Hazelton, Garrett PhD; Williams, Jennifer W. PhD; Wakefield, Jessica MA; Perlman, Adam MD, MPH; Kraus, William E. MD; Wolever, Ruth Q. PhD
Psychosocial distress contributes to and exacerbates cardiovascular disease.1 Fortunately, cardiac rehabilitation (CR) effectively improves both medical and psychosocial endpoints, reducing cardiac risk, improving functional capacity, and improving the quality of life of men and women.2–13
Compared with men, however, women with cardiac disease often experience greater medical and psychosocial burden, including higher rates of morbidity, disability, and early death after coronary events.14 These disparities may be related to some combination of factors. First, women report higher rates of preexisting psychosocial distress.15 Second, women show notable differences in the pathophysiology of heart disease.16 Third, there is evidence of gender bias, with society assuming that heart disease is a disease of men.17 Finally, compared with men, cardiac disease occurs among women at relatively older ages, when women have lower overall exercise capacity and health status, have lower social support, and are more likely to be widowed.18,19
Gender differences on the psychosocial outcomes of CR have been most commonly reported for depression, anxiety, and the behavioral variables included in cardiac risk.11,13,20,21 Less has been reported on the differential impact across gender of CR on variables such as anger, panic, or relationship satisfaction. Moreover, there have only been limited reports that discuss clinically relevant changes in psychosocial functioning.
The purpose of the current study was to further our understanding of the unique psychosocial experience of women completing CR for the first time by examining the statistical and clinical relevance of potential changes in a broad set of psychosocial variables. The aims were to (1) compare medical, demographic, and psychosocial characteristics of men and women pre-CR; (2) assess for change in psychosocial status and cardiac risk from pre-CR to post-CR; and (3) describe clinically meaningful change in psychosocial status among men and women.
This retrospective observational study evaluated a convenience sample of 380 CR participants who had completed psychosocial questionnaires pre-CR and immediately post-CR. The patients were referred from local inpatient and outpatient settings to a CR program at a southeastern US academic medical facility. Inclusion criteria required that participants were adults at initial enrollment into CR, able and willing to give informed consent, and provide biomarkers as well as completed questionnaires at both time points. Given our focus on psychosocial outcomes, participants were excluded if their psychosocial data were completely missing for 1 of the 2 time points. Participants were not excluded for missing medical data.
Patients were assigned by clinic staff to a supervising exercise physiologist or nurse who was responsible for conducting entrance assessments. During this first meeting, informed consent for descriptive research was obtained, demographic and medical information was collected, and surveys were completed. The clinician was also responsible for monitoring patient progress throughout the program and completing exit assessments. The psychosocial care of patients in this CR program is similar to the standard of care as set forth by the position statement of the American Association of Cardiovascular and Pulmonary Rehabilitation22 and provided by most US CR programs. Patients are provided with three to six 60-minute psychoeducational classes each month that cover topics including the following: psychosocial risk factors (depression, anxiety, anger hostility, and social isolation), their potential influence on coronary disease and treatment adherence, interventions for distress, skills for building social support, effective behavioral change strategies, communication strategies to support treatment adherence, and behavioral change.
In addition, a licensed psychotherapist provides 10 to 20 minutes of individual attention for all participants (unless participants refuse or are never available) while they are on the exercise floor, which is intended to establish rapport and provide general psychoeducation. The content of conversations vary, but typically the initial interaction includes discussion of psychosocial risk factors for heart disease (ie, stress management, social support, depression, anxiety), asking them about these factors in their lives, educating them about how these factors can affect the heart, and then suggesting resources (ie, classes, individual therapy, and learning skills such as relaxation techniques) intended to help reduce or modify potential risk factors or situations that contribute to the person's health and quality of life. Second and sometimes third conversations (also 10-20 minutes) occur with approximately 35% to 40% of participants. More visits may occur depending on patient need (ie, experiencing high levels of distress or are requesting support) and patient openness to talk with the provider.
