Frequency and costs of diagnostic imaging in office practice--a comparison of self-referring and radiologist-referring physicians.

BACKGROUND

To assess possible differences in physicians' practices with respect to diagnostic imaging, we compared the frequency and costs of imaging examinations as performed by primary physicians who used imaging equipment in their offices (self-referring) and as ordered by physicians who always referred patients to radiologists (radiologist-referring).

METHODS

Using a large, private insurance-claims data base, we analyzed 65,517 episodes of outpatient care by 6419 physicians for acute upper respiratory symptoms, pregnancy, low back pain, or (in men) difficulty urinating. The respective imaging procedures studied were chest radiography, obstetrical ultrasonography, radiography of the lumbar spine, and excretory urography, cystography, or ultrasonography.

RESULTS

For all four clinical presentations, the self-referring physicians obtained imaging examinations 4.0 to 4.5 times more often than the radiologist-referring physicians (P less than 0.0001 for all four). For chest radiography, obstetrical ultrasonography, and lumbar spine radiography, the self-referring physicians charged significantly more than the radiologists for imaging examinations of similar complexity (P less than 0.0001 for all three). The combination of more frequent imaging and higher charges resulted in mean imaging charges per episode of care that were 4.4 to 7.5 times higher for the self-referring physicians (P less than 0.0001). These results were confirmed in a separate analysis that controlled for the specialty of the physician.

CONCLUSIONS

Physicians who do not refer their patients to radiologists for medical imaging use imaging examinations more frequently than do physicians who refer their patients to radiologists, and the charges are usually higher when the imaging is done by the self-referring physician. From our results it is not possible to determine which group of physicians uses imaging more appropriately.