PRESIDENTIAL ADDRESS Myths and Realities of the Menopause KAREN A. MATTHEWS, PHD
Menopause is a reproductive milestone in a woman's life (1). To paraphrase MacDonald (1), on the one hand, it marks the end of reproductive life, with unfulfilled dreams of childbearing for some, and the beginning of increased risk for the health consequences of estrogen deprivation. On the other hand, it signifies a time when women no longer need to experience the inconvenience associated with menses, when they are free from dysmenhorrhea, and when they need not fear unwanted pregnancies. Menopause occurs in midlife at a time when many women are experiencing changes in roles, responsibilities, and relationships that accompany aging generally and the maturation of children and their departure from home in particular. These changes may create considerable stress for some women, affecting their identity, self-esteem, and social and family relationships. For others, these might mark the beginning of more fulfilling re-
Presented at the Annual Meeting of the American Psychosomatic Society, Sante Fe, New Mexico, March 1991. From the University of Pittsburgh, School of Medicine, Pittsburgh, Pennsylvania. Address reprint requests to: Karen A. Matthews, Ph.D., Department of Psychiatry, University of Pittsburgh, 3811 O'Hara Street, Pittsburgh, PA 15213. Received for publication August 7, 1991; revision received August 30,1991
Psychosomatic Medicine 54:1-9 (1992) 0033-3174/92/5401-000l$03.00/0 Copyright © 1992 by the American Psychosomatic Society
lationships and new challenges for positive psychological growth. Thus, menopause is a critical transition in women's lives not only because of the biological changes they are experiencing, but also because of the co-occurring social and psychological changes. Understanding the independent and interactive effects of the social, psychological, and biological changes during the menopause is crucial for understanding midlife development and provides the basis for understanding later adult health and dysfunction. Furthermore, understanding the menopausal transition can provide a window on basic social, psychological, and biological processes in the life course of a woman. Stated differently, a psychosomatic approach to the menopause has substantial heuristic and practical value. This paper examines the basic biological changes that occur during the menopausal years and how those impact, both alone and in interaction with social and psychological factors, on the quality of women's lives and on their risk for coronary heart disease (CHD). The discussion is organized according to the myths that society has held and continues to hold with regard to the meaning of the menopause. As will be noted, myths about the menopause have been remarkably negative.
KAREN A. MATTHEWS, PhD
MYTHS ABOUT THE MENOPAUSE
Historical Perspective The menopause was referred to in early cultures and in many texts (2). By the sixth century A.D., the cessation of menstruation was well documented to occur around the age of 50 (3) and that average age has not changed (4), despite the facts that the onset of menses has shifted to an earlier age, and that women live longer now than ever before. In fact, women now live about one third of their adult years beyond the menopause. Some observers recognized that the risk for chronic diseases increased after the menopause (2). In 1777, John Leake in his book, Chronic or Slow Diseases Peculiar to Women, noted that: "At this criticaJ time of life, the female sex are often visited with various diseases of the chronic kind" (5). In 1814, John Burns indicated that "the cessation of menses does of itself seem, in some cases, to excite cancer of the breast" (6). Some observers also indicated that psychological complaints and symptoms increased during the menopausal transition (2). Again, John Leake noted that "some are subject to pain and giddiness of the head, hysteric disorders, colic pains, and a female weakness . . . often very troublesome to others" due to "the many excesses introduced by luxury and the irregularities of the passion" (5). This negative expectation regarding psychological changes during the menopause has persisted into the 20th century (7). Two female psychoanalysts, Deutsch and Benedek, writing in the midcentury, believed that psychological symptoms at the time of the menopause were grief reactions to the loss of menstruation and to all it represents (8, 9). Deutsch noted
that "woman has ended her existence as bearer of future life,. ..—as servant of the species. She is now engaged in an active struggle against her decline" (8, p. 459). Nonetheless, Benedek saw opportunities for growth in the process of coping with the menopause, in that it can result in the "emancipation from sexual competition and from the fear of sexual rejection (which) often releases talents and qualities unsuspected before" (9, p. 23). Contemporary Perspective More recent writers, most notably Wilson who has championed hormone replacement therapy, have expressed negative attitudes toward the menopause: "A large percentage of women . . . acquire a vapid cowlike feeling called a 'negative state.' It is a strange endogenous misery . . . the world appears as though through a grey veil, and they live as docile, harmless creatures missing most of life's values" (10, p. 347). "The menopausal woman is not normal; she suffers from a deficiency disease with serious sequelae and needs treatment" (11, p. 110). Despite these views expressed throughout the centuries as well as in more recent writings, many women currently have some positive expectations and beliefs about the menopausal experience, especially for themselves. We have been studying how the menopause affects women's mental and physical health in a sample of 541 women, who at the beginning of the study, were premenopausal, relatively healthy, and ages 42 to 50 years old (12). At entry into this study, The Healthy Women Study, women were asked about their attitudes toward the menopause and their expectations about its psychological and somatic effects on Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
women, using an adaptation of a measure developed by Neugarten and colleagues (13). After completion of the Neugarten measure, the women were asked about the same expectations, but with regard to their own menopausal experience. About 60% agreed with the statement, "Going through the menopause does not really change a woman in any important way," whereas 79% agreed with the statement, "Going through the menopause will not really change me in any important way." Furthermore, almost all women (94%) believed that if they knew what to expect, and did not expect trouble during the menopause, they would do well. Despite these positive attitudes, a substantial number of the women did have rather negative expectations about some effects of the menopause. About 80% thought women are liable to get depressed during the menopause, and 55% thought they themselves would be liable to become depressed then. About % of the sample thought women in general have hot flashes during the menopause, and 70% thought they would have these. Sixtyeight percent agreed that "During the menopause, women are liable to fly off the handle more easily than at other times"; and 48% thought that "During the menopause, I am liable to fly off the handle more easily than at other times." Are these expectations myths or realities? Often expectations that develop throughout the centuries are based on reality, but sometimes expectations can act like self-fulfilling prophesies and determine the reality that society accepts. Perhaps these expectations are true for a small subset of women, but because these occurrences are extremely salient and negative, society overgeneralizes their experience to all women. In the next sections, we will examine the extent to Psychosomatic Medicine 54:1-9 (1992)
