Does Anxiety Increase the Risk of Coronary Heart Disease?: A Meta-Analysis
Charlotte van den Hengel, BSc.
University of Amsterdam, The Netherlands
Introduction
Even though many studies have investigated the relationship between psychological factors such and stress or depression and the development of coronary heart disease (CHD) and outcomes, not many research has yet focused on the relationship between anxiety and coronary heart diseases (Brotman, Golden & Wittstein, 2007; Rosengren, Hawken & Ounpuu, 2004; Wulsin & Singal, 2003). Notwithstanding, prior research suggests that anxiety has an effect on the prognosis of coronary heart disease in patients, independent of depression, but the strength and role of anxiety remains unclear (Rothenbacher, Hahmann, Wüst, Koenig & Brenner, 2007; Shibeshi, Young-Xu & Blatt, 2007; Strik, Denollet, Lousberg & Honig, 2003; Grace, Abbey, Irvine, Shnek & Stewart, 2004). This meta-analysis focuses on the relationship between anxiety and the development of coronary heart diseases in initially healthy persons.
Methods
The aim of this meta-analysis is to assess the relationship between different types of anxiety (e.g. general anxiety, panic, phobia, worry, PTSD) and incidence of coronary heart diseases. All studies included in this meta-analysis are cohort studies measuring both the incidence of coronary heart diseases and anxiety in initially heathy people, and are conducted between 1980 and 2016. Methods for identification, evaluation, synthesis, statistical aggregation of information, appropriate calculations or conversions, and reporting of results were chosen according to predetermined methods for meta-analyses (Roest, Martens, de Jonge & Denollet, 2010, Stroup et al., 2000; Liberati et al., 2009; Hedges & Vevea, 1998; Parmar, Torri & Stewart, 1998).
Identification of studies
PubMed, EMBASE and PsychINFO computerised databases were searched with publication year range restriction being 1980 to 2016. Search criteria were the following terms: “Search points were the following terms: ”(mortality or coronary mortality or myocardial infarction or coronary heart disease or sudden cardiac death or cardiac death or myocardial infarction or cardiac events) and (community or cohort or healthy persons or risk) and (anxiety or post traumatic stress disorder or tension or anxiety symptoms or anxiety disorder or panic or panic attacks or phobic anxiety or phobia or worry)”. When possible, a filter was used to select only humans. There were no language restrictions applied, but all the included articles were reported in English. Both published and unpublished data were included in this meta-analysis.
Study Selection
All studies meeting the following pre-specified inclusion criteria were included, mostly based upon the inclusion criteria of a leading meta-analysis concerning the relationship between anxiety and CHD (Roest, Martens, de Jonge & Denollet, 2010):
Studies had to be prospective in nature, following a non-psychiatric cohort of initially healthy persons over time.
Studies had to include at least 1 measure (i.e. self-report or interview-based assessment) of anxiety symptoms or anxiety disorders (e.g. PTSD, worry, phobia, panic).
Studies had to include end points including cardiac mortality, cardiac events, cardiovascular events, or (non-fatal) myocardial infarction.
All studies meeting the following pre-specified exclusion criteria were excluded:
Studies explicitly focusing on persons over the age of 75.
Studies following a psychiatric cohort, non-cohort and/or not being prospective in nature.
Studies following patients with coronary heart disease at baseline or following treatments for one of the measures (i.e. anxiety and/or coronary heart diseases).
Initially 4629 studies were found in the search results, whereafter duplicates were removed and studies were evaluated on eligibly based upon study characteristics, title, abstracts and where applicable, full-text. The studies were included or excluded based upon the pre-specified inclusion and exclusion criteria. After this selection, a total of twenty-four studies were included in this meta-analysis. All these twenty-four studies reported on one or multiple outcomes of coronary heart diseases.
