The role of sex/gender in coronary heart disease (CHD) has been extensively studied since its importance was recognized in the 1980s.1–3 Similarly, the role of depression in CHD has been studied since the 1990s.4–6 Depression is now recognized as a risk factor for poor outcomes in CHD.5,7,8 Both CHD and depression place a tremendous burden on women: CHD is the leading cause of death among women, women have poorer outcomes after a cardiac event compared with men,9,10 and women are affected by depression at a 2:1 rate compared with men.11,12 Despite the frequent occurrence of both CHD and depression in women, interpreting studies about women with heart disease and depressive symptoms is challenging for several reasons. Women continue to be underrepresented in research despite legislation requiring their inclusion, making interpretation of studies with small numbers of women problematic. Often, data are not presented separately by sex/gender, so examination and interpretation of women’s data apart from men’s are not possible. Many studies do not look at depressive symptoms over time, leaving our understanding of what happens to women over time incomplete. Finally, the use of multiple depressive symptom assessment instruments makes comparisons between studies problematic.
This state of the science review examined depressive symptom data from longitudinal studies of CHD, which included women, collected depressive symptom data at more than 1 time point, and reported and/or analyzed data by sex. When possible, studies were examined by comparing the prevalence of elevated depressive symptoms (percentage of individuals meeting cutoff criteria on assessment instruments) as different instruments’ results cannot be easily compared. Studies were also examined based on which instrument was used. For the purposes of this review, “sex” and “gender” were used interchangeably given the psychosocial influences that “gender”—a social construct rather than a biological state—may confer in the context of illness development and symptom experience.
This literature review process follows the PRISMA 2009 checklist (see Figure 1).13 A search of databases including MEDLINE via PubMed, CINAHL, EMBASE, and PsycINFO was performed on February 1, 2016, and updated on April 26, 2018, for any recent additions. Terms searched included “coronary artery disease,” “coronary arteriosclerosis,” “depression,” “depressions,” “depression+,” “depression/exp,” “major depression,” “anaclitic depression,” “dysthymic disorder,” “endogenous depression,” “postpartum depression,” “reactive depression,” “recurrent depression,” “treatment resistant depression,” and “depressi*.” Search limiters or filters included “female,” “not male,” “English,” “humans,” “adult: 19+ years,” “longitudinal,” “longitudinal study,” “peer reviewed,” and “1990–2017.”
Because the exploration of the role of depression in CHD began in the 1990s, the year 1990 is used as a starting point for article inclusion. Journal articles were included if they met the following criteria: the study examined ischemic CHD and had a descriptive, longitudinal design; the study included women; depressive symptoms were assessed at more than 1 time point; results were reported and/or analyzed by sex; and the article was published in a peer-reviewed journal. This search retrieved 1098 articles. Removal of duplicates left 723 articles. Articles were then excluded for the following reasons: nonresearch articles (such as reviews), qualitative articles, noncardiac populations (such as diabetes), controlled trials (where an intervention may have affected depressive symptoms), studies of other cardiac conditions (transplants, heart failure, dysrhythmia/pacer/implantable cardioverter-defibrillators, etc), studies that focused on risk for CHD, different phenomenon within CHD (such as medications, biomarkers, or ST depression), populations known to have high depression levels (abused women), or studies where minimal (ie, 1 sentence) examination or reporting of data by sex was performed. One hundred thirteen full-text articles were screened, and 28 hand-searched articles were identified and screened at that time. No gray literature was searched. Twenty articles met inclusion criteria and were reviewed and included in Table 1 (see Supplemental Digital Content, available at http://links.lww.com/JCN/A66).
Prevalence and Cutoff Criteria
When possible, studies were examined by comparing the prevalence of elevated depressive symptoms. Prevalence is the proportion of a population affected by a certain condition. In the reviewed articles, prevalence was expressed as the percentage of participants meeting cutoff criteria on the depressive symptom measurement instrument used by the author. Cutoff scores are determined by receiver operating characteristic curve analysis, with the area under the curve representing the score with the greatest sensitivity and specificity for correctly identifying the presence of a condition. These scores are not diagnostic but rather serve as a proxy for a clinically significant level of impaired functioning in the individual.
