Diagnostic imaging (DI) tests (for example, x-ray tests, ultrasounds, CT scans, or MRIs) are used by health care providers to help diagnose patient illness, but decisions regarding when these tests should be ordered are subjective. As a result, some physicians order these tests more than others. Ordering rates between clinicians seeing similar kinds of patients have been shown to be considerably different, suggesting that many of the tests are unnecessary. DI currently accounts for about 6.6% of Canadian hospital budgets, but this percentage may be much higher in Newfoundland where test ordering rates are almost twice the national average. However, cost is not the only concern. Over-testing can lead to further unnecessary testing to follow-up on harmless findings, and in the case of CT, large doses of potentially harmful radiation. We suspect that many clinicians are not aware that they order more DI tests than their peers. We will therefore develop a "report card" for family physicians in the province that shows them how many tests they are ordering compared to other physicians in the region. We expect that physicians who are over-ordering DI tests will reduce the number of tests they order after receiving their report cards. This is a low-cost way to potentially prevent expensive over-ordering of DI tests that can easily be implemented in the province's other health regions and elsewhere. The Quality of Care Newfoundland and Labrador (QCNL) organization in the province currently provides feedback reports and in-person detailing sessions to physicians, but the effectiveness of these interventions has not been studied.

The purpose of this trial is to compare the effectiveness of two versions of a DI utilization feedback report for family physicians, and determine whether in-person detailing sessions offer additional benefit in changing ordering practices. Group practices (i.e. all physicians practicing at the same address) of family physicians and general practitioners will be stratified by community of practice within Eastern Health, then randomized into one of four groups: 1. A usual QCNL feedback report alone, 2. A usual QCNL feedback report plus in-person detailing, 3. A new feedback report alone, 4. A new feedback report plus in-person detailing. Communities with fewer than five physicians will be grouped with similar communities for stratified randomization purposes.

Physicians will be provided a semi-annual (every 6 months) report card outlining the number of CT scans, ultrasounds and plain x-rays they ordered per 100 patients for whom they were the primary provider. We will assume that the primary provider is the most frequent biller of primary care services. Participants will receive a link to the report card via email from the Newfoundland and Labrador Medical Association (NLMA) by a two-step process: The initial email gives a brief description of the report and contains a link which then redirects the clinician to a secure web page that displays the individual clinician diagnostic imaging utilization in a prior one year period compared to the aggregate of their peers in the same region. Physicians in the detailing groups will be contacted to arrange a single in-person session with one of their colleagues to discuss the detailing reports.

There are no restrictions on the age of participants.

Inclusion Criteria:

• 1) All family physicians or general practitioners practicing within Eastern Health. In order to obtain an accurate assessment of the report in real-world implementation, we are requesting the right to enroll everyone who meets the inclusion criteria, and to waive the requirement to obtain informed consent
• 2) Be the most frequent provider of primary care billings for a minimum of 20 patients during the period.

Exclusion Criteria:

• 1) Physicians who will be within the Eastern Health network for a relatively short period, e.g. clinicians engaged in an exchange program for training, or locums.