Nonclinical Factors Substantially Affect Detection of Obstructive CAD

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Apart from clinical characteristics, physician and system factors account for almost one-quarter of variation among hospitals regarding the detection of obstructive coronary artery disease (CAD) with elective coronary angiography, according to a study published online September 2, 2014, ahead of print in Circulation: Cardiovascular Quality and Outcomes.

The study “sets the stage for further work into understanding the drivers of the differences observed between physicians and hospitals, so as to identify foci for quality improvement,” the authors write. 

Overall, 39.7% of patients were classified as self-referral, ranging from 4.8% to 74.6% across hospitals. Interventional physicians performed 44.3% of the angiograms; most were done at full-service centers (70.8%), while 16.9% were done at cath-only hospitals and 12.3% at stand-alone PCI hospitals.

Patients who had angiograms at diagnostic-only hospitals were more likely to undergo preprocedural testing (59.8% vs 55.8% for full-service) Also, self-referral patients had fewer functional tests before angiography compared with patients who did not self-refer (52.9% vs 56.9%), as did patients whose angiograms were performed by invasive cardiologists (52.2%) rather than interventional cardiologists (57.8%).

Consistent differences were observed among all physician and hospital groups in the rates of obstructive coronary stenosis, severe disease, and revascularization (table 1). Nonclinical factors accounted for 23.8% of the variation between hospitals. 

Table 1. Unadjusted Outcomes^a

^a P < .001 for all comparisons.

On multivariable analysis, self-referral (OR 0.89; 95% CI 0.86-0.93) and angiograms performed by invasive physicians (OR 0.85; 95% CI 0.81-0.90) were associated with a lower likelihood of obstructive disease. Operator experience, age, and sex were not significant predictors of obstructive CAD, but operator annual volume was related (OR 1.02 per 50 cases; 95% CI 1.01-1.03). Lastly, angiograms performed at diagnostic centers were associated with a 38% lower likelihood of showing obstructive disease compared with full-service centers (OR 0.62; 95% CI 0.39-0.98). 

Drivers of Variation Unknown

While the researchers acknowledge that the reasons for the observed discrepancies are unclear, they suggest access to preprocedural testing, education, and expectations may vary among physician and hospital types. Additionally, “we cannot discount that secondary financial incentives may play a role in the variations we observed, especially in regard to self-referral or operator volume,” they write, explaining that although Ontario has universal health care, “physicians are remunerated on a fee-for-service basis and institutions receive funding based on meeting specific procedural volumes.” 

Dr. Wijeysundera and colleagues also point out several limitations of their study. First, since the referral mechanism for angiography differs among the 18 study hospitals, “there is the potential that higher-risk patients will be preferentially booked with interventional physicians by the coordinator to facilitate same-sitting PCI,” they say. Also, there is a high likelihood of reporting errors in any registry-based study. Lastly, “there may be systematic under or over reading of coronary artery severity by physician groups,” the authors observe. 

Ultimately, understanding the reasons associated with varying thresholds for angiography “should be a focus of quality improvement initiatives to make more efficient use of this invasive procedure,” they conclude. “This will translate into both cost savings, as well as improved care for patients by eliminating exposure to the risks of unnecessary testing.”

Source:
Wijeysundera HC, Qiu F, Bennell MC, et al. Impact of system and physician factors on the detection of obstructive coronary disease with diagnostic angiography in stable ischemic heart disease. Circ Cardiovasc Qual Outcomes. 2014;Epub ahead of print.

Related Stories

• Diagnostic Yield of Catheterization in Canada Much Higher Than in New York State
• Consensus Document Outlines Guidelines for Cardiac Cath Use
• PCI More Common in NY State than Ontario—for Better and Worse