Prevalence of Depressive Symptoms
Prevalence in Women at Baseline
Of the 20 studies that met inclusion criteria, only 8 provided prevalence data on women participants separately from men.14–21 The prevalence of elevated depressive symptoms in women during hospitalization for an index cardiac event ranged from 27.9%16 to 40.3%,19 with a calculated average of 35.75% across the 8 studies (see Figure 2). The studies that used the Beck Depression Inventory-II (BDI-II) showed the highest prevalence: 38.8% and 40.3% in patients with acute coronary syndrome (ACS) and myocardial infarction (MI), respectively.19,21 The single study that used the Depression Interview and Structured Hamilton (DISH) found a prevalence of 36% before hospital discharge.15 A much lower prevalence of elevated depressive symptoms (27.9%) was found in the study that used the Hospital Anxiety and Depression Scale (HADS) instrument.16 Two studies17,20 used the measurement of depressive symptoms before beginning cardiac rehabilitation (typically 2–3 weeks after hospital discharge) as the baseline time point, although 1 additional study measured symptoms at this time as a second measurement time point—“postdischarge.”15 Among these studies, the Beck instruments (BDI and BDI-II) demonstrated the highest prevalence: 36%17 and 31%,20 respectively. The “postdischarge” study, which used the DISH, had a much lower prevalence of only 16.4%.15
Longitudinal Prevalence Among Women
Of the 20 studies that met inclusion criteria, only 5 provided prevalence data for more than 1 time point.15,16,18,19,21 The average prevalence of elevated depressive symptoms among women across studies decreased from 35.75% during hospitalization to 27.8% 2 to 3 weeks postdischarge, to a calculated average of 22.71% over the next 24 months (see Figure 2). The trend overall was for improvement in depressive symptoms through the first 6 months, with leveling off between 6 and 24 months. However, 1 study demonstrated a small, statistically nonsignificant increase in prevalence from hospitalization (40.3%) to 12 months (40.4%) using the BDI-II.19
An examination of the available longitudinal data shows conflicting findings from hospitalization to 1 month. Prevalence of elevated depressive symptoms dropped significantly, from 36% to 16.4%, in a study15 of patients who have undergone post–coronary artery bypass graft (CABG) using DISH but only slightly (and statistically nonsignificantly) from 38.8% to 26.3% in a study of patients with ACS that used the BDI-II.21 A drop from 23% at 1 month to 12% at 4 months post-hospitalization in women referred to cardiac rehabilitation (CR) was noted using the Kellner questionnaire.18 Doering et al’s15 study showed a nearly 3-fold drop in prevalence from 36% at hospitalization to 12.7% at 6 months using the DISH. Hunt-Shanks et al’s16 study showed a smaller decline between hospitalization and 6 months: from 27.9% to 20.6% using the BDI. However, Hunt-Shanks et al16 reported a nonsignificant drop from 6 months (20.6%), to 19.2% at both 12 and 24 months. Norris et al’s19 study noted a small increase in prevalence from 40.3% during hospitalization to 40.4% at 12 months using BDI-II. No differences were noted based on the year the study was conducted, despite significant changes to diagnostic and treatment approaches over the intervening years (1999–2010).
Prevalence in Men vs Women
Five of the 20 studies reviewed provided prevalence data for both men and women.14,16,17,19,21 All 5 studies, regardless of the assessment instrument used, showed a higher baseline prevalence of elevated depressive symptoms in women than in men, and all but 1 study21 also demonstrated a statistically significant difference between sexes (see Figure 3). The average calculated baseline prevalence in men across these 5 studies was 23.46%, versus 34.8% in women. Of the depressive symptom assessment instruments used, the HADS showed the lowest baseline prevalence in both men (18.7%) and women (27.9%).16 The highest prevalence for both sexes was found using the BDI-II: 32.7% in men and 40.3% in women.19 Three of the 5 studies provided prevalence data for both sexes beyond baseline,16,19,21 with length of follow-up time ranging from 4 weeks21 to 2 years.16 All 3 studies showed a higher prevalence of elevated depressive symptoms among women than men regardless of the assessment instrument used, although not all were statistically significant. For both sexes, the HADS yielded the lowest prevalence, whereas BDI-II yielded the highest prevalence.
