem to reduce athero~clerosis.~~ The data of Gertler and White4’ suggest that androgen levels of male coronary patients do not differ from androgen levels in a control group. Thus male hormones do not appear to increase the risk of coronary heart disease. Do female hormones lower the risk of coronary heart disease? Several studies have found that oophorectomy of young women is associated with increased atherosclerosis and CHD43.4B-s’ but other studies have not found a relationship.43.52.53One investigation with negative findings is particularly interesting because women with a simple hysterectomy were used as the control group, and the prevalence of arteriosclerotic heart disease was as high for these women with only their uterus removed as it was for women whose ovaries had been removed.” For both groups the prevalence of arteriosclerotic heart disease was higher than the prevalence for the general female population. Since there appears to be no physiological reason why the simple removal of the uterus should lead to increased coronary heart d i s e a ~ e , ’ ~this . ’ ~ observation suggests that hysterectomy may be associated with behavioral characteristics, such as cigarette smoking, or psychological characteristics, such as anxiety and n e u r o t i c i ~ mwhich , ~ ~ in turn are associated with elevated risk of coronary heart disease.56This speculation is particularly plausible since many hysterectomies are elective (for purposes of sterilization) or unnecessary. (The rate of hysterectomies has been reduced by two*We will analyze the cultural causes of the international variation in a subsequent paper.

March, 1976


WALDRON thirds as a result of the initiation of either a medical audit or a requirement for consultati~n.~ These ~ ) considerations are of particular importance for the interpretation of the correlates of oophorectomy since many of the oophorectomies in the studies cited were performed in association with hysterectomies. Thus, if we tentatively accept the weight of evidence as indicating that oophorectomy of young women is associated with increased atherosclerosis and coronary heart disease, we still are left with the question of whether this increase is due to the removal of female hormones or to some behavioral or psychological characteristic of women who undergo these operations. Studies of the effect of replacement therapy are the ideal method for resolving such questions, but in this case such studies have yielded conflicting results. One investigation of estrogen therapy in castrated women found a decreased prevalence of arteriosclerotic cardiovascular disease,S1 and another found trends suggesting a reduced prevalence of electrocardiographic abnormalitie~.~~ However, a third study found no effect on death rates or prevalence of CHD.52 Furthermore, most studies of estrogen therapy in men have found an increased risk of recurrence of myocardial infarction,41,59-62 although in two studies, some treatment groups appeared to have a reduced risk.43.59 The use of oral contraceptives apparently is associated with an increased risk of myocardial infarction, and the estrogen component is suspected as the c a u ~ e . The ~ ~ .effects ~ ~ of estrogen therapy on the risk of coronary heart disease appear to vary with dosage, type of estrogen, and type of patient. The failure of estrogen therapy to produce a consistent decrease in CHD is not surprising in view of the fact that estrogens have biological effects which tend to increase the risk of CHD as well as effects that tend to lower the risk. Evidence for this conclusion is presented in the following paragraphs, which summarize the effects of sex hormones on risk factors for CHD. The data must be interpreted with caution, since some of the studies used synthetic hormones or doses higher than normal physiological levels. Only studies of humans have been cited, with one exception as noted. Important risk factors for CHD include high March. 1976

