In our large retrospective study consisting of 2,577 consecutive patients undergoing CCTA, 83.5% of CCTAs performed were deemed appropriate. Following the publication of the 2013 AUC, the percentage of appropriate CT utilization increased significantly while the percentage of rarely appropriate utilization decreased significantly. There was no significant difference in patient characteristics when compared pre and post AUC publication. The decline in rarely appropriate scans translated to potential cost savings of approximately $57 million per year, when extrapolated to the Medicare population of the United States. Our results were driven by CCTAs performed on symptomatic patients, who accounted for approximately 89% of the total cohort. In the approximately 11% of patients who were asymptomatic in our cohort, the proportion of appropriate CCTAs declined after publication of the AUCs.
There have been a small number of previous studies examining the effect of AUC publication on appropriateness of CCTAs [
9‐
13]. For example, a meta-analysis reported that while there was improvement in appropriate utilization of CCTA after publication of the AUC (from 37 to 55%), the overall rate of appropriate utilization remained relatively low [
11]. The data included in the meta-analysis was generated from many parts of the world including the United States. One study included in the meta-analysis from the Mayo Clinic reported that only 27% of patients referred for cardiac CT were considered appropriate based on AUC criteria at that time [
20]. In contrast, more contemporary data for cardiac magnetic resonance appropriateness by Kaushal et al
. reported that 95.5% of cardiac MRIs were deemed appropriate. Similar to the Kaushal study, this paper also reported a significantly higher percentage of appropriate tests following AUC publication, in addition to a decline in rarely appropriate cases and an overall high number of appropriate CCTAs (83.5%) when compared to the older data. In this study, we found a higher proportion of appropriate CCTAs than what was reported in previous studies, although we found similar improvements in appropriateness in response to AUC publication. The relatively high number of appropriate CCTAs when compared to other studies may be in part related to the controlled access of the technology in our jurisdiction [
21,
22]. Our findings may also be indicative of ongoing incremental improvement in physician ordering patterns, potentially as a result of iterative AUC publications. Although there are multiple factors that impact physician ordering patterns of non-invasive diagnostic testing, these results add to the growing body of evidence that supports the notion that publication of AUCs is associated with increased appropriateness of testing [
23‐
25]. Furthermore, our study found potential cost savings of approximately $57 million per year arising from publication of the AUC. No study to our knowledge has previously determined cost savings associated with differing appropriate CCTA utilization in response to publication of the AUC. Our findings reporting potential cost savings associated with publication of the AUC are consistent with similar research performed on other imaging modalities, such as MRI [
14,
25].
Clinical importance
Ordering physicians in the United States will be required to consult the AUC when ordering advanced imaging tests such as CCTAs and cardiac MRI scans starting January 1st, 2021[
5]. Our work highlights that after publication of AUC, appropriate utilization of CCTA is high in our large quaternary care centre. Interestingly, while our overall cohort and the symptomatic subgroup report higher appropriate utilization after publication of the AUC, appropriate utilization declined after AUC publication in our small asymptomatic subgroup. These results suggest that despite the impressive overall results, there remain areas for future improvement in the education of referring physicians and triage staff regarding the appropriateness of CCTAs in asymptomatic patients. Surveys aimed at evaluating physician consciousness and/or knowledge of the AUC before and after implementation of the mandatory consultation requirement, mentioned above, would be a valuable avenue for future research. Furthermore, our findings of $57 Million of annual potential costs savings that may be attributable to AUC publication indicate a potential real-world financial impact due to the publication of the AUC.
Limitations
It is important to interpret the results of this study in the context of its limitations. First, in this study, CCTAs were scanned at one institution, potentially limiting results from being transferable to other jurisdictions. However, the institution is a quaternary care centre which receives patient referrals from four other community and academic hospitals and a wide array of outpatient clinics. Second, there are inherent limitations to this study due to its retrospective nature. For example, we did not evaluate appropriateness prospectively. The classification for each test was completed using retrospective patient data and occasionally by contacting the referring physician when needed. However, our classification methods were similar to those used in other studies with reported success rates for classifying patients > 95% [
14,
15]. The retrospective nature of the study also translated to a lack of granularity in some clinical parameters, such as the processes used by referring physicians to access the AUC guidelines. Finally, traditional cost-effectiveness modelling was not appropriate for the design of this study. However, we used cost estimations using methods similar to those utilized in prior similar studies [
19,
26]. Nonetheless, it is important to state that the cost savings reported in our study are estimates and not based on exact data (for example, the number of CCTAs performed annually in the United States is an estimated number because the exact number is not known).