Depression is a global health concern, affecting millions of people worldwide (1). Depending on the geographic region, estimates of lifetime prevalence of major depressive disorder vary between 2 and 30% (2–5). Depression is more prevalent among women than men (about 2:1 ratio), and the median age of onset is around 25 years (6, 7). Depression strongly contributes to higher Disability Adjusted Life Years (DALYs) (8). Projections assume that depression will be the leading cause of disability worldwide by 2030 (9), which might be partly due to the high comorbidity with cardiovascular diseases (CVDs) (10, 11).

Close links between the mind and the heart have been suggested already in the first half of the 17th century, but empirical evidence for this assumption only appeared around 1930 when researchers found that patients who were depressed at the time they experienced an acute myocardial infarction had a significantly higher mortality risk compared to non-depressed counterparts (6, 12). Since then, research on the associations between symptoms of depression, CVDs and mortality steadily increased (13).

With regard to excess mortality, compared to the general population, individuals with severe mental illnesses (SMI) die on average 10–25 years earlier, most often in close association with CVDs (10, 14, 15). This mortality gap appeared to increase over the past years (16). Even in Scandinavia, where high-quality and most equitable healthcare is provided, this gap has not substantially narrowed during the past decades (17). One reason for the persisting or widening gap might be that secondary prevention is much less successful in people with mental illnesses than in the general population including programs that target on reduction of overweight, smoking, unhealthy nutrition and physical inactivity (10).

In line with this, compared to the general population, individuals with depression have an increased risk for nearly all ‘traditional’ cardiovascular risk factors, including higher prevalence of overweight/obesity, type II diabetes, hypertension, dyslipidaemia, inflammation, autonomic dysfunction, unbalanced diet, higher alcohol consumption, tobacco smoking, and low compliance to pharmacological and non-pharmacological treatments as well as higher levels of sedentary behavior (7, 18). This notion was supported by a recent review (13), in which a large number of meta-analyses were identified on the relationship between depression and excess mortality (k = 14), CVDs (k = 15), diabetes (k = 25), overweight/obesity (k = 15), and metabolic syndrome (k = 5). In summary, the association between depression and excess mortality is considered as a robust epidemiological finding, which is partly attributable to the fact that people with depression are at increased risk for CVDs and accumulate more cardiovascular risk factors. However, the causal associations between depression and CVDs are complex and to some extent bidirectional. For instance, while depression is associated with incident CVD, depression can also be a consequence of cardiac events and worsen prognosis.

Among individuals with depression, biological and behavioral factors may simultaneously contribute to the increased risk of CVD (7, 10, 16, 17). Biological mechanisms include alterations in heart function and rhythm, increased coagulability and platelet aggregation, dysregulations of the autonomic nervous system, alterations of the hypothalamic–pituitary–adrenocortical (HPA) axis, increased inflammation, and endothelial dysfunction (6, 7, 12, 19–21). Further, psychopharmacological treatments can lead to weight gain and metabolic changes (6, 21), as they contribute to obesity, dyslipidaemia, type II diabetes and other metabolic alterations (22, 23). Another factor is seen in the decreased likelihood of receiving adequate medical care (19, 23). Psychiatrists and family physicians might be more likely to oversee and/or not treat physical conditions in individuals with depression, because they might wrongly assume that physical symptoms are psychological (diagnostic overshadowing) or because individuals with depression are less able or likely to communicate their physical needs (17, 24). Although screening recommendations have existed for almost 20 years, many patients with psychiatric disorders remain unscreened with regard to cardiovascular risk factors (17, 25). Moreover, given that socioeconomic disadvantage is more common among people with SMI (16), costs associated with insurance and treatment may be further barriers (24). Finally, poor compliance with healthy lifestyle recommendations such as balanced diet, avoiding tobacco, and sufficient and regular physical activity might also contribute to pathophysiology of CVDs in people with depression (6, 7, 19, 21).

So far, most public health efforts to reduce mortality in people with SMI have been directed towards reducing suicide rates, although a large number of excess deaths are the result of chronic physical health conditions (16, 26). Nevertheless, awareness for more systematic screening, monitoring and management of existing CVDs and their risk factors in this patient population is increasing (10, 25). This may help to identify individuals with particularly high risks who need different monitoring and/or intervention schemes. Moreover, systematic assessment of modifiable risk factors (see below) that moderate the CVD risk may help to identify areas in which there is a particularly great need for action (10).