levels of serum cholesterol, /3-lipoproteins, probably triglycerides, pre-/3-lipoproteins and possibly also c h y l o m i ~ r o n s . ~In ~ - ~some ~ studies, but not all, oophorectomy of young women is associated with elevated serum c h o l e ~ t e r o l . ~ ~ ~ ~ Estrogen therapy generally causes a decrease in serum cholesterol and /3-lipoproteins.5 8 . 7 0 - ' 2 In contrast, estrogens cause an increase in serum triglycerides and pre-/3-lipoproteins 4 3 . 7 0 - 7 ' Androgens generally cause an increase in /3-lipoproteins and a decrease in serum triglycerides, pre-a-lipoproteins and chylomicrons; they have variable effects on serum cholesterol level^.^^,^' Progestogens in general have little or no effect on serum lipids and lipoproteins, although sometimes they produce a small decrease in serum trigly~erides.~~.'~ Female sex hormones enhance thrombotic processes. Oral contraceptives (probably the estrogenic component) cause increased platelet reactivity, increases in various coagulation fact o r and ~ ~ fibrotic ~ vascular lesions.75 In consequence, women who use oral contraceptives have an increased risk of death due to myocardial i n f a r ~ t i o n , cerebral ~ ~ . ~ ~ thrombosis, deep vein thrombosis and pulmonary e m b ~ l i s m . ~ ~ . ~ ~ High doses of estrogen given to men produce increased rates of pulmonary embolism and thrombophlebiti~~~.~" and cerebral thrombosis.61 We did not find data on the effects of male hormones on thrombosis. The use of oral contraceptives results in higher blood pressures, higher angiotensinogen, plasma renin activity and aldosterone level^.^^.'^ Experiments with rats suggest that female hormones may enhance vasoconstriction due to sympathetic activity, which also could lead to elevated blood pressures in females.80-8* However, there is some doubt whether natural female hormones cause elevated blood pressures, since oophorectomy or early menopause has been found to be associated with no significant change52,69.83 or a slight fal130 in systolic blood pressure and no significant ~ h a n g e ~or" , ~ ~ a slight rise69in diastolic blood pressure. Male hormones may cause elevated blood pressures, since testosterone elevates h e r n a t ~ r i t ,and ~~ higher hematocrits are associated with higher blood pressures, possibly due to the increase in viscosity of the blood.84However, testosterone in male castrates apparently produces only small Journal of Human Stress

9

SEX DIFFERENTIAL I N MORTALITY


changes in blood pressure.’OThe average systolic blood pressure of women is lower than that of men until an age which varies between 30 and 50,depending on the sample; at older ages men have lower systolic blood pressures.69 Sex differences in diastolic blood pressure show somewhat similar trends but are much smaller and more variable. This pattern gives no clear evidence for differential effects of sex hormones and can as easily be attributed to psychogenic effects on blood pressure.86 RESPIRATORY DISEASES AND SMOKING For “malignant neoplasm of the respiratory system” (primarily lung cancer), men’s death rates are six times higher than women’s. For “other bronchopulmonic diseases” (primarily emphysema), men’s death rates are five times higher than women’s. These mortality ratios are higher than for any other major cause of death (Table 2). Men have higher mortality for these diseases primarily because men smoke more and cigarette smoking is the major cause of both lung cancer and emphysema.b22If men and women who have never smoked regularly arle compared, the sex mortality ratios for lung cancer and emphysema are drastically reduced (Table 3). These data suggest that cigarette smoking is the primary cause of men’s excess lung cancer and emphysema mortality. Comparing those who have ever smoked cigarettes to nonsmokers of the same sex. lung cancer death rates are elevated ninefold for men, but only twofold for women. Similarly, emphysema death rates are elevated sevenfold among men smokers, but only fivefold for wornen smokers. The elevation of death rates is less for women smokers than for men in large part because women smokers inhale less, smoke fewer cigarettes and less of each cigarette and, in the past, women have begun smoking at older ages.” 24 In addition, industrial hazards aggravate the effects of cigarette smoking f c r many men (as discussed below). The total pathological effect of smoking, pairticularly the elevation of coronary heart diseasle, lung cancer and emphysema, makes a major contribution to the sex differential in total death rates. For middle-aged adults who never have smoked regularly, men’s mortality exceeds


