Does self-referral for imaging actually yield “same-day service”?

Objective: Diagnostic imaging has been growing particularly rapidly in recent years, and the share in imaging of non-radiologist physicians themselves performing high-tech, high-cost imaging has grown especially rapidly. For a given medical condition, self-referral has been found to result in much higher rates of imaging utilization than radiologist referral.. However, defenders of self-referred imaging assert it provides same-day, one stop service which, in turn, contributes to better quality in three ways: through the patient convience of one-stop service; by getting treatment started immediately instead of only after a visit to an imaging facility and then a repeat visit to the treating physician; and eliminating non-compliance that would result from some patients never completing the three-stop pathway of initial visit, trip to an imaging facility, and return.

The goal of this study is to examine in what percent of patients imaging self-referral actually provides same-day service. No empirical information has been available.

Method: Imaging claims from the 2005 Medicare 5% Research Identifiable Files are studied. For each imaging claim by a non-radiologist, we observe whether there was an evaluation and management (E&M) service (visit) on the same day with the same physician. We conduct the analysis separately for different types of imaging.

Results: Of the self-referred images performed in offices, only 14% of MRIs, 25% of CTs, and 12% of nuclear medicine procedures were accompanied by an office visit on the same day with the same physician as the imaging procedure. For the most straightforward X-rays (chest X-rays and orthopedic X-rays), nearly three-fourths of the in-office self-referred imaging was performed on the same day as an office visit with the same physician. For less straightforward X-rays and for ultrasounds, the rates of same-day imaging were between 28% and 35%.

Conclusions: Except for the simplest X-rays, same-day service is uncommon. Thus, improved quality and or/patient convenience do not, realistically, provide a rationale for self-referral except possibly for the most straightforward X-rays. Policies aimed at restricting financial self-interest in imaging because of its effects on utilization might usefully take note of this distinction. Self-referral for MR, CT, and nuclear medicine could be banned with little effect on access or patient convenience.