Nonprevalence Findings in Men and Women at Baseline
Twelve of the 20 studies provided mean depressive symptom scores rather than prevalence data.22–33 Of these studies, 1 measured baseline 2 weeks before CABG,24 7 measured baseline during hospitalization,22,23,25,27,30,31,33 3 measured baseline before CR,28,29,32 and 1 measured baseline 45 days post–cardiac event.26 Six studies (2 post-CABG studies and 4 post-CHD hospitalization studies) found women to have significantly higher mean scores than men at baseline, regardless of the instrument used.24,26,29–31,33 One study27 reported only a slightly lower mean score for men compared with women on the BDI-II pre-CR. Three studies found no significant difference between men and women at baseline post-MI23,29 or pre-CR32 using the HADS23,32 and Kellner29 questionnaires. Grace and colleagues25 found no significant difference between CR participants and nonparticipants at baseline in their all-female study using the HADS. Brink et al23 found women to have slightly (although statistically insignificantly) lower levels of depressive symptoms than men at baseline using the HADS.
Nonprevalence Findings in Men and Women After Baseline
Postbaseline time points ranged from 1 week24 to 2 years.31 More than half of the studies found improvement in symptoms for both genders within 3 to 6 months after hospitalization,22,24,26,27,29,32,33 regardless of the instrument used. Two studies had contrary findings: Brink and colleagues23 found significantly increased depressive symptom scores for both sexes between baseline and 5 months on the HADS after a first-time MI, and Lavie and Milani28 found a significant decrease in depressive symptom scores in men only after CR using the Kellner questionnaire. For the remaining 3 studies, the measurement time frame was much broader. Phillips Bute et al’s30 12-month study using the Center for Epidemiological Studies-Depression (CES-D) scale found statistically significantly higher levels of depressive symptoms both pre-CABG and 1 year postoperatively among women compared with men, although both sexes also experienced significant improvements in depressive symptoms over the 1-year interval. Grace et al’s25 study found an improvement on the BDI at 18 months compared with baseline in women who had not participated in CR compared with women who had. Zaninotto and colleagues31 found that, among those who had experienced angina or an MI in the past 2 years, women demonstrated a 13% higher probability of meeting cutoff criteria compared with men both at baseline and at 2-year follow-up; this decreased to 5% at 4-year follow-up using the CES-D-8 (an abbreviated version of the CES-D containing only 8 items).
This review of 20 longitudinal studies examining elevated depressive symptoms in women with CHD yielded several important findings. It demonstrated that the prevalence of elevated depressive symptoms among women at the time of CHD diagnosis was high, with a calculated average of 35.75%—higher than men (23%) and higher than the commonly perceived prevalence in the population with CHD (20%)—and this is true both at baseline and over time. Prevalence varied by measurement instrument rather than participant inclusion criteria (ACS, MI, PCI, general CHD), with the exception of CABG; this also held true over time. This review also supported the perception that most women’s symptoms do improve and that this improvement largely occurs within the first 6 months after the index cardiac event.
However, the 20 articles summarized in this review varied widely in terms of time span covered (1 month to 2 years) and participant inclusion criteria. Previous research suggests that those who have undergone CABG tend to experience fewer depressive symptoms than those who have not, particularly compared with those who are medically managed after MI without intervention34 or those with unstable angina.35,36 Only 2 studies in this review focused exclusively on patients who have undergone post-CABG. Of these, only one provided prevalence data, and this was an all-women study. Exclusion of this study made a little difference to the calculated average prevalence among women at hospitalization (35.75% with the study, 35.66% without), but it did correlate to the lowest prevalence among the studies at each of its follow-up time points (2–3 weeks and 6 months).
