Can experiencing mental illness literally cause heartache?

The heart has long served as a metaphor for emotions, the soul, and mental suffering. Idioms, proverbs and quotes linking the heart to emotional pain can be found across many languages and cultures. American author Zelda Fitzgerald (1900-1948), who experienced mental illness herself, once wrote that “nobody has ever measured, not even poets, how much the heart can hold.” (Sarah, 2024).

While this may still be true, it is becoming increasingly clear that the heart is affected by what it holds. Most of us will be familiar with the feeling of our heart racing when we are afraid, anxious or excited. Beyond these everyday experiences, there is growing evidence that psychological stress is closely linked to poorer cardiovascular health (Vancheri et al., 2022). As mental illness often goes hand in hand with emotional distress and persistent stress, it may not only break hearts metaphorically but also affect them physically.

So, how strong is the connection between mental and cardiovascular health? Gupta and colleagues set out to answer this question by systematically reviewing the available scientific literature.

The heart has long served as a literary symbol of emotional distress, but growing evidence suggests its link with cardiovascular health may not be purely metaphorical.

Methods

In this paper, Gupta et al. asked themselves “What is the association between mental disorders and acute coronary syndrome (ACS)?”. To answer this question, they searched MEDLINE, Embase, and PubMed for published studies on the influence of mental health conditions on different ACSs including acute myocardial infarction (AMI; a heart attack), and angina (tightness or pressure in the chest). The authors included studies according to the following criteria:

Published in any language
Observational or randomised controlled trials in design
Investigated the frequency of ACS in relation to prior mental illness
Reported adjusted effect sizes.

Once they identified all eligible studies, the references of said studies were screened for any other potentially relevant publications. Each study’s quality was evaluated with the National Institutes of Health Study Quality Assessment Tool. Subsequently, Gupta et al. conducted meta-analyses to calculate pooled hazard ratios or odds ratios for different groups of mental health conditions. The certainty of the evidence was assessed using the GRADE framework.

Results

Twenty-five studies were included, most of which (72%) were retrospective cohort studies, meaning that pre-existing data from a group of people was analysed. Across all included studies, more than 22 million people were represented. The studies examined a range of mental health conditions, including anxiety disorders, panic disorder, depression, mood disorders, bipolar disorder, psychotic disorders, sleep disorders, post-traumatic stress disorder (PTSD), and substance use disorders. Some analyses also combined conditions into broader diagnostic categories, such as mood disorders. The most frequently studied conditions were AMI (84%) and mood disorders (36%) respectively. Study quality was mostly fair on a scale of poor, fair, and good.

Several mental health conditions were associated with an increased risk of ACS or AMI:

Individuals with anxiety disorders had a 1.63-fold risk of experiencing AMI (5 studies),
those with substance use disorders a 2.41-fold risk of experiencing AMI (3 studies), and
those with post-traumatic stress disorder (PTSD) a 2.73-fold risk of experiencing AMI (2 studies).

For acute coronary syndrome, the authors found:

A 1.40-fold risk in individuals with depression (6 studies),
a 1.42-fold risk for mood disorders in general (9 studies), and
a 1.60-fold risk among participants with sleep disorders (3 studies).

The calculated hazard ratios for bipolar disorder, psychotic disorders and severe mental illness did not reach statistical significance.

Several mental disorders were found to be associated with an increased risk of acute coronary syndrome, with PTSD showing the strongest association and relatively consistent evidence across included studies.

Conclusions

This systematic review and meta-analysis found that several mental health conditions, including depression, anxiety, PTSD and sleep disorders, were associated with an increased risk of acute coronary syndrome. The authors conclude that their findings:

reaffirm mental disorders as a potential risk factor for cardiovascular disease, namely ACS and AMI.

However, they also acknowledge that the evidence base was largely observational and heterogeneous, and associations for some conditions, including bipolar and psychotic disorders, were not statistically significant.

The authors conclude that their findings “reaffirm mental disorders as a potential risk factor for cardiovascular disease, namely ACS and AMI.”

Strengths and limitations

This systematic review and meta-analysis used sound methodology by searching three large literature databases, thoroughly reporting search terms and inclusion and exclusion criteria, and following the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines. The chosen criteria were also reasonably selected. The review protocol was also registered in advance, and study screening, data extraction and quality assessment were conducted independently by multiple reviewers, reducing the risk of reviewer bias.

Not applying a language restriction is also a strength of this study because it reduces the risk of publication bias. Focusing on observational studies made sense as both studied variables (mental illness and ACS) are not feasible or ethical to be randomised in a research setting.

The review included a very large overall sample of more than 22 million participants, which is a clear strength. However, several disorder-specific analyses were based on only two or three studies. This reflects limited availability of eligible studies for some disorders, and makes it difficult to draw firm conclusions, particularly where confidence intervals were wide and certainty ratings were low or very low.