Data Extraction
Data was extracted from the studies by using a pre-set data extraction form to identify the study characteristics and thereby identifying studies meeting the following pre-specified inclusion or exclusion criteria. Studies were evaluated and data was extracted on sample size, sample characteristics, study design, anxiety type, anxiety assessment, type of coronary heart diseases (i.e. end-point), adjusted and unadjusted relative risks and their confidence intervals.
Quality Assessment
The contribution of each study to the meta-analysis based upon quality scoring was not weighed, because there are no valid measures of quality for observational studies, and the use of subjective rating scales might lead to bias according to the authors of a leading meta-analysis regarding this topic (Roest, Martens, de Jonge & Denollet, 2010).
Data synthesis
The data from all studies were taken together. In the case of a different measure of relative risk (i.e. relative risk not presented as a hazard ratio), outcomes were calculated into a new score with confidence interval. All relative risk outcomes were converted to hazard ratios (HR’s). In the case of multivariable analysis, outcomes were combined. In the case of multiple anxiety measures or multiple scales used, the most representative measure or scale was used. Given that the studies vary a lot in both anxiety types, measures and sample characteristics, a relatively large heterogeneity in the results was expected (Hedges & Vevea, 1998). Therefore, a random effect method was used to generate a summary estimate of effect. The amount of heterogeneity was measured with the Q-test and I2 test.
Results
Study Characteristics
Together the twenty-four studies reported on coronary heart disease in 371,325 people. The follow-up periods ranged between 2 and 37 years, with a mean of 12.1 years. The age of the participants at baseline assessment ranged from 18 to 72 years. All included studies and the corresponding characteristics are listed in Table 1. Twelve studies reported on men only, seven studies reported on women only, and five reported on a combined sample of men and women. Sixteen of the twenty-four studies showed a significant association between anxiety and coronary heart disease outcomes. Twenty-two studies adjusted their hazard ratio’s for several demographic variables (e.g. age and gender), biological risk factors (e.g. diabetes and family history) and lifestyle related behaviours (e.g. smoking, physical activity and alcohol consumption). After these adjustments, twelve out of the twenty-two studies still
showed a significant association between anxiety and incidence of coronary heart diseases. If not otherwise stated, results are for the adjusted hazard ratio’s.
Findings of the review
The mean effect size for coronary heart diseases measured in hazard ratio is Hedges’s g = 1.504 [1.337 – 1.670], p <.0001, according to the random effect analysis. The mean effect size for coronary heart diseases measured in hazard ratio, independent of several demographic variables, biological risk factors and lifestyle related behaviours is Hedges’s g = 1.197 [1.099 – 1.294], p < 0.0001, according to the random effect analysis. An overview of the adjusted effect sizes and corresponding variances and 95% confidence intervals for each study is presented in Figure 1. A forest plot of the effect sizes and corresponding 95% confidence intervals of each individual study is presented in Figure 2.
Heterogeneity
Given that the studies vary a lot in both anxiety types, measures and sample characteristics (e.g. age, gender), a relatively large heterogeneity in the results was expected. The amount of heterogeneity was Q (24) = 54.792, p = 0.0003, according to the Q-test of the random effect method. The proportion of the total variance explained by heterogeneity was I2 (24) = 0.562, which equals 56%, meaning that
this may represent moderate to substantial heterogeneity (Higgins, Thompson, Deeks & Altman, 2003).
Discussion
This meta-analysis focusing on the association between anxiety and incidence of coronary heart diseases in initially healthy individuals shows that anxiety can increase the risk of incidence of coronary heart diseases from 20% until 50%. Given that half of the studies still reported a significant association between anxiety and the incidence of coronary heart diseases after the adjusting the hazard ratio to the effect of anxiety independently of covariates for several demographic variables, biological risk factors and lifestyle related behaviours, suggests that anxiety could be a independent risk factor for incidence of coronary heart diseases and cardiac mortality (Roest, Martens, de Jonge & Denollet, 2010).
