Author + information
- †Cardiology Section, VA Eastern Colorado Health Care System, University of Colorado School of Medicine, Denver, Colorado
- ‡East Carolina Heart Institute Department of Cardiovascular Sciences, Brody School of Medicine at East Carolina University, Greenville, North Carolina
- ↵∗Reprint requests and correspondence:
Dr. Thomas M. Maddox, Cardiology Section, Mailstop 111B, VA Eastern Colorado Health Care System, 1055 Clermont Street, Denver, Colorado 80220.
The link between type 1 diabetes mellitus (T1DM) and cardiovascular disease is strong. Multiple studies have demonstrated that T1DM results in atherosclerosis and inflammation and confers significant risk for cardiac events and that these risks are directly related to glycemic control, as measured by glycosylated hemoglobin (HbA1c) levels (1–4). Accordingly, the American Diabetes Association clinical guidelines call for tight glycemic control among these patients to minimize cardiovascular morbidity and mortality (5).
The current study by Nyström et al. (6) supports this admonition. Using observational data from Swedish patients, the investigators examined outcomes among patients with T1DM who were undergoing coronary artery bypass graft (CABG) procedures between 1997 and 2012. These investigators found that patients with progressively higher preoperative HbA1c values had worse cardiac outcomes compared with patients with normal glycemic levels. The methods were rigorous, and the findings were convincing. These insights will provide useful information to clinicians regarding prognosis of post-CABG patients with T1DM and to researchers needing baseline risks to inform sample size calculations in interventional trials.
Several studies have examined the effect of diabetes and preoperative diabetic control on outcomes after CABG, so what does this study add to the published data? We believe that it has 2 important contributions. First, this study investigates the incidence of long-term adverse events in T1DM after CABG. Second, this study illustrates the potential of health data to fuel both clinical practice and research in an iterative fashion. This approach—a fundamental tenet of “learning health care systems”—can overcome the historical divide between practice and research, fuel quicker translation of research insights into clinical practice, and bring front-line clinical insights into the conduct of clinical research (7).
The study by Nyström et al. (6) used data from the SWEDEHEART system (Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies) (8). Established in 2009, the national Web-based clinical registry collects data on patients with coronary artery disease. It merged existing registries of coronary care units, coronary angiography and interventions, cardiac surgical procedures, and cardiac secondary prevention into a cohesive, longitudinal registry for patients hospitalized with acute coronary syndrome or undergoing coronary or valvular intervention. Importantly, this was facilitated by the unique personal identity number assigned to every Swedish citizen. The SWEDEHEART data are also combined with national databases of vital status, hospitalizations, and pharmacy data, thus allowing for a comprehensive view of patient care and outcomes. A central coordinating center provides data and quality oversight.
SWEDEHEART advances care improvements in coronary artery disease by informing both practice and research. Participating caregivers receive interactive process and outcomes data reports for their patients. These data are also used for intrahospital and interhospital comparisons, for public reporting, to identify areas for quality improvement, and to monitor the impact of improvement projects over time. SWEDEHEART also makes these same data available for research; this study by Nyström et al. is but one example. SWEDEHEART also serves as an infrastructure for pragmatic randomized controlled trials. As demonstrated by the recent TASTE (Thrombus Aspiration in ST-Elevation in Myocardial Infarction in Scandinavia) trial (9), the registry identified potential enrollees for the trial and collected endpoints and other data during the trial. These measures dramatically improved the trial’s efficiency and reduced its costs.
For practitioners in the United States, the SWEDEHEART registry also exemplifies many of the “learning health care system” concepts codified in the Institute of Medicine report, Best Care at Lower Cost (7). Improvements in digital infrastructure to capture clinical data more accurately to inform practice and evidence, accelerated incorporation of the best clinical knowledge into care decisions by cross-platform data integration, improvements in care coordination and communication across different care settings, and an increase in transparency of health system performance are all aspects of the SWEDEHEART registry approach.
Do efforts currently exist in the United States to develop a similar system? In the cardiovascular domain, the Society of Thoracic Surgeons National Database and the American College of Cardiology’s National Cardiovascular Data Registry (NCDR) programs both host hospital-based cardiac clinical registries that provide valuable performance and benchmarking information for quality improvement and public reporting (10). Similarly, the Veterans Affairs (VA) Clinical Assessment, Reporting, and Tracking (CART) program uses software integrated into the VA’s electronic health record (EHR) to collect information on all its coronary intervention procedures and uses the data for quality improvement and research activities (11). Other efforts exist outside the cardiovascular realm. Large, national registries, such as the Health Maintenance Organization Research Network (HMORN), the National Patient-Centered Clinical Research Network (PCORnet), and the Food and Drug Administration’s Mini-Sentinel program use EHR-based data from participating health systems to conduct clinical research, pragmatic trials, and safety monitoring programs (12). Various commercial efforts also collect clinical data for quality improvement, clinical research, and cost reduction.
Ideally, these registries would fuel a cohesive data stream of clinical, quality, operational, administrative, and cost data to facilitate care quality and clinical research (13). Unfortunately, they are not currently designed to deliver this infrastructure. For example, most of the registries are not yet integrated into the clinical workflows of EHR systems and capture only a subset of patients over a limited time frame. In addition, cross-platform registry integration for clinical research is hampered by registry infrastructure limitations and the lack of a unique patient identifier in the United States. Finally, much additional work is needed to establish a supportive leadership and regulatory environment for learning health care systems.
The study by Nyström et al. uniquely documents the importance of preoperative glycemic control before CABG in patients with T1DM. However, its larger importance is in drawing our attention to the very real potential for learning health care systems here in the United States. This Swedish study provides a promising example of the power of the ability of learning health care systems to generate new insights, to translate these insights into quality improvement programs for certain populations, and to trigger new research to investigate optimal treatment strategies and targets. U.S. learning health care system initiatives modeled on SWEDEHEART could fuel a “virtuous cycle” of care improvement and research in cardiovascular disease.
↵∗ 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. Maddox is the associate director of the Veterans Affairs Clinical Assessment, Reporting, and Tracking (VA CART) program. Dr. Ferguson has served on the Institute of Medicine panel that authored Best Care at Lower Cost. The views represented are not necessarily those of the U.S. government. Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Nathan D.M.,
- Bayless M.,
- Cleary P.,
- et al.
- ↵(2015) Standards of medical care in diabetes—2015: summary of revisions. Diabetes Care 38(Suppl):S4.
- Nyström T.,
- Holzmann M.J.,
- Eliasson B.,
- Kuhl J.,
- Sartipy U.
- Committee on the Learning Health Care System in America; Institute of Medicine
- Jernberg T.,
- Attebring M.F.,
- Hambraeus K.,
- et al.
- Messenger J.C.,
- Ho K.K.L.,
- Young C.H.,
- et al.
- Ferguson T.