Abstract
Regular exercise lowers the risk of cardiovascular death in coronary heart disease (CHD) patients. We aimed to investigate regular exercise behaviour and intention in relation to symptoms of anxiety and depression in CHD patients across Europe.
This study was based on a multicentre cross-sectional survey.
In the EUROpean Action on Secondary and Primary Prevention through Intervention to Reduce Events (EUROASPIRE) III survey, 8966 CHD patients <80 years of age from 22 European countries were interviewed on average 15 months after hospitalisation. Whether patients exercised or intended to exercise regularly was assessed using the Stages of Change questionnaire in 8330 patients. Symptoms of anxiety and depression were evaluated using the Hospital Anxiety and Depression Scale. Total physical activity was measured by the International Physical Activity Questionnaire in patients from a subset of 14 countries.
Overall, 50.3% of patients were not intending to exercise regularly, 15.9% were intending to exercise regularly, and 33.8% were exercising regularly. Patients with severe symptoms of depression less frequently exercised regularly than patients with symptoms in the normal range (20.2%, 95% confidence interval (CI) 14.8–26.8 vs 36.7%, 95% CI 29.8–44.2). Among patients not exercising regularly, patients with severe symptoms of depression were less likely to have an intention to exercise regularly (odds ratio 0.62, 95% CI 0.46–0.85). Symptoms of anxiety did not affect regular exercise intention. In sensitivity analysis, results were consistent when adjusting for total physical activity.
Lower frequency of regular exercise and decreased likelihood of exercise intention were observed in CHD patients with severe depressive symptoms. Severe symptoms of depression may preclude CHD patients from performing regular exercise.
Introduction
Physical inactivity is a leading cause of the global disease burden and, along with smoking, one of the most important modifiable risk factors.1 Convincing epidemiological evidence supports the conclusion that regular exercise decreases the risk of cardiovascular disease.2 The benefit of regular exercise in the secondary prevention of coronary heart disease (CHD) is well documented: aerobic exercise training reduces the medium to long-term risk of cardiovascular death by 25% when compared to usual care.3 Indeed, exercise and drug interventions may be equally effective in the prevention of mortality outcomes among CHD patients.4 The long-term benefit of regular exercise among CHD patients increases with the number of weekly sessions.5 Higher exercise levels elicit greater cardio-protective effects and more strongly reduce the cardiovascular risk of CHD patients.6 The low risk of cardiovascular events during high- and moderate intensity exercise training supports its adoption for secondary prevention.7 Accordingly, patients with a previous acute coronary event or procedure should undergo regular exercise of moderate-to-vigorous intensity.8
Elevated symptoms of anxiety and depression worsen the clinical course and prognosis of cardiovascular disease, but may not represent independent cardiovascular risk factors.9,10 Psychobiological mediators possibly linking anxiety and depression with cardiovascular outcomes in CHD patients are unfavourable lifestyle factors: in particular, lack of regular exercise has been shown to largely explain the adverse association between depressive symptoms and cardiovascular events.11 Moreover, regular exercise and symptoms of anxiety and depression seem to be interrelated: notably, depression and anxiety symptoms have been prospectively associated with reduced levels of physical exercise.12,13 However, we are not aware of any prior multicentre study that has systematically evaluated to what extent symptoms of anxiety and depression impact regular exercise among CHD patients. The aim of the present study, therefore, was to investigate regular exercise behaviour and intention in relation to symptoms of anxiety and depression in CHD patients across Europe. For this purpose, we took the opportunity to analyse data from the EUROpean Action on Secondary and Primary Prevention through Intervention to Reduce Events (EUROASPIRE) III survey conducted in patients previously hospitalised for coronary events and procedures.
Methods
Study participants
The EUROASPIRE III survey among CHD patients is a cross-sectional study conducted in 2006–2007 in 76 study centres in 22 European countries. Patients hospitalised for coronary artery bypass graft (CABG), percutaneous coronary intervention (PCI), acute myocardial infarction (AMI) and myocardial ischaemia were retrospectively identified from hospital records.14 Trained research staff reviewed 13,935 medical records and interviewed 8966 patients <80 years of age at least six months after hospitalisation for their index event or procedure. Information on study centres and organisations is provided in the Supplementary Material, Material S1. The study was approved by local ethics committees. Study participants provided written informed consent.