women’s by only 30 percent, compared to a male excess of 120 percent for the total sample (Table 3). For older nonsmokers, men’s mortality exceeds women’s by 60-70 percent, compared to 100-150 percent for the total sample. Retherford5 (in a similar analysis which was published while this manuscript was in the final stages of preparation) estimates that as much as half of the sex differential in life expectancy from the ages of 37 to 87 may be due to the effects of higher rates of cigarette smoking in men. Why do more men than women smoke? Smoking by women was strongly discouraged by the social mores of the early twentieth century. The conventions ofthat period continue to influence the smoking patterns of people who were teenagers at that time, since relatively few people begin smoking cigarettes after age 20. As a consequence, the sex differential in cigarette smoking is largest for older people who were over 60 in 1970 and who thus were teenagers before 1930.87 Many other social and motivational factors have been shown to influence cigarette ~ m o k i n g . ~ ~ . ~ Among these, the factor which probably contributes most to the sex differential in smoking is the strong component of rebelliousness which cigarette smoking has had for many teenagers. I n general, girls tend to be less rebellious and more conforming to adult standards, probably in part because parents and teachers of schoolaged children allow boys more independence and expect girls to be more ~ b e d i e n t . ~ ~ . ~ O Girls’ lesser rebelliousness is probably one reason why, until very recently, fewer teenage girls than boys had begun smoking. Although cigarette smoking is the major cause of the higher rates of lung cancer in men, industrial carcinogens also make a substantial contribution. Men who work with asbestos have up to eight times higher a risk of (bronchogenic) lung cancer than other men.93This elevated risk affects primarily cigarette smokers. Asbestos is widely used in construction and insulation materials, and about one man in 100 is now or has been exposed to asbestos dust at his Thus, asbestos may be responsible for one in 20 male lung cancer deaths in this country. Metallic dusts and fumes elevate lung cancer risk between 20 percent and 130 percent for various categories of metal worker^.^' About one man in 30 works or has worked in such an ~ c c u p a t i o n . ~ ~ March, 1976


WALDRON Thus, metallic dusts and fumes may be responsible for one in 50 male lung cancer deaths. Taken together, the established and suspected industrial carcinogens appear to be a factor in roughly one out of every 10 male lung cancer death^.^^^^' I n Part IZ, which will appear in the next issue, Dr. Waldron and Susan Johnston analyze the causes o f higher male rates o f j a t a l accidents, cirrhosis of the liver and suicide. Part I1 also contains brief discussions of sex di3erences in total cancer mortality, sex chromosome-linked heritable conditions, the contribution o f health care, and recent trends in the sex dijferential in mortality. These analyses. together with those presented above. lead to their concluding suggestions j o r the reduction of male mortality. They suggest that this may be accomplished by changes in the social and economic pressures which .foster the behaviors and attitudes that currentlv elevate male mortality. The author is happy to thank Joseph Eyer, lean Gerth, Deborah Heebner. and Kimberly Schmidt for their help in finding useful materials. She is grateful to many friends and colleagues. particularly C.D. Jenkins. for their helpful comments on an earlier version of the manuscript.

INDEX TERMS sex dif€erenca, comauy bart dbeue, smoking, Coronary Prone Behvior P8ttem, wx hormones.

REFERENCES 1. Keyfitz, N., and W. Flieger. World Population - A n Analysis of Vital Data. University of Chicago Press, Chicago, 1968. 2. United States Department of Health, Education and Welfare, Public Health Service. Vital Statistics of the United States. 1970, VoL II-Mortality. Government Printing Office, Washington, D.C., 1974. 3. Yerushalmy. J. “Factors in Human Longevity.” Amer. J. Public Health, Vol. 53, 1963, pp. 148-162. 4. Enterline. P.E. “Causes of Death Responsible for Recent Increases in Sex Mortality Differentials in the United States,” Milbank Mem. Fund Q.. Vol. 39. 1961. pp. 312-338. 5. Retherford. R. D. The Changing Sex DSfferential in Mortality. Studies in Population and Urban Demography X1. Greenwood Press, Westporf Conn., 1975. 6. Burbank, F. “U. S. Lung Cancer Death Rates Begin to Rise Proportionately More Rapidly for Females than for Males: A Dose-Response Effect?” J. Chron. Dis., Vol. 25, 1972, pp. 473-479. 7. Stolnitz, G.J. “A Century of International Mortality Trends: 11,” Population Studies, Vol. 10, 1956, pp. 1742 8. United States Department of Health, Education and Welfare, National Center for Health Statistics. Mortali-

March. 1976

9. 10.

