Background/Purpose:

The US Department of Veterans Affairs (VA) has developed a national continuing professional development program to train primary care providers (PCPs) in the care of patients with common musculoskeletal (MSK) conditions. Utilization of advanced imaging technology (e.g., magnetic resonance imaging (MRI)) in evaluating knee pain is an important concern because inappropriate ordering of MRI adds to costs in health care systems without increasing benefits, and limits access to this technology for other patients who may have greater need. Several recent reports suggest that PCPs overuse MRI in evaluating knee pain, and have called for the creation of educational programs and establishment of performance measures to address this issue. The aim of this study was to investigate the impact of this educational program on providersÕ utilization of MRI in the evaluation of knee pain.

Methods:

Two hundred twenty seven providers from 13 VA medical centers participated in the MSK ÒMini-ResidencyÓ between April 2012 and October 2014. All orders for knee MRIs submitted by these providers over the 12-months prior to their participation in the mini-residency (pre-training) were reviewed, as well as all orders submitted over the 12-months following their participation (post-training). MRIs were categorized as follows:

ÒInappropriateÓ:  No prior weight-bearing x-rays done within the 12 months preceding the MRI order

ÒProbably InappropriateÓ: Findings of osteoarthritis (OA) described on x-ray report

ÒPossibly AppropriateÓ: No findings of OA described on x-ray report.

The number of MRIs in each category was recorded for the pre-training and post-training period specific to each provider, and the number of MRIs in each category was tallied. Differences in the numbers of MRIs that were ordered post-training as compared to pre-training for each of the three categories were evaluated using paired StudentÕs t-test (2-tailed).

Results:

Numbers of MRIs ordered in the 12 months preceding training and following training for each of the categories described above are presented in the Table:

Conclusion:

Following the MSK Mini-Residency program, the total number of MRIs ordered by participants decreased by 27%. This reduction was greatest in the number of studies classified as inappropriate. The total number of knee MRI orders that were categorized as either inappropriate or probably inappropriate decreased by 100 (32%), while the number of those that were possibly appropriate were not significantly changed. These findings provide further evidence that the VA MSK Mini-Residency program is effective in changing provider behavior and improving access to appropriate care for patients, though the high percentage of MRI scans classified as inappropriate indicate that additional work is needed in this area.