Evaluation of magnetic resonance imaging regulatory methods in the Israeli Air Force.

BACKGROUND: The use of magnetic resonance (MR) imaging (MRI) among many medical professions is growing. Many health care systems have formed control mechanisms to ensure proper utilization of MRI. This western world trend is also valid in the Israeli Air Force (IAF). At the time of the study, two methods existed for consideration of MR requests in the IAF: (1) consideration by a primary reviewer, no clinical guidelines (applied to all MR examination requests, knee MR excluded). (2) Consideration by a primary reviewer according to basic clinical guidelines established by the Israeli Defense Forces medical section and by communication with an orthopedic specialist (applied to knee MR requests). Both methods did not include consultation with established criteria (such as American College of Radiology (ACR) appropriateness criteria). OBJECTIVE: To evaluate the appropriateness of the current regulatory methods of MRI utilization in the IAF by comparing approval/rejection decisions to established ACR criteria. METHODS: The study is a retrospective analysis of written records of air force personnel, for whom MRI was requested by a specialist. We gathered information regarding the clinical problem and the final decision concerning approval or rejection of the MRI request. We then consulted with the ACR appropriateness criteria. In case a matching ACR clinical variant was found, an appropriateness value was assigned to the request. Otherwise, the request was noted as "ACR irrelevant". We predetermined to label all studies with an ACR value of 1 to 3 as inappropriate, 4 to 6 as "gray zone," and 7 to 9 as appropriate. We then compared the ACR-based decision to the original outcome of the request. RESULTS: The overall approval rate for MRI requests evaluated by a primary reviewer only was 96%. The overall approval rate for MRI requests evaluated by a primary reviewer, basic clinical guidelines, and specialist consultation was 51%. Among the four most prevalent MR requests types (brain, knee, spine, and shoulder), requests in the 7 to 9 scale (appropriate requests) accounted for 52%. Regarding appropriate requests, there was a 100% approval rate by a primary reviewer only compared with 17% for requests considered by a primary reviewer, basic guidelines, and specialist consultation (83% of appropriate requests were rejected by this method). Requests in the 1 to 3 scale (inappropriate requests) accounted for 3% of all requests. In this group, there was a 100% approval rate by both methods of consideration. Requests in the 4 to 6 scale (gray zone requests) also accounted for 3% of the total and requests which could not be assigned an ACR appropriateness value (ACR-irrelevant requests) accounted for 42% of total requests. The rate of approval of these requests by a primary reviewer only and by a primary reviewer, guidelines, and consultant was 97% and 83%, respectively. CONCLUSIONS: Both MR approval mechanisms that were applied in the IAF have not shown a strong correlation with ACR appropriateness criteria, with significant rates of both overuse and underuse of MRI. The high rate of requests that could not be assigned an ACR appropriateness value may indicate a need to broaden the appropriateness criteria coverage of clinical conditions and variants.

Nonradiologist utilization of American College of Radiology Appropriateness Criteria in a preauthorization center for MRI requests: applicability and effects.

OBJECTIVE: The purpose of this study was to evaluate the effects and applicability of use of the American College of Radiology (ACR) Appropriateness Criteria by nonradiologist physicians in an MRI preauthorization center. MATERIALS AND METHODS: All MRI requests received at our preauthorization center during a 19-month period were included in the study. The study period was divided into preintervention and postintervention phases, indicating before and after introduction of the ACR criteria to the general practitioners staffing our center. ACR appropriateness values were classified into three groups: appropriate, indeterminate, and inappropriate. Requests for which a matching ACR value could not be assigned were labeled ACR-noncodable. Multiple parameters evaluated and compared for the two phases included rate of request receipt, total approval and denial rates, and approval and denial rates according to the ACR Appropriateness Criteria and by anatomic region to be evaluated. RESULTS: There was no significant change in rate of request receipt and total approval and denial rates. However, there was an increase in the rate of approval of appropriate requests (phase 1, 71/96 [74%]; phase 2, 74/76 [97%]; p < 0.001) and the rate of denial of inappropriate requests (phase 1, 0/12 [0%]; phase 2, 9/13 [69%]; p < 0.001). More than 40% of requests were marked "ACR-noncodable" because of a lack of a matching clinical condition or variant. CONCLUSION: Introduction of the ACR Appropriateness Criteria resulted in an increase in the rate of performance of appropriate MRI examinations and a decrease in the rate of performance of inappropriate MRI examinations. ACR Appropriateness Criteria were applicable to approximately 50% of MRI requests.