Study follow-up time frames may have also affected results. Because of the small number of studies that examined symptoms beyond 6 months, the apparent improvement associated with this time frame may be a reflection of the paucity of longer studies rather than actual increased improvement in this early stage of recovery. Stapelberg and colleagues37 indicated that one of the main gaps in understanding the link between depression and CHD is failing to examine the follow-up time frame. The authors are not aware of any large studies where quantitative changes in the severity of major depressive disorder are followed at regular intervals over time. It is recommended that frequent measures of patient mental state be correlated over time with measures of cardiac health to provide more accurate longitudinal estimates of patient progress.37
The examination of prevalence data in this review demonstrated a difference in findings based on instrument, with the BDI and BDI-II demonstrating a much higher prevalence than the HADS or Kellner questionnaire. A critical difference between instruments is the inclusion or exclusion of somatic depressive symptoms (such as fatigue, alterations in sleep, or changes in appetite). Instruments that included somatic symptoms yielded higher scores. Ormel and de Jonge38 made a case for 2 different types of depression in patients with CHD: “typical depression” (characterized by vulnerability and stressful life events) and “somatic depression” (triggered by “the presence of vascular disease, systemic inflammation, and atherosclerosis”). A study39 testing their integrative model supported their conclusion: those with somatic depression were at risk for increased mortality. They noted that those with persistent somatic depressive symptoms were more likely to be older women. The exception in this review regarding a higher prevalence on instruments with somatic symptoms was the results for the DISH instrument. It detected a much lower prevalence than the Beck instruments at baseline and pre-CR and HADS at 6 months despite including somatic symptoms. This may, however, reflect the CABG population who it was used with rather than its inclusion of somatic symptoms.
Although it seems that there is a relationship between somatic symptoms and an increased risk of mortality in patients with CHD, it remains unclear whether these somatic symptoms are depressive in nature or are related to the underlying heart disease itself. Both depression and heart disease in women can manifest as fatigue, sleep disturbances, and/or gastrointestinal symptoms such as poor appetite or nausea.15,40–42 The HADS instrument deliberately does not contain questions about somatic symptoms because the authors were concerned about confounding symptoms of physical illness with symptoms of depression.43 DeVon et al,42 who evaluated patients with ACS using the HADS, stated that these overlapping symptoms “confound [the] clinical picture.” DeVon used the HADS specifically because it does not measure physical symptoms, which could be heart disease rather than depression (DeVon, personal communication, October 14, 2016). Thus, although it seems that these symptoms are significant and should be assessed, it remains unclear whether they indicate depressive symptoms or rather reflect a physical disease process. Because management will differ depending on the symptom’s source, further investigation of the source of these somatic symptoms is warranted.
The reviewed studies also demonstrated difficulty in trying to compare findings between studies that use different measures of depressive symptoms. Use of item response theory and the development of common metrics for assessing depressive symptoms as a means to generate easily comparable findings across studies have begun. Patient-Reported Outcomes Measurement Information System, a set of 90+ assessment instruments, was developed using item response theory and proposed as a common metric, but the depressive symptom assessment instrument does not contain somatic symptoms. Stapelberg and colleagues37 recommend a combination of interview and Diagnostic and Statistical Manual of Mental Disorders criteria-based self-report instruments to assess depression. This would facilitate a more accurate diagnosis of major depressive disorder, increase accuracy in assessing symptom severity, and potentially allow for a comparison between the 2 methods. Until an agreed-upon common metric for measuring depressive symptoms is available, reporting prevalence data either with or without an accompanying diagnostic interview will allow a simple, direct comparison between studies.
This review demonstrated that women continue to be underrepresented in mixed-sex studies of CHD, with women comprising an average of only 28.9% of study participants in this review. The lack of data regarding women has been reflected in other reviews.10 By not separating data by sex, the differences between men and women’s symptoms continue to be obscured because of this underrepresentation. To understand women’s unique experiences (and to create interventions that meet women’s needs), data must be separately analyzed and reported by sex, a practice now expected by the National Institutes of Health.44 Ouyang and colleagues45 also suggest encouraging, if not requiring, data reporting by gender by both research sponsors and journal editors.
Although much has been learned over the past 30 years about women with heart disease and comorbid depressive symptoms, this review highlights just how much remains unknown. Further longitudinal studies and secondary analyses of existing longitudinal data that focus on sex differences are needed to further elucidate women’s experiences and symptom trajectories. The reporting of data by sex needs to become standard practice, particularly in light of the difficulty in recruiting equal numbers of women into clinical trials alongside the National Institutes of Health’s recognition of sex as an important biological variable. Although use of item response theory and the development of common metrics for assessing depressive symptoms are underway, a simple solution currently for reporting these data in a manner easily comparable is to report such data as prevalence. These steps will provide a fuller picture of women with CHD and depressive symptoms and allow development of more efficacious interventions for women.
