The Importance of Specialization in Radiology

Doctors at the Shouldice Hospital in Ontario, Canada, offer only one service: hernia repairs. At first glance this seems like a flawed business model, but it may in fact be ideal. 

The recurrence rate of inguinal hernias after a repair performed at Shouldice is 0.5 percent. This figure, based on data collected from more than 300,000 operations, distinguishes Shouldice as the world leader in hernia repairs. Their success has been featured in Harvard Business School (HBS) case studies as “a model of business excellence.” According to HBS, Shouldice is operating as a “focused factory,” a strategy of pursuing a narrow product niche, which has allowed the hospital to achieve efficiencies and expertise that is unparalleled. Leaders at HBS have proposed that this strategy will almost always out perform conventional operations with a broad mission.

Most hospitals in the United States, however, pursue a strategy that offers every service line imaginable, even ones in which they do not excel. I believe radiologists, especially those in private practice, may encounter a similar dilemma.

If you are a frequent reader of this blog you are aware that I am currently pursuing a fellowship in abdominal imaging. During this year-long fellowship I will focus exclusively on abdominal imaging, reading hundreds of abdominal and pelvic scans utilizing all currently available modalities. I will also participate in multidisciplinary conferences that allow me to gauge the accuracy of my findings and tailor my interpretations to best serve my patients and referring providers.

To become a board certified radiologist, I was required to pass the breast imaging section, among others, of the written and oral board examinations and have been deemed qualified to interpret mammograms by the American Board of Radiology. Yet, with only a few months of exposure to mammography (via residency training) and the prospect of being one year removed from reading any mammograms, my confidence to deliver quality care to these patients is rapidly waning.

I know I add the most value to my patients and colleagues when reading studies within my expertise. Anecdotally I know many radiologists feel the same way.

If I choose to join a private practice, I will likely be asked to interpret mammograms. Similarly, those who have completed breast imaging fellowships will be asked to interpret abdominal scans. Familiar with the specialization that takes place within our field, I would certainly insist that my or a loved one’s study be interpreted by a radiologist specializing in that area. Recent publications in the peer reviewed literature support this claim.

The Breast Cancer Surveillance Consortium (BSCS) study entitled “Variability in interpretive performance at screening mammography and radiologists' characteristics associated with accuracy,” found that the only characteristic associated with improved accuracy and sensitivity in interpreting mammograms was fellowship training in breast imaging¹. Moreover, the BSCS has concluded from their prior work that fellowship-trained breast imagers entered clinical practice with no learning curve as opposed to non-fellowship-trained radiologists who gained significant clinical experience interpreting mammograms during the first three years after residency². I am confident that a study evaluating other subspecialists in radiology would come to similar conclusions.

A radiologist who lacks fellowship training is not deficient in the knowledge or skill set needed to interpret a wide variety of studies, but there is greater value in a fellowship-trained radiologist who specializes in one field. As an abdominal imager, for example, I have a better understanding of when a liver lesion should undergo percutaneous sampling and when an additional imaging exam may obviate a biopsy. Breast imagers possess similar knowledge of their field and can better guide the care of patients with breast pathology.

Many other industries understand the importance of specialization and divesting of service lines in which they do not excel. It’s the reason that Lexus, in their “relentless pursuit of perfection,” does not make computers or budget automobiles, but rather focuses on luxury cars. Why should the practice of radiology be any different? However, while this arrangement is feasible and sustainable in the academic setting it is not clear if the private practice world of radiology is equipped to follow this model.

1 Elmore JG, Jackson SL, Abraham L, Miglioretti DL, Carney PA, Geller BM, Yankaskas BC, Kerlikowske K, Onega T, Rosenberg RD, Sickles EA, Buist DS. Variability in interpretive performance at screening mammography and radiologists' characteristics associated with accuracy. Radiology. 2009 Dec;253(3):641-51. Epub 2009 Oct 28.

2 Diana L. Miglioretti, Charlotte C. Gard, Patricia A. Carney, Tracy L. Onega, Diana S. M. Buist, Edward A. Sickles, Karla Kerlikowske, Robert D. Rosenberg, Bonnie C. Yankaskas, Berta M. Geller, and Joann G. Elmore. When Radiologists Perform Best: The Learning Curve in Screening Mammogram Interpretation Radiology December 2009 253:632-640