which each of the above beliefs about the menopause was supported in the Healthy Women Study. Do Women Experience Psychological and Vasomotor Symptoms During the Menopause? To address this question, women's selfreported psychological characteristics and symptoms at entry into the Healthy Women Study when all women were premenopausal were compared to those same characteristics reported after the women became postmenopausal. Study protocol was that all 541 women were evaluated extensively when they were premenopausal at study entry. This examination was repeated when women became postmenopausal, i.e., had ceased menstruating for 12 months and had elevated follicular stimulating hormone (FSH) levels. For each postmenopausal woman, a premenopausal control, i.e., who had menstruated within the previous 3 months and was not taking hormone replacement therapy, was also re-evaluated at the same time. Analyses compared the psychological characteristics and symptoms of the first 69 postmenopausal women with that of their premenopausal controls. Changes that occurred in both groups were interpreted as being due to aging, whereas changes that were different in the menopausal group were interpreted as being due to aging plus change to postmenopausal status. Our results showed women who became postmenopausal and their premenopausal controls reported the same increase in depressive symptoms on the Beck Depression Inventory and the same increase in the total number of symptoms occurring during the previous 2 weeks
KAREN A. MATTHEWS, PhD
(14). However, postmenopausal women did report experiencing more frequently a subset of symptoms at the postmenopausal examination, relative to the premenopausal examination. More specifically, postmenopausal women reported more hot flashes, joint pain, and troubles sleeping. Despite experiencing more vasomotor symptoms, postmenopausal women perceived lower levels of stress in their daily lives at the postmenopausal examination, relative to the premenopausal examination (14). Taken together, these findings suggest that natural menopause does not lead to enhanced psychological distress, even though it does lead to an increase in vasomotor symptoms. Indeed, these findings are consistent with the results of two other recent longitudinal studies of the menopause (15, 16), as well as cross-sectional studies of large samples of women (see 17 for a review). Although the complete cessation of menses does not cause adverse changes for the majority of women, perhaps the initial cessation of menses does. The perimenopause might be often accompanied by the first occurrences of uncomfortable vasomotor symptoms, but as women learn to cope with them, they might experience acute adverse change in depressive symptoms that do not persist. Indeed, that appears to be the case among the first 105 women who ceased cycling for 3 months. Perimenopausal women increased significantly, albeit slightly in their Beck Depression scores, relatively to women who remained premenopausal. The number of women who had Beck Depression Inventory scores 9 or above increased sixfold over the course of the follow-up period. Note, however, that this six-fold increase represented only about 10% of the perimenopausal women. More striking is the fact that many
women reported at their perimenopausal examination experiencing a variety of disturbing symptoms: There was a tenfold increase in the number of women reporting hot flashes and a three-fold increase in the number of women reporting joint pain, being forgetful, worrying about their body, being constipated, and having dizzy spells. No woman reported experiencing cold sweats at her premenopausal examination only, whereas 14 women reported them at their perimenopausal examination (and hot at the premenopausal exam). Our hypothesis that the psychological symptoms that do occur during the early perimenopause are related to the experience of having vasomotor symptoms for the first time are supported elsewhere. Campbell and Whitehead (18) demonstrated that 4 months of conjugated estrogen therapy during the perimenopause led to a dramatic improvement in hot flashes, vaginal dryness, insomnia, irritability, and anxiety, compared to a placebo condition. They noted that these symptoms were interrelated and that the benefit was initiated by improvement in hot flashes. One reason for their latter conclusion is that of nine symptoms that improved with hormone replacement therapy, five of them did not improve unless the patient had hot flashes at outset. Women who did not have hot flashes at outset failed to show improvement in the more psychological symptoms, like irritability, optimism, and good spirits. In sum, the myth that women experience negative symptoms during the menopause has a grain of truth. Women do experience uncomfortable physical symptoms during the perimenopause, some of which are relieved by the time of the complete cessation of menses. Even so, the vast majority of women do not bePsychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
come depressed either during the perimenopause or afterwards.
tive expectations about the menopause would facilitate a good experience during the menopause had lower levels of depression and fewer symptoms at their If Women Expect Trouble During the postmenopausal evaluation, relative to Menopause, Will They Have It? their premenopausal evaluation. Thus, it does appear that women's expectations Do expectations about the menopause about aspects of the menopause affect affect the quality of the menopausal ex- their psychological experience during the perience? To examine this question, we menopause. first factor analyzed ratings of the attiWe can suggest at least two possible tudes toward menopause that the 541 explanations of these results. First, women reported at their baseline exami- women at baseline who expected adverse nation. This analysis led to the develop- symptoms during the menopause may ment of five scales, three of which are have already begun to experience them relevant here: a) expectations of vasomo- at that time and merely were accurate tor symptoms; b) expectations of benefits reporters. This explanation, however, of the menopause, e.g., no more worry seems incompatible with the fact that we about unwanted pregnancy; and c) expec- statistically controlled for the baseline tations that information and positive ex- levels of symptoms. Or, second, women pectancies will help during the meno- who expect negative consequences of the pause. We correlated these scales meas- menopause may be behaving in such a ured at study entry with the changes from way that at least in part contributes to the study entry to postmenopausal examina- development or maintenance of symption in Beck Depression Inventory scores, toms. For example, women who expect total number of symptoms experienced in adverse vasomotor symptoms, when they the previous 2 weeks, Spielberger Trait are encountered, may become depressed Anger Scale scores, and level of social at the prospect of facing the inconvenisupport. The change scores were adjusted ence and discomfort of those symptoms, for the baseline premenopausal level of rather than taking positive action in copthe characteristic in question. ing with them. Stated differently, expectThese analyses showed several intrigu- ing negative menopausal symptoms may ing findings: a) Those women who ex- lead to a self-fulfilling prophesy. pected to experience vasomotor symptoms during the menopause did increase in their level of depressive symptoms and Are There No Important Changes angry feelings at their postmenopausal During the Menopause? evaluation, relative to their premenopausal evaluation, b) Those women who exAs noted above, it has long been recogpected the menopause to have some ben- nized that postmenopausal women have efits did indeed report that their levels of more chronic disease than do premenosocial support increased at their postmen- pausal women, but the role of declining opausal evaluation, relative to their pre- reproductive hormones in contributing to menopausal evaluation, c) Those women increasing prevalence of chronic disease who expected that information and posi- has not been clear. Tracy (19) noted that Psychosomatic Medicine 54:1-9 (1992)
