We read with interest the systematic review by Jenkins and colleagues.1 Yes, imaging is sometimes overused and is a substantial financial cost for our publicly funded health care system.2,3 However, we question the utility of assessing interventions by measuring reductions in imaging referral rate. This yields information pertinent to cost-savings and does little toward balancing patient care. Referral rates are a poor surrogate for the more important and relevant outcomes of imaging appropriateness and patient-important outcomes (i.e., diagnostic yield and subsequent treatment outcomes).
An Ontario Ministry of Health bulletin, disseminated to all physicians in June 2012 stated that:
OHIP payment eligibility will continue to be limited to services that are medically necessary to the individual patient’s circumstances, informed by the best available evidence. New language has been added to the Schedule of Benefits for Physician Services noting that studies of the lumbar spine should not be routinely ordered or rendered without suspected or known pathology … If the diagnostic services were found to be not medically necessary in accordance with s. 18.2(1) and 18.2(2) of the Health Insurance Act, the physician requesting the diagnostic services will be responsible for repayment.4
Such an aggressive approach to reducing referral rates remains unparalleled in Canada. Threats of financial retribution against referring physicians are a dangerous precedent and could make physicians reluctant to order imaging regardless of whether it is needed. Such decisions would not be guided by concern for the patient but by the desire for cost-savings. This approach was shown to be ineffective in reducing referral rates.5
Though costly, imaging undoubtedly has value. We should not be guided by evidence conceived with cost-savings in mind but by evidence based on patient-important outcomes.