58 Acute low back pain in australian general practice: image requests according to medicineinsight

Objectives The aim of this study was to use general practice data from the MedicineInsight program to determine the proportion and type of imaging of the low back region requested for people with acute low back pain (LBP) in Australian general practice. Many GPs are aware of the limited role of imaging in non-specific LBP, yet a large proportion of patients presenting to GPs with acute LBP receive an image request. Methods MedicineInsight operates under ethics approval and a robust data governance framework. De-identified patient data are extracted routinely from the clinical information systems (CIS) of consenting general practices participating in the MedicineInsight program. To ensure high quality data, MedicineInsight applies acceptability criteria before extracting from the CIS. A cohort of patients aged 18 years or older with acute LBP was identified by searching the MedicineInsight database. Both coded and free-text condition information were analysed from designated CIS fields including: diagnosis, reason for encounter and reason for prescription. Imaging requests were included only if the imaging was specific to the low back region. Requests made through the CIS, and results of imaging requests, including those ordered by other external clinical providers, received back to the practice were captured. Results National MedicineInsight data at 1 July 2018, show that from a total of 1,865,565 patients aged 18 years or older, 86,377 (4.6%) were managed for the condition of acute LBP in MedicineInsight practices in the previous 12 months. Of these patients, more than one third (34%) received a lumbosacral image request in the previous 12 months (either Medicare Benefits Schedule [MBS] or privately rendered). The highest proportion of imaging requested in these patients was for computed tomography (CT) scans (57%), followed by X-rays (42%) and then magnetic resonance imaging (MRI) (19%). Unique patients were counted more than once only if they received more than one type of image request. Conclusion Lumbosacral imaging is not indicated for the management of acute non-specific low back pain and should be reserved for the very small amount (up to 1%) of patients with acute low back pain where a potentially serious underlying condition is suspected. Outcomes of this study showed that more than one third of MedicineInsight patients with acute low back pain received a lumbosacral image request in the previous 12 months. MedicineInsight data also indicated CT scans were the most common request types when imaging was requested for these patients, which was comparable to MBS lumbosacral imaging data for the financial year 2017–18. This study shows the type and extent of imaging requested for patients with acute low back pain in Australian primary care, which is clearly too high. The effectiveness of presenting MedicineInsight data to GPs to influence rates of image request for patients with LBP is yet to be evaluated.