KAREN A. MATTHEWS, PhD
sex differences in diseases of the heart and arteriosclerotic diseases began to decline around the time of the menopause, especially among whites, and suggested that the menopause may be a risk factor for CHD in women. Nonetheless, it has been difficult to prove that menopause is a risk factor for CHD in epidemiological studies. For example, Colditz et al. (20) reported on data from the Harvard Nurses Study showing that among newer users of hormone replacement therapy, postmenopausal women have about the same risk of CHD as do premenopausal women. In their analyses, they adjusted for both age and cigarette smoking status because these are risk factors for coronary heart disease. Note, however, that there is a natural confounding between menopause and cigarette smoking because women who are smokers have menopause earlier than nonsmokers (for a review, see 17). Cigarette smoking is thought to affect ovarian function by both decreasing production of estrogens and by enhancing rates of metabolism and utilization (21). In any event, by controlling for cigarette smoking, one may be inadvertently controlling for menopausal status. Nonetheless, these types of data led investigators to think that natural menopause may not be an important risk factor for CHD (22). The natural history of CHD would suggest, however, that the menopause should not immediately increase rates of CHD. Rather, estrogen deficiency should accelerate the rates of atherosclerosis, which should eventually be manifested as clinical CHD only years later. If so, menopause should be linked with biological variables that contribute to the development of atherosclerosis in the perimenopausal and postmenopausal years. Epidemiological and experimental tests of this hypothesis
have preoccupied us for the last several years. In the Healthy Women Study, we compared the magnitude of change in biological risk factors from the baseline examination, when all women were premenopausal, to a follow-up examination for a) postmenopausal women who were not using hormone replacement therapy, b) hormone users who had ceased cycling and were using hormone replacement therapy that in combination totaled 12 months, and c) age-matched premenopausal women. Postmenopausal women not using hormone replacement therapy exhibited greater increases in levels of low density lipoprotein (LDL-C), and declines in high density lipoprotein (HDL-C) levels, relative to both hormone users and premenopausal women (23). Internal analyses provided support for the notion that these lipid changes were related to changes in estrogen levels. These analyses showed that women who had the lowest levels of estradiol during perimenopause had the highest level of LDL-C, whereas women with the highest levels of estradiol postmenopause had the highest levels of HLD-C and HDL2 (24). Taken together, these data suggest that lipids are substantially altered during the menopausal period, which may set the stage for later clinical CHD, and that changing levels of estrogens play a key role in determining lipid levels during the menopausal years. Low levels of estrogens due to the menopause may also affect CHD risk by a pathway other than alteration in lipid metabolism. That is, individual differences in cardiovascular and neuroendocrine responses to mental stress have been proposed as a risk factor for CHD; and female reproductive hormones are thought to effect the magnitude of those responses during stress (25). Consistent with this notion Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
are the results of a study by our group (26). We compared age-matched premenopausal and postmenopausal women's responses to several stressors, serial subtraction, star tracer, tilt bed, and public speaking. We reasoned that public speaking would be conceptually relevant to women due to its emphasis on social skills and self-representation, and that that stimulus would elicit the most substantial differences between pre- and postmenopausal women's stress responses. That in fact was the case. Postmenopausal women exhibited larger increases in systolic blood pressure, heart rate, and plasma epinephrine levels during the stress of public speaking than did premenopausal women. Our most recent study compared the stress responses of postmenopausal and premenopausal women and men. Again, participants were asked to deliver a speech. Postmenopausal women exhibited greater increases in diastolic blood pressure and plasma norepinephrine levels during the speech than did premenopausal women. To our surprise, postmenopausal women exhibited larger increases in systolic and diastolic blood pressure than did men. Taken together, these findings suggest that the decline in ovarian hormones associated with menopausal status does lead to an enhancement of cardiovascular and neuroendocrine responses to social stressors during midlife. To the extent that middle-aged women are exposed to such stressful situations and that exaggerated cardiovascular and neuroendocrine responses are a risk factor for CHD, increasingly larger stress responses should accelerate their risk for CHD during the postmenopausal years. Perhaps, however, stress responses associated with estrogen deficiency during the postmenopausal years are relatively unimportant because women may be exPsychosomatic Medicine 54:1-9 (1992)
posed to little social stress. Some evidence suggests that women are exposed to considerable interpersonal stress, perhaps more so than men. Kessler and McLeod (27) collapsed the data from five epidemiological investigations of the prevalence of life events in 6919 women and men and compared the annual rates of life events categorized as income-related, divorce/separation, other love loss, death of a loved one, ill health, and negative events to someone in their network. Results showed that employed women and homemakers more often reported experiencing death of a loved one and negative events to others in their network than did men. Furthermore, another study by Bolger and colleagues (28) showed that certain types of interpersonal stress leads to more negative mood states among women than men. Married couples daily completed ratings of minor life events and negative mood states. Regression analyses demonstrated that the strongest correlate of negative mood states were arguments with others. More importantly, seven out of 10 comparisons between men and women showed that the effects of minor life events on mood were stronger among women than men, with arguments with spouse and with multiple others significantly more strongly associated with negative mood among women than men. This suggests that the covariation between interpersonal stress and negative mood might be stronger in women than in men. Figure 1 graphically shows how these findings might be interconnected. Perhaps menopausal women are as likely as or more likely than similarly aged men to be exposed to frequent interpersonal stress (arrow "a"), to which they tend to respond with more negative mood states (arrow "c"). In response to heightened social stress, they should be likely to exhibit
KAREN A. MATTHEWS, PhD Gender/ Reproductive Hormones
Exposure to Interpersonal Stress
Negative Mood
Cardiovascular and Neuroendocrine Fig. 1. Hypothesized relationships among reproductive hormones, interpersonal stress, negative mood states, and physiological responses.
