Quality in Radiology: Science or Science Fiction?

In times of turmoil, I have always turned to science fiction. Hard science is, well, too hard for this intellect, and science fiction offers refuge when the world around is, as Wordsworth said, too much with us. I nearly missed a flight back to Los Angeles recently when I couldn’t pull myself out of a startling exhibit, sadly now closed, called Design and the Elastic Mind, at New York’s Museum of Modern Art. The exhibit chronicled the remarkable synergies between design and science in adapting hard science to our world. An obvious example of this—one considered too old to be included in the exhibit—is the Internet, nothing but strings of code until designers added buttons, structure, and graphical user interfaces. A section on nanotechnology  explored the attempts of scientists to emulate the inherent design in nature by applying the atomic palette to a variety of constructions, including the manufacture of ubiquitous solar cells to meet the world’s energy requirements. Another investigated the efforts of social scientists to design methods of critical visualization that construct understanding from the deluge of information available to 21st-century dwellers. An example is a project from Columbia University to map the source of prisoners residing in upstate New York prisons, resulting in the identification of million-dollar blocks, so called because these individual New York City blocks produced enough inmates to cost the state at least $1 million a year to house. This reciprocal relationship between science and design has applications on both the clinical and business sides of radiology. Cass Schaedig’s article on digital dashboards (page 22) represents the application of critical visualization to the many data points generated by a busy radiology service. On the clinical side, a role for design emerged from a session at the Society for Imaging Informatics in Medicine meeting in Seattle. As a panel grappled with the present-day workflow challenges of banging through an 8,000-image dataset, Paul Chang, MD, suggested that there is no one way to do this, and that PACS vendors will need to add a map of the radiologist’s method of interrogation for both teaching and medicolegal purposes. Design is not science, though, nor is science fiction science. The Webster II New Riverside University Dictionary (Houghton Mifflin) defines science as “the observation, identification, description, experimental investigation, and theoretical explanation of natural phenomena.” In the next listing, science fiction is defined: “Fiction in which actual or potential discoveries and developments form part of the plot.” That precisely describes the approach of MedSolutions, Nashville, Tenn, in building its case that “quality is a problem in radiology, maybe even an epidemic.” That is the premise on which its new company Premerus is based (see article, page 12). The company is marketing a network of subspecialist teleradiologists to insurers and managed care companies, with the promise that its designated subspecialist teleradiologists will improve quality and lower cost by producing more accurate interpretations. The company, however, has taken the approach of building a business model before it has any vetted data to support the need for a third party to route studies to the network it has begun building. Nonetheless, when medical director Gregg Allen, MD, a family practice physician, asks,  “Who would you like to read your MRI of the brain or your mother’s MRI of the brain, when there is a question of a tumor?” he lays bare the fuzzy and emotional side of the issue of quality in medicine. We all want the best, especially when it comes to our families, but what is the best? What proof is there that the faceless Premerus teleradiology subspecialist is any better than the subspecialist from the local radiology group, or, for that matter, the good generalist who reads a lot of brain MRIs? My bet is that if there were a question, the general radiologist would ask a specialist. A friend whose 84-year-old husband has Parkinson disease spends quite a bit of time taking him to his neurologist, urologist, cardiologist, and endocrinologist, whose prescriptions for him add up to 17 pills a day. On a recent visit to her general practice physician for a checkup, they spent much of the time discussing her husband, who was so busy going to specialists that he hadn’t been in to see the physician in years. Prefacing his remarks with a wish not to offend, the doctor asked her, “Why are you spending so much time taking him to doctors? You don’t want him to die of Parkinson disease, do you? It’s a terrible way to die.” The perspective of a generalist was gravely in need here and should not be trivialized. The quality issue is a critical one for all physicians at the moment. Suits have been filed by physicians and others challenging payors’ definitions of quality (usually linked, in some way, to cost), most recently by the Massachusetts Medical Society, Boston. Physician rankings are popping up in every state. Clearly, we have too much design in the payor’s quality constructs and not enough science.   Like it or not, the issues of quality and cost will not go away. I hope you will keep an open mind when reading Jim Kieffer’s article on the history and future of radiology compensation on page 38 of this issue. He has baited me over the years with his imagineering of the future of radiology, but I think it is time to give him an audience. Clearly, he is not the only one thinking along these lines. With respect to quality in radiology, in the absence of science, science fiction will prevail.