TO THE EDITORS
We have concerns about the "Methods" and "Discussion" sections in the study reported in the December 2005 issue by Weiner,1 "Radiology by Nonradiologists: Is Report Documentation Adequate?"
In the "Methods" section, the comparison radiologists' reports were not selected using the same methodology as the self-referring clinicians' reports. Two hundred bone, joint, and chest x-ray reports were requested from 50 self-referring clinicians. However, only 50 reports from an unknown number of radiologists were randomly selected from reports on file in a quality management department. Only 43 were bone, joint, and chest x-rays, and 7 were abdominal x-rays, including upper gastrointestinal and barium enema examinations. This is not a valid comparison group. In addition, it is true that the Centers for Medicare & Medicaid Services requires a separate radiograph interpretation report to bill for the professional component of radiologic services,2 yet the author does not comment whether the 2 participating health plans have a similar requirement. Radiologists must generate a separate report, as they are never the treating physician. If the health plans do not require a separate report from the treating physician, then this comparison study is inappropriate. Furthermore, a single radiologist reviewed the reports for the "required elements" yet was not blinded to the specialty of the reporting physician. This introduces the potential for significantly biased results.
In the "Discussion" section, under the subsection titled "Inadequate Imaging Reports Compromise Care," the author states that "myriad documentation errors among records from nonradiologists" were found. We are not sure how many errors constitute a myriad; however, the radiologists made many errors of omission as well, admittedly not as many as the self-referring clinicians. The author also makes the unsubstantiated comment that "communication breakdown may result in repeat examinations, which increase costs [and] expose patients to unnecessary radiation." Our study3 of 1393 radiograph reports interpreted initially by family physicians found the opposite result. When there was disagreement between the radiologist and family physician, we found more unnecessary repeat and additional examinations that did not change the management or outcome for the patient. Finally, the author also states that poor communication "may potentially delay patient care," when in fact the interpretation of radiographs at the point of care is always faster than waiting for a radiologist's interpretation hours or days later.
Point-of-care interpretation of ambulatory radiographs has shown few if any substantive changes in care when a radiologist provides a second reading sometime after the original patient encounter. Studies4-6 of orthopedists' initial radiograph reading and clinical management with a second reading by radiologists have uniformly found no change in clinical care resulting from the second reading. Similarly, our study3 and a study by Halvorsen et al7 reported zero substantial changes in care as a result of the radiologists' second reading.
We agree with Weiner that radiologists and other specialists should improve the quality of their radiology reports and should be held to the same standards for reporting and communication of results. However, this fundamentally flawed study does not support the assertion that inadequate imaging reports compromise care by self-referring clinicians.
Paul D. Smith, MD
Marlon Mundt, MA, MS
University of Wisconsin Medical School, Madison
Am J Manag Care.
1. Weiner SN. Radiology by nonradiologists: is report documentation adequate? 2005;11:781-785.
2. Centers for Medicare & Medicaid Services. The Carriers Manual, Part 3, Chapter XV, Â§ 15022 A and Â§ 15023 A. Available at: http://new.cms.hhs.gov/Manuals/PBM/list.asp#TopOfPage. Accessed March 7, 2006.
J Am Board Fam Pract.
3. Smith PD, Temte J, Beasley JW, Mundt M. Radiographs in the office: is a second reading always needed? 2004;17:256-263.
4. Clark R, Anderson MV, Johnson BH, Moore DE, Herbert FD. Clinical value of radiologists' interpretations of perioperative radiographs of orthopedic patients. 1996;19:1003-1007.
5. Anglen J, Marberry K, Gehrke J. The clinical utility of duplicate readings for musculoskeletal radiographs. 1997;20:1015-1019.
J Pediatr Orthop.
6. Parmar VS, Stanitski DF, Stanitski CL. Interpretation of radiographs in a pediatric limb deformity practice: do radiologists contribute? 1999;19: 732-734.
J Fam Pract.
7. Halvorsen JG, Kunian A, Gjerdingen D, et al. The interpretation of office radiographs by family physicians. 1989;28:426-432.
RESPONSE TO LETTER TO THE EDITORS
In response to the concerns raised by Smith et al1 about my article "Radiology by Nonradiologists: Is Report Documentation Adequate?",2 the study was not designed to address the accuracy of interpretation but rather the adequacy of communication of results and the accessibility to the imaging record. Therefore, the type of study evaluated was not considered to be critical. Nonradiologists are privileged by our health plans to self-refer a limited number of x-ray examinations. The most commonly self-referred studies are chest and bone films. As stated in the article, reports were requested from 200 different self-referring providers selected at random from paid claims data for 2 different health plans. Fifty chest x-rays and 50 bone or joint films from each health plan were requested; 147 responses were received. These were compared with 50 different radiologists' reports.
Current Procedural Terminology.
In addition, health plans do not specifically address the issue of a separate imaging report to bill for the professional component of radiologic services, to my knowledge. However, as indicated in the article, the American Medical Association addresses the reporting of results in the 2004 edition of 3 Good documentation and communication are essential when caring for health plan members who may be seen by several different healthcare providers for the same clinical problem.
Regarding the comment by Smith et al about the single radiologist who reviewed the reports for the required elements, the article clearly states that the study was nonblinded. The elements evaluated were objective; therefore, the potential for biased results was limited.
I strongly agree that all physicians, including radiologists, need to improve communication of results and documentation of records. Again, the study was not designed to address the accuracy of interpretation of x-rays by primary care physicians and orthopedists. This has been well documented in the literature. Rather, the study addressed the problem of communication of results. Failure to have a stand-alone report may limit access to the results by other healthcare providers. The interpretation may be buried in office notes that are difficult to read and sometimes poorly organized. The potential problem is not at the initial point of care but rather when the member is referred to another healthcare provider who does not have access to the report or to the films and, to expedite treatment, repeats the study.
Shelley Nan Weiner, MD
Wappingers Falls, NY
Am J Manag Care.
1. Smith PD, Temte J, Beasley JW, Mundt M. To the editor. 2006;12:352.
Am J Manag Care.
2. Weiner SN. Radiology by nonradiologists: is report documentation adequate? 2005;11:781-785.
CPT 2004 Physician's Current Procedural
3. American Medical Association. Chicago, Ill: American Medical Association Press; 2004:208.