enhanced cardiovascular and neuroendocrine responses (arrow "b"], relatively to their premenopausal counterparts. This pattern of responses should accelerate their rates of atherosclerosis and place them at increased risk of clinical CHD in their later years. SUMMARY AND CONCLUSIONS
that holding negative expectations about the menopause affects the quality of the menopausal experience. Indeed, that appears to be the case, perhaps because myths can function as self-fulfilling prophecy. The third myth is that there are no important changes that occur during the menopause. That is incorrect. Estrogen deficiency during the menopause sets the stage for substantial changes in risk for CHD, which becomes clinically apparent later in life. We discussed how estrogen deficiency may influence both lipids and lipoprotein levels and the magnitude of neuroendocrine and cardiovascular respond to mental stress. That latter pathway is of particular interest because middle-aged women may be exposed more often to interpersonal stress and may respond more emotionally to it, relative to men, suggesting a potential interactive effect of the decline in reproductive hormones and co-occurring social and psychological changes during the menopausal period. This discussion of the myths and realities of the menopause has deliberately not been exhaustive. Rather, it has been intended as an illustration of how a psychosomatic approach to the menopausal transition is especially useful for understanding basic social, psychological, and biological processes in midlife. It also demonstrates the complexity of the interplay of environmental, behavioral, and biological factors in determining the quality of women's lives.
Menopause is a reproductive milestone in a woman's life around which many different myths have developed. We reviewed three sets of myths that middleaged premenopausal women hold and evaluated those myths according to scientific data from our own work and that of others. First, middle-aged women expect to experience depression, irritability, and vasomotor symptoms during the menoThis work was supported by grants HL pause. It appears that the vast majority of 28266 and HL 38712 from the National postmenopausal women do not experi- Institutes of Health. ence depression, but do experience vasoThe author gratefully acknowledges the motor symptoms that are uncomfortable and may have secondary effects on psy- collaboration of Drs. Lewis Kuller, Rena chological well being, especially during Wing, and Elaine Meilahn of the Healthy Women Study, on which much of this pathe perimenopause. Second, middle-aged women believe per is based.
Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
REFERENCES 1. MacDonald PC: Estrogen plus progestin in postmenopausal women—Act II. N Engl J Med 313:959-961, 1985 2. Utian WH: Menopause in Modern Perspective: A Guide to Clinical Practice. New York, AppletonCentury-Crofts, 1980 3. Amundsen DW, Diers CJ: Age of menopause in medieval Europe. Hum Biol 45:605, 1973 4. McKinlay SM, Bifano NL, McKinlay JB: Smoking and age at menopause in women. Ann Intern Med 103:350-356, 1985 5. Leake J: Chronic or Slow Diseases Peculiar to Women. London, Baldwin, 1777 6. Burns, J: Diseases of Women and Children. London, Longman, 1814 7. Swartzman LC, Leiblum SR: Changing perspectives on the menopause. In J Psychosom Obstet Gynecol 6:11-24, 1987 8. Deutsch H- Psychology of Women: A Psychoanalytic Interpretation, Vol 11. Motherhood. New York, Grune and Stratton, 1945, 456-491 9. Benedek T: Climacterium: A developmental phase. Psychoanal Q 19:1-27, 1950 10. Wilson RA, Wilson TA: The fate of the nontreated postmenopausal woman. A plea for the maintenance of adequate estrogen from puberty to the grave. J Am Geriatr Soc 11:347, 1963 11. Wilson RA, Brevetti RE, Wilson TA: Specific procedures for the elimination of the menopause. West J Surg Obstet Gynecol 7:110, 1963 12. Matthews KA, Kelsey SF, Meilahn EN, Kuller LH, Wing RR: Educational attainment and behavioral and biologic risk factors for coronary heart disease in middle-aged women. Am J Epidemiol 129:1132-1144, 1989 13. Neugarten BL, Wood V, Kraines RJ, Loomis B: Women's attitudes toward the menopause. Vita Hum 6:140-151, 1963 14. Matthews KA, Wing RR, Kuller LH, Meilahn EN, Kelsey SF, Costello EJ, Caggiula AW: Influences of natural menopause on psychological characteristics and symptoms of middle-aged healthy women. J Consult Clin Psychol 58:345-351, 1990 15. Hallstrom T, Samuelsson S: Mental health in the climacteric: The longitudinal study of women in Gothenburg. Acta Obstet Gynecol Scand 130(Suppl):13-18, 1985 16. McKinlay JB, McKinlay SM, Brambilla DJ: Health status and utilization behavior associated with menopause. Am J Epidemiol 125:110-121, 1987 17. Matthews KA, Bromberger J, Egeland G: Behavioral antecedents and consequences of the menopause. In SG Korenman (ed), The Menopause. Norwell, MA, Serono Symposia, 1990, 1-15 18. Campbell S, Whitehead M: Oestrogen therapy and the menopausal syndrome. Clin Obstet Gynecol 4:31-47, 1977 19. Tracy RE: Sex differences in coronary disease: Two opposing views. J Chronic Dis 19:1245-1251,1966 20. Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH: Menopause and the risk of coronary heart disease in women. N Engl J Med 316:1105-1110 21. Surgeon General: The health consequences of smoking. Nicotine addiction. Washington, DC. US Government Printing Office, 1988 22. Eaker ED, Packard B, Wenger NK, Clarkson TB, Tyroler HA (eds): Coronary Heart Disease in Women. New York, Haymarket Doyma, 1987 23. Matthews KA, Meilahn E, Kuller LH, Kelsey SF, Caggiula AW, Wing RR: Menopause and risk factors for coronary heart disease. N Engl J Med 321:641-646, 1989 24. Kuller LH, Gutai )P, Meilahn E, Matthews KA, Plantinga P: Relationship of endogenous sex steroid hormones to lipids and apoproteins in postmenopausal women. Arteriosclerosis 10:1058-1066, 1990 25. Matthews KA: Interactive effects of behavior and reproductive hormones on sex differences in risk for coronary heart disease. Health Psychol 8:373-387, 1989 26. Saab PG, Matthews KA, Stoney CM, McDonald RH: Premenopausal and postmenopausal women differ in their cardiovascular and neuroendocrine responses to behavioral stressors. Psychophysiology 26:270280, 1989 27. Kessler R, McLeod JD: Sex differences in vulnerability to undesirable life events. Am Sociol Rev 49:620631, 1984 28. Bolger N, DeLongis A, Kessler RC, Schilling EA: Effects of daily stress on negative mood. J Pers Soc Psychol 57:808-818, 1989
Psychosomatic Medicine 54:1-9 (1992)
Menopause is a reproductive milestone in a woman's life (1). To paraphrase MacDonald (1), on the one hand, it marks the end of reproductive life, with unfulfilled dreams of childbearing for some, and the beginning of increased risk for the health consequences of estrogen deprivation. On the other hand, it signifies a time when women no longer need to experience the inconvenience associated with menses, when they are free from dysmenhorrhea, and when they need not fear unwanted pregnancies. Menopause occurs in midlife at a time when many women are experiencing changes in roles, responsibilities, and relationships that accompany aging generally and the maturation of children and their departure from home in particular. These changes may create considerable stress for some women, affecting their identity, self-esteem, and social and family relationships. For others, these might mark the beginning of more fulfilling re-
Presented at the Annual Meeting of the American Psychosomatic Society, Sante Fe, New Mexico, March 1991. From the University of Pittsburgh, School of Medicine, Pittsburgh, Pennsylvania. Address reprint requests to: Karen A. Matthews, Ph.D., Department of Psychiatry, University of Pittsburgh, 3811 O'Hara Street, Pittsburgh, PA 15213. Received for publication August 7, 1991; revision received August 30,1991
