Reference to specialist care after magnetic resonance imaging of the lumbosacral spine in primary care
MetadataShow full item record
OBJECTIVE: To examine the extent of health care services useafter a magnetic resonance imaging (MRI) scan of the lumbosacral spine ordered by a primary care physician. BACKGROUND:The use of MRI of the spine in the primary care setting is increasing, yet little is known about the relationship between MRI scan findings and subsequent patterns of health care utilization. MATERIAL AND METHODS: We obtained all MRI scans of the lumbosacral spine performed throughout the year of 2008 in Sunnmøre, a province in Norway. The abstracted data regarding these MRI scans, ordered by primary care physicians, were linked to administrative hospital databases. RESULTS: Of the 2034 patients who underwent a lumbosacral spine MRI scan ordered by a primary care physician, 311 (15,3%) were seen in a consultation by an orthopedic surgeon or a practitioner in physical medicine and rehabilitation, and 84 (4,1%) received spine surgery during 2-3 years of follow-up. Of the 1415 that had nonspecific findings on MRI, 136 (9,4%) were referred to a physician specialist. Similarly 169 of 534 (31,6%) patients with radiological nerve root affection and 6 of 49 (12,2%) with possibly serious spinal pathology were referred to an orthopedic surgeon or a practitioner in physical medicine and rehabilitation. The latter patients were likely referred to other physician specialists not accounted for in this study. CONCLUSION: Patients receiving MRI scans of the lumbosacral spine in a primary care setting do not appear to generate considerable use of specialist care. Observations suggest that primary care physicians do discern between MRI findings before referring patients to specialist care. Spinal disorders are common clinical problems in the general adult population and cause considerable impairment and disability. In terms of work absenteeism and health care utilization it imposes a substantial socioeconomic burden. About three-quarters of adults are affected by low back pain at some point during their lifetime (1). Prevalence ranges from 9-17% (2). In Canada annual incidence of low back pain was found to be 18,6% (3) and in Norway daily prevalence is 15-20% (4). Magnetic resonance imaging (MRI) is frequently usedto evaluate low back pain. When used in the assessment of these patients it can provide valuable information and help guide subsequent care. However we do know that MRI scans often detect spine abnormalities that are not the cause of symptoms(5). Clinical guidelines have been developed, including indications for MRI, to assistthe management of low back pain (6, 7). Still, MRI availability continues to increase, and concerns have been raised about the potential overuse of MRI for the evaluation of these patients. Low back pain is most commonly treated in primary care settings (6). Thus, the primary care physician must usually discern the most appropriate candidate for specialist care and for surgical referral. Assessing these patientscan be challenging and imaging is often performed in the absence of clear indications(8). Although MRI scanning has been in clinical use for decades, little is known about the results at a population level. In a primary care setting there is still less published data about patterns of subsequent care among patients receiving MRI scans of the lumbosacral spine. We implicate that obtained data regarding this should inform discussions about MRI use among primary care physicians and subsequent healthcare utilization. If only a small proportion of patients receiving MRI scans are referred to physician specialists it may suggest that most patients are adequately managed in a primary care setting. Alternatively, if many patients are referred to specialist care it could seem that MRI scans contribute to increased referrals and that optimizing the management should be pursued. The objective of this study was to examine the extent of health care services utilization after a performed MRI scans of the lumbosacral spine in a primary care setting, by linking these scans to administrative hospital databases. Another objective was to detect MRI scan findings and see how they distribute amongthe three categories: nonspecific findings, nerve root affection and possible seriousspinal pathology.