An adaptation of the Brief Mood Survey22 consists of a 25-item self-report inventory that includes multiple 5-item rating scales to assess symptoms of depression, anxiety, panic, anger, and relationship satisfaction over the past week. The 2 depression symptom questions covering suicidal ideation were removed before administration. For depression, anxiety, panic, and anger, individual response options range from 0 (not at all) to 4 (extremely), such that the overall scores on these tests range from 0 (no symptoms) to 20 (the most severe), with higher scores indicating greater symptoms. Scores of 3 and more on subscales of depression, anger, anxiety, and panic symptoms are indicative of clinical levels for each respective scale.23–25 The inventory includes a 5-item Likert scale assessing relationship satisfaction, with responses ranging from 0 (very dissatisfied) to 6 (very satisfied). Scores of 25 or less are indicative of relationship dissatisfaction. The Brief Mood Survey has been shown to be valid and reliable, and the measures have been shown to compare favorably with the more well-established and longer tests of psychological status (eg, the Symptom Checklist-90-Revised).23–25
The Framingham Risk score26 is an individualized cardiac risk score that incorporates a number of demographic and behavioral cardiovascular risk factors, including age, gender, smoking status, weight, blood pressure, and cholesterol. Framingham Risk score was included in the current study for 2 reasons: to assess change in cardiac status after the course of CR, and because of the inclusion of psychosocial/behavioral factors (eg, smoking status) in the cardiac risk score calculation.
Descriptive statistics provided medical and sociodemographic status of men, women, and the total sample. For normally distributed continuous or ordinal data, gender differences were assessed using 1-way analysis of variance (ANOVA) for pre-CR measures, and 2 (men/women) × 2 (pre/post) repeated-measures ANOVA procedures to assess for gender differences across time. Specifically, gender served as the between-groups factor, and time served as the within-groups factor in this 2 × 2 design. For variables that could not be adequately transformed to normal distributions by using natural log or square root transformations, Kruskal-Wallis 1-way ANOVA procedures were used to assess pre-CR differences. To test for group × time interactions for such variables, we again used the Kruskal-Wallis 1-way ANOVA procedure to assess gender differences at pre-CR, at the end of CR, and on the delta scores (post-CR minus pre-CR) between genders. We used χ2 test to assess clinically significant change in psychosocial variables and cardiac risk. Percentages of men and women with elevated scores pre-CR were compared with percentages of men and women with elevated scores post-CR.
The sample (N = 380) was composed of 32.1% (n = 122) women and 67.9% (n = 258) men. Women were significantly older than men, t(374) = 2.06, P= .04; and there was a higher proportion of African American participants among women. As would be expected, the genders differed with respect to some clinical characteristics and performance on the 6-Minute Walk Test. They were statistically equivalent on other clinical measures, percentage using common cardiovascular drug classes, and importantly, age, authorized and completed CR sessions, and length of time spent in CR. Means and standard deviations for pre-CR sociodemographic and clinical characteristics of the sample are reported in Table 1. Pre-CR, women presented with more favorable Framingham Risk scores than men (P= .002). Both genders, however, experienced improvements in Framingham Risk scores across CR (P< .001). The means and standard deviations of these results are reported in Table 2.
In terms of pre-CR psychosocial profiles, the overall sample was in the reference range for symptoms of depression, anxiety, panic, and relationship satisfaction. The overall sample was in the elevated range for symptoms of anger. Evaluating separately by gender, women had anger scores in the elevated range while men did not. Similarly, compared with men, women's pre-CR psychosocial symptom scores were worse for panic (P= .004), depression (P= .023), and relationship satisfaction (P= .008). The means and standard deviations of these results are reported in Table 2.
Overall, the sample improved significantly from pre-CR to post-CR on psychosocial measures. More specifically, patients lowered their scores on symptoms of depression, anxiety, anger, and panic, all at P< .001. Improvement across time was not seen in relationship satisfaction (P= .703). The means and standard deviations of these results are also reported in Table 2.
No gender × time interactions were observed in the ANOVA tests on psychosocial variables, indicating that the amount of change from pre-CR to post-CR was similar for both genders. Among the 2 psychosocial variables tested with nonparametric statistics, the only variable demonstrating differential change across time was panic. Women demonstrated larger delta scores on panic than did men (P= .024). The means and standard deviations of these results are reported in Table 2.
