If a macro were available to ease the task of writing articles about contemporary medical imaging, it would begin like this: “Technology is evolving more rapidly than ever before, challenging those in radiology and other imaging specialties to stay informed and adequately trained and to be ready for the next round of change. Nowhere is this more true than in _________” (fill in the blank with the subject of your article). So serviceable is this introductory template that it could be applied in every modality and across all organ systems and disease classifications. In one area, however, practice has remained curiously static for more than a century: the radiologic report. The process of reporting, of course, progressed from the earliest handwritten buckslips to today’s sophisticated speech recognition systems. Yet the form and content of the radiologic report itself have not evolved along with the technology that facilitates its delivery to the clinician.
The fact that report content has remained static for decades is surprising, given its fundamental importance to the status and position of the radiologist in the medical setting. In the earliest years of the field, the report provided merely descriptive information from which the clinician drew diagnostic conclusions. Recognizing that this threatened to make imaging a hospital service on a par with laboratory work or any other patient study, early radiologists worked hard to make sure they were recognized as members of a distinct medical specialty. Throughout the 20th century, improved imaging techniques, enhanced training, and the clear value of medical imaging to all aspects of patient care made the radiologist one of the most highly compensated physician specialists. At the same time, face-to-face interactions with clinicians who came to the radiology department to consult on patient studies raised the level of collegial trust and added value to the written report.
Today, as PACS and radiology information systems are increasingly integrated into medical enterprise networks that speed images and results throughout the hospital, to satellite facilities, and even to distant locations, the report once again takes center stage. Some clinicians and private practice physicians, in fact, now would not recognize the person who reads their mammogram or CT series if they passed him on the street. It seems an appropriate time to look at the evolution of the radiology report to determine where, in the emerging relationships that characterize digital/all access imaging, value can be added for the future. The answer would appear to lie in fundamentally reworking both the process and content of radiologic reporting.
In this first of a two-part series, we look back at the beginning years in which the radiology report we know today was created. The second article will explore ways in which we are likely to see revolutionary changes in reports over the next decade.
THE FIRST RADIOLOGY REPORTS
The earliest radiology report was the slim reprint publication that went out across Europe and began to make its way to the United States in the first days of 1896: Eine Neue Art von Strahlen , Wilhelm Konrad Röntgen’s account of his discovery of the x-ray less than 2 months before. The stunning announcement that “a new form of light” could see through living human flesh caused a rush of public and scientific excitement and almost overnight changed the way that medicine was practiced. Hospitals and private practices embraced the new technology and soon incorporated it into routine use.
|Figure 1. Often cited as among the earliest, this report was made by William James Morton, MD, a neurologist from New York, in May 1896. The report that accompanied the images elevated the interaction with the referring physician to the level of a professional consultation.|
In the United States, the earliest imaging studies of patients were made not in hospitals but in nearby “x-ray laboratories,” usually photographic studios refitted with the simple apparatus that generated x-rays. Most such laboratories were run not by physicians but by photographers and electricians who had the know-how to run the bulky and sometimes dangerous apparatus. Patients (regardless of their status) were sent out of the hospital to the laboratory for imaging. Existing reports from this earliest period are from individuals with varying degrees of medical and/or technical training. One report often cited as among the earliest was made by William James Morton, MD, a neurologist from New York, in May 1896 (Figure 1). The half-sheet reads:
“Dear Dr Stieglitz: The X ray shows plainly that there is no stone of an appreciable size in the kidney. The hip bones are shown & the lower ribs and lumbar vertebrae, but no calculus. The region of the kidneys is uniformly penetrated by the X ray & there is no sign of an interception by any foreign body. I only got the negative today and could not therefore report earlier. I will have a print made tomorrow. The picture is not so strong as I would like, but it is strong enough to differentiate the parts.”