Another important limitation is that many of the analyses had substantial heterogeneity, indicating that the studies varied considerably. Differences in participant characteristics, follow-up periods, study design and adjustment for confounding factors may all have contributed to this variability. In several cases, the small number of studies included also limited the ability to explore sources of heterogeneity in more depth, again reflecting limitations in the available evidence base.

The observational nature of the included studies also limits causal interpretation. While mental health conditions were typically measured before ACS occurred, this study cannot show whether mental illness directly causes heart problems. Most studies adjusted for important factors such as age and sex, but differences in how studies accounted for other risk factors mean that some of the observed associations may still be explained by confounding.

Lastly, as noted by the authors, most of the studies were only of fair quality. Though this is preferable to poor-quality evidence, fair-quality studies remain vulnerable to important sources of bias and therefore provide only moderate confidence in the findings.

This systematic review provides a comprehensive synthesis of the available evidence on mental disorders and ACS, is limited by heterogeneity, the small number of studies in some analyses, and generally low certainty of evidence.

This systematic review provides a comprehensive synthesis of the available evidence on mental disorders and ACS, but is limited by heterogeneity, the small number of studies in some analyses, and generally low certainty of evidence.

Implications for practice

The results are based on observational evidence with low to very low certainty for several outcomes and so, they should be interpreted cautiously in medical and psychotherapeutic practice. However, the findings do add to a growing body of evidence suggesting an association between mental health conditions and cardiovascular outcomes, such as ACS.

Compiling the available evidence and shining a light on the lack of research in this area may help guide future research towards a more clinically relevant and underexplored field. At present, the evidence is not strong enough to support major changes to routine clinical practice, but it does reinforce the importance of considering physical health in people with mental illness.

In terms of current practice, structured cardiovascular risk assessment in people with mental health conditions is not yet consistently implemented in routine care, despite increasing interest in integrated physical and mental healthcare (Griffiths, 2026). However, once the evidence base is stronger and the mechanisms are better understood, it might be possible to develop targeted prevention programs for both cardiovascular risk reduction in people with mental health conditions and for improved mental health support in individuals with established cardiovascular risk factors. Interventions like this have been explored in recent years with promising results, such as the PRIMROSE intervention in England (Glasgow MSc Students, 2020) and a behavioural intervention in the USA (Mishu, 2020). However, the evidence remains limited, and we cannot yet recommend widespread implementation of specific screening or intervention programmes based on mental health status alone.

It appears that Zelda Fitzgerald remains correct for now, that “nobody has ever measured, not even poets, how much the heart can hold.”

Evidence linking mental health and cardiovascular outcomes is growing, but it remains insufficient to support consistent implementation of routine cardiovascular risk assessment in people with mental illness.

Statement of interests

Hannah Bielefeld has no conflicts of interest to report and did not use AI. Hannah wrote the first full draft of the blog. Éimear Foley assisted with the writing of the “Strengths and limitations” and “Implications for practice” sections and used ChatGPT to refine language. Éimear is an editor for The Mental Elf and has no conflicts of interests to declare.

Editor

Edited by Éimear Foley. AI tools assisted with language refinement and formatting during the editorial phase.

Links

Primary paper

Arnav Gupta, Tushar Tejpal, Chanhee Seo, Nicholas Fabiano, Selina Zhao, Stanley Wong, Yuan Qiu, Jenna MacNeil, Dain Kim, Natasha Aleksova, Sara Siddiqi, Marco Solmi & Jess Fiedorowicz (2026). Mental Disorders as a Risk Factor of Acute Coronary Syndrome: A Systematic Review and Meta-Analysis. JAMA Psychiatry, 83(3), 259. https://doi.org/10.1001/jamapsychiatry.2025.4253

Other references

Glasgow MSc Students. (2020, March 12). Cardiovascular risk in severe mental illness. National Elf Service. https://www.nationalelfservice.net/other-health-conditions/cardiovascular-disease/cardiovascular-risk-in-severe-mental-illness/

Griffiths, D. C. (2026, January 16). Cardiovascular screening for people with severe mental illness. National Elf Service. https://www.nationalelfservice.net/other-health-conditions/cardiovascular-disease/cardiovascular-screening-severe-mental-illness/

Mishu, M. P. (2020, October 21). Reducing cardiovascular risk in people with severe mental illness. National Elf Service. https://www.nationalelfservice.net/other-health-conditions/cardiovascular-disease/reducing-cardiovascular-risk-in-people-with-severe-mental-illness/

Sarah. (2024, June 12). “Nobody has ever measured, not even poets, how much the heart can hold.”—Zelda Fitzgerald. Medium. https://medium.com/@saraism/nobody-has-ever-measured-not-even-poets-how-much-the-heart-can-hold-zelda-fitzgerald-02775add684b

Vancheri, F., Longo, G., Vancheri, E., & Henein, M. Y. (2022). Mental Stress and Cardiovascular Health—Part I. Journal of Clinical Medicine, 11(12), 3353. https://doi.org/10.3390/jcm11123353

The Mental Elf