11.

12. 13.

ty from Selected Causes by Marital Status: US. Vital and Health Statistics. Series 20, Nos. 8a and 8b. Government Printing Office, Washington, D.C., 1963. Cove. W.R. “Sex, Marital Status, and Mortality,” Amer. J. SocioL. Vol. 79, 1973, pp. 45-67. Hamilton, 1.B. “The Role of Testicular Secretions as Indicated by the Effects of Castration in Man and by Studies of Pathological Conditions and the Short Life Span Associated with Maleness.” Recent Prog. Horm. Res.. Vol. 3. 1948, pp. 251-322. Ricklefs, R.E. “Fecundity. Mortality and Avian DemcF graphy,” Breeding Biologv of Birds, D.S. Farner, ed., pp. 370-390. National Academy of Science, Washing ton, 1973. Caughley. G. “Mortality Patterns in Mammals.” Eco1 0 0 , Vol. 47. 1966. pp. 906-918. Tricomi, V., 0. Serr, and C. Solish. “The Ratio of Male to Female Embryos as Determined by the Sex Chromatin.” A m . J. Obstet. Gynecol.. Vol. 79, 1960, pp. 504-

509. 14. Donaldson, R. S.. and S. G. Kohl. “Perinatal Mortality in Twins by Sex,” A m . J. Public Health. Vol. 55, 1965. pp. 1411-1418. 15. Hammoud. E. 1. “Studies in Fetal and Infant Mortality.“ Am. J. Public Health. Vol. 55. 1965, pp. 11521163. 16. Teitelbaum, M. S. “Male and Female Components of Perinatal Mortality: International Trends, 1901 -63.” Demography. Vol. 8, 1971, pp. 541-548. 17. Scheinfeld, A. “The Mortality of Men and Women,” Sci Am.. Vol. 198, Feb., 1958, pp. 22-27. 18. Madigan, F.C. “Are Sex Mortality Differentials Biologically Caused?”, Milbank Mem. Fund Q., Vol. 35.1957. pp. 202-223. 19. Sowder. W. T.. 1. 0. Bond, E. H. Williams, Jr.. and E. L. Flemming. Man to Man Talk about Women . . . and Men. Florida State Board of Health. Monograph Series No. 10, Jacksonville, 1966. 20. Potts, D.M. “Which is the Weaker Sex?”. J. Biosoc. Sci., Suppl. 2, 1970. pp. 147-157. 21. ”Sex Differentials in Mortality,” Stat. BulL Metropol Life Ins. Co.. August, 1974, pp. 2-5. 22. Report of the Royal College of Physicians. Smoking and Health Now. Pitman Medical and Scientific Publ., London, 1971. 23. Lelbach, W.K. “Organic PatholoaJ Related to Volume and Pattern of Alcohol Use.” Research Advances in AIcoholand Drug Programs, Vol. I, 1974, pp. 93-198. 24. Hammond, E.C. “Smoking in Relation to the Death Rates of One Million Men and Women,” Natl. Cancer Inst. Monogr., Vol. 19, 1966, pp. 127-204. 25. Jenkins, C.D. “The Coronary-Prone Personality.” Psychological Aspects of Myocardial Infarction and Coronary Care, W.D. Gentry and R.B. Williams. eds.. pp. 5-23. V. Mosby. St Louis, 1975. 26. Bonami. M., and B. Rime. “Approche Exploratoire de la Personalite Pre-Coronarienne par Analyse Standardisee de Donnks Projectives Thematiques,” J. P s y chmom. Res.. Vol. 16,1972, pp. 103-113. 27. Rosenman, R.H. “The Role of Behavior Patterns and Neurogenic Factors in the Pathogenesis of Coronary

Journal of Human Stress 11

SEX DIFFERENTIAL I N MORTALITY Heart Disease.” Stress and the Heart. R.S. Eliot, ed.. pp. 123-141. FuturaPubl., Mount Kisco, N.Y.. 1974. 28. Rosenman, R.H.. R.J. Brand, C.D. Jenkins, et al. ‘ T o r onary Heart Disease in the Western Collaboraitive Group Study.” J.A.M.A.. Vol. 233, 1975, pp. 872877. 29. Kenigsberg, D., S. I. Zyzanski. C. D. Jenkins, el: al. “The Coronary-Prone Behavior Pattern in Hospitalized Patients With and Without Coronary Heart Disease.” Psychosom. Med.. Vol. 36, 1974, pp. 344-351.