Many of the elements in this report are familiar to modern eyes: a succinct statement of findings, attempts to orient the referring physician to readily recognizable landmarks, comments on the quality of the image, and apologies for unexpected delays. Morton’s service was the provision of the technological expertise needed to make the images. The written report, however, elevated the interaction with the referring physician to the level of a professional consultation. His fee was $5, a not inconsiderable sum at a time when hospital care typically cost $1 per day and even physicians to the well-to-do charged $2 for a house call, medication, and follow-up.
|Figure 2. Printed blanks were used for reporting the results of early hospital-based imaging.|
By 1900, many hospitals had incorporated x-ray equipment into their facilities. Both in and outside hospitals, printed blanks were used for reporting the results of imaging (see Figure 2). In part, this was facilitated by the relatively limited number of possible diagnoses: fractures, calculi, foreign bodies, and various anatomical anomalies. The earliest of these standardized forms functioned more like laboratory request slips than actual reports. At Guys Hospital in London, patients to be radiographed in 1901 were sent to the x-ray laboratory with “skiagraphy requests.” Requesting physicians were asked to detail “the character of the lesion of which a skiagraph is requested, and of any particular point which it is desired to elucidate.” When complete, the fill-in-the-blank report that accompanied the glass plate images contained the basic patient information, as well as data on the current, voltage, and distance required to make the image. The “results” were often an assessment of the quality of the image rather than a diagnostic assessment. “Very good negative” was a typically terse and unassertive “report.”
The general assumption at Guys and at many hospitals where x-rays were welcomed as a beneficial new technology was that radiographs were self-evident. Specialist physicians and surgeons who were shown a “penetrating photograph” of their patients believed they needed no one else to interpret the meaning of those images. At Boston Hospital in 1901, x-ray pioneer Francis Williams described the “standardized” x-ray reporting and medical record process:
“The blank (filled out by the house officer and signed by the visiting physician and surgeon) gives the name of the patient, the ward and bed number, also the volume and page of the medical or surgical record. The negatives are seen by the surgeon or the physician, as the case may be, and a print of each is inserted in the record book as part of the record of the patient.”
Modern imaging specialists reading this account of standard operating procedure are painfully aware of the missing ingredient: the radiologist.
“A GOOD FEE TO A GOOD SPECIALIST”
It was during the first 25 years of the x-ray, from 1896 until just after the first World War, that radiology would face the uphill task of establishing itself as a distinct medical specialty. The written report would play a large role in convincing other physicians that specific and cumulative skills were needed to interpret radiographs. Most large academic hospitals soon had at least one full-time medical doctor devoted to reading imaging studies. Often, however, the radiologist was in danger of being unseen and unheard. The interpretive skills of the x-ray specialist had to be proven both to doubting clinicians and to hospital administrators.
Several factors worked in the radiologists’ favor. The first was the sheer bulk of the product. Glass plate roentgenographs, sometimes as large as 11 x 18 inches, were heavy and fragile. Most hospital departments established a policy that, except in surgical emergencies, the x-ray department should maintain custody of the plates and the referring physician would go to the department to view the images. This provided an opportunity for radiologists to demonstrate specific and unique knowledge and to illustrate fine points of the technology of which clinicians might not be aware. This direct contact, which persisted in many institutions until quite recently, went far in establishing radiologists as colleagues rather than suppliers.
The second factor that kept the radiologist in the position of consultative colleague was the dogged determination of these early professionals to avoid becoming salaried employees of institutions. By the second decade after the discovery of the new technology, many hospitals had worked out fee-splitting arrangements with their radiologists, dividing up profits from imaging services on a prearranged basis. In a survey taken in 1909, several larger hospitals reported paying their radiologists salaries of $1,000 per year – not a large sum by the standards of the time. The difficulty arose when hospitals attempted to treat salaried radiologists as technical employees of the hospital rather than as practitioners of a medical discipline. Early professional radiology organizations supported the efforts of their practitioners to maintain private practices and hospital privileges just like their medical colleagues in other specialties. As part of this effort, radiologists found themselves “selling” their skills and professional abilities both to their colleagues in the hospital and to referring physicians in the community.
There were clearly those who doubted that the “product” offered by private practice and fee-for-service radiologists was worth the high price paid for “merely looking at pictures and writing down what we can see plainly for ourselves.” For these doubters, George C. Johnston, MD, president of the American Roentgen Ray Society in 1909, had this advice:
“The general practitioner sees the roentgenologist charge $15 or $25 for an examination when he is accustomed to spending a night at a labor case for $10, and he demurs. He fails to see that every man who pays a good fee to a good specialist appreciates all the more the value of medical services.”
