Author + information
- Published online July 3, 2017.
- Kristopher P. Kline, DO, MPH,
- Jared Plumb, MD,
- Linda Nguyen, BS,
- Leslee J. Shaw, PhD,
- Rebecca J. Beyth, MD, MSc,
- Tianyao Huo, MS and
- David E. Winchester, MD, MS∗ ()
- ↵∗Division of Cardiovascular Medicine, University of Florida College of Medicine, 1600 SW Archer Road, P.O. Box 100277, Gainesville, Florida 32610-0277
Appropriate use criteria (AUC) have been developed for a variety of cardiovascular tests and services. AUC are based on the best available evidence with the goal of identifying which clinical scenarios are likely to result in net patient benefit or harm. Little is known about how well physicians on the clinical care frontlines know about and apply AUC when making decisions about care. Patients do not have a role in developing AUC; their reflections on appropriateness of care have not been well studied.
We conducted an anonymous survey of patients and health care providers and asked them to rate the appropriateness of myocardial perfusion imaging (MPI) for a series of clinical scenarios. We also gathered data on attitudes and awareness of issues related to AUC. We hypothesized that patients and providers would overestimate the appropriateness of MPI compared with the published AUC. We asked respondents to use the same scale used to develop the AUC; they rated 5 clinical scenarios from 1 (“bad idea”/inappropriate) to 9 (“good idea”/appropriate). Responses were compared with the 2009 AUC for the same scenarios (1).
The survey was completed by 456 respondents: 342 patients (56% veterans, 69% male) and 114 providers (62.3% academic, 27.2% Veterans Affairs, 7.9% private practice). Most of the providers were physicians (86.8%), with the remainder being advanced practice providers. Of the 5 clinical scenarios, the patient and provider ratings agreed with the published AUC for 2 criteria; there was a categorical difference from published AUC for the remaining 3 criteria (Table 1). In 2 of 3 cases, the patient rating was higher than the provider or published rating (AUC indication #1: 5.0 vs. 3.5 and 3.0; p < 0.0001; AUC indication #12: 5 vs. 2 and 1; p < 0.0001). In the third case, the provider rating was lower than both the patient and published rating (AUC indication #60: 2 vs. 6 and 6; p < 0.0001). For all 5 scenarios, the patient ratings ran the entire 9-point scale; 3 scenarios had a rating of 5 as the mode. In contrast, provider ratings were more closely clustered around the median response. Using a logistic regression model, we did not find any correlation between the baseline characteristics of the patients and the overestimation for the clinical scenarios.
Most providers (55.4%) believed that having 0% to 5% of tests rated as inappropriate would be acceptable. Some providers (36.6%) had never heard of AUC, whereas 12.5% reported using AUC regularly. Providers reported that they believed that the use of guidelines from specialty societies improved decision-making (80.2%), lowered cost (69.4%), and enhanced quality of care (65.8%); however, only 34.2% reported that AUC improved the health of their patients.
In this novel investigation of the opinions that patients and providers had about AUC, we observed that patients tended to overestimate the appropriateness of testing. The reasons for patients to overestimate appropriateness are unclear. Although this could reflect patient values that differed from the published AUC, we suspected, based on the wide distribution of responses, that patients had low confidence in this exercise and/or were poorly informed about the appropriate use of MPI. This raises the question as to whether patients should play any role in future iterations of AUC. Although the choices and values of the patients are important considerations when deciding on a course of action, they are most applicable when there is clinical equipoise or no clear indication and/or contraindication. Patient opinion may be more valuable in areas such as choice of elective percutaneous coronary intervention.
Overall, the provider-based appropriateness ratings were in agreement with the published AUC. Because >10% of MPIs are rarely appropriate in real-world data (2), we suspect that the responses of the providers were at least partially driven by response bias. We were disappointed to observe that, despite the fact that the first AUC for nuclear MPI were published a decade ago, awareness and regular use of AUC was low among our providers, with more than one-third having never heard of the guidelines at all.
Our investigation was limited because it was conducted at a single site; survey responses might be biased and not representative of the actual clinical decision making and ordering habits of providers. This was offset by the anonymous nature of the survey. The survey was designed around the 2009 AUC for MPI because regulatory submission occurred before publication of the most contemporary version of the AUC.
Please note: This study was funded by University of Florida Gatorade Endowment Fund, Gainesville, Florida; National Institutes of Health’s Clinical and Translational Science Awards program, Award Number UL1TR001427 (REDCap Program); the Discovery Pathways Program, Medical Student Research Program, Award Number 5T35HL007489-33, the Malcom Randall VAMC, and the Florida Heart Research Institute. The contents of this paper do not represent the views of the U.S. Department of Veterans Affairs or the United States Government. This study is the result of work supported with the resources of and the use of facilities at the Malcom Randall Veterans Affairs Medical Center. Dr. Winchester has received a research grant from and served on the Advisory Board for Roche Diagnostics. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Todd Miller, MD, served as the Guest Editor for this article.
- 2017 American College of Cardiology Foundation
- Hendel R.C.,
- Berman D.S.,
- Di Carli M.F.,
- et al.
- Elgendy I.Y.,
- Mahmoud A.,
- Shuster J.J.,
- Doukky R.,
- Winchester D.E.