What’s New and Important
- Most studies show that women with CHD experience more depressive symptoms than men both immediately after a cardiac event and over the following 2 years.
- Most women improve within the first 3 to 6 months after a cardiac event, but this may also reflect fewer studies greater than 6 months.
- Depression assessment instruments that included somatic symptoms yielded higher prevalence rates than those that did not, except in the population who have undergone CABG.
- Very few studies report prevalence of elevated depressive symptoms: the longitudinal prevalence data reviewed here represent less than 416 women.
1. Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population. Am Heart J
2. Wenger NK. Coronary disease
. Annu Rev Med
3. Wenger NK. Gender, coronary artery disease, and coronary bypass surgery. Ann Intern Med
4. de Miranda Azevedo R, Roest AM, Hoen PW, de Jonge P. Cognitive/affective and somatic/affective symptoms of depression
in patients with heart disease and their association with cardiovascular prognosis: a meta-analysis. Psychol Med
5. Leung YW, Flora DB, Gravely S, Irvine J, Carney RM, Grace SL. The impact of premorbid and postmorbid depression
onset on mortality and cardiac morbidity among patients with coronary heart disease: meta-analysis. Psychosom Med
6. Nicholson A, Kuper H, Hemingway H. Depression
as an aetiologic and prognostic factor in coronary heart disease: a meta-analysis of 6362 events among 146 538 participants in 54 observational studies. Eur Heart J
7. Barth J, Schumacher M, Herrmann-Lingen C. Depression
as a risk factor for mortality in patients with coronary heart disease: a meta-analysis. Psychosom Med
8. Lichtman JH, Froelicher ES, Blumenthal JA, et al. Depression
as a risk factor for poor prognosis among patients with acute coronary syndrome: systematic review and recommendations: a scientific statement from the American Heart Association. Circulation
9. Doering LV, Eastwood JA. A literature review of depression
, anxiety, and cardiovascular disease in women
. J Obstet Gynecol Neonatal Nurs
10. McSweeney JC, Rosenfeld AG, Abel WM, et al. Preventing and experiencing ischemic heart disease as a woman: state of the science: a scientific statement from the American Heart Association. Circulation
11. Mazure CM, Keita GP, Blehar MC. Summit on women and depression: proceedngs and recommendations
. Wye River Conference Center: Paper presented at: Summit on Women
12. Shanmugasegaram S, Russell KL, Kovacs AH, Stewart DE, Grace SL. Gender and sex differences in prevalence
of major depression
in coronary artery disease patients: a meta-analysis. Maturitas
13. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ
14. 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
15. Doering LV, Magsarili MC, Howitt LY, Cowan MJ. Clinical depression
after cardiac surgery. J Cardiovasc Nurs
16. Hunt-Shanks T, Blanchard C, Reid RD. Gender differences in cardiac patients: a longitudinal investigation of exercise, autonomic anxiety, negative affect and depression
. Psychol Health Med
17. Josephson EA, Casey EC, Waechter D, Rosneck J, Hughes JW. Gender and depression
symptoms in cardiac rehabilitation: women
initially exhibit higher depression
scores but experience more improvement. J Cardiopulm Rehabil
18. Lavie CJ, Milani RV, Cassidy MM, Gilliland YE. Effects of cardiac rehabilitation and exercise training programs in women
. Am J Cardiol
19. Norris CM, Hegadoren K, Pilote L. Depression
symptoms have a greater impact on the 1-year health-related quality of life outcomes of women
post-myocardial infarction compared to men. Eur J Cardiovasc Nurs
20. Sanderson BK, Bittner V. Women
in cardiac rehabilitation: outcomes and identifying risk for dropout. Am Heart J
21. Shin NM, Hagerty B, Williams R. Gender comparison in depressive symptoms and use of antidepressant medications after acute coronary syndrome. Appl Nurs Res
22. Bogg J, Thornton E, Bundred P. Gender variability in mood, quality of life and coping following primary myocardial infarction. Coronary Health Care