Level of lifestyle physical activity and cardiorespiratory fitness (CRF) might be such moderating factors (7). Caspersen et al. (27) defined physical activity as “any bodily movement produced by skeletal muscles that results in energy expenditure” (p. 126), whereas CRF corresponds to one specific component of health-related physical fitness (an attribute that people have or achieve). According to Caspersen et al. (27), CRF relates to the “ability of the circulatory and respiratory systems to supply fuel during sustained physical activity and to eliminate fatigue products after supplying fuel” (p. 128). The reasons why physical activity and CRF might qualify as moderating factors are as follows: first, people with depression are less likely to engage in regular physical activity, and tend to engage in more sedentary activities (7). Second, CRF is significantly decreased in people with depression (28), and significantly reduced heart rate recovery further points towards autonomic dysfunction in these patients (29). Third, poor CRF is a risk factor for CVDs and all-cause mortality (30), and improvements are associated with a reduced risk, independent of other relevant factors such as age, body composition and smoking (31). Importantly, CRF improves in a clinically meaningful way in people with depression after participation in exercise training (32, 33). Fourth, empirical evidence suggests that the CVD risk of depression is mitigated in people who engage in regular physical activity or who present with better CRF. To provide some examples, the prospective Cooper Center Longitudinal Study with 47’702 adults showed that both history of depression (HR = 1.24) and metabolic syndrome (HR = 1.28) were independent risk factors of increased mortality risk, with the highest risk found in individuals combining both conditions (HR = 1.59). By contrast, moderate (HR = 0.64) or high CRF (HR = 0.50) were associated with a significantly decreased mortality risk. Importantly, those who suffered from both depression and metabolic syndrome and showed moderate or high CRF levels had a similar mortality risk as controls with low CRF (34). Similarly, a longitudinal study with 5’888 individuals from US communities showed that individuals presenting with either depressive symptoms or low physical activity were at greater risk for premature cardiovascular mortality. The probability was even higher in those who combined both risk factors. After adjustment for further confounders, physical activity was associated with a 26% reduction of cardiovascular mortality. This rate was similar in people without and with previously established coronary heart disease (35).

Given the background presented above, the purpose of the present study was five-fold. First, to find out whether common physiological cardiovascular risk factors (waist circumference, body mass index, body fat, blood pressure, cholesterol levels, triglycerides, and blood glucose) differ between patients with depression and healthy (non-depressed) controls. Second, to examine whether patients and controls differ in CRF, and whether higher CRF is associated with a lower cardiovascular risk. Third, to examine whether cardiovascular risk is moderated by participants’ CRF level. Fourth, to examine whether within the patient sample, cardiovascular risk factors would differ between patients with mild, moderate and severe depression, after controlling for sociodemographic factors, depression history and antidepressant intake. Fifth, to examine whether the relationship between depressive symptom severity and cardiovascular risk is moderated by patients’ CRF levels.

Based on current empirical evidence, we formulated the following hypotheses. First, cardiovascular risk is increased in patients with depression compared to healthy controls (Hypothesis 1) (13). Second, healthy controls present with higher CRF than patients with depression (Hypothesis 2a); participants with better CRF have lower cardiovascular risk (Hypothesis 2b) (28). Third, differences in cardiovascular risk factors between patients with depression and healthy controls are more pronounced in participants with low CRF levels (Hypothesis 3) (34, 35). Fourth, patients with higher depressive symptom severity report a less favorable cardiovascular risk profile than counterparts with lower symptom severity (Hypothesis 4) (13). Fifth, depressive symptom severity and CRF interact, in the sense that stronger (positive) associations appear between symptom severity and cardiovascular risk in patients with poor CRF compared to counterparts with better CRF (Hypothesis 5) (34, 35).

The key findings of the present study are that compared to healthy controls, patients with depression had a higher cardiovascular risk as evident from about half of the examined indicators. Moreover, participants with good CRF had more favourable scores across nearly all risk markers than counterparts with poor CRF. For most variables, no interaction occurred between group and fitness, indicating that in patients and controls, similar differences existed between participants with poor and good CRF. Finally, within the patient sample, no major differences in the presented cardiovascular risk markers were found between patients with mild, moderate and severe depression.