Regular exercise behaviour and intention
Whether patients exercised or intended to exercise regularly was assessed using the Stages of Change (SoC) short-form questionnaire. Figure 1 shows the questionnaire including the definition of regular exercise, question and answer choices, and the scoring scheme. Study centres in all 22 countries that participated in EUROASPIRE III contributed SoC data for analysis. We categorised patients not intending to exercise regularly (precontemplation stage), patients intending to exercise regularly in the next six months or 30 days (contemplation and preparation stage, respectively) and patients exercising regularly for less or more than six months (action and maintenance stage, respectively).15,16
Stages of Change short-form questionnaire.
Symptoms of anxiety and depression
Symptoms of anxiety and depression during the last seven days were evaluated using the Hospital Anxiety and Depression Scale (HADS). The HADS is a four-point Likert scale (range 0–3) with seven items being related to anxiety and seven items to depression. Item scores are added to obtain the summary scores on anxiety (HADS-A) and depression (HADS-D) separately, ranging between 0–21. A score <8 is considered to be in the normal range, scores from 8–10 (moderate symptoms) indicate a possible disorder and a score ≥11 (severe symptoms) indicates a probable disorder.17
Covariates
Covariates were socio-demographic and clinical characteristics (age at interview, sex, educational level, current smoking, diabetes, obesity) and factors possibly influencing physical performance (diagnostic group, history of stroke, recurrent CHD, cardiac rehabilitation, longstanding illness, disability or infirmity, time since hospitalisation). Ten-year age groups were used in the analysis. Educational levels considered were primary, secondary, intermediate and higher education. Smoking was self-reported and validated by carbon monoxide in breath (Bedfont Scientific, Model Micro 4 Smokerlyser, Kent, UK) using ≤10 ppm as cut-off value. Diagnosis of diabetes was reported by the patient. Body height and weight were measured using calibrated instruments. Obesity was defined as a body mass index ≥30 kg/m2. Information on cardiac rehabilitation was assessed by asking: ‘Were you advised to follow a cardiac prevention/rehabilitation programme within three months of discharge following the index event or procedure?’ and ‘If yes, did you attend the cardiac rehabilitation programme offered?’ Diagnostic groups were patients’ index events or procedures. History of stroke was self-reported stroke prior and/or following hospital admission for the index event or procedure. Recurrent CHD was self-reported hospitalisation for PCI, CABG, AMI or myocardial ischaemia since the index event. Longstanding illness, disability or infirmity was assessed by asking: ‘Do you have any long-standing illness, disability or infirmity? By long- standing it is meant anything that has troubled you over a period of time, or that is likely to affect you over a period of time?’ Total physical activity in the last seven days was assessed using the International Physical Activity Questionnaire (IPAQ) short-form in patients from 14 countries.18
Statistical analysis
Mean percentages of SoC by HADS-A and HADS-D and odds ratios (ORs) of intending to exercise regularly with 95% confidence intervals (CIs) were estimated using logistic regression analysis. Models allowed for random variation in regression intercepts between countries. To estimate the effects of HADS-A and HADS-D on regular exercise intention, a minimal sufficient adjustment set that comprised the above-mentioned covariates was determined using directed acyclic graphs (Supplementary Material, Figure S1). In sensitivity analysis, multivariable associations of HADS-A and HADS-D with exercise intension were examined in patients from the subset of 14 countries with IPAQ data. All patients without missing data on HADS-A and HADS-D were considered in univariate analyses and patients without missing data on all covariates entered the multivariable analysis. Statistical tests were two-tailed, and an alpha level of 0.05 was chosen to indicate statistical significance. SAS software version 9.4 (SAS Institute Inc., Cary, North Carolina, USA) was used.
Results
Study population
Overall, the number of patients with available data on SoC was 8330 (Supplementary Material, Figure S2, shows the study population by data availability). Compared to patients with missing/unclassifiable SoC data, patients with available SoC data showed lower HADS-A (median (interquartile range (IQR)), 5 (3–8) vs 6 (4–10)) and HADS-D (median (IQR), 4 (2–7) vs 5 (3–8)) scores.
