Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
Big data analytical applications have tremendous potential to improve the overall quality of cardiovascular (CV) care, but translating that potential into clinical practice comes with significant challenges.
The amount of data being collected from patients’ electronic health records is enormous and ultimately could be used for purposes such as predictive risk modeling, drug and medical device safety, disease and treatment heterogeneity, precision medicine, and public health and research applications.1 Smart devices are also yielding large amounts of important health-related data.
Thus far, companies such as Amazon and others have incorporated learned digital information into the overall user experience, but fully processing and analyzing big data to improve healthcare outcomes is not nearly as advanced.
Challenges for implementing the potential applications of big data in CV care include the need for evidence of effectiveness and safety; methodologic issues, such as data quality and validation; and clinical integration and viability.1 Visualization of the data will ultimately be needed so that clinicians and practitioners can access and use the information.2
Successful implementation of big data for CV population health management will require a multidisciplinary approach, including investing in big data platforms, harnessing technology to create novel digital applications, developing digital solutions that can inform the actions of clinical and policy decision makers, and optimizing engagement strategies with the public and patients, according to Nasir et al.3
Some large cardiology datasets include the American College of Cardiology’s National Cardiovascular Data Registry, the Agency for Healthcare Research and Quality’s Healthcare Cost and Utilization Project databases, the Society of Thoracic Surgeons’ National Database, the American Heart Association’s Get With the Guidelines databases, and the European Society of Cardiology’s EuroObservational Research Program.4
Database research in cardiology has yielded valuable information in perioperative medicine and can lead to quality improvement initiatives, such as evaluating the adoption and use of evidence-based guidelines, as in a recent study by Gouda et al.5
“Knowledge translation takes patience, time, and energy,” Michelle M. Graham, MD, FRCPC, FCCS, told Cardiology Advisor. “The reality is that practitioners and organizations can’t turn on a dime.”
Graham is the principal investigator of the study by Gouda et al and is a professor of medicine, division of cardiology, at the University of Alberta in Alberta, Canada.
Using healthcare administrative data from 5 databases, Graham and colleagues assessed the use of biomarkers in patients who underwent elective noncardiac surgery in Alberta from January 1, 2013 to December 31, 2017.
The researchers focused on implementation of the Canadian Cardiovascular Society recommendations for preoperative brain natriuretic peptide (BNP) or the N-terminal prohormone of BNP screening to enhance cardiac risk prediction and postoperative electrocardiogram and troponin screening in patients with elevated cardiac risk to detect myocardial injury after noncardiac surgery. The guidelines were published online in 2016 and in print in early 2017.
Among 59,506 patients, 6.8% had pre-operative natriuretic peptide screening. The rates of appropriate pre-operative natriuretic peptide testing increased from 5.7% in January 2013 to 8.8% in December 2017, and postoperative troponin was measured at least once in 19.5% of patients. The use of these biomarkers “remains low,” reported the investigators.
“The biggest issue with perioperative screening is that there are multiple practitioners involved, and different ones depending on context in different places,” Dr Graham told Cardiology Advisor. “For example, while anesthesia is almost always involved, pre-operative assessment sometimes also includes an internal medicine specialist, a cardiologist, or both. Furthermore, the BNP assays are often restricted, limiting who is able to order a BNP.”
For postoperative troponin screening, surgeons, hospitalists, and other consultants — and not necessarily the same ones who were involved preoperatively — are all factors in the process, according to Graham.
“There are lots of barriers,” she said. “What often works is a local champion who changes how things are done, demonstrates improvement, and then there is scale and spread as others adopt processes.”
Translating scientific literature or evidence-based recommendations into clinical practice is “notoriously slow,” echoed Joel L. Parlow, MD, FRCPC, and Michael McMullen, MD, FRCPC, in an editorial6 accompanying the study by Gouda et al.
“Multivariate analysis of data can identify subgroups that can be targeted to identify barriers to implementation, such as educational or structural initiatives among specific groups or institutions,” stated Parlow and McMullen. “Gouda et al illustrate the potential for tracking guideline implementation guidelines over time across varied clinical environments as a quality improvement tool, and, as new evidence emerges, tailored strategies for education and communication can be designed.”
“Cardiac disease has some of the best evidence-based medicine around,” Graham commented. “Therefore, it is easier to build robust guidelines around care, and develop knowledge to action strategies to support implementation of new evidence.”
Overall, big data applications and tools have great potential to optimize point-of-care management, enhance CV healthcare quality and performance, and improve patient outcomes.3 If big data analytics can lead to improvement in the quality of care and patient outcomes and be successfully implemented in CV practice, it will fulfil its potential as an important component of a learning healthcare system.1
References
1. Rumsfeld JS, Joynt KE, Maddox TM. Big data analytics to improve cardiovascular care: promise and challenges. Nat Rev Cardiol. 2016;13(6):350-359.
2. Nazir S, Khan MN, Anwar S, et al. Big data visualization in cardiology—a systematic review and future directions. IEEE Access. 2019;7:115945-115958.
3. Nasir K, Javed Z, Khan SU, et al. Big data and digital solutions: laying the foundation for cardiovascular population management. Methodist Debakey Cardiovasc J. 2020;16(4):272-282.
4. Lima FV, Fahed AC. Innovation at ACC. Harnessing the power of big data in cardiovascular disease. Cardiology Magazine. April 18, 2018. Accessed March 21, 2021. https://www.acc.org/latest-in-cardiology/articles/2018/04/17/12/42/harnessing-the-power-of-big-data-in-cardiovascular-disease.
5. Gouda P, Wang X, McGillion M, Graham MM. Underutilization of perioperative screening for cardiovascular events after noncardiac surgery in Alberta. Can J Cardiol. 2021;37(1):57-65.
6. Parlow JL, McMullen M. Big data for a big problem: how can we enhance the implementation of perioperative cardiovascular guidelines? Can J Cardiol. 2021;37(1):11-13.