30. Bengtsson. C. “Ischaemic Heart Disease in Women,” Acta Med. Scand.. Suppl. 549. 1973, pp. 1-128. 31. Rosenman, R.H., and M. Friedman. “Association of Specific Behavior Pattern in Women with Blood and Cardiovascular Findings.” Circulation, Vol. 24, 1061 pp. 1173-1 184. 32. Friedman, M.. and R.H. Rosenman. “Association of Specific Overt Behavior Pattern with Blood and Cardiovascular Findings.” J.A.M.A. Vol. 169, 1959. pp. 1286-1296.


I

33. U.S. Department of Labor. Bureau of Labor Statistics. Emplovmenr and Earnings. Vol. 19,1973. p. 30. 34. Maccoby. E.E., and C.N. Jacklin. The Psychology of’ Sex Dzperences. Stanford University Press, Stanford. 1974. 35. Dentan. R.K. The Semai-A Nonviolent People of‘ Adala-va. Holt, Rinehart and Winston, New York, 1968. 34. Saario. T.N.. C.N. Jacklin. and C.K. Tittle. “Sex Role Stereotyping in the Public Schools,” Harvard Educational Review. Vol. 43. 1973, pp. 386-416. 37. Bart, P. ”Why Women See the Future Differently from Men.” Learning,for Tomorrow: The Role of the Future in Education. A. Toffler, ed., pp. 33-55. Random House, New York. 1972. 38. U’Ren. M.B. “The Image of Woman in Textbooks.” Woman in Sexist Society. V. Gornick. and B.K. Moran, eds., pp. 318-328. New American Library, New York. 1971. 34. Chinoy. E. The Automobile Worker and the American Dream. Doubleday & Co., New York. 1955. 40. Report of a Special Task Force to the Secretary of Health, Education and Welfare. Work in America. MIT Press, Cambridge, 1973. 41. ”U.S. Churchgoing 40% in ‘71 Poll,” New York T i m w January 9, 1972. p. 59. 42. Comstock. G.W.. and K.B. Partridge. “Church Attcndance and Health,” J. Chronic Dis.. Vol. 25. 1972. pp. 665-672. 43 Furman, R. H. “Endocrine Factors in Atherogenesis,“ Arherosclerosis. F. G. Schettler, and G. S. Boyd. eds.. pp. 375-409. Elsevier Publ.. Amsterdam, 1969. 44 Segi. M.. M. Kurihara, and Y. Tsukahara. Mortal‘ity ,for Selected Causes in 30 Countries 11950-1961). Kosei Tokei Kyokai. Tokyo. 1966. 45. Hamilton, J.B.. and G.E. Mestler. “Mortality and Survival: Comparison of Eunuchs with Intact Men and Women in a Mentally Retarded Population,” J. Gertmt01.. Vol. 24, 1969, pp. 395-411. 46. London, W. T.. S. E. Rosenberg, 1. W. Draper, and T. P. Almy. “The Effect of Estrogens on Atherosclerosis.”