We next explored the degree of clinically significant improvement within each gender. Again, scores of 3 and more on symptoms of depression, anger, anxiety, and panic subscales are indicative of clinical levels for each respective scale; and scores of 25 or less are indicative of relationship dissatisfaction. Regarding symptoms of depression and anxiety, the percentage of participants reporting clinically relevant symptoms significantly decreased from pre-CR to post-CR among both genders. Symptoms of anger and panic presented different pictures. The percentage of men reporting clinically relevant symptoms of anger decreased significantly (P= .03), while the percentage of women did not. Most men did not endorse panic symptoms at either time point and change was not significant (P= .13). In women, however, a notable decrease occurred in the percentage that reported clinical levels of panic (P= .005). Neither gender showed a change in relationship satisfaction. These results are reported in Table 3.
Compared with men, women generally experienced psychosocial and medical benefit after CR at similar rates; however, among participants with clinically relevant symptoms of anger pre-CR, women did not benefit from CR to the same degree as men. Cardiac rehabilitation did not appear to increase relationship satisfaction, especially notable since women reported relatively low levels of relationship satisfaction.
The impact of CR on clinically relevant anger symptoms appears distinct between genders. Although the percentage of men reporting significant levels of anger decreased significantly across CR, the percentage of women reporting significant levels of anger did not significantly change. Anger is important to address since there is evidence of a causal relationship with heart disease,27 and prospective studies from Framingham have shown that anger is specifically associated with heart disease risk in women.28 We found no evidence in the literature suggesting that anger in women is more resistant to improvement during CR, which reveals a need to further explore this unique finding.
Evidence has been available for a number of years, suggesting that men are more likely to demonstrate type A personality, a behavioral pattern involving free-floating hostility, impatience, competitiveness, and a propensity toward stress and anger.28 Those with type A personalities often experience frequent angry emotional states. While personality is often resistant to intervention, management of angry emotional states has a substantial evidence base for successful treatment.29 Few studies have reported specifically on anger in CR samples, and none describe gender-specific concerns in anger across CR. Previous CR studies have assessed hostility or type A personality rather than anger. Cognitive-behavioral approaches toward treating anger could be integrated into CR programs in a way that specifically attends to the unique patterns women more commonly have for expressing anger. With a substantial portion of women in the current sample not experiencing relief from their anger symptoms, more intervention may be needed to target these needs.
A second notable finding from this study is that in both genders, the level of relationship satisfaction did not change significantly from pre-CR to post-CR; men remained generally satisfied while women remained generally dissatisfied with their relationships. While CR is not designed to improve relationships, it is important to assess relationship satisfaction as a baseline variable (predictor variable) to help determine other types of support that may be needed to optimize outcomes. Given that research suggests that women may be more sensitive than men to negative features of relationships, this may be important to assess in the CR population. For example, marital stress combined with work stress is associated with relatively worse health outcomes in women. This interpersonal sensitivity in women may potentially impact coronary events, participation in CR, and their recovery and healing.30,31
While women often report greater psychosocial distress, evidence suggests that men may experience greater cardiovascular effect at similar levels of reported distress. Specifically, women appear to be at greater risk of experiencing anxiety than men after a coronary incident,11 while anxiety among men is predictive of 1-year cardiac mortality, though it is not for women.32
There are several limits to the generalizability of our findings. First, this observational study used a single-site, clinical sample with no control group, making it impossible to separate the potentiation contributions of the Hawthorne Effect (being monitored while participating in a study) and of natural history (time) from the contribution of the CR program itself in leading to symptom improvement. However, the fact that not all psychosocial variables improved does ameliorate this concern. Also, the use of a convenience sample that had psychosocial data available at both time points limits the accuracy of the findings. It is quite possible that those with more severe symptoms did not complete the CR program, thus not providing post-CR data. This scenario would have underestimated pre-CR symptoms while leaving open the accuracy of change in symptoms.