Five hypotheses were formulated and each of them will now be addressed in turn. Our first hypothesis was that patients with depression have higher cardiovascular risk compared to healthy controls. This hypothesis was supported for five markers (waist circumference, body mass index, body fat, diastolic blood pressure, HbA1c), whereas no differences were found for systolic blood pressure, total, HDL and LDL cholesterol, as well as triglycerides. The present findings are in line with previous studies showing that patients with depression have a markedly higher burden of physical comorbidities compared to other age-gender-matched hospital patients (53). The increased presence of cardiovascular risk markers may explain why cardiovascular and all-cause mortality are higher in patients with depression than in the general population (6, 12). While the higher all-cause mortality in depressed people is to some extent due to higher death rates from unnatural causes such as suicides, accidents, homicides and alcohol misuse, the mortality rate is also increased for natural causes (7, 14, 24, 54). Some scientists have criticized that the persisting mortality gap denotes a cynical disregard for the lost lives among people with mental illnesses (17). Others have highlighted that the high prevalence of depression worldwide is alarming and should make people with depression a priority target population for public health strategies to prevent CVDs and mortality (7). In our study, the largest differences between patients and controls were found for BMI, waist circumference and body fat, which is in line with recent epidemiological studies in depression and bipolar disorder, especially among female participants (55, 56). This is important because depression and obesity are widely prevalent issues with considerable implications for public health (57). However, given the cross-sectional nature of our data, we need to be careful with inferring causal relationships. Thus, while it is possible that people with depression might be more likely to develop overweight through dysregulated stress systems, through unhealthy lifestyles or medication, it is also conceivable that overweight favours the development of depression through negative effects on self-image or somatic consequences (57).

Our second hypothesis was that patients would present lower CRF than healthy controls (Hypothesis 2a), and that better CRF would be associated with a more favorable cardiovascular risk factor profile (Hypothesis 2b). This hypothesis was supported by our data. Thus, our results are in line with both cross-sectional and longitudinal studies showing that people with higher CRF have lower risk of reporting/developing mental illnesses (28, 43, 58, 59). For instance, Kandola et al. (58) showed in a meta-analysis of longitudinal cohort studies that participants with low or medium CRF have a 47 and 23% increased likelihood of incidence of mental health disorders. Moreover, our findings corroborate previous studies showing that higher CRF levels are associated with decreased overall cardiovascular risk (60, 61), as well as presence of risk markers such as overweight (62, 63), hypertension (64, 65), dyslipidemia (66, 67), and diabetes (68, 69).

Only limited support was found for our third hypothesis, stating that differences in cardiovascular risk markers between patients and healthy controls would be more pronounced in participants with low CRF levels. This finding is at odds with previous studies showing that simultaneously suffering from mental disorders and having low CRF would exacerbate the risk for CVDs and CVD-related mortality (34, 35). Rather, our findings show that having poor CRF is similarly associated with higher presence of cardiovascular risk markers in both healthy people and patients with depression. This is encouraging as CRF turned out to be modifiable via regular exercise training in both healthy populations (70, 71) and people with mental illnesses (72, 73).

Our fourth hypothesis, that patients with more severe depression would report a less favorable cardiovascular risk profile than counterparts with mild or moderate depression, was not supported. Again, this is at odds with prior investigations in which patients with more severe depression were more likely to develop a CVD (9). Given this result, the prerequisites for our fifth hypothesis, that the association between depressive symptom severity and cardiovascular risk would be moderated by patients’ CRF level, were not given. We assume that this unexpected result is attributable to the fact that our sample was relatively homogeneous (all in-patients), whereas previous cohort studies were based on broader populations. However, this finding could also be due to the fact that symptom severity assessed in this study reflects a momentary state. Maybe different results would have appeared if the longitudinal course of the disease had been considered.

While the focus of this paper was on biological risk markers, it should be noted that in our study the prevalence of smoking was higher in patients with depression (41.0%) than in healthy controls (20.8%). Accordingly, smoking was considered as a potential covariate in the subsequent (M)ANCOVAs. This finding supports prior research showing that depressed people are more likely to smoke (6, 7, 19, 21–23) and to take other substances (7, 23). Most probably, the direction of the relationship is reciprocal. Thus, whereas depression increases the likelihood of smoking, depression can also reduce the probability of short-and long-term smoking cessation (7). While researchers have emphasized that smoking should be an important target for prevention because it is so common among patients with depression (15, 22, 25), there is still little evidence whether exercise training has a positive effect on smoking among people with depression (7).

