By Kristina Orth-Gomér, MD, PhD
Although acute myocardial infarction and unstable angina are almost as common in women as in men, women get their first manifestations about 10 years later in life than men do. Before age 50, coronary disease seldom occurs in women, but by age 70, it is almost as common as in men.1 The reason for this age discrepancy long has been thought to be related to female sex hormones; however, the only two randomized, placebo-controlled trials of hormone replacement therapy (HRT) published to date have failed to confirm a protective effect of estrogen against recurrent coronary events.2
Social and psychological influences may help to explain why observational, but not placebo-controlled studies have noted a benefit in cardiovascular disease risk among HRT users. Women with higher socioeconomic status (SES) are prescribed HRT more often than women in lower strata. Women in higher SES also are less likely to get coronary disease and they have a better prognosis once they get the disease (in general, women with coronary disease, especially those younger than age 70, have a poorer prognosis than do men). Studies of psychosocial and behavioral factors in women are increasingly important in understanding these complex associations. In men, psychological risk factors for coronary disease (including hostility, depression, and personality type) have been identified. Social factors, including job stress, lack of social support, and low SES have been investigated thoroughly and identified as risk factors.3 Among women, psychosocial stress in relation to coronary disease rarely has been studied.
We initiated the Stockholm Female Coronary Risk (FemCorRisk) Study both to explore gender-related models of psychosocial risk and to integrate known psychosocial and physiological risk factors for coronary disease in women. In the initial phase, we investigated several psychosocial risk factors that have been established in men, and found that low socioeconomic position was linked in many ways to coronary disease in women. Compared to women who had graduated from college, women with less education had a coronary disease risk twice as high and women with low-status occupations had a risk four times as high.4 Smoking, poor diet, and sedentary lifestyle were more frequent in low SES women, but the largest proportion of social gradient was ascribed to psychosocial stress factors. Stress at work (the ratio between high work demand and low work control) was an important factor; lack of social support, depressed mood, and poor capacity for mastery and coping with hardships also were more common in low SES women. When controlled for education and occupation, psychosocial factors alone still increased risk significantly and worsened prognosis of coronary disease in women.5 The cluster of depression, social disadvantage, work stress, and social isolation increases the risk of developing coronary disease and worsens prognosis. These findings parallel those from studies of men.3
Because women’s lives are different from men’s in many respects, we hypothesized that women might have unique stressors in their daily lives. Therefore, we conducted a structured comprehensive interview about chronic stressful influences among the nearly 600 women of the FemCorRisk Study. The interview covered topics ranging from education, work life, family situation, and relationships to opportunities for personal recreation and development. The methodology and psychometrics of the interview procedure have been described elsewhere.6 The validity of the interview as a measure of stress exposure in retrospective analyses can be questioned; women who have had an acute heart attack may be more likely than healthy women to report stress and to ascribe their heart disease to an external stressor, such as their work situation. However, the interview method was validated by comparing it to the widely used, standardized scales of stress experiences from the work and social environment.7 High work demand and low control characterize a stressful work situation. Lack of social ties, insufficient social networks, and inadequate social support are characteristics of a stressful social situation and are known to have negative effects on health. A high work-stress ratio predicts coronary risk in both men and women.7 The interview method had an almost identical predictive power as the standardized scale. Therefore, we assumed that the other parts of the interview procedure might have similar validity and psychometric reliability as the work stress scales.6
In addition to work stress, we found that emotional stress, originating from the relationship with a spouse or cohabitant, was a significant risk factor in the Stockholm women. Those women who had experienced serious, longstanding problems in their relationships had both a higher risk of developing coronary disease and a worse prognosis once they were sick. Marital stress nearly tripled the risk of developing coronary disease; the multivariable hazard ratio for cases vs. controls was 2.9 (95% confidence interval [CI] 1.5-5.5). Similarly, the hazard ratio for a recurrent event due to marital stress during five years of follow-up was 2.9 (95% CI 1.3-6.5), while that due to work stress was 1.6 (95% CI 0.8-3.3). Controlling for standard risk factors and clinical prognostic markers, including heart failure (ejection fraction), diabetes, hypertension, dyslipidemia, and smoking, only marginally affected the risk.
The Stockholm Marital Stress Scale (SMSS), a validated scale developed in our research laboratory, addressed the quality of the emotional, social, and sexual relationship with the spouse or cohabitant. The questions focused on the quality of the relationship, whether it was loving and friendly or problematic and filled with crises; whether the spouse was also a confidante and supporter; whether the couple engaged in activities outside the home together; and whether the woman had opportunities to pursue her own activities. The specific marital problems (ranging from violence, infidelity, or substance abuse to a spouse’s chronic illness) that the women described were substantial and concrete and unlikely to be subject to serious reporting bias due to the experience of life-threatening disease.8
During the course of these interviews, participants often asked us why rehabilitative and supportive interventions and opportunities were not available for women as they were for men. In fact, only half of women patients invited to a rehabilitation program participate, compared to 80% of the men.6 Confronted with these figures, many of the women said that they couldn’t adhere to physical training or other types of group-based programs, which were designed for and involved mostly men. Women who had tried these programs felt that they too easily became the advisors or confidantes of the male patients, but that there were fewer of the same resources available to them.