Ann. Intern. Med.. Vol. 55, 1961, pp. 63-69. 47. Gertler. M.M., and P.D. White. Coronary Heart Disease in Young Adults. Harvard University Press, Cambridge, 1954. 48. Berkson. D.M., J. Stamler, and D.B. Cohen. “Ovarian Function and Coronary Atherosclerosis.” Clin Obsret. Gynecol.. Vol. 7, 1964, pp. 504-530. 49. Ask-Upmark, E. “Life and Death Without Ovaries,” ActaMed. Scand.. Vol. 172. 1962, pp. 129-135. 50. Parrish, H., C.A. Carr, D.G. Hall, and T.M. King. ”Time Interval from Castration in Premenopausal Women to Development of Excessive Coronary Atherosclerosis,” Amer. J. Obstet. Gynecol.. VoS. 99, 1967, pp. 155-162. 51. Higano, N., R. W. Robinson, and W. D. Cohen. “lncreased Incidence of Cardiovascular Disease in Castrated Women,” N. Engl. J. Med.. Vol. 268, 1963, pp. 1 123-1125. 52. Ritterband. A.B.. I.A. Jaffe, P.M. Densen. et al. “Gonadal Function and the Development of Coronary Heart Disease.” Circulation. Vol. 27, 1963, pp. 237-251. 53. Williams, T.J., and E.R. Novak. “Effect of Castration and Hysterectomy on the Female Cardiovascular System.” Geriatrics. Vol. i8. 1963. 852-859. 54. Beavis. E.L.G.. J.B. Brown, and M.A. Smith. “Ovarian Function after Hysterectomy with Conservation of the Ovaries in Premenopausal Women,” J . Obstet. Gynaecol. Br. Commonw.. Vol. 76, 1969, pp. 969-978. 55. Barker, M.G. “Psychiatric Illness after Hysterectomy,” Br. Med. J.. Vol. 2, 1968, pp. 91-95. 56. Jenkins, C.D. “Psychologic and Social Precursors of Coronary Disease, Part 1,” N. Engl. J . Med.. Vol. 284, 1971. pp. 244-255. 57. Bunker, J.P. “Surgical Manpower - A Comparison of Operations and Surgeons in the United States and in England and Wales,” N. Engl. J. Med.. Vol. 282, 1970. pp. 135-144. 58. Davis. M.E.. R.J. Jones. and C. Jarolim. “Long-term Estrogen Substitution and Atherosclerosis,” A m . J. Obstet. Cynecol.. Vol. 82, 1961. pp. 1003-1018. 59. Stamler. J., M.M. Gest, and J.P. Turner. “The Status of Hormonal Therapy for the Primary and Secondary Prevention of Atherosclerotic Coronary Heart Disease,” Prog. Cardiovasc. Dis.. Vol. 6, 1963, pp. 220-235. 60. Veterans Administration Cooperative Urological Research Group. “Treatment and Survival of Patients with Cancer of the Prostate.” Surgery, Gynecol. and Obsfer.. Vol. 124. 1967, pp. 1011-1017. 61. Blackard, C.E., R.P. Doe, G.T. Mellinger. and D.P. Byar. “Incidence of Cardiovascular Disease and Death in Patients Receiving Diethylstilbestrol for Carcinoma of the Prostate.” Cancer, Vol. 26, pp. 249-256. 62. Coronary Drug Project Research Group, “The Coronary Drug Project-Initial Findings Leading to Modifications of its Research Protocol,” J.A.M.A. Vol. 214, 1970, pp. 1303-1313. 63. Mann, J. I.. M. P. Vessey. M. Thorogood, and R. Doll. “Myocardial Infarction in Young Women with Special Reference to Oral Contraceptive Practice,” Br. Med. J., VOI. 2, 1975. pp. 241-245. 64. Mann, 1. I., and W. H. W. Inman. “Oral Contracep

March. 1976

WALDRON tives and Death from Myocardial Infarction,” Br. Med. J., Vol. 2,1975. pp, 245-248.

