Minding the heart: Why are we still not closer to treating depression and anxiety in clinical cardiology practice?

Psychological comorbidities are common in patients with ischaemic heart disease (IHD) and other cardiovascular conditions, such as peripheral arterial disease, atrial fibrillation and cardiomyopathies, with as many as 20% of patients suffering from depression and/or anxiety.1 Not only a clinical diagnosis but also subthreshold levels have prognostic implications, as depression is a barrier for appropriate risk factor management, increases risk of non-adherence, refusal or drop-out from cardiac rehabilitation, readmission and premature death despite state-of-the-art treatment.2–4 Unfortunately, depression and anxiety tend to be overlooked and undertreated in cardiac patients,5,6 although depression alone in cardiac patients adds an extra cost of 33% compared with patients without depression.7 Both IHD8 and depression9,10 are associated with a high economic burden and will be among the top 10 contributors to the disease-burden worldwide in 2020.11,12 Paradoxically, despite this consistent evidence and recognition of depression, anxiety and psychosocial factors as risk factors by European and National Societies1,13 already in 2012,14 this has had little impact on changing clinical practice.15 Given that the number of patients with IHD has increased considerably due to better treatment options and ageing of the population, there is an urgent need to treat heart and mind in concert and not only the underlying somatic disease but also psychological comorbidity, in order to bridge this gap in clinical practice.16 The state-of-the-art Cochrane review by Richards et al. published in this issue may comprise an important piece of the puzzle in terms of helping us to understand why we are not closer to implementing screening for depression and anxiety combined with interventions targeting depression and anxiety in clinical practice. Although no randomized controlled trial (RCT) to date has examined the effectiveness of screening combined with intervention targeting depression and anxiety,17,18 the review by Richards et al. pinpoints other challenges, such as uncertainty about treatment effects due to low quality of evidence across different endpoints from cardiovascular morbidity and mortality to psychological outcomes. In addition, the heterogeneity between studies is significant and the level of quality varies from moderate to low and very low. Although the review demonstrates an effect of psychological interventions on cardiovascular mortality, no effects are seen on total mortality and major adverse cardiac events, such as myocardial infarction and revascularization procedures.19 In addition, only small to moderate improvements are seen in symptoms of depression, stress and anxiety, and these estimates are plagued by some uncertainty. This despite the inclusion of 35 randomized studies and 10,703 participants (median follow-up of 12 months) with data added from 14 new studies as compared with the previous Cochrane review in 2011.20 It is disappointing and of great concern to see that little has changed in the cardiovascular behavioural field in terms of demonstrated effects of psychological interventions since 2011,20 with the fear that the next update of this Cochrane review will show similar results. Although challenged health care budgets and hospital resources across the world combined with an increasing number of patients with complex chronic disease and multi-morbidities may constitute another piece of the puzzle why clinical practice has not changed, the time may be ripe for behavioural scientists and mental health professionals engaged in research and working in the cardiovascular field to take a step back and evaluate whether we are on the right path. If we take a critical look at the field, there has been a tendency to use a simplistic one-size-fits-all approach.21 However, the Cochrane review clearly demonstrates that this approach does not work, as patients are different and also have different preferences with respect to the type of intervention they are able and willing to engage in.22,23 Our trials also show an over-representation of well educated and motivated patients,24 while patients with low socio-economic status and limited resources are those who are most likely to benefit from interventions if we can get them into the trial and retain them. There is an urgent need to develop interventions for these vulnerable patients with low health literacy.25 A one-size-fits-all approach is also likely to increase the risk of non-compliance and drop-out, deflate the effects of interventions and lead to underpowered studies. Let us not forget that when designing new studies, we use the effect sizes from these ‘underpowered studies’ to calculate an expected effect size for our new study, which means that we are perpetuating ‘bad studies’ and producing more of the same! A one-size-fits-all approach will also not help us understand what works for the individual patient19 and, overall, none of these outcomes are going to take us closer to changing clinical practice and improving patient care. Richards et al. suggest that we might need to go back to evaluating the individual components to ascertain what works when delivering broader psychosocial interventions. However, risk factors, including psychosocial risk factors, cluster together, and we are living in an era where we have moved towards more complex interventions due to evidence showing that many of these risk factors and their pathways may likely interact in intricate ways that we have only just begun to unravel.2 To add to these challenges, we need to ask ourselves if something is wrong with the frameworks and paradigms that we use and whether it is possible to fit in human complexity at the cardiovascular behavioural and psychological level to paradigms that are used to demonstrate the effectiveness of a new drug. The answer to this question is not straightforward, as the RCT design is the gold standard and provides the strongest evidence. However, one way forward could be to supplement the RCT design with a user-centred phase – to enhance later implementation in practice – where patients and health care professionals are involved in designing the intervention, followed by a feasibility study to evaluate procedures, retention and attrition prior to conducting the RCT.26 Hence, rather than investing in new, large-scale, one-size-fits-all trials, we need to develop interventions that work for the individual patient in a more precision medicine approach that can be integrated with care provided by a multi-disciplinary team in routine practice.16 An open eHealth trial of tailored treatment for anxiety based on patients’ preferences shows promising results with average within-group effect sizes (Cohen’s d) of 0.80 and a very low drop-out rate of one out of 27 patients (i.e. 3.7%).23 A patient-preferred and tailored approach could be used in the framework of a stepped-care model, as in the COPES and CODIACS intervention trials targeting depression in patients post myocardial infarction. The latter studies have shown promising results in terms of larger effect sizes (Hedges’ G of 0.6) as compared with most other trials in the field, although also here the associated confidence intervals around the effect sizes are wide.27,28 Given that the stepped-care model is based on the principles that treatment should have the best chance of improving outcomes while providing the least possible burden to patients and that treatment will only be stepped up to a more intensive level if the previous level is insufficient, this model ensures that treatment is given to those patients that need it the most without over treating, increasing the likelihood of being cost-effective at a time when health care budgets are challenged. In conclusion, Richards et al. are to be commended for their efforts with respect to writing and publishing this important Cochrane review. The review not only provides an overview of current evidence and unravels gaps and challenges in the field, it also serves as a reminder to the field to take a critical look at the paths we have taken and whether we need to make adjustments in order to enhance the quality of care for patients. While we decide where to go from here, in the interim it is paramount that we do our utmost with the tools and resources that we have available to treat psychological comorbidities in patients with heart disease. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article.