In our research on life stress in women, conducted since the early 1990s, we have found it useful to apply a simple integrative model of pathways and mechanisms to the effects of various stressors on disease. This model assumes that environmental, social, or physical stressors, mediated by physiological adaptation mechanisms, evoke behaviors and/or emotions that amplify the disease process. Depression and anger are common reactions to life. In a cross-sectional analysis of emotional dimensions within a subgroup of these women in FemCorRisk, depressive symptoms were related both to the experience of work stress and marital stress. In an analysis of women who were both working full time and married or cohabiting at the time of the investigation, women with coronary disease and healthy women were analyzed separately for emotional responses to their work and marital situations. Standardized depression and anger scales were applied. In both groups, depressive reactions significantly were related to the experience of longstanding marital stress in a graded fashion. Women with both coronary disease and severe marital stress averaged four depressive symptoms (on a standardized scale9). Healthy women with low levels of marital stress reported the fewest (< 1) depressive symptoms. Depressive reactions were not significantly associated with work stress in either group.
In all groups, marital stress was more closely and directly associated with depression than was work stress. This finding is most important because practically all women in the study group worked; only two of 600 women characterized themselves as housewives. This concurs with national statistics about the Swedish labor market, i.e., around 85% of both sexes are gainfully employed. We also investigated the role of lifestyle factors, including smoking, diet, and exercise habits, in relation to marital stress; it is conceivable that stressed women might smoke more, eat less healthy diets, and exercise less. However, there were no, or only weak, associations between marital stress and health habits.
In summary, women’s marital stress exposure should be considered a serious and potentially hazardous risk factor, both for development and recurrence of coronary disease. Although most women worked, work stress was not a strong coronary health hazard. Standard behavioral risk factors did not explain the association between marital stress and disease outcome. With the cautions inherent in conclusions based on cross-sectional analyses, it appears that depressive symptoms are the primary factors for coronary disease development and progression.
During interviews, many women with coronary disease expressed a need for rehabilitation programs. To meet some of these needs, an intervention program has been developed and is being evaluated in a clinical trial. Based on principles of cognitive behavior therapy, including both educational cognitive components and training sessions for improved emotional and social handling of stress reactions, the basic features of the program have been tested in clinical situations for many years and been found to be well-received, easily applied, and clinically useful.10 The original program was modified to be more suitable for urban women. There was a particular emphasis on techniques to manage stress from multiple roles, lack of time, resources for multiple tasks in home and work, and the sensation of being inadequate and insufficient when dealing with the stressors of daily life. The program provides education and training about biology and pathophysiology, emphasizing the cardiovascular system, and about cardiovascular risk factors and their roles in atherosclerosis. Subjects that are covered include social stress due to women’s multiple roles and conflicting demands, physiological and emotional stress reactions, and possible strategies for better coping and increased social support. More importantly, it provides training in healthier reactions to social stressors and techniques for creating and maintaining social support, healthier habits, and lifestyle. A stress-management program encourages exercise, healthy diet, and smoking cessation.
Women from diverse social backgrounds, educational levels, and ages (30-75 years) meet in groups of 6-8 for half a day once a week for 20 sessions to deal with the one factor they have in common: coronary disease. This meeting occurs within a structured educational program that includes both homework and training in new behaviors and reactions. Retention in the program is extremely high. Near the end of a course, requests for opportunities to continue the contacts with other participants became frequent. Asked about what they felt was most useful, one woman said: "I have learned many useful things and acquired many new skills. When I go to the grocery store, I may choose the longest cashier line, just for the enjoyment of being able to stand in a line thinking about pleasant things. But my most important insight is that [what I worry about] isn’t worth dying for."
This program now has been examined in a randomized, controlled trial with 270 female patients up to age 75 who had been hospitalized for acute coronary events. Controls received usual counseling and care, while the intervention patients also received the behavioral program. Systematic evaluation of health effects are not yet final, but subjective effects on emotional and physical self-rated health have been reported and are promising.10 Scores on mastery and coping, self-esteem, and self-care improved in both groups, but more so in the intervention group than in the controls. Thus, it seems likely that the quality of these women’s daily lives and their social, emotional, and physical well-being are positively influenced by this program.
These results should be sufficient motive for practical implementation. However, ongoing analyses include effects on recurrent cardiac events, hospitalization rates, and the costs and benefits of the program in relation to health care utilization. If these aspects are similarly positively influenced, such programs should be considered as public health interventions for rehabilitation and prevention of coronary heart disease.
Dr. Orth-Gomér is Professor of Community Medicine, Department of Public Health Sciences, Division of Psychosocial Factors and Health, Karollinska Institutet, Stockholm, Sweden.
1. Wenger NK, et al. Cardiovascular health and disease in women. N Engl J Med 1993;329:247-256.
2. Welty FK. Who should receive hormone replacement therapy? J Thromb Thrombolysis 1996;3:13-21.
3. Rozanski A, et al. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 1999;99:2192-2217.
4. Wamala SP, et al. Job stress and occupational gradient in coronary heart disease risk in women. Soc Sci Med 2000; 51:481-489.
5. Horsten M, et al. Depressive symptoms and lack of social integration in relation to prognosis of CHD in middle-aged women. Eur Heart J 2000;13:1072-1080.
6. Moser V, et al. Psykosocial stress hos kvinnor med hjärtsjukdom (Psychosocial risk factors for coronary heart disease in women, II). Stressforskningsrapporter (Stress Research Reports) no. 268. Stockholm: Statens Institut för Psykosocial Miljömedicin (IPM); 1996.
7. Karasek R, Theorell T. Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books; 1990.
8. Orth-Gomér K, et al. Marital stress worsens prognosis in women with coronary heart disease. JAMA 2000;284: 3008-3014.
9. Pearlin LI, et al. The stress process. J Health Soc Behav 1981;22:337-356.
10. Orth-Gomér K. Women and heart disease: New evidence for psychosocial, behavioral, and biological mediators of risk and prognosis. Int J Behav Med 2001;8:251-270.