Strength for this meta-analysis comes from the adjustment for covariates in most of the studies that were included, as these covariates might have mediated the relationship between anxiety and coronary heart diseases. Twenty-two out of the twenty-four studies adjusted there effect sizes for covariates for a variety of demographic variables, biological risk factors and lifestyle related behaviours. After these adjustments the overall effect size decreased, but still remained significant.
In line with results of previous meta-analysis in this topic (Roest, Martens, de Jonge & Denollet, 2010), the increased risk of anxiety (i.e. 20-55%) is comparable with the increased risk of depression, which lies between 46-55% (Van der Kooy, van Hout, Marwijk, Stehouwer & Beekman, 2007; Roest, Martens, de Jonge & Denollet, 2010). Given that anxiety and depression have a moderate-to-strong correlation, it is possible that anxiety and depression are both part of a larger and more stable psychological factor influencing heart diseases, such as negative affect, nevertheless further research is necessary to investigate the different negative variables or components and its association with influence on heart diseases (Suls & Bunde, 2005; Denollet, 2008; Roest, Martens, de Jonge & Denollet, 2010).
Two important limitations in conducting a meta-analysis brought by Roest and colleagues (Roest, Martens, de Jonge & Denollet, 2010) is the prone to publication bias and the inevitability of combining data from studies that are not equally designed. The prone to publication bias is applicable for almost all meta-analyses and is partly for the reason that journals do not easily publish articles with insignificant results and unpublished articles are mostly not included in the electronic databases that are used the most to search the articles. This results in an overdosis of articles that are published and include significant results, and a shortage on unpublished articles. However, this meta-analysis also selected some unpublished articles during the selection process. Notwithstanding, publication bias could bring about a more strengthen association between anxiety and coronary heart disease than it actually is, yet even though actual numbers might be smaller, the association found was significant, even after adjustments made. Next, since the studies varied a lot – even in the most important measures (i.e. anxiety type, anxiety assessment and outcomes) – relatively large heterogeneity in the results was expected and extant.
Another limitation of this meta-analysis, also stated by Roest and colleagues (Roest, Martens, de Jonge & Denollet, 2010), is that almost all of the studies were conducted in Western countries and/or populations, and this might limit the generalisation of the findings.
Further research
Since this meta-analysis focused on anxiety in the general, non-psychiatric, population other than people coping with anxiety disorders, a recommendation for future research would be to investigate the association between anxiety disorders and coronary heart disease since one could expect that higher levels of anxiety (as present in anxiety disorders) might increase the strength of the association and increase the relative risk of developing heart diseases. It might also be interesting to inquisite what specific aspects or symptoms of anxiety strengthen the association and/or increase the relative risk, in view of one of the included studies which reported that the somatic symptoms cause a higher risk of incidence of coronary heart disease (47%) than the psychological symptoms (24%). But again, future research should focus on the specific aspects of anxiety that contribute to the development of heart diseases.
Conclusions
This meta-analysis shows that anxiety was found to be an independent risk factor for the incidence of coronary heart diseases in initially healthy individuals. It was found that anxiety could increase the incidence of coronary heart diseases with 20-50%. Recommendations for future research are investigating the specific aspects of anxiety that contribute to the development of heart diseases and the role of anxiety disorders.
Funding
This meta-analysis was not funded, but articles included may be established with the help of funding. There are no indications of specific profit or loose dependently on the outcome of this meta-analysis.
Conflicts of interest
The author has no (biomedical) financial, professional, or personal interest or conflict relevant to the study and consequences of publishing.
References
Albert CM, Chae CU, Rexrode KM, Manson JE, Kawachi I (2005). Phobic anxiety and risk of coronary heart disease and sudden cardiac death among women. Circulation ;111:480–7.
Boyle SH, Michalek JE, Suarez EC (2006). Covariation of psychological attributes and incident coronary heart disease in U.S. Air Force veterans of the Vietnam war. Psychosom Med;68:844–50.
