Author + information
- William Whang, MD, MS∗ ()
- ↵∗Address for correspondence:
Dr. William Whang, Icahn School of Medicine at Mount Sinai, Helmsley Electrophysiology Center, One Gustave L. Levy Place, Box 1030, New York, New York 10029.
In their book based on behavioral economics, Thaler and Sunstein write that “Doctors are crucial choice architects, and with an understanding of how Humans think, they could do far more to improve people’s health and thus to lengthen their lives” (1). As a cardiologist, this quote brings to mind the daunting set of choices that patients with atrial fibrillation (AF) face given the expanding array of treatments. Discussions with AF patients may involve everything from lifestyle changes to anticoagulation to antiarrhythmic medication to catheter ablation to appendage closure to cardiac surgery. As health care providers, we frame the context in which patients make life-impacting decisions based on our perspective of the evidence.
In this issue of the Journal, Perino et al. (2) examine variations in outcomes of AF treatment associated with cardiology versus primary care. They studied a large Veterans Health Administration database of patients with newly diagnosed nonvalvular AF between 2004 and 2012. They used an indicator for early cardiology care consisting of an appointment in a cardiologist clinic within 90 days of diagnosis. This approach may or may not reflect the extent to which cardiology treatment was received; for instance, several visits with a cardiologist that start at month 6 after diagnosis would not be considered cardiology care in this analysis. Nonetheless, being seen in a cardiology clinic was associated with an impressively lower risk of stroke (hazard ratio: 0.91; 95% confidence interval: 0.86 to 0.96) and death (hazard ratio: 0.89; 95% confidence interval: 0.88 to 0.91) compared with being seen only by primary care during the first 90 days’ post-diagnosis.
Overall, 60% received primary care “only,” and 40% received cardiology care (2). The cardiology-treated patients were sicker, with higher comorbidity indices, and greater proportions with heart failure (19.8% vs. 13.3%) and previous myocardial infarction (35.6% vs. 29.7%). Because the study design was observational, the authors adjusted for potential confounders, including patients’ distance to their clinic, Veterans Affairs priority status, comorbidity index, diabetes, heart failure, hypertension, myocardial infarction, stroke/transient ischemic attack, and cardiovascular medications. In addition, they performed a separate analysis with propensity matching that confirmed the essential findings in the multivariable models.
In interpreting this study (2), caveats to consider include that the analysis was performed in the context of the integrated delivery system that comprises the Veterans Health Administration, among a mostly male cohort, and in the setting of very little use of novel oral anticoagulant agents (≤1%). Nonetheless, the findings in this study are thought-provoking and prompt questions about the mechanisms. In fact, the authors performed a mediation analysis that examined the extent to which anticoagulation prescription could explain the differences in outcomes. Cardiology-treated patients were more likely to have received oral anticoagulation therapy by 90 days’ post-diagnosis (70.3% vs. 58.8%). However, the indirect effect of oral anticoagulation receipt on stroke was rather modest (hazard ratio: 0.957; 95% confidence interval: 0.909 to 1.007) and in the article’s Online Table 3 (2), there seems to have been little change in the hazard ratio for stroke associated with cardiology care after addition of anticoagulation to the model (0.819 to 0.823), suggesting that other variables are more likely to explain the decreased stroke risk. Also, analyses of mortality did not find that greater anticoagulation prescription mediated the mortality differences, again suggesting that a different mechanism was behind this association.
If sicker patients were seen by cardiology, what else besides early anticoagulation therapy could explain the better outcomes after diagnosis with AF (2)? There are possible clues in Table 1 in the comparison between the cardiology and primary care groups, including the greater use of medications such as beta-blockers, statins, and angiotensin-converting enzyme inhibitors. Also, hospitalization for supraventricular tachycardia/AF was more common among cardiology-treated patients, suggesting that overall aggressiveness of care was greater in this group. It would be interesting to understand whether rhythm control attempts in the cardiology-treated group played a mediator role in any of the outcome differences. Another important factor that may explain the findings involves unmeasured patient characteristics that correlate with specialty care. Patients who were more motivated to follow treatment recommendations may have sought out referral to cardiologists and self-selected for better prognosis.
Perino et al. (2) suggest that pragmatic trials evaluating AF specialty clinics and novel care models for newly diagnosed AF should be considered as a next step. Although anticoagulation treatment was not necessarily a primary factor in the outcome improvement in the present data, it is still a proven therapy to prevent stroke and may be a source of low-hanging fruit to improve outcomes. For instance, a cooperative pathway delivered in 56 primary care practices in the United Kingdom offered consultation to AF patients in the primary care setting and improved the proportion of eligible patients who received anticoagulation therapy (3). AF clinics in an integrated management model may also offer greater dissemination of guideline-based therapies (4). Such efforts may ultimately make an important difference, although they have yet to be studied fully (5).
In the meantime, the study by Perino et al. (2) makes an important statement that cardiologists play a critical role in framing the decisions that patients face when diagnosed with AF. Despite the growing administrative hurdles and outside noise for us and our patients, we cannot forget that as choice architects, our efforts may make all the difference.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Whang has reported that he has no relationships relevant to the contents of this paper to disclose. P.K. Shah, MD, served as Guest Editor-in-Chief for this paper. Ivan Ho, MD, served as Guest Editor for this paper.
- 2017 American College of Cardiology Foundation
- Thaler R.H.,
- Sunstein C.R.
- Perino A.C.,
- Fan J.,
- Schmitt S.K.,
- et al.
- Das M.,
- Panter L.,
- Wynn G.J.,
- et al.
- Carter L.,
- Gardner M.,
- Magee K.,
- et al.
- Kirchhof P.,
- Breithardt G.,
- Bax J.,
- et al.