From a practical point of view, different strategies seem promising to improve cardiovascular health among patients with depression. Recommended measures include intensified screening for cardiovascular risk factors. However, it is currently not well-known whether existing decision-making tools and screening models (e.g., Cox-Framingham model) work well in people with mental illnesses (74), particularly as people with mental illnesses often are younger, have abnormal (both higher and lower) blood pressure (55, 56, 75) and are more likely to smoke than the general population (25). It is therefore recommended to use specific risk prediction models for people with mental illnesses to establish better suited thresholds for offering CVD interventions (74). It has also been emphasized that improvements in health outcomes for people with mental illnesses seem unlikely if no system-wide efforts to achieving equality in health service delivery and access are undertaken (16). Possible solutions to address systemic barriers to healthcare provision could include integrated care models such as cooperation between and co-location of physical and mental health services, the use of case managers or other liaison staff to undertake a coordination role between services, or facilitated sharing of electronic health records between physical and mental health care systems (22, 25). Another strategic approach is the promotion of lifestyle interventions targeted towards promoting healthy diet or increasing lifestyle physical activity (18, 19). Maintaining a healthy body weight through a healthy diet and regular physical activity is a key component of lowering CVD risk (25). Exercise interventions should be given high priority in clinical practice, as they are not only beneficial for cardiovascular health outcomes, but also improve patients’ mental health and cognitive functioning (7, 32, 76). For instance, low fitness proved to be more closely associated with depression than fatness (77). Accordingly, it seems important to raise awareness in psychiatrists that low CRF might be a more important predictor of morbidity and mortality than overweight and obesity (62). Thus, while reducing body weight is challenging (particularly in people with mental illnesses), improvements in CRF are achievable in relatively short time (73). In line with this notion, exercise training is included as a treatment in the context of some clinical guidelines for depression (78–82). However, given that pleasant and positive feelings during exercise have an important impact on adherence (7), it is important that exercise for individuals with depression is delivered by professionals with specific experiences in mental health care (76, 83).

The strengths of the present study were that all patients were diagnosed with depression via structured clinical interview by a psychiatrist. Although structured clinical interviews are the only well-validated method to establish a clinical diagnosis of depression, in many studies, researchers used self-report questionnaires to assess depressive symptoms (12). Additionally, detailed information was collected about participants’ use of antidepressant medication, and intake was considered as potential covariate. However, the impact on antidepressants on cardiovascular risk is not entirely clear. For instance, while tricyclic anti-depressants have been shown to increase CVD risk, some uncontrolled studies found that SSRIs (serotonin reuptake inhibitors) may reduce cardiac risk (6). Randomised placebo-controlled trials, however, failed to replicate these findings (6). Another advantage was that we not only focused on differences between patients and controls, but also examined differences within the patient sample. This seemed important as some studies showed that already mild symptoms of depression might be associated with increased cardiac risk (12). Furthermore, we controlled for a wide range of potential covariates because the association between depression and CVDs is complex and influenced by various sociodemographic factors (7) and because patients and controls differed in several sociodemographic factors. Despite these strengths, some aspects need to be considered that might limit the generalizability of our findings. First, although we controlled for antidepressant intake, we were not able to control for specific type and dose of antidepressants as there was a high heterogeneity and as some patients were taking more than one antidepressant drug (84, 85). For the same reason, it was not possible to control for pre-existing diseases although data was systematically collected in the present study. As mentioned previously, the cross-sectional analysis of the data does not allow a causal interpretation of between-group differences. Furthermore, as the present study was done in an in-patient setting, findings cannot be generalized to outpatients. Finally, it should be noted that baseline data assessment took place 2–3 weeks after admission to the hospital. Within these first weeks, depression symptom severity has decreased already substantially in the in-patient sample. Thus, improvements in symptom severity may have happened at a faster pace than changes in some of the cardiovascular risk factors.

Patients with depression and healthy controls differ in several cardiovascular risk markers, putting patients at increased risk for CVDs. In contrast, people with good CRF show more favourable scores across all cardiovascular risk indicators, a relationship which was observed in both healthy controls and patients with depression. Physical health of psychiatric patients should receive the clinical attention that it deserves. Lifestyle interventions targeting healthy diet and/or physical activity are recommended as a physically active and healthy lifestyle contributes equally to patients’ mental well-being and cardiovascular health.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethikkommission Nordwest-und Zentralschweiz. The patients/participants provided their written informed consent to participate in this study.

JB, MH, CI, UL, SM, TM, AO, and NS-K supported the patient screening and recruitment processes on the four study sites. RC and J-NK recruited the participants and collected the data. RC, JB, SB, LD, AE, MH, CI, J-NK, UL, SL, SM, TM, AO, NS-K, LZ, and OF offered thematic support. MG was responsible for conceptualizing the manuscript, conducted the statistical analyses, and wrote the first draft of the manuscript. RC carried out the final language check of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the Swiss National Science Foundation (grant number 321003B-179353).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.