Frequency distributions of SoC
Overall, 50.3% of patients (in precontemplation stage) were not intending to exercise regularly, 15.9% (in contemplation and preparation stage) were intending to exercise regularly, and 33.8% (in action and maintenance stage) were exercising regularly. The mosaic plots in Figure 2 show the frequency distribution of SoC by HADS-A and HADS-D score. The frequency of exercising regularly decreases and the frequency of not intending to exercise regularly increases with increasing HADS-D score (Figure 2(b)). This distribution pattern is less pronounced with increasing HADS-A score (Figure 2(a)). Compared to patients with symptoms of depression in the normal range (HADS-D < 8), patients with severe symptoms (HADS-D ≥ 11) less frequently exercised regularly (20.2%, 95% CI 14.8–26.8 vs 36.7%, 95% CI 29.8–44.2). Table S1 in the Supplementary Material provides further mean percentages of SoC by HADS-A and HADS-D.
Mosaic plots showing the frequency distribution of Stages of Change by (a) Hospital Anxiety and Depression Scale-Anxiety (HADS-A) (n = 8000) and (b) Hospital Anxiety and Depression Scale-Depression (HADS-D) (n = 8039) score. 1 = Precontemplation stage (not intending to exercise regularly); 2 = contemplation and preparation stage (intending to exercise regularly); 3 = action and maintenance stage (exercising regularly).
Associations of HADS-A and HADS-D with exercise intention
Among the subgroup of patients not exercising regularly, Table 1 presents univariate and multivariable associations of HADS-A and HADS-D with exercise intention. In the univariate analysis, severe symptoms of anxiety (OR 0.76, 95% CI 0.62–0.93) and moderate (OR 0.71, 95% CI 0.58–0.85) and severe (OR 0.53, 95% CI 0.41–0.68) symptoms of depression were inversely associated with intending to exercise regularly. Table S2 in the Supplementary Material shows univariate associations of covariates with intending to exercise regularly.
Univariate and multivariable associations of Hospital Anxiety and Depression Scale-Anxiety (HADS-A) and Hospital Anxiety and Depression Scale-Depression (HADS-D) with intending to exercise regularly in patients not exercising regularly.
| Univariate associations | Multivariable associations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | n | OR | 95% CI | p-value | N | n | OR | 95% CI | p-value | ||
| HADS-A | <8 | 3510 | 891 | 1.00 | Reference | – | 3351 | 858 | 1.00 | Reference | – |
| 8–10 | 980 | 226 | 0.90 | 0.75–1.07 | 0.228 | 936 | 216 | 1.06 | 0.87–1.29 | 0.562 | |
| ≥11 | 789 | 163 | 0.76 | 0.62–0.93 | 0.009 | 739 | 149 | 0.99 | 0.77–1.27 | 0.951 | |
| HADS-D | <8 | 3879 | 1016 | 1.00 | Reference | – | 3683 | 974 | 1.00 | Reference | – |
| 8–10 | 889 | 180 | 0.71 | 0.58–0.85 | 0.001 | 847 | 172 | 0.82 | 0.67–1.01 | 0.067 | |
| ≥11 | 538 | 85 | 0.53 | 0.41–0.68 | <0.001 | 496 | 77 | 0.62 | 0.46–0.85 | 0.003 | |
| Univariate associations | Multivariable associations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | n | OR | 95% CI | p-value | N | n | OR | 95% CI | p-value | ||
| HADS-A | <8 | 3510 | 891 | 1.00 | Reference | – | 3351 | 858 | 1.00 | Reference | – |
| 8–10 | 980 | 226 | 0.90 | 0.75–1.07 | 0.228 | 936 | 216 | 1.06 | 0.87–1.29 | 0.562 | |
| ≥11 | 789 | 163 | 0.76 | 0.62–0.93 | 0.009 | 739 | 149 | 0.99 | 0.77–1.27 | 0.951 | |
| HADS-D | <8 | 3879 | 1016 | 1.00 | Reference | – | 3683 | 974 | 1.00 | Reference | – |
| 8–10 | 889 | 180 | 0.71 | 0.58–0.85 | 0.001 | 847 | 172 | 0.82 | 0.67–1.01 | 0.067 | |
| ≥11 | 538 | 85 | 0.53 | 0.41–0.68 | <0.001 | 496 | 77 | 0.62 | 0.46–0.85 | 0.003 | |
CI: confidence interval; OR: odds ratio.