Psychosomatic Medicine 54:1-9 (1992) 0033-3174/92/5401-000l$03.00/0 Copyright © 1992 by the American Psychosomatic Society
lationships and new challenges for positive psychological growth. Thus, menopause is a critical transition in women's lives not only because of the biological changes they are experiencing, but also because of the co-occurring social and psychological changes. Understanding the independent and interactive effects of the social, psychological, and biological changes during the menopause is crucial for understanding midlife development and provides the basis for understanding later adult health and dysfunction. Furthermore, understanding the menopausal transition can provide a window on basic social, psychological, and biological processes in the life course of a woman. Stated differently, a psychosomatic approach to the menopause has substantial heuristic and practical value. This paper examines the basic biological changes that occur during the menopausal years and how those impact, both alone and in interaction with social and psychological factors, on the quality of women's lives and on their risk for coronary heart disease (CHD). The discussion is organized according to the myths that society has held and continues to hold with regard to the meaning of the menopause. As will be noted, myths about the menopause have been remarkably negative.
KAREN A. MATTHEWS, PhD
MYTHS ABOUT THE MENOPAUSE
Historical Perspective The menopause was referred to in early cultures and in many texts (2). By the sixth century A.D., the cessation of menstruation was well documented to occur around the age of 50 (3) and that average age has not changed (4), despite the facts that the onset of menses has shifted to an earlier age, and that women live longer now than ever before. In fact, women now live about one third of their adult years beyond the menopause. Some observers recognized that the risk for chronic diseases increased after the menopause (2). In 1777, John Leake in his book, Chronic or Slow Diseases Peculiar to Women, noted that: "At this criticaJ time of life, the female sex are often visited with various diseases of the chronic kind" (5). In 1814, John Burns indicated that "the cessation of menses does of itself seem, in some cases, to excite cancer of the breast" (6). Some observers also indicated that psychological complaints and symptoms increased during the menopausal transition (2). Again, John Leake noted that "some are subject to pain and giddiness of the head, hysteric disorders, colic pains, and a female weakness . . . often very troublesome to others" due to "the many excesses introduced by luxury and the irregularities of the passion" (5). This negative expectation regarding psychological changes during the menopause has persisted into the 20th century (7). Two female psychoanalysts, Deutsch and Benedek, writing in the midcentury, believed that psychological symptoms at the time of the menopause were grief reactions to the loss of menstruation and to all it represents (8, 9). Deutsch noted
that "woman has ended her existence as bearer of future life,. ..—as servant of the species. She is now engaged in an active struggle against her decline" (8, p. 459). Nonetheless, Benedek saw opportunities for growth in the process of coping with the menopause, in that it can result in the "emancipation from sexual competition and from the fear of sexual rejection (which) often releases talents and qualities unsuspected before" (9, p. 23). Contemporary Perspective More recent writers, most notably Wilson who has championed hormone replacement therapy, have expressed negative attitudes toward the menopause: "A large percentage of women . . . acquire a vapid cowlike feeling called a 'negative state.' It is a strange endogenous misery . . . the world appears as though through a grey veil, and they live as docile, harmless creatures missing most of life's values" (10, p. 347). "The menopausal woman is not normal; she suffers from a deficiency disease with serious sequelae and needs treatment" (11, p. 110). Despite these views expressed throughout the centuries as well as in more recent writings, many women currently have some positive expectations and beliefs about the menopausal experience, especially for themselves. We have been studying how the menopause affects women's mental and physical health in a sample of 541 women, who at the beginning of the study, were premenopausal, relatively healthy, and ages 42 to 50 years old (12). At entry into this study, The Healthy Women Study, women were asked about their attitudes toward the menopause and their expectations about its psychological and somatic effects on Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
women, using an adaptation of a measure developed by Neugarten and colleagues (13). After completion of the Neugarten measure, the women were asked about the same expectations, but with regard to their own menopausal experience. About 60% agreed with the statement, "Going through the menopause does not really change a woman in any important way," whereas 79% agreed with the statement, "Going through the menopause will not really change me in any important way." Furthermore, almost all women (94%) believed that if they knew what to expect, and did not expect trouble during the menopause, they would do well. Despite these positive attitudes, a substantial number of the women did have rather negative expectations about some effects of the menopause. About 80% thought women are liable to get depressed during the menopause, and 55% thought they themselves would be liable to become depressed then. About % of the sample thought women in general have hot flashes during the menopause, and 70% thought they would have these. Sixtyeight percent agreed that "During the menopause, women are liable to fly off the handle more easily than at other times"; and 48% thought that "During the menopause, I am liable to fly off the handle more easily than at other times." Are these expectations myths or realities? Often expectations that develop throughout the centuries are based on reality, but sometimes expectations can act like self-fulfilling prophesies and determine the reality that society accepts. Perhaps these expectations are true for a small subset of women, but because these occurrences are extremely salient and negative, society overgeneralizes their experience to all women. In the next sections, we will examine the extent to Psychosomatic Medicine 54:1-9 (1992)
which each of the above beliefs about the menopause was supported in the Healthy Women Study. Do Women Experience Psychological and Vasomotor Symptoms During the Menopause? To address this question, women's selfreported psychological characteristics and symptoms at entry into the Healthy Women Study when all women were premenopausal were compared to those same characteristics reported after the women became postmenopausal. Study protocol was that all 541 women were evaluated extensively when they were premenopausal at study entry. This examination was repeated when women became postmenopausal, i.e., had ceased menstruating for 12 months and had elevated follicular stimulating hormone (FSH) levels. For each postmenopausal woman, a premenopausal control, i.e., who had menstruated within the previous 3 months and was not taking hormone replacement therapy, was also re-evaluated at the same time. Analyses compared the psychological characteristics and symptoms of the first 69 postmenopausal women with that of their premenopausal controls. Changes that occurred in both groups were interpreted as being due to aging, whereas changes that were different in the menopausal group were interpreted as being due to aging plus change to postmenopausal status. Our results showed women who became postmenopausal and their premenopausal controls reported the same increase in depressive symptoms on the Beck Depression Inventory and the same increase in the total number of symptoms occurring during the previous 2 weeks