82. Weil-Malherbe. H. “The Adrenergic Amines of Human Blood,” Lancet I, 1953, pp. 974-977.

65. Rosenman, R.H., M. Friedman, R. Straus, C.D. Jen-

83. Kannel, W.B., and P. Sorlie. “Hypertension in Fra-


kins, S.J. Zyzawki, and M. Wurm. “Coronary Heart Disease in the Western Collaborative Group Study: A Follow-up Experience of 4% Years,” J. Chronic Dis., Vol. 23,1970, pp. 173-190. 66. Carlson, L.A., and L.E. Bottiger. “lschaemic HeartDisease in Relation to Fasting Values of Plasma Triglycerides and Cholesterol,” Lancet, Vol. 1, 1972. pp. 865-868. 67. Zilversmit, D.B. “A Proposal Linking Atherogenesis to the Interaction of Endothelial Lipoprotein Lipase with Triglyceride-rich Lipoproteins,” Circ. Res.. Vol. 33, 1973, pp. 633-638. 68. Tydskrif, S.-A.M. “Cholesterol. Coronary Heart Disease and Oestrogens.” S. A f . Med. J., Vol. 45, 1971, pp. 359-361. 69. Weiss, N.S. “Relationship of Menopause to Serum Cholesterol and Arterial Blood Pressure: The United States’ Health Examination Survey of Adults,” Amer. J. Epidemiol. Vol. 96.197Z pp. 237-241. 70. Coronary Drug Project Research Group. “The Coronary Drug Project - Findings Leading to Discontinuation of the 2.5-mg/day Estrogen Group.” J.A.M.A., Vol. 226,1973, pp. 652-657. 71. Furman. R.H. “Gonadal Steroid Effects on Serum Lipids,” Metabolic Effects of Gonadal Hormones and Contraceptive Steroids, H.A. Salhanick, D.M. Kipnis, and R.L. Vande Wiele, eds., pp. 247-264. Plenum Press, New York, 1%9. 72. Alfin-Slater, R. B., and L. Aftergood. “Lipids and the Pill,” Lipids, Vol. 6. 1971. pp. 693-705. 73. Glueck, C.J., and R. Fallat “Gonadal Homes and Triglycerides,” Proc. R. Soc. M e d . Vol. 67, 1974, pp. 667-669. 74. Dugdale, M., and A.T. Masi. “Hormonal Contraception and Thromboembolic Disease: Effects of the Oral Contraceptives on Hemostatic Mechanisms,’’J. Chron. Dis.. Vol. 23,1971. pp. 775-790. 75. Irey, N.S.. W.C. Manion, and H.B. Taylor. “Vascular Lesions in Women Taking Oral Contraceptives.” Arch. PathoL, Vol. 39, 1970, pp. 1-8. 76. Vessey, M.P. “Thromboembolism, Cancer and Oral Contraceptives.” Clin Obstet. GynecoL. Vol 17, 1974. pp. 65-78. 77. Stolley. P. D., 1. A. Tonascia. M. S. Tockman, et al. “Thrombosis with Low-Estrogen Oral Contraceptives.” Amer. 1. Epidemiol. Vol. 102. 1975. pp. 197-208. 78. Fregly. M. J.. and M. S. Fregly. eds., Oral Contmcep tives and High Blood Pressure. Dolphin Press, Gainesville, Fla., 1974. 79. Fisch, I.R., S.H. Freedman, and A.V. Myatt “Oral Contraceptives, Pregnancy. and Blood Pressure.” J.A.M.A.. Vol. 222, 1972. pp. 1507-1510. 80. Altura. B.N. “Sex as a Factor Influencing the Responsiveness of Arterioles to Catecholamines,” Eur. J. Pharmacol., Vol. 20. 1972. pp. 261-265. 81. Green. R.D.. J.W. Miller. ”Catecholamine Concentrations: Changes in Plasma of Rats During Estrous Cycle and Pregnancy,” Science, Vol. 151,1966. pp. 825-826.

March. 1976

mingham,” In Manuscript

84. Kannel, W.B., T.R. Dawber. “Hypertensive Cardiovascular Disease: The Framingham Study,” &pertension: Mechanisms and Management, C. Onesti, K.E. Kim. and J.H. Moyer, eds.. pp. 93-110. Grune and Stratton. 1973. 85. McDonough, J.R., G.E. Garrison, and C. G. Hames. “Blood Pressure and Hypertensive Disease Among Negroes and Whites.” A n n Intern. Med.. Vol. 61, 1964. pp. 208-228. 86. Benson, H.,and M.C. Gutmann. “The Relation of Environmental Factors to Systemic Arterial Hypertension.” Stress and the Heart. R.S. Eliot, ed.. pp. 13-31. Futura Publ. Co., Mount Kisco, New York, 1974. 87. United States Department of Health, Education and

88.