23. Brink E, Grankvist G, Karlson BW, Hallberg LR. Health-related quality of life in women
and men one year after acute myocardial infarction. Qual Life Res
24. Duits AA, Duivenvoorden HJ, Boeke S, et al. The course of anxiety and depression
in patients undergoing coronary artery bypass graft surgery. J Psychosom Res
25. Grace SL, Grewal K, Arthur HM, Abramson BL, Stewart DE. A prospective, controlled multisite study of psychosocial and behavioral change following women
’s cardiac rehabilitation participation. Journal of Womens Health (Larchmt)
26. Gravely-Witte S, De Gucht V, Heiser W, Grace SL, Van Elderen T. The impact of angina and cardiac history on health-related quality of life and depression
in coronary heart disease patients. Chronic Illn
27. Gupta R, Sanderson BK, Bittner V. Outcomes at one-year follow-up of women
and men with coronary artery disease discharged from cardiac rehabilitation: what benefits are maintained? J Cardiopulm Rehabil Prev
28. Lavie CJ, Milani RV. Effects of cardiac rehabilitation and exercise training on exercise capacity, coronary risk factors, behavioral characteristics, and quality of life in women
. Am J Cardiol
29. McGrady A, McGinnis R, Badenhop D, Bentle M, Rajput M. Effects of depression
and anxiety on adherence to cardiac rehabilitation. J Cardiopulm Rehabil Prev
30. Phillips Bute B, Mathew J, Blumenthal JA, et al. Female gender is associated with impaired quality of life 1 year after coronary artery bypass surgery. Psychosom Med
31. Zaninotto P, Sacker A, Breeze E, McMunn A, Steptoe A. Gender-specific changes in well-being in older people with coronary heart disease: evidence from the English Longitudinal Study of Ageing. Aging Ment Health
32. Barth J, Volz A, Schmid JP, et al. Gender differences in cardiac rehabilitation outcomes: do women
benefit equally in psychological health? Journal of Women’s Health (Larchmt)
33. Grace SL, Abbey SE, Pinto R, Shnek ZM, Irvine J, Stewart DE. Longitudinal course of depressive symptomatology after a cardiac event: effects of gender and cardiac rehabilitation. Psychosom Med
34. Pająk A, Jankowski P, Kotseva K, de Smedt D, De Bacquer D. EUROASPIRE Study Group. Depression
, anxiety, and risk factor control in patients after hospitalization for coronary heart disease: the EUROASPIRE III Study. Eur J Prev Cardiol
35. Geovanini GR, Gowdak LHW, Pereira AC, et al. OSA and depression
are common and independently associated with refractory angina in patients with coronary artery disease. Chest
36. Lotufo P, Nunes MA, Brunoni A, Barreto SM, Ribeiro ALP, Bensenor IM. Psychiatric morbidity and stable angina pectoris. The ELSA-Brasil. Eur Heart J
37. Stapelberg NJC, Neumann DL, Shum DHK, McConnell H, Hamilton-Craig I. The impact of methodology and confounding variables on the association between major depression
and coronary heart disease: review and recommendations. Current Psychiatry Reviews
38. Ormel J, de Jonge P. Unipolar depression
and the progression of coronary artery disease: toward an integrative model. Psychother Psychosom
39. Roest AM, Wardenaar KJ, de Jonge P. Symptom and course heterogeneity of depressive symptoms and prognosis following myocardial infarction. Health Psychol
40. Kohlmann S, Gierk B, Hummelgen M, Blankenberg S, Lowe B. Somatic symptoms in patients with coronary heart disease: prevalence
, risk factors, and quality of life. JAMA Intern Med
41. Sanner JE, Frazier L, Udtha M. Self-reported depressive symptoms in women
hospitalized for acute coronary syndrome. J Psychiatr Ment Health Nurs
42. DeVon HA, Pettey CM, Vuckovic KM, Koenig MD, McSweeney JC. A review of the literature on cardiac symptoms in older and younger women
. J Obstet Gynecol Neonatal Nurs
43. Snaith RP, Zigmond AS. The hospital anxiety and depression
scale. Br Med J (Clin Res Ed)
45. Ouyang P, Wenger NK, Taylor D, et al. Strategies and methods to study female-specific cardiovascular health and disease: a guide for clinical scientists. Biol Sex Differ