KAREN A. MATTHEWS, PhD
(14). However, postmenopausal women did report experiencing more frequently a subset of symptoms at the postmenopausal examination, relative to the premenopausal examination. More specifically, postmenopausal women reported more hot flashes, joint pain, and troubles sleeping. Despite experiencing more vasomotor symptoms, postmenopausal women perceived lower levels of stress in their daily lives at the postmenopausal examination, relative to the premenopausal examination (14). Taken together, these findings suggest that natural menopause does not lead to enhanced psychological distress, even though it does lead to an increase in vasomotor symptoms. Indeed, these findings are consistent with the results of two other recent longitudinal studies of the menopause (15, 16), as well as cross-sectional studies of large samples of women (see 17 for a review). Although the complete cessation of menses does not cause adverse changes for the majority of women, perhaps the initial cessation of menses does. The perimenopause might be often accompanied by the first occurrences of uncomfortable vasomotor symptoms, but as women learn to cope with them, they might experience acute adverse change in depressive symptoms that do not persist. Indeed, that appears to be the case among the first 105 women who ceased cycling for 3 months. Perimenopausal women increased significantly, albeit slightly in their Beck Depression scores, relatively to women who remained premenopausal. The number of women who had Beck Depression Inventory scores 9 or above increased sixfold over the course of the follow-up period. Note, however, that this six-fold increase represented only about 10% of the perimenopausal women. More striking is the fact that many
women reported at their perimenopausal examination experiencing a variety of disturbing symptoms: There was a tenfold increase in the number of women reporting hot flashes and a three-fold increase in the number of women reporting joint pain, being forgetful, worrying about their body, being constipated, and having dizzy spells. No woman reported experiencing cold sweats at her premenopausal examination only, whereas 14 women reported them at their perimenopausal examination (and hot at the premenopausal exam). Our hypothesis that the psychological symptoms that do occur during the early perimenopause are related to the experience of having vasomotor symptoms for the first time are supported elsewhere. Campbell and Whitehead (18) demonstrated that 4 months of conjugated estrogen therapy during the perimenopause led to a dramatic improvement in hot flashes, vaginal dryness, insomnia, irritability, and anxiety, compared to a placebo condition. They noted that these symptoms were interrelated and that the benefit was initiated by improvement in hot flashes. One reason for their latter conclusion is that of nine symptoms that improved with hormone replacement therapy, five of them did not improve unless the patient had hot flashes at outset. Women who did not have hot flashes at outset failed to show improvement in the more psychological symptoms, like irritability, optimism, and good spirits. In sum, the myth that women experience negative symptoms during the menopause has a grain of truth. Women do experience uncomfortable physical symptoms during the perimenopause, some of which are relieved by the time of the complete cessation of menses. Even so, the vast majority of women do not bePsychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
come depressed either during the perimenopause or afterwards.
tive expectations about the menopause would facilitate a good experience during the menopause had lower levels of depression and fewer symptoms at their If Women Expect Trouble During the postmenopausal evaluation, relative to Menopause, Will They Have It? their premenopausal evaluation. Thus, it does appear that women's expectations Do expectations about the menopause about aspects of the menopause affect affect the quality of the menopausal ex- their psychological experience during the perience? To examine this question, we menopause. first factor analyzed ratings of the attiWe can suggest at least two possible tudes toward menopause that the 541 explanations of these results. First, women reported at their baseline exami- women at baseline who expected adverse nation. This analysis led to the develop- symptoms during the menopause may ment of five scales, three of which are have already begun to experience them relevant here: a) expectations of vasomo- at that time and merely were accurate tor symptoms; b) expectations of benefits reporters. This explanation, however, of the menopause, e.g., no more worry seems incompatible with the fact that we about unwanted pregnancy; and c) expec- statistically controlled for the baseline tations that information and positive ex- levels of symptoms. Or, second, women pectancies will help during the meno- who expect negative consequences of the pause. We correlated these scales meas- menopause may be behaving in such a ured at study entry with the changes from way that at least in part contributes to the study entry to postmenopausal examina- development or maintenance of symption in Beck Depression Inventory scores, toms. For example, women who expect total number of symptoms experienced in adverse vasomotor symptoms, when they the previous 2 weeks, Spielberger Trait are encountered, may become depressed Anger Scale scores, and level of social at the prospect of facing the inconvenisupport. The change scores were adjusted ence and discomfort of those symptoms, for the baseline premenopausal level of rather than taking positive action in copthe characteristic in question. ing with them. Stated differently, expectThese analyses showed several intrigu- ing negative menopausal symptoms may ing findings: a) Those women who ex- lead to a self-fulfilling prophesy. pected to experience vasomotor symptoms during the menopause did increase in their level of depressive symptoms and Are There No Important Changes angry feelings at their postmenopausal During the Menopause? evaluation, relative to their premenopausal evaluation, b) Those women who exAs noted above, it has long been recogpected the menopause to have some ben- nized that postmenopausal women have efits did indeed report that their levels of more chronic disease than do premenosocial support increased at their postmen- pausal women, but the role of declining opausal evaluation, relative to their pre- reproductive hormones in contributing to menopausal evaluation, c) Those women increasing prevalence of chronic disease who expected that information and posi- has not been clear. Tracy (19) noted that Psychosomatic Medicine 54:1-9 (1992)
KAREN A. MATTHEWS, PhD
sex differences in diseases of the heart and arteriosclerotic diseases began to decline around the time of the menopause, especially among whites, and suggested that the menopause may be a risk factor for CHD in women. Nonetheless, it has been difficult to prove that menopause is a risk factor for CHD in epidemiological studies. For example, Colditz et al. (20) reported on data from the Harvard Nurses Study showing that among newer users of hormone replacement therapy, postmenopausal women have about the same risk of CHD as do premenopausal women. In their analyses, they adjusted for both age and cigarette smoking status because these are risk factors for coronary heart disease. Note, however, that there is a natural confounding between menopause and cigarette smoking because women who are smokers have menopause earlier than nonsmokers (for a review, see 17). Cigarette smoking is thought to affect ovarian function by both decreasing production of estrogens and by enhancing rates of metabolism and utilization (21). In any event, by controlling for cigarette smoking, one may be inadvertently controlling for menopausal status. Nonetheless, these types of data led investigators to think that natural menopause may not be an important risk factor for CHD (22). The natural history of CHD would suggest, however, that the menopause should not immediately increase rates of CHD. Rather, estrogen deficiency should accelerate the rates of atherosclerosis, which should eventually be manifested as clinical CHD only years later. If so, menopause should be linked with biological variables that contribute to the development of atherosclerosis in the perimenopausal and postmenopausal years. Epidemiological and experimental tests of this hypothesis
