Egos and Experience: Personalized Risk Assessment Works, but Entrenched Beliefs Hinder Implementation

Why do some operators shy away from using precision medicine in their PCI decision-making process? A group of researchers set out to find out, uncovering a tangled web of reluctance backed up by assertions of experience and ego. 

Take Home. Egos and Experience: Personalized Risk Assessment Works, but Entrenched Beliefs Hinder Implementation

In an earlier study, John A. Spertus, MD (St. Luke’s Mid America Heart Institute, Kansas City, MO) and colleagues looked at what happened when a risk-stratification tool called ePRISM (Health Outcomes Sciences, Overland Park, KS) was used prior to PCI at nine US hospitals. The tool proved to be associated with high patient satisfaction, increased use of bleeding avoidance strategies, and reductions in bleeding. At the same time, patterns of ePRISM use by operators were highly variable.

“Through word-of-mouth we understood that there were some cardiologists who appreciated the information and thought it was valuable, and there were others who really disregarded the information,” said Adnan K. Chhatriwalla, MD (Saint Luke’s Mid America Heart Institute), who served as senior author on that paper as well as the new one—published recently in the American Heart Journal—which sought to assess physician attitudes and better understand the physician-level barriers to use of a tool such as ePRISM.

Judgment, Ego Drive Reluctance

Led by Carole Decker RN, PhD, the researchers invited physicians who participated in the earlier study to be interviewed in-person or by phone. The average length of interventional cardiology experience for the group of 27 operators who were interviewed was 13 years (range, 0.5-30).

Three distinct themes emerged from the interviews. One was that clinical judgment, based upon experience and training, was a big reason for reluctance. Decker and colleagues termed this ‘Experience versus Evidence.’ In the interviews, physicians who gave this as a reason not to use ePRISM said things such as:

“There’s just so much more information that I process not just based on the clinical history of the patient but on the angiogram of the patient . . . none of that stuff is reflected in the [ePRISM] tool. But those for me are many times more important variables.”

“I want to be rational, I really do, but I don’t think it captures all the nuance that’s required in stent selection.”

A second reason given by some physicians for their reluctance to use the tool was a belief that risk models support rationing of care, or withdrawing of therapy, in low-risk patients. Decker and colleagues termed this ‘Omission of Therapy.’ Physicians who took this view said things such as:

At the end of the day, you want to do the best thing you can do for each and every patient, not just the high-risk patients.”

“I don’t buy into it as a health care provider, as an intellectual, as a physician.”

Finally, a third reason given by some was that they had no need for the information in the model. This was termed ‘Unnecessary Information.’ Physicians who fell into this group said things like:

“Maybe it’s my own personality or maybe it’s age, but I would say, compared to some of the other physicians in my practice, I don’t worship at the altar of evidence-based medicine to the degree that they do.”

“I don’t need [those] data to tell me what I already know. . . . To me this is more for the patient’s education, not for me. I already know this.”

To TCTMD, Chhatriwalla said he was struck by the egotistical nature of many of the statements, adding that it was “surprising to hear so bluntly that ego enters the equation at all.” More importantly, he added, many of them are missing out on a main goal of risk-assessment tools, which is to complement the operator’s ability as opposed to replacing or overriding their judgment.

‘Cultural Shift’ Needed in Thinking and Training

In an accompanying editorial, Michael B. Rothberg, MD, and Kathryn A. Martinez, PhD (Cleveland Clinic, Cleveland, OH), say the operator statements “carry the underlying assumption that physicians value risk estimations but feel that they can estimate risk as well or better than a model. Unfortunately, the results of the ePRISM trial do not support this assertion.”

By not using personalized risk assessments, they argue, these physicians are failing to present patients with probabilities, thus depriving them of the opportunity to make an informed decision as well as undermining their autonomy. What is needed, Rothberg and Martinez suggest, is “a cultural shift away from clinical intuition toward evidence-based decision making.”

But another important question that arises is whether such a shift is possible for established physicians. Chhatriwalla said his group is planning a future study to explore this issue by looking at the characteristics of those who embraced ePRISM versus those who did not.

“Maybe, as physicians, we need to be educated early on, not only in how to make good decisions for patients, but how to help the healthcare system as a whole,” he added.

Similarly, Sunil V. Rao, MD (Duke University Medical Center, Durham, NC), who was not involved in the ePRISM studies, told TCTMD in an email that risk-assessment tools only work when you have “a coalition of the willing.”

“In other words, clinicians have to ‘buy in’ that these tools are robust, important, and necessary,” he said. Rao also pointed out that with some exceptions, medical education has remained largely unchanged over time.

“The focus has been on work hours and trainees working hard, rather than implementing programs that allow trainees to work smart,” he observed. “We do need to have an overhaul of how we train students, residents, and fellows to include programs on evidence-based medicine and electronic tools that improve the quality of care.”

Rothberg and Martinez also point out that practice guidelines need to be more explicit and helpful, noting as one example that the current American Heart Association guidelines for PCI recommend assessment of bleeding risk, but make no recommendation for how to do it or what threshold to use.

Related Stories:



  • Decker C, Garavalia L, Garavalia B, et al. Understanding physician-level barriers to the use of individualized risk estimates in PCI. Am Heart J. 2016;Epub ahead of print.

  • Rothberg MB, Martinez KA. Risky business: personalizing the approach to percutaneous coronary intervention. Am Heart J. 2016;Epub ahead of print.

  • Decker and Chhatriwalla report no relevant conflicts of interest. 
  • Rao reports consulting for Astra Zeneca, Medtronic, Merck, and Terumo. 
  • The editorial contains no disclosure information for Rothberg or Martinez. 

We Recommend