N corresponds to the total number of patients, n corresponds to the number of patients intending to exercise. ORs with 95% CIs from logistic regression analysis with not intending to exercise regularly as the reference category and country modelled as a random effect. Multivariable associations from a single logistic regression model adjusted for age at interview, sex, diagnostic group, educational level, current smoking, diagnosis of diabetes, obesity, cardiac rehabilitation, history of stroke, recurrent coronary heart disease, longstanding illness, disability or infirmity, and time since hospitalisation.
Univariate and multivariable associations of Hospital Anxiety and Depression Scale-Anxiety (HADS-A) and Hospital Anxiety and Depression Scale-Depression (HADS-D) with intending to exercise regularly in patients not exercising regularly.
| Univariate associations | Multivariable associations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | n | OR | 95% CI | p-value | N | n | OR | 95% CI | p-value | ||
| HADS-A | <8 | 3510 | 891 | 1.00 | Reference | – | 3351 | 858 | 1.00 | Reference | – |
| 8–10 | 980 | 226 | 0.90 | 0.75–1.07 | 0.228 | 936 | 216 | 1.06 | 0.87–1.29 | 0.562 | |
| ≥11 | 789 | 163 | 0.76 | 0.62–0.93 | 0.009 | 739 | 149 | 0.99 | 0.77–1.27 | 0.951 | |
| HADS-D | <8 | 3879 | 1016 | 1.00 | Reference | – | 3683 | 974 | 1.00 | Reference | – |
| 8–10 | 889 | 180 | 0.71 | 0.58–0.85 | 0.001 | 847 | 172 | 0.82 | 0.67–1.01 | 0.067 | |
| ≥11 | 538 | 85 | 0.53 | 0.41–0.68 | <0.001 | 496 | 77 | 0.62 | 0.46–0.85 | 0.003 | |
| Univariate associations | Multivariable associations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | n | OR | 95% CI | p-value | N | n | OR | 95% CI | p-value | ||
| HADS-A | <8 | 3510 | 891 | 1.00 | Reference | – | 3351 | 858 | 1.00 | Reference | – |
| 8–10 | 980 | 226 | 0.90 | 0.75–1.07 | 0.228 | 936 | 216 | 1.06 | 0.87–1.29 | 0.562 | |
| ≥11 | 789 | 163 | 0.76 | 0.62–0.93 | 0.009 | 739 | 149 | 0.99 | 0.77–1.27 | 0.951 | |
| HADS-D | <8 | 3879 | 1016 | 1.00 | Reference | – | 3683 | 974 | 1.00 | Reference | – |
| 8–10 | 889 | 180 | 0.71 | 0.58–0.85 | 0.001 | 847 | 172 | 0.82 | 0.67–1.01 | 0.067 | |
| ≥11 | 538 | 85 | 0.53 | 0.41–0.68 | <0.001 | 496 | 77 | 0.62 | 0.46–0.85 | 0.003 | |
CI: confidence interval; OR: odds ratio.
N corresponds to the total number of patients, n corresponds to the number of patients intending to exercise. ORs with 95% CIs from logistic regression analysis with not intending to exercise regularly as the reference category and country modelled as a random effect. Multivariable associations from a single logistic regression model adjusted for age at interview, sex, diagnostic group, educational level, current smoking, diagnosis of diabetes, obesity, cardiac rehabilitation, history of stroke, recurrent coronary heart disease, longstanding illness, disability or infirmity, and time since hospitalisation.
In the multivariable analysis, only severe depressive symptoms decreased the likelihood of intending to exercise regularly (OR 0.62, 95% CI 0.46–0.85), whereas the univariate associations of severe anxiety and moderate depressive symptoms were attenuated (Table 1). No statistically significant interaction was observed between HADS-A and HADS-D.
Sensitivity analysis
The sensitivity analysis included patients from the subset of 14 countries with IPAQ data (total physical activity, n = 4438). Overall, 23.1%, 37.6% and 39.4% respectively had a low, moderate and high IPAQ level. Patients with severe symptoms of depression more frequently had a low IPAQ level than patients with depressive symptoms in the normal range (41.4%, 95% CI 30.2–53.8 vs 19.3%, 95% CI 13.2–27.3, Supplementary Material, Table S3).
