Interventions to reduce imaging use in low-back pain


Rates of investigation by way of imaging (X-ray, computed axial tomography (CAT) and magnetic resonance imaging (MRI)) for low-back pain are high and can lead to radiation exposure. Not only does unnecessary imaging lead to higher healthcare costs but it has also been associated with poorer patient outcomes.

Estimates in 2008 were that imaging accounts for 7% of direct costs associated with low-back pain (Dagenais,,2008). At this time this percentage equated to £114 million in the UK, which would obviously now translate to much higher current costs.

As an Elf working within the spinal division of an Orthopaedic service I was keen to investigate the effectiveness of interventions designed to reduce the use of imaging and, consequently, improve both my practice and my patients’ outcomes. I found a systematic review which did just this.

Here’s what they did

The authors searched MEDLINE, Embase, CINAHL and the Cochrane Central Register of Controlled Trials from earliest records to June 2014. Eligible studies included RCTs, controlled clinical trials and interrupted time series studies (only if they included at least three individual data-collection points before and after the selected intervention). The interventions were designed to decrease the use of imaging in any clinical setting – primary (general practitioners) or secondary (emergency and specialist care/clinics in a hospital setting) care.

Interventions included:

  • A modified referral form used in secondary care
  • Targeted reminders to GPs which specified appropriate interventions for imaging
  • Individual practitioner audit and feedback
  • Practitioner education (face to face workshop which included guideline dissemination and implementation strategy
  • Guideline dissemination via post

The review protocol was based on guidelines produced by the Cochrane Effective Practice and Organisation of Care (EPOC) Review Group and the PRISMA statement.

Here’s what they found

Of the 8500 records identified only seven were included in the review.

  • The modified referral form produced the largest reductions in imaging (absolute change -44.3, 95% CI -48.7 to – 39.9 where the reduction was in referrals submitted the month after introduction of the form).
  • Targeted reminders produced an absolute change of -1.5 radiographs per 1000 patients (95% CI -2.5 to -0.6).
  • Individual practitioner audit and feedback, practitioner education (face to face workshops) and postal guideline dissemination were not found to significantly reduce imaging rates.

The authors concluded

Clinical decision support involving a modified referral form in a hospital setting and targeted reminders to primary care doctors of appropriate indications for imaging were interventions that specifically decreased the use of imaging for low-back pain by 36.8% and 22.5% respectively.

A modified referral form in a hospital setting and targeted reminders to primary care doctors decreased the use of imaging for low-back pain by 36.8% and 22.5% respectively.

A modified referral form in a hospital setting and targeted reminders to primary care doctors decreased the use of imaging for low-back pain by 36.8% and 22.5% respectively.

The Musculoskeletal Elf’s view

The Musculoskeletal Elf

The authors acknowledge limitations,  most notably the small number and quality of the studies, that meta analysis was not possible and that one person performed the initial screening of articles which were only in English.

The statistics produced for the percentage reduction in referrals subsequent to providing clinical decision support were not a direct finding from the relevant study. Rather, the authors conducted time series regression analysis with the statistics to produce their findings. Associated change in the time series regression lines, however, exhibited wide confidence intervals making it unclear whether the initial promising decrease in imaging referrals would be maintained over time. This 36.8% reduction in imaging was based on only one study published in 1987 where number of participants was not reported and the strength of these findings, therefore, has to be limited.

Outcome measures varied between studies. Two merely reported on whether imaging was done or not. Five of the studies did not confine imaging only to low-back pain but within the ‘Limitations’ section it appears that there was at least one other study (assessing guideline dissemination) that was not included in the review as it failed to differentiate imaging for low-back pain only.

What do you think?

  • Do you carry out regular request for imaging audits? If so, have you found that your referral rates have decreased? If you have shared your reflections with colleagues have there been any similar trends with regard to why rates have decreased?
  • Do you receive regular updates regarding clinical decision support from colleagues in radiology? If not, is this something which requires to be addressed given the potential for financial savings?

Send us your views on this blog and become part of the ever expanding Musculoskeletal Elf community. Post your comment below, or get in touch via social media (FacebookTwitterLinkedInGoogle+).

Do you know that there is an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses? This is called the Preferred Reporting Items for Systematic Reviews and Meta-Analyses or PRISMA statement and can be accessed through the website of the EQUATOR Network. The Elves use the PRISMA statement for critical appraisal of systematic reviews, although it is not a quality assessment instrument to gauge the quality of a systematic review.


Jenkins, H. J., Hancock, M. J., French, S. D., Maher, C. G., Engel, R. M., & Magnussen, J. S. (2015). Effectiveness of interventions designed to reduce the use of imaging for low-back pain: a systematic review. Canadian Medical Association Journal, 187(6), 401-408 [Abstract]

Dagenais, S., Caro, J., & Haldeman, S. (2008). A systematic review of low back pain cost of illness studies in the United States and internationally. The spine journal, 8(1), 8-20 [Abstract]

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+