have preoccupied us for the last several years. In the Healthy Women Study, we compared the magnitude of change in biological risk factors from the baseline examination, when all women were premenopausal, to a follow-up examination for a) postmenopausal women who were not using hormone replacement therapy, b) hormone users who had ceased cycling and were using hormone replacement therapy that in combination totaled 12 months, and c) age-matched premenopausal women. Postmenopausal women not using hormone replacement therapy exhibited greater increases in levels of low density lipoprotein (LDL-C), and declines in high density lipoprotein (HDL-C) levels, relative to both hormone users and premenopausal women (23). Internal analyses provided support for the notion that these lipid changes were related to changes in estrogen levels. These analyses showed that women who had the lowest levels of estradiol during perimenopause had the highest level of LDL-C, whereas women with the highest levels of estradiol postmenopause had the highest levels of HLD-C and HDL2 (24). Taken together, these data suggest that lipids are substantially altered during the menopausal period, which may set the stage for later clinical CHD, and that changing levels of estrogens play a key role in determining lipid levels during the menopausal years. Low levels of estrogens due to the menopause may also affect CHD risk by a pathway other than alteration in lipid metabolism. That is, individual differences in cardiovascular and neuroendocrine responses to mental stress have been proposed as a risk factor for CHD; and female reproductive hormones are thought to effect the magnitude of those responses during stress (25). Consistent with this notion Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
are the results of a study by our group (26). We compared age-matched premenopausal and postmenopausal women's responses to several stressors, serial subtraction, star tracer, tilt bed, and public speaking. We reasoned that public speaking would be conceptually relevant to women due to its emphasis on social skills and self-representation, and that that stimulus would elicit the most substantial differences between pre- and postmenopausal women's stress responses. That in fact was the case. Postmenopausal women exhibited larger increases in systolic blood pressure, heart rate, and plasma epinephrine levels during the stress of public speaking than did premenopausal women. Our most recent study compared the stress responses of postmenopausal and premenopausal women and men. Again, participants were asked to deliver a speech. Postmenopausal women exhibited greater increases in diastolic blood pressure and plasma norepinephrine levels during the speech than did premenopausal women. To our surprise, postmenopausal women exhibited larger increases in systolic and diastolic blood pressure than did men. Taken together, these findings suggest that the decline in ovarian hormones associated with menopausal status does lead to an enhancement of cardiovascular and neuroendocrine responses to social stressors during midlife. To the extent that middle-aged women are exposed to such stressful situations and that exaggerated cardiovascular and neuroendocrine responses are a risk factor for CHD, increasingly larger stress responses should accelerate their risk for CHD during the postmenopausal years. Perhaps, however, stress responses associated with estrogen deficiency during the postmenopausal years are relatively unimportant because women may be exPsychosomatic Medicine 54:1-9 (1992)
posed to little social stress. Some evidence suggests that women are exposed to considerable interpersonal stress, perhaps more so than men. Kessler and McLeod (27) collapsed the data from five epidemiological investigations of the prevalence of life events in 6919 women and men and compared the annual rates of life events categorized as income-related, divorce/separation, other love loss, death of a loved one, ill health, and negative events to someone in their network. Results showed that employed women and homemakers more often reported experiencing death of a loved one and negative events to others in their network than did men. Furthermore, another study by Bolger and colleagues (28) showed that certain types of interpersonal stress leads to more negative mood states among women than men. Married couples daily completed ratings of minor life events and negative mood states. Regression analyses demonstrated that the strongest correlate of negative mood states were arguments with others. More importantly, seven out of 10 comparisons between men and women showed that the effects of minor life events on mood were stronger among women than men, with arguments with spouse and with multiple others significantly more strongly associated with negative mood among women than men. This suggests that the covariation between interpersonal stress and negative mood might be stronger in women than in men. Figure 1 graphically shows how these findings might be interconnected. Perhaps menopausal women are as likely as or more likely than similarly aged men to be exposed to frequent interpersonal stress (arrow "a"), to which they tend to respond with more negative mood states (arrow "c"). In response to heightened social stress, they should be likely to exhibit
KAREN A. MATTHEWS, PhD Gender/ Reproductive Hormones
Exposure to Interpersonal Stress
Negative Mood
Cardiovascular and Neuroendocrine Fig. 1. Hypothesized relationships among reproductive hormones, interpersonal stress, negative mood states, and physiological responses.
enhanced cardiovascular and neuroendocrine responses (arrow "b"], relatively to their premenopausal counterparts. This pattern of responses should accelerate their rates of atherosclerosis and place them at increased risk of clinical CHD in their later years. SUMMARY AND CONCLUSIONS
that holding negative expectations about the menopause affects the quality of the menopausal experience. Indeed, that appears to be the case, perhaps because myths can function as self-fulfilling prophecy. The third myth is that there are no important changes that occur during the menopause. That is incorrect. Estrogen deficiency during the menopause sets the stage for substantial changes in risk for CHD, which becomes clinically apparent later in life. We discussed how estrogen deficiency may influence both lipids and lipoprotein levels and the magnitude of neuroendocrine and cardiovascular respond to mental stress. That latter pathway is of particular interest because middle-aged women may be exposed more often to interpersonal stress and may respond more emotionally to it, relative to men, suggesting a potential interactive effect of the decline in reproductive hormones and co-occurring social and psychological changes during the menopausal period. This discussion of the myths and realities of the menopause has deliberately not been exhaustive. Rather, it has been intended as an illustration of how a psychosomatic approach to the menopausal transition is especially useful for understanding basic social, psychological, and biological processes in midlife. It also demonstrates the complexity of the interplay of environmental, behavioral, and biological factors in determining the quality of women's lives.
