I Used to Interpret Images for a Living

By Stuart E. Mirvis, MD, FACR
pdf path

Dr. Mirvis is the Editor-in-Chief of this journal and a Professor of Radiology, Diagnostic Imaging Department, University of Maryland School of Medicine, Baltimore, MD.

Remember a long time ago when you were a radiologist and interpreted studies? If you are relatively recently trained you will no tremember such a time. But there was actually a past era when you spent the vast majority of your day and career looking at images and trying your best to interpret them correctly.

A few years ago someone came up with a statistic that the medical profession commits as many as 98,000 fatal errors a year.1 Clearly, if you accept this as truth, we needed overseers to make sure we did not make so many mistakes. In fact, an entire infrastructure evolved around this goal. So what strategies were created to protect our patients from us?

Well, most of us already had the monthly case review conference so that just continued on, but now in the presence of a risk-management official. Every 2 years or so, one would need to be recredentialed. Even if absolutely nothing changed in your story, you still needed to fill out all the forms as if you were a virgin to the system. Those of us working at many different institutions had to fill out lots of different forms, often requiring unique information for each center. Far be it from anyone to create one form that the vast majority of hospitals could use that could be copied and updated electronically as needed.

At our institution, we now review every tenth case in the picture archiving and communications system for accuracy of the interpretation by another department radiologist. The key is to identify significant misses. Usually, these cases pop up when trying to rapidly review an urgent case. I suspect most people just click “agree with interpretation” and move on. Most commonly, I am the one who misses findings my colleagues made, but I don’t have to mention those. Clearly, my judgment of them has some limitations. It’s amazing how few significant interpretation errors are found each month in my department. That 98,000 errors thing must be wrong.

Another consumer of our valuable time is that every doctor being recredentialed needs letters from colleagues to attest to their value, safety, and lack of socially disagreeable behavior. Personally, I find this difficult because my colleagues have such a wide variety of social behaviors, which I think adhere to the broad range of normal. It’s not likely I will “out” the person in my section who works like a dog just because of a few aberrant personality traits that some people may see as psychotic. Sitting in a dark room for 20 years or so may induce this behavior and is a known risk factor. So, in my view, everyone I work with is beautiful, brainy, and as proper as a high-class English butler.

I am frequently asked to evaluate a former co-faculty, fellow, or resident applying for credentialing outside our system. I have actually had the audacity to say what I really believe when the public health is endangered. I have noticed, when imparting what are certainly negative comments, that apparently no one cares how I respond. In all such situations, to date, the applicant has been hired. Even when I commented (somewhat tongue in cheek) that the applicant comes with their own body bags or has a graveyard named after them, my critique is ignored. So what’s the point in asking? Also, some of these requested evaluations are for people I have not worked with or even talked to in 10 years or more. I think this makes any assessment of dubious value. I would encourage a statute of limitations o f5 years for evaluating former associates. After that time only documentation of service is supplied along with all written evaluations.That should provide an adequate picture.

Letters of recommendation are, in general, another great time consumer. I have seen a slew of these while serving time on promotions committees. They are always fantastic and thus have no discriminating value. Very few people would agree to write a support letter that is anything but wonderful. “This resident walks on water, gives Mother Teresa advice on caring about people and entertains the Emergency residents with hysterical interpretations.” Some of my colleagues believe they can “read between the lines” of these letters. Oh great, someone’s professional life rests on supposition and intuition about what is actually meant but not stated.

Moving along, each year there are more online educational surveys that must be read and questions answered to validate our learning of the material. Most people immediately answer the questions without looking at the teaching material and correct them when they are wrong. We have mandatory airway management every 2 years. I have done this for at least 10 years. I still can’t figure out how to hold the bag to get an air seal. Sometimes I pull the chin so far back that I would surely create a hyperextension fracture-dislocation in a real live person. Going into radiology was a good decision in my case. Another video we need to view concerns recognizing, understanding, and dealing with the hostile physician. Well, at least it’s easy to understand why they’re hostile.

We also take our time meeting general medical skill requirements. Some hospitals require Advanced Cardiac Life Support training for all their physicians. At least for a non-interventional radiologist, having ACLS certification is like a license to kill. Let’s face it; that little course completion card in your wallet means nothing compared to actually “running a code.” In our institution, most radiologists are required to pass Basic Cardiac Life Support every 2 years. This is a requirement for employment. Therefore, the instructors find away to pass you no matter how spastic you are. A favorite moment in my professional life was watching one of our senior staff knock the baby model’s head off doing back slaps in CPR training. He was passed anyway, but with the advice to avoid small children whenever possible.

Here’s another time waster; reviewing our residents’ performance. We do this every 4 weeks for any resident we’ve worked with. If you give the resident a “falls below standard” score it is you, not the resident, who are brought before the Inquisition to justify your heresy. Many of the rating questions are ridiculous. You assess the resident on hygiene. This guy smells like Burberry London Eau DeToilette for Men: A+; this guy smells like a gym locker at the end of the year: E-. We are asked to rate the residents’ ethics. If they agree with my ethics, as anemic as they are, they get a high score. My favorite rating is whether the resident interacts well with the referring physicians. You’ve got to be kidding! What’s the right answer here?

I could carry on almost ad infinitum about how many things we need to do to try to ensure the accuracy of our work. In fact, I have come to realize, with no small amount of frustration, how much of my professional life is now being taken up by these sorts of activities. I also realize I am reading cases faster than ever and possibly missing more findings as “credentialing needs” and similar extraneous mandates increase exponentially. I believe this is a case of the alleged cure being far worse than the disease. I am going to sneak off to the reading room now and try to knock off some cases before the risk manager calls.


  1. National Academy of Sciences. To Err Is Human. 2000. Washington, D.C: The National Academic Press. Available at: http://www.nap.edu/catalog.php?record_id=9728.Accessed October 24, 2010.
Back To Top

I Used to Interpret Images for a Living.  Appl Radiol. 

November 17, 2010

Copyright © Anderson Publishing 2020