Use of Cardiac Imaging in Canada—Missing the Trees for the Forest

The American Heart Association recently published a scientific statement defining quality in cardiovascular imaging.¹ The statement describes key indicators of high-quality imaging as effective, safe, patient-centered, timely, equitable, and efficient. Braga et al² provide important data on the use of imaging tests for patients with heart failure in Canada, and they lay the groundwork for future studies on imaging quality. The authors used provincewide administrative data from Ontario, Canada, to examine trends in the use of cardiac imaging among a large cohort of patients with heart failure from 2002 to 2016. Transthoracic echocardiography was by far the most common imaging modality and was the largest contributor to imaging costs throughout the study period. There was a marked increase in the use of echocardiography in the early 2000s, followed by a decrease in the rate of growth from 2011 to 2016, which coincided with the introduction of advanced imaging modalities such as cardiac magnetic resonance imaging and computed tomography, as well as the implementation of a province-wide accreditation program for echocardiography laboratories.

Furthermore, Braga et al² found that the proportion of patients receiving 1 or more echocardiograms annually increased over time and that there was a slight decrease in the interval between repeated echocardiograms over time. The study has several strengths, including (1) examination of imaging trends among a large cohort of patients with a stable prevalence of heart failure, (2) examination of the potential reasons for the decrease in the rate of growth of echocardiography, (3) inclusion of payer-perspective costs, and (4) consideration of potentially duplicative imaging studies.

Despite these notable findings, much remains unknown about the quality of cardiovascular imaging since the early 2000s. The findings by Braga et al² mirror the rapid increase and decrease in the rate of growth of diagnostic cardiovascular imaging in the United States,^3,4 and available evidence suggests that this trajectory occurred as a result of a complex interplay between patients, clinicians, payers, and health care systems.⁵ Descriptive studies of imaging trends such as the one by Braga et al² provide important information on real-world changes in the use of cardiac imaging services and costs, but they have 2 main limitations. First, these studies generally do not include information on whether imaging was beneficial to patients (commonly measured by guideline concordance, appropriateness, or the health care value formula of quality over costs). For example, Braga et al² found that the use of echocardiography for patients with heart failure increased rapidly until 2011, but we do not know whether the growth was due to changes in beneficial testing (eg, assessment of left ventricular ejection fraction in patients hospitalized with incident heart failure) or testing that provides minimal patient benefit (eg, routine annual reassessment of left ventricular ejection fraction in the absence of a clinical change).

The second limitation is that these studies are generally unable to establish causal connections to the myriad factors that may have affected changes in the use of testing. These factors include policy changes such as payments that incentivize testing (eg, fee-for-service payment models) or constrain testing (eg, capitated payment models), structural changes such as the vertical integration of cardiologists from private practices into hospital-based practices, guideline changes intended to improve ordering practices such as appropriate use criteria and the Choosing Wisely campaign, and technological changes such as the introduction of new imaging modalities. Although Braga et al² found that the decrease in the growth of echocardiography coincided with the introduction of advanced imaging modalities and the implementation of an echocardiography accreditation program, the number of advanced imaging examinations performed was small (<40 tests per 1000 patients in 2016), and a causal connection of the decrease in echocardiography rates with the accreditation program could not be clearly established. Descriptive studies have generally been unable to ascertain what caused the observed changes in the use of imaging and whether these changes were good or bad for patients.

The American Heart Association’s scientific statement¹ describes a research hierarchy in imaging quality. The hierarchy ranges from single-center reports on appropriateness that have limited generalizability to large randomized clinical trials that require considerable expense to conduct. Within that hierarchy, there remains a clear role for observational studies on imaging trends such as the study by Braga et al.² Future studies could consider using additional methods of observational data analysis to build on these findings and reveal new insights. These methods might include exploring the sequence and progression of testing (eg, identifying patients who receive echocardiography as an initial test for heart failure and subsequently receive cardiac magnetic resonance imaging), understanding trends in certain high- and low-value imaging tests that can be identified from administrative data with reasonable sensitivity and specificity,⁶ or using difference-in-difference analytic models with clearly defined exposures and outcomes (eg, changes in the use of stress testing after implementation of medical liability reform).⁷ To guide future policies on the use of cardiovascular imaging, a better understanding of the causal factors affecting cardiac imaging use and the effect of changes in use on imaging quality is needed.

Published: August 9, 2019. doi:10.1001/jamanetworkopen.2019.8700

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Kini V. JAMA Network Open.

Corresponding Author: Vinay Kini, MD, MSHP, Division of Cardiology, University of Colorado Anschutz Medical Campus, 12631 E 17th Ave, Mail Stop B130, Academic Office One, Aurora, CO 80045 (vinay.kini@ucdenver.edu).

Conflict of Interest Disclosures: Dr Kini reported receiving grant support from career development award K23HL143208 from the National Institutes of Health/National Heart, Lung, and Blood Institute to improve quality in the use of noninvasive cardiovascular testing.