Brotman DJ, Golden SH, Wittstein IS (2007). The cardiovascular toll of stress. Lancet;370:1089–100.
Denollet J (2008). Depression, anxiety and trait negative affect as predictors of cardiac events: ten years after. Psychosom Med 2008;70:949–51.
Denollet J, Maas K, Knottnerus A, Keyzer JJ, Pop VJ (2009). Anxiety predicted premature all-cause and cardiovascular death in a 10-year follow-up of middle-aged women. J Clin Epidemiol 2009;62:452–6.
Eaker ED, Pinsky J, Castelli WP (1992). Myocardial infarction and coronary death among women: psychosocial predictors from a 20-year follow-up of women in the Framingham Study. Am J Epidemiol;135:854–64.
Eaker ED, Sullivan LM, Kelly-Hayes M, D’Agostino RB Sr., Benjamin EJ (2005). Tension and anxiety and the prediction of the 10-year incidence of coronary heart disease, atrial fibrillation, and total mortality: the Framingham Offspring Study. Psychosom Med;67: 692– 6.
Einvik G, Ekeberg O, Klemsdal TO, Sandvik L, Hjerkinn EM (2009). Physical distress is associated with cardiovascular events in a high risk population of elderly men. BMC Cardiovasc Disord;9:14.
Gafarov VV, Gromova HA, Gagulin IV, Ekimova YC, Santrapinskiy DK (2007). Arterial hypertension, myocardial infarction and stroke: risk of development and psychosocial factors. Alaska Med;49:117–9.
Grace SL, Abbey SE, Irvine J, Shnek ZM, Stewart DE (2004). Prospective examination of anxiety persistence and its relationship to cardiac symptoms and recurrent cardiac events. Pyschother Psychosom 2004; 73:344 –52.
Haines A, Cooper J, Meade TW (2001). Psychological characteristics and fatal ischaemic heart disease. Heart 2001;85:385–9.
Haines AP, Imeson JD, Meade TW (1987). Phobic anxiety and ischaemic heart disease. Br Med J 1987;295:297–9.
Hedges, L. V., & Vevea, J. L. (1998). Fixed-and random-effects models in meta- analysis. Psychological methods, 3(4), 486.
Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. Bmj, 327(7414), 557-560.
Kawachi I, Colditz GA, Ascherio A, et al., (1994). Prospective study of phobic anxiety and risk of coronary heart disease in men. Circulation 1994; 89:1992–7.
Kubzansky LD, Cole SR, Kawachi I, Vokonas P, Sparrow D. (2006). Shared and unique contributions of anger, anxiety, and depression to coronary heart disease: a prospective study in the normative aging study. Ann Behav Med 2006;31:21–9.
Kubzansky LD, Koenen KC, Jones C, Eaton WW (2009). A prospective study of posttraumatic stress disorder symptoms and coronary heart disease in women. Health Psychology 2009;28:125–30.
Liberati A, Altman DG, Tetzlaff J, et al. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med 2009;151:W65–94.
Mykletun A, Bjerkeset O, Dewey M, Prince M, Overland S, Stewart R. (2007). Anxiety, depression, and cause-specific mortality: the HUNT study. Psychosom Med 2007;69:323–31.
Nicholson A, Fuhrer R, Marmot M (2005). Psychological distress as a predictor of CHD events in men: the effect of persistence and components of risk. Psychosom Med 2005;67:522–30.
Parmar, M. K., Torri, V., & Stewart, L. (1998). Extracting summary statistics to perform meta‐analyses of the published literature for survival endpoints. Statistics in medicine, 17(24), 2815-2834.
Phillips AC, Batty GD, Gale CR, et al. (2009). Generalized anxiety disorder, major depressive disorder, and their comorbidity as predictors of all-cause and cardiovascular mortality: the Vietnam experience study. Psychosom Med 2009;71:395–403.