Another limitation is that data were not recorded regarding use of individual psychotherapeutic attention. When multiple contacts occur, this is typically related to the patient having some kind of need for more support for affective distress (ie, is depressed, anxious, highly stressed, or does not have healthy coping mechanisms, etc). Men generally bring up more anger issues than do women, while the women bring up concerns around anxiety and their gender roles. Depression is by far the most common reason for additional attention or referral for therapy; however, we are unsure how this might have affected followup compared with other affective factors. On the floor, men seem to talk to the psychotherapist about anger issues, and women most commonly talk about anxiety, gender roles, and relationship issues. On a similar note, an estimated 15% to 20% of participants are referred for individual counseling, and less than half of them follow through with treatment. Still, it is a limitation and a potential confounding factor that we cannot control for statistically, since we are unsure of who follows through with more intensive psychosocial treatment. However, the impact is likely to be fairly small, given that most patients get some attention and there is very limited follow through with individual psychotherapy.
In addition, we were limited to available data that did not include current relationship status, curbing our ability to interpret the relationship dissatisfaction findings. For example, divorced participants could rate relationship satisfaction less favorable that married participants, and in some cases as better. Finally, while it is encouraging that the overall sample of women appeared to benefit equitably from CR compared with men, our study still had relatively lower female participation; just less than one-third of our sample was female.
This study supports findings from a number of smaller studies by providing further evidence that CR offers psychosocial benefit for women. The sample experienced significant improvement in psychosocial distress and cardiovascular risk from pre-CR to immediately post-CR. The level of clinical impact on symptoms of depression, anxiety, and panic was notable for women, as significant proportions of women experienced symptom reduction from clinical into nonclinical ranges. However, greater clinical attention and changes in CR programming may be warranted to address symptoms of anger and relationship dissatisfaction among women.
The authors thank Jennifer Robbins, MS for institutional review board management; Sherri L. Westbrook, MA, RCEP, senior exercise physiologist; Karen Craig, MA, director, Cardiac Rehabilitation for Database Management; and Daniel Webber, MS, for data preparation and preliminary analyses.
1. Artham SM, Lavie CJ, Milani RV. Cardiac rehabilitation programs markedly improve high-risk profiles in coronary patients with high psychological distress. South Med J. 2008;101:262–267.
2. Lavie CJ, Thomas RJ, Squires RW, Allison TG, Milani RV. Exercise training and cardiac rehabilitation in primary and secondary prevention of coronary heart disease. Mayo Clin Proc. 2009;84:373–383.
3. Mirowsky J, Ross CE. Social patterns of distress. Annu Rev Sociol. 1986;12:23–45.
4. Buchner B, Kleiber C, Stanske B, Hermann-Lingen C. Stress and heart disease in women. Herz. 2005;30:416–428.
5. Hardy SE, Gill TM. Strengthening the evidence base to support the use of cardiac rehabilitation with older individuals. J Cardiopulm Rehabil. 2004;24:245–247.
6. McGrady A, McGinnis R, Badenhop D, Bentle M, Rajput M. Effects of depression and anxiety on adherence to cardiac rehabilitation. J Cardiopulm Rehabil Prev. 2009;29:358–364.
7. Beckie TM, Beckstead JW. The effects of a cardiac rehabilitation program tailored for women on global quality of life: a randomized clinical trial. J Womens Health. 2010;19:1977–1985.
8. Dunlay SM, Witt BJ, Allison TG, et al. Barriers to participation in cardiac rehabilitation. Am Heart J. 2009;158:852–859.
9. Marzolini S, Brooks D, Oh PI. Sex differences in completion of a 12 month cardiac rehabilitation programme: an analysis of 5922 women and men. Eur J Cardiovasc Prev Rehabil. 2008;15:698–703.
10. Caulin-Glaser T, Maciejewski PK, Snow R, LaLonde M, Mazure C. Depressive symptoms and sex affect completion rates and clinical outcomes in cardiac rehabilitation. Prev Cardiol. 2007;10:15–21.
11. Shen BJ, Wachowiak PS, Brooks LG. Psychosocial factors and assessment in cardiac rehabilitation. Eur Med Phys. 2005;41:75–91.