89.

90.

91.

92.

93.

94. 95.

%.

97.

98.

Welfare, Public Health Service, National Clearinghouse for Smoking and Health. Adult Use of Tobaccc+l970. U.S. Government Printing Office. Washington, D.C.. 1973. Borgatta. E.F., and R.R. Evans. “Social and Psychological Concomitants of Smoking Behavior and Its Change Among University Freshmen,” Smoking, Health and Behavior, E. F. Borgatta and R.R. Evans, eds.. pp. 206-219. Aldine Publ.. Chicago, 1968. Mausner, B., E.S. P l a n with assitance of J.S. Mausner. Smoking: A Behavioral Analysis. Pergamon Press, New York, 1971. Fischer, J.L., and A.F. Fischer. “The New Englanders of Orchardtown,” Six Cultures-Studies of ChildRearing. B.B. Whiting, ed.. John Wiley and Sons, New York, 1963. Hoffman, L.W. “Early Childhood Experiences and Women’s Achievement Motives.” J. Social Issues. Vol. 28,1972, pp. 129-155. Levitin, T.E.. and J.D. Chananie. “Responses of Female Primary School Teachers to Sex-typed Behaviors in Male and Female Children.” Child Dev.. Vol. 43, 1972, pp. 1309-1316. Selikoff. I.J.. E.C. Hammond, and J. Churg. “Asbestos Exposure, Smoking, and Neoplasia,” J.A.M.A.. Vol. 204,1968, pp. 104-110. Stellman. J.M.. and S.M. Daum. Work is Dangerous to YourHealth. Vintage Books, New York, 1973. Hueper. W.C. “Occupational and Environmental Cancers of the Respiratory System.” Recent Results Cancer Res.. Vol. 3, 1966, pp. 1-130. U.S. Department of Commerce, Social and Economic Statistics Administration. Subject Reports. Final Report PCL3 7A. Occupational Characteristics. Government Printing Office. Washington. D.C., 1973. Breslow. L., L. Hoaglin. G. Rasmussen, and H.K. Abrams. “Occupations and Cigarette Smoking as Factors in Lung Cancer,” Amer. J. Pub. Health, Vol. 44. 1954,pp. 171-181. United States Department of Health. Education and Welfare, Public Health Service. Vital Statistics of the United States, 1967. VoL I1 -Mortality. Government Printing Oftice, Washington, D.C.. 1%9.

Journal of Human Stress 13

Why do women live longer than men?

The gender gap: why do women live longer than men?

Why Do Women Justify Violence Against Wives More Often Than Do Men in Vietnam?

Do men perform better than women in trauma?

Do women develop alcoholic brain damage more readily than men?

Do Men Really Benefit More From Marriage Than Women?

Why do psychiatric patients in Korea stay longer in hospital?

Do women live longer or do men die earlier? Reflections on the causes of sex differences in life expectancy.

Longevity after mechanical aortic root replacement--do men live longer?

Why is Cancer More Depressing for Men than Women among Older White Adults?

Psychosocial Variables Related to Why Women are Less Active than Men and Related Health Implications.

Could androgens be relevant to partly explain why men have lower life expectancy than women?

Why more women than men have cholesterol gallstones: studies of biliary lipids in pregnancy.

Why do women have trouble getting credit?

Do women benefit more than men from post-MI beta-blockade?

Steeplechase barriers affect women less than men.

What do men want? Re-examining whether men benefit from higher fertility than is optimal for women.

Metabolically healthy obese women have longer telomere length than obese women with metabolic syndrome.

We live longer than politicians.

Breast cancer: why do women opt for contralateral prophylactic mastectomy?

Women as family caregivers: why do they care?

Why are the Japanese living longer?

Women are better at selecting gifts than men.

Fasting hepatic glucose uptake is higher in men than women.