Ringbäck Weitoft G, Rosén M. (2005). Is perceived nervousness and anxiety a predictor of premature mortality and severe morbidity? Longitudinal follow up of the Swedish survey of living conditions. J Epidemiol Community Health 2005;59:794–8.
Roest, A.M., Martens, E.J., de Jonge, P. and Denollet, J., (2010). Anxiety and risk of incident coronary heart disease: a meta-analysis. Journal of the American College of Cardiology, 56(1), pp.38-46.
Rosengren A, Hawken S, Ounpuu S, et al., (2004). INTERHEART Investigators. Association of psychosocial risk factors with risk of acute myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study): case-control study. Lancet 2004;364:953– 62.
Rosengren A, Tibblin G, Wilhelmsen L. (1991). Self-perceived psychological stress and incidence of coronary artery disease in middle-aged men. Am J Cardiol 1991;68:1171–5.
Rothenbacher D, Hahmann H, Wüst B, Koenig W, Brenner H. (2007). Symptoms of anxiety and depression in patients with stable coronary heart disease: prognostic value and consideration of pathogenetic links. Eur J Cardiovasc Prev Rehabil 2007;14:547–54.
Shibeshi WA, Young-Xu Y, Blatt CM. (2007). Anxiety worsens prognosis in patients with coronary artery disease. J Am Coll Cardiol 2007;49: 2021–7.
Smoller JW, Pollack MH, Wassertheil-Smoller S, et al. (2007). Panic attacks and risk of incident cardiovascular events among postmenopausal women in the Women’s Health Initiative Observational Study. Arch Gen Psychiatry 2007;64:1153–60.
Strik JJMH, Denollet J, Lousberg R, Honig A. (2003) Comparing symptoms of depression and anxiety as predictors of cardiac events and increased health care consumption after myocardial infarction. J Am Coll Cardiol 2003;42:1801– 7.
Stroup DF, Berlin JA, Morton SC et al. (2000). Meta- analysis of observational studies in epidemiology: a proposal for reporting. JAMA 2000;283:2008–12.
Suls J, Bunde J. (2005). Anger, anxiety, and depression as risk factors for cardiovascular disease: the problems and implications of overlapping affective dispositions. Psychol Bull 2005;131:260–300.
Thurston RC, Kubzansky LD, Kawachi I, Berkman LF. Do depression and anxiety mediate the link between educational attainment and CHD? Psychosom Med 2006;68:25–32.
Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. Int J Geriatr Psychiatry 2007;22: 613–26.
Vogt T, Pope C, Mullooly J, Hollis J. Mental health status as a predictor of morbidity and mortality: a 15-year follow-up of members of a health maintenance organization. Am J Public Health 1994;84: 227–31.
Wulsin LR, Singal BM. Do depressive symptoms increase the risk for the onset of coronary disease? A systematic quantitative review. Psychosom Med 2003;65:201–10. Yasuda N, Mino Y, Koda S, Ohara H. The differential influence of distinct clusters of psychiatric symptoms, as assessed by the General Health Questionnaire, on cause of death in older persons living in a rural community of Japan. J Am Geriatr Soc 2002;50:313–20.
Additional URL’s:
Link to the Downloaded Search Results: https://docs.google.com/spreadsheets/d/1IC7rgd-HG8WX-o8m4GssZoS3e07Jaw0tWfaRbDGO-4Q/edit?usp=sharing
Link to the Data Extraction Form Questions: https://docs.google.com/forms/d/14VrLDB8hQp64C0D7ZtOh9v_-KvUih94nTeX9fU8bSFQ/viewform
Link to the Data Extraction Form Responses: https://docs.google.com/spreadsheets/d/1lHsA_adt6i7qudnPX6i5xMbif7DdABwftFyo7Ae8VII/edit?usp=sharing
Link to the Data of the Studies: https://docs.google.com/spreadsheets/d/1SjPPDwAeLfPFad6gwiGcFEeDp2vDRC4nuV9HLtwzcE8/edit?usp=sharing