Although one interpretation of this comment could be the interesting observation that the simple act of paying a high fee raised the level of appreciation, it is more probable that Johnston meant that the services provided by a capable radiologist proved their own worth to referring physicians.
OF DRESSING UP AND ROUTINIZING
|Figure 3. Along with cordial, typed letters detailing his findings to referring clinicians, prominent New York roentgenologist Lewis Gregory Cole sent photographic prints of the features of interest.|
The most effective way of pointing up the value of the radiologists’ services lay in the preparation and delivery of the radiology report. In the 1910s, several private practitioners took full service reports to innovative and complex lengths. Like many of his colleagues, Lewis Gregory Cole, a prominent New York roentgenologist, sent cordial, typed letters detailing his findings to referring clinicians (see Figure 3). But this Park Avenue practitioner went several steps further. He also sent along special packets that contained photographic prints of the features of interest. Each print was nestled in its own cutout cartouche, with additional highlighting and arrows to explain points mentioned in the letter. Background information, other observations, and sometimes references from the literature were included. The entire packet, made of powder blue rag paper from one of New York’s finest stationers, was then folded and tied with a matching grosgrain ribbon. Knowing the importance of patient requests in future referrals, Cole included a second copy of the packet for those physicians who wished to send each patient home with copies of his or her own radiographs. Cole’s practice, like that of many radiologists who went to a more detailed reporting style during this period, was quite successful. He charged as much as $100 per consultation (at a time when the average annual salary in the United States was less than $750).
Cole and other private radiologists could not have produced their voluminous reports and supporting material without the aid of three factors: clerical assistance, certain routinized elements in reporting, and mechanical aids to transcription. The dictation machine (usually either a Dictaphone or Ediphone) was already in use in the 1910s, imprinting recordings of the spoken word onto a wax cylinder for playback and transcription. In large practices such as Cole’s and at urban hospitals, cases were dictated and then typed by pools of transcribers or “typewriters.” Reviews of carbon copies of Cole’s office records make it clear that almost every report contained some boilerplate material that probably was not respoken with every dictation. Lists of descriptive prompts and phrases such as “From a study of these plates one is justified in stating that…” are repeated in almost every report.
STICKING WITH WHAT WORKS
By 1920, radiology had established itself as a separate specialty, although it would continue to struggle to maintain parity with other specialties for many years. The basic elements of the radiology report were firmly in place as well. The prevailing style of report would go back and forth between prepared form and “letter of consultation” (sometimes supplemented by line drawings and/or marks made directly onto film) over the course of the century, determined largely by type of examination, size of institution, and (to be frank) by the financial and/or insurance status of the patient. The content of these reportsthe order of elements and the level of analysisremained virtually unchanged for decades. If the results of a posterior/anterior and lateral study made in 1930 and one made in 1990 were read aloud, a modern listener would be hard pressed to differentiate between the two. Except perhaps for typeface, the reports would likely be physically quite similar, regardless of the dictation/transcription/sign-off routine used to bring them to completion.
In the past decade, the impetus for revolutionary change in radiology reporting has come not from medical innovations but from technological breakthroughs in computing and engineering. The rapid growth of computing power and the advent of practical, effective speech recognition technology in the late 1990s challenged imaging specialists to create the radiology record in an entirely new way. Yet most of these new technologies have been used to add value in turnaround time and economies associated with transcription personnelno fundamental changes have been made in the actual process of reporting (words-to-text) or the form of the report (much the same in content as it has been for well over a century). New modalities, such as magnetic resonance imaging, CT, and PET, may have changed the wording in some reports but not the basic form or content.
Some industry observers believe that we are ready for a radical shift in thinking about the entire process of radiologic reporting. The technology is already available to radically restructure the process of creating reports and also to open up dramatic new possibilities for enhancements, links, addenda, and information highlighting that will continue to prove the value of the radiologist in the increasingly complex digital medical environment. In the second part of this series, we will look at innovations that are just around the corner and at others that hold promise for the next decade and beyond.
Nancy Knight, PhD, is a medical writer based in Washington, DC.
Bruce I. Reiner, MD, is director of research, VA Maryland Health Care System, Baltimore . Parts of this article were taken from the authors’ previously published introduction to Electronic Reporting in the Digital Medical Enterprise. Great Falls, Va: Society for Computer Applications in Radiology