Forty percent of radiology practices in the U.S. perform outside reading for other institutions, with smaller and non-suburban practices taking on significantly larger workloads of outside reading than other institutions, according to a study published in the November edition of the American Journal of Roentgenology.
The authors sent 2,400 members of the American College of Radiology (ACR) surveys asking respondents whether their practice performed outside readings, what fraction of their workloads consisted of outside readings and various other demographic and billing information about the department. The survey was part of ACR's 2007 Survey of Radiologists.
"As best we could ascertain, quantitative information on outside reading by radiology practices had not been gathered previous to the ACR's 2007 survey," wrote Ryan I. Huffman, of Yale University School of Medicine, and colleagues. Huffman and colleagues sought to shed light on what they presumed to be the expanding area of outside reading, in which teleradiology and imaging conducted at nonradiology facilities that send the images to radiology practices for review are two of the most common forms.
Twenty percent, or 480 of the 2,400 radiologists, queried completed the survey. Given the survey's relatively low response rate, the findings were weighted to more authentically represent the entire population of U.S. radiology practices.
Forty percent of respondents confirmed that their department performed outside reading, with outside readings constituting an average of 11 percent and a median value of 5 percent of these practice's workloads. Forty-five percent of radiologists reported being part of a practice that performed outside readings.
By multiplying the proportion of radiologists in departments conducting outside readings by the average workload for which outside readings account, the authors estimated that outside reading totals 4 percent of the workload of all U.S. radiologists.
Outside reading varied substantially between practice type and size. Radiology departments at academic, private non-academic with subspecialties and government institutions were significantly less likely to perform outside readings than private radiology-only practices, with 22 percent of academic departments, 25 percent of private radiology departments with subspecialties and just 2 percent of government facilities performing outside readings. These numbers were compared with 44 percent of private radiology groups and 51 percent of solo radiology practices that performed outside readings.
The authors argued that these findings "suggest that performing outside reading is an entrepreneurial activity more than a service activity. That seems the most plausible explanation for the finding that academic practices, government practices, and the radiology departments of multispecialty practices are substantially less likely to perform outside readings than nonacademic private radiology-only practices."
Larger practices were also found to be more likely to perform outside readings, while outside reading constituted a larger proportion of radiologists' workloads in smaller practices.
Patterns of outside reading also varied dramatically by region, with practices in the Midwest being almost twice as likely (54 percent versus 28 percent) as departments in the Northeast to perform outside readings. Practices located in suburbs also had markedly higher rates of outside reading performance, though these findings were not statistically significant.
The mode of payment for outside reading also varied considerably, with the average practice utilizing 1.14 methods. The most common billing method was for the reading practice to bill for professional charges while the institution for whom the reports were read would bill for the technical component; 48 percent of practices employed this method either exclusively or along with other billing methods. The next most common form of billing was for reading radiology departments to receive a flat fee per study from the referring institution.
The authors acknowledged that the study's low response rate might have introduced nonresponse bias into the statistical analysis of outside billing when the authors weighted the findings for more encompassing representation of U.S. radiology as a whole. Huffman and co-authors also emphasized the importance of establishing a clear understanding of outside reading. "The survey