12. Wenger NK. Current status of cardiac rehabilitation. J Am Coll Cardiol. 2008;51:1619–1631.
13. Grace SL, Abbey SE, Shnek ZM, Irvine J, Franche R-L. Cardiac rehabilitation I: review of psychosocial factors. G Hosp Psychiatry. 2002;24:121–126.
14. Davidson P, DiGiacomo M, Zecchin R, et al. A cardiac rehabilitation program to improve psychosocial outcomes of women with heart disease. J Womens Health. 2008;17:123–134.
15. Roger VL, Go AS, Lloyd-Jones DM, et al. Executive summary: heart disease and stroke statistics—2012 update: a report from the American Heart Association. Circulation. 2012;125:188–197.
16. Vaccarino V, Badimon L, Corti R, et al. Ischemic heart disease in women: are there sex differences in pathophysiology and risk factors? Position paper from the working group on coronary pathophysiology and microcirculation of the European Society of Cardiology. Cardiovasc Res. 2011;90:9–17.
17. Wenger NK, Froelicher ES, Smith LK, et al. Cardiac Rehabilitation. Clinical practice guideline number 17. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research, National Heart, Lung, and Blood Institute; 1995. AHCPR publication no. 96-0672.
18. Heath GW, Hagberg JM, Ehsani AA, et al. A physiologic comparison of young and older athletes. J Appl Physiol. 1991;51:634–640.
19. McCarthy MM, Vaughan Dickson V, Chyun D. Barriers to cardiac rehabilitation in women with cardiovascular disease: an integrative review. J Cardiovasc Nurs. 2011;26:E1–E10.
20. Barth J, Volz A, Schmid J, Kohls S. Gender differences in cardiac rehabilitation outcomes: do women benefit equally in psychological health? J Womens Health. 2009;18:2033–2039.
21. Carhart RL Jr, Ades PA. Gender differences in cardiac rehabilitation. Cardiol Clin. 1998;16:37–43.
22. Hamm LF, Sanderson BK, Ades PA, et al. Core competencies for cardiac rehabilitation/secondary prevention professionals: 2010 update (position statement of the American Association of Cardiovascular and Pulmonary Rehabilitation). J Cardiopulm Rehab Prev. 2011;31:2–10.
23. Burns DD. Therapist's Toolkit: Comprehensive Assessment and Treatment Tools for the Mental Health Professional. Gladwyne, PA: Author; 2009.
24. Sekirnjak GC, Beal D. The concurrent validity of the Burns Depression Checklist. Paper presented at: Annual Meeting of the Association for the Advancement of Behavior Therapy; November 1999; Toronto, ON.
25. Sekirnjak GC, Beal D. The concurrent validity of the Burns Anxiety Checklist. Paper presented at: Annual Meeting of the Association for the Advancement of Behavior Therapy; November 1999; Toronto, ON.
26. D'Agostino RB, Grundy S, Sullivan LM, Wilson P. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. JAMA. 2001;286:180–187.
27. Low CA, Matthews KA, Kuller LH, Edmundowicz D. Psychosocial predictors of coronary artery calcification progression in postmenopausal women. Psychosom Med. 2011;73:789–794.
28. Haynes SG, Feinleib M. Women, work and coronary heart disease: prospective findings from the Framingham Heart Study. Am J Public Health. 1980;70:133–141.
29. Matthews KA, Haynes SG. Type A behavior pattern and coronary disease risk: update and critical evaluation. Am J Epidemiol. 1986;123:923–960.
30. Butler AC, Chapman JE, Forman EM, Beck AT. The empirical status of cognitive-behavioral therapy: a review of meta-analyses. Clin Psychol Rev. 2006;26:17–31.
31. Terrill AL, Garofalo JP, Soliday E, Craft S. Multiple roles and stress burden in women: a conceptual model of heart disease risk. J App Biobehav Research. 2012;17:4–22.
32. Emery CF, Frid DJ, Engebretson TO, et al. Gender differences in quality of life among cardiac patients. Psychosom Med. 2004;66:190–197.
cardiac rehabilitation; psychological benefits; women