Menopause is a reproductive milestone in a woman's life around which many different myths have developed. We reviewed three sets of myths that middleaged premenopausal women hold and evaluated those myths according to scientific data from our own work and that of others. First, middle-aged women expect to experience depression, irritability, and vasomotor symptoms during the menoThis work was supported by grants HL pause. It appears that the vast majority of 28266 and HL 38712 from the National postmenopausal women do not experi- Institutes of Health. ence depression, but do experience vasoThe author gratefully acknowledges the motor symptoms that are uncomfortable and may have secondary effects on psy- collaboration of Drs. Lewis Kuller, Rena chological well being, especially during Wing, and Elaine Meilahn of the Healthy Women Study, on which much of this pathe perimenopause. Second, middle-aged women believe per is based.
Psychosomatic Medicine 54:1-9 (1992)
PRESIDENTIAL ADDRESS
REFERENCES 1. MacDonald PC: Estrogen plus progestin in postmenopausal women—Act II. N Engl J Med 313:959-961, 1985 2. Utian WH: Menopause in Modern Perspective: A Guide to Clinical Practice. New York, AppletonCentury-Crofts, 1980 3. Amundsen DW, Diers CJ: Age of menopause in medieval Europe. Hum Biol 45:605, 1973 4. McKinlay SM, Bifano NL, McKinlay JB: Smoking and age at menopause in women. Ann Intern Med 103:350-356, 1985 5. Leake J: Chronic or Slow Diseases Peculiar to Women. London, Baldwin, 1777 6. Burns, J: Diseases of Women and Children. London, Longman, 1814 7. Swartzman LC, Leiblum SR: Changing perspectives on the menopause. In J Psychosom Obstet Gynecol 6:11-24, 1987 8. Deutsch H- Psychology of Women: A Psychoanalytic Interpretation, Vol 11. Motherhood. New York, Grune and Stratton, 1945, 456-491 9. Benedek T: Climacterium: A developmental phase. Psychoanal Q 19:1-27, 1950 10. Wilson RA, Wilson TA: The fate of the nontreated postmenopausal woman. A plea for the maintenance of adequate estrogen from puberty to the grave. J Am Geriatr Soc 11:347, 1963 11. Wilson RA, Brevetti RE, Wilson TA: Specific procedures for the elimination of the menopause. West J Surg Obstet Gynecol 7:110, 1963 12. Matthews KA, Kelsey SF, Meilahn EN, Kuller LH, Wing RR: Educational attainment and behavioral and biologic risk factors for coronary heart disease in middle-aged women. Am J Epidemiol 129:1132-1144, 1989 13. Neugarten BL, Wood V, Kraines RJ, Loomis B: Women's attitudes toward the menopause. Vita Hum 6:140-151, 1963 14. Matthews KA, Wing RR, Kuller LH, Meilahn EN, Kelsey SF, Costello EJ, Caggiula AW: Influences of natural menopause on psychological characteristics and symptoms of middle-aged healthy women. J Consult Clin Psychol 58:345-351, 1990 15. Hallstrom T, Samuelsson S: Mental health in the climacteric: The longitudinal study of women in Gothenburg. Acta Obstet Gynecol Scand 130(Suppl):13-18, 1985 16. McKinlay JB, McKinlay SM, Brambilla DJ: Health status and utilization behavior associated with menopause. Am J Epidemiol 125:110-121, 1987 17. Matthews KA, Bromberger J, Egeland G: Behavioral antecedents and consequences of the menopause. In SG Korenman (ed), The Menopause. Norwell, MA, Serono Symposia, 1990, 1-15 18. Campbell S, Whitehead M: Oestrogen therapy and the menopausal syndrome. Clin Obstet Gynecol 4:31-47, 1977 19. Tracy RE: Sex differences in coronary disease: Two opposing views. J Chronic Dis 19:1245-1251,1966 20. Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH: Menopause and the risk of coronary heart disease in women. N Engl J Med 316:1105-1110 21. Surgeon General: The health consequences of smoking. Nicotine addiction. Washington, DC. US Government Printing Office, 1988 22. Eaker ED, Packard B, Wenger NK, Clarkson TB, Tyroler HA (eds): Coronary Heart Disease in Women. New York, Haymarket Doyma, 1987 23. Matthews KA, Meilahn E, Kuller LH, Kelsey SF, Caggiula AW, Wing RR: Menopause and risk factors for coronary heart disease. N Engl J Med 321:641-646, 1989 24. Kuller LH, Gutai )P, Meilahn E, Matthews KA, Plantinga P: Relationship of endogenous sex steroid hormones to lipids and apoproteins in postmenopausal women. Arteriosclerosis 10:1058-1066, 1990 25. Matthews KA: Interactive effects of behavior and reproductive hormones on sex differences in risk for coronary heart disease. Health Psychol 8:373-387, 1989 26. Saab PG, Matthews KA, Stoney CM, McDonald RH: Premenopausal and postmenopausal women differ in their cardiovascular and neuroendocrine responses to behavioral stressors. Psychophysiology 26:270280, 1989 27. Kessler R, McLeod JD: Sex differences in vulnerability to undesirable life events. Am Sociol Rev 49:620631, 1984 28. Bolger N, DeLongis A, Kessler RC, Schilling EA: Effects of daily stress on negative mood. J Pers Soc Psychol 57:808-818, 1989
Psychosomatic Medicine 54:1-9 (1992)