When further adjusting multivariable associations of HADS-A and HADS-D with exercise intention for IPAQ categories in patients not exercising regularly (n = 2632), similar results were observed as in the main analysis: in patients with severe depressive symptoms, the likelihood of intending to exercise regularly was lower as compared to patients with symptoms in the normal range (OR 0.60, 95% CI 0.39–0.91, p = 0.017).
Discussion
In the present study, we report on the frequency of regular exercise behaviour and intention and on associations of exercise intention with regard to symptoms of anxiety and depression among patients with CHD across Europe. Only one-fifth of patients with severe symptoms of depression exercised regularly, compared to almost two-fifths of patients with symptoms in the normal range. Severe depressive symptoms had a detrimental effect on planned regular exercise, lowering the likelihood of intending to exercise regularly by 40%. Symptoms of anxiety did not independently affect intention to exercise regularly.
Previous studies
To the best of our knowledge, this is the first multicentre study investigating the association of anxiety and depression with regular exercise in CHD patients. Our results support the findings of a previous study that prospectively investigated depressive symptoms and regular exercise in 502 patients hospitalised for coronary syndromes in the UK: HADS-D score prior to hospital discharge was inversely associated with taking regular exercise 12 months later.12 Our results further concur with findings of a prospective analysis in the Whitehall II study, where symptoms of anxiety and/or depression at baseline decreased the probability of exercising regularly as assessed twice during an eight-year follow-up period.13 Of note, this study also showed that regular exercise at baseline reduced the likelihood of depressive symptoms at follow-up examinations, suggesting that the association between depressive symptoms and regular exercise is bi-directional. Indeed, other studies in general population samples similarly observed prospective associations between regular exercise and incident depressive symptoms.19,20 Accordingly, we refrained from examining cross-sectional associations of symptoms of anxiety and depression with current regular exercise, and analysed associations of HADS with intention to exercise regularly, excluding patients who already exercised. Our data show a marked difference in regular exercise between patients with severe symptoms of depression and patients with symptoms in the normal range as well as a considerable effect of severe depressive symptoms on intention to exercise regularly. This suggests that depressive symptoms represent a major barrier to exercise behaviour change in CHD patients. Investigating the directionally opposite association than the one investigated in the present study, our group previously reported higher HADS-A and HADS-D scores in overweight/obese patients intending to exercise regularly when compared to patients not intending to become physically active.21 Thus, in line with general population studies, previous results from EUROASPIRE together with the findings of the present study provide further evidence for a bi-directional association between depressive symptoms and regular exercise.
Stages of change
Among patients with stable CHD, aerobic exercise training is recommended ≥3 times a week and 30 min per session by current European guidelines on cardiovascular disease prevention.8 This recommendation is largely consistent with the assessment of regular exercise by the SoC questionnaire (3–5 times per week for 20–60 min per session). According to the transtheoretical model, individuals advance through the ‘stages of change’ (from which the name of the SoC questionnaire is derived) until they successfully adapt a new health behaviour.22 A study validating SoC against exercise behaviour demonstrated accurate stage distinction by moderate and strenuous exercise.23 A meta-analysis of studies investigating predictors of exercise behaviour showed a large effect size for the relationship between exercise intention and actual exercise uptake.24 A meta-analysis of studies on regular exercise SoC showed that stage membership well defines the progress of change, concluding that future studies should investigate the moderators and mediators of stage transition.25 In this regard, the present study identified severe depressive symptoms as a barrier to advancement from the precontemplation stage (not intending) to the contemplation and preparation stage (intending to exercise).
The level of total physical activity (not only exercise, but all types of activities) is a possible confounding factor in the association of symptoms of anxiety and depression with intention to exercise regularly. As outlined above, it was measured using the IPAQ, which covers a wide range of physical activity domains and is recommended for use in large study samples.26 In sensitivity analysis, a low IPAQ level was more frequently observed in patients with severe depressive symptoms. Importantly, however, the multivariable association of severe depressive symptoms with intention to exercise regularly persisted after further adjustment for IPAQ level, indicating that the effect is independent from total physical activity.
Implications and perspectives
Our results show that a majority of CHD patients are not intending to exercise regularly and emphasise the need for novel preventive strategies to enhance motivational readiness and finally lifestyle change. Randomised controlled trials on the efficacy of SoC-matched exercise interventions have shown positive effects on regular exercise in healthy individuals and CHD patients.27,28 A systematic review of intervention studies to increase adherence to regular exercise in cardiac rehabilitation concluded – due to heterogeneous results – that practice recommendations could not be made.29 The authors of this study, however, inferred from their findings that ‘individually tailored approaches may increase the likelihood of success’. Based on a large sample of CHD patients from 22 countries, our study results could be helpful in designing future behavioural interventions for exercise promotion that may need to be specifically tailored to patients with severe symptoms of depression. As an important take-home message for cardiologists, general practitioners, specialist nurses and other allied professionals, we would like to emphasise the adverse association of depressive symptoms with regular exercise in CHD patients.
Limitations
As study limitations we note the following. Firstly, regular exercise was assessed simultaneously with symptoms of anxiety and depression. Secondly, the thresholds used by the instruments for assessing regular exercise and total physical activity ignore the benefit of low-volume exercise among individuals at risk of cardiovascular disease. Thirdly, we lacked medical data on functional status and relied on information provided by the patients on long-standing illness, disability or infirmity. Fourthly, our study sample can be looked at as survivors of a hypothetical cohort of CHD patients. Back in time, it is likely that these survivors exercised and intend to exercise more frequently and exhibited severe symptoms of depression less frequently than non-survivors. Thus, our study may overestimate the frequency of regular exercise and exercise intention, but underestimate the association of HADS-D with exercise intention in that hypothetical cohort.
Conclusion
In conclusion, our results indicate that severe symptoms of depression may preclude CHD patients from performing regular exercise. Strategies for regular exercise promotion may need to specifically target CHD patients with severe depressive symptoms across Europe.
Author contribution
CP, JW, JH, DDB, DDS, GDB, ZR, ZF, KK, DW and UK contributed to the conception or design of the work. CP, JW, JH, DDB, DDS, GDB, ZR, JPE, ZF, DG, CJ, KK and UK contributed to the acquisition, analysis or interpretation of data for the work. CP drafted the manuscript. JW, JH, DDB, DDS, GDB, ZR, JPE, ZF, DG, CJ, KK and UK critically revised the manuscript. All gave final approval. CP, JW, JH, DDB, DDS, GDB, ZR, ZF, DG, KK, DW and UK agree to be accountable for all aspects of work ensuring integrity and accuracy.
Acknowledgements
The EUROASPIRE Study Group is grateful to all the hospitals in which the study was carried out. Their administrative staff, physicians, nurses and other personnel helped the authors in many ways and they very much appreciate this. The authors are particularly grateful to the patients who participated in the study.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: D De Smedt was financially supported by the Research Foundation Flanders. Z Reiner reports personal fees from AstraZeneca, personal fees from Abbott, personal fees from Aegirion, and other fees from Sanofi, outside the submitted work; K Kotseva reports grants from the European Society of Cardiology, during the conduct of the study; grants from Hoffman La Roche, from Boehringer Ingelheim, outside the submitted work; D Wood reports grants from AstraZeneca, Bristol Myers Squibb, Glaxo SmithKline, Pfizer, Sanofi-Aventis, Servier Laboratories, Merck, Sharp and Dohme (MSD)/Schering-Plough, personal fees from Bristol Myers Squibb Merck, Sharp and Dohme/Schering Plough, Pfizer, Zentiva, during the conduct of the study; grants from Pfizer, grants from Hoffman La Roche, Merck Sharp and Dohme, Bristol-Myers Squibb Emea Sarl, AstraZeneca, GlaxoSmithKline, personal fees from AstraZeneca, Merck Sharp and Dohme, Kowa Pharmaceuticals, Zentiva, Ranexa, Sanofi, Pfizer, personal fees from Merck Sharp and Dohme, outside the submitted work; U Keil reports grants from Pfizer, MSD and AstraZeneca, for projects generated within EUROASPIRE III. All other authors declare that there is no conflict of interest.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by unconditional educational grants provided to the European Society of Cardiology by the following companies. Main sponsors: AstraZeneca; Bristol-Myers Squibb; GlaxoSmithKline; Pfizer; Sanofi-Aventis; and Servier; sponsors: Merck/Schering-Plough; and Novartis.
Comments