Letters

RAPID RESPONSES FROM BMJ.COM

Agreed! Refrain from lumbar spine x ray

The successful use of audits and education

Back pain patients' misconceptions of x rays

X rays can reassure and change patients' behavior

Authors' reply

This article originally appeared in BMJ USA

As of March 15, eight e-letters had been posted on bmj.com in response to the paper by Kendrick et al, along with four replies from the authors. Several of these are published below, in whole or in part.Editor

Agreed! Refrain from lumbar spine x ray

EDITORAlthough the conclusions of this paper are hardly a surprise, the research does provide support for those wishing to discourage the use of this virtually worthless examination, which imposes a significant population radiation dose. Disguised as reassurance for patient and doctor, lumbar spine x ray images are obtained to give the impression to patients that something is being done, thereby removing them from the consultation.

It can hardly be a surprise to hospital doctors and general practitioners (GPs) that radiography for routine back pain virtually never leads to any benefit to patients to balance the slightly increased risk of cancer (from the added radiation) that may result from radiation exposure. While it is difficult to write down what routine back pain is, we all have a high degree of certainty in knowing it when we see it.

It is hard to conceive of what benefit there could be. Setting aside any nebulous psychological effect (consideration of which should be strongly discouraged), something would have to happen to the back pain sufferer as a result of the radiographic procedure. Whatever one's views of the relative merits of surgery, pain clinics, aromatherapy, etc, it cannot be argued that plain x ray films have any role in the selection of patients for them. Clinical features: yes; x ray: no.

There were 420 patients in the trial reported by Kendrick et al. This is only a few months' worth of lumbar spine patients for me, many of whom would fit the inclusion criteria. I report over 90% of their x ray films as normal, and the vast majority of the rest as essentially normal. I struggle to think of any occasion when the radiographic result affected management.

William Stevenson, consultant radiologist. 
Burnley, Lancashire, UK wtjs{at}ouvip.com

The successful use of audits and education

EDITORI would like to report the results of an audit undertaken by a group of established GP trainers. The audit looked retrospectively at the number of lumbar spine x ray films requested by the individual practitioner (or his deputy). We then delivered an educational intervention and produced our own recommendations with the advice of one of our consultant radiologists and guidelines from the Royal College of Radiologists.¹ In most cases of low back pain, plain radiology will only show degenerative changes, yet it exposes the patient to about 50 times the radiation dose of a standard chest film. There may be more appropriate investigations to exclude secondary cancers (bone scan), a prolapsed disc or cord pathology (MRI), and osteoporosis (bone densitometry).

We agreed as a group to follow our guidelines and repeat our audit after one year. We set our goal as a 50% reduction in requests. The results showed a reduction in the number of x ray films ordered from 91 to 46 (49.5% reduction). The requests fell for all participants, dramatically in some cases.

One suggestion by Kendrick et al is to increase patient satisfaction without recourse to radiology, but clearly another challenge to primary care physicians is to use investigations more appropriately and more cost-effectively. The delivery of an educational package may be the way to achieve this.

Alexander Williams, general practitioner. 
St Thomas Health Centre, Exeter, UK alex.jane{at}virgin.net

Back pain patients' misconceptions of x rays

EDITORIn the study by Kendrick et al, there does not appear to have been any monitoring of the differential use of other diagnostic tests as co-interventions. Is it possible that, because doctors were asked not to refer patients for radiography in the control group, they were referred for alternative tests. Such as magnetic resonance imaging, computed tomography, or nuclear medicine? In addition, those in the intervention group may have been referred for other diagnostic tests subsequent to lumbar radiography. This potentially important confounding factor could also have huge cost implications.¹

The trial findings, however, are consistent with a recent on-the-street survey of public perceptions of over 500 people who responded to statements based on The Back Book and the Royal College of General Practitioners guidelines about their expectations and understanding of back pain and its management.² Forty percent had experienced back pain in the previous year. The survey showed that most people would expect their GP to send them for an x ray film and that the great majority believe that the most important thing the GP can do for them is to reassure them and advise them to return to normal activities.

In the trial by Kendrick et al, although those who did not have radiography apparently had better outcomes, 80% of patients still wanted to undergo radiography, presumably to get a diagnostic label. Both the trial and survey findings imply that the GP is often not successful in reassuring the patient. This misconception of the usefulness of x ray images needs to be addressed.

It is also interesting to note that 88% (421/476) of patients agreed to be randomized, accepting that they might not have radiography. Furthermore, for only 14% (26/199) of control patients did doctors judge it "clinically necessary" to request radiography. The challenge is for GPs to match these figures in the clinical setting. Surely, similar explanations can be given to the patients and now applied with greater confidence by GPs in light of these findings.

Stephen Brealey, research fellow. 
University of York, York, UK sb143{at}york.ac.uk

X rays can reassure and change patients' behavior

EDITORWe know that x ray images of the back do not improve clinical outcome. Perhaps the major value of this study is the finding that patients were satisfied with their care. So why do the authors believe that patients should not undergo radiography for reassurance? Anything that acts as a trigger to get people back to better function is good. One shouldn't underestimate the power of an x ray image to influence change in patients' behavior and lifestyle. For the average low back pain patient in the pain management unit, nothing less than a plain x ray film will do, although (shock! horror!) some cases require an advanced form of imagingcompleted, of course, at huge expense to the taxpayer.

As a physician treating this difficult group of patients and as a taxpayer too, I feel this can be money well spent. Plain x ray films of the back represent a cost-effective management tool, one that can lead to a change in activity and behavior patterns that gets patients out of the sick role and back to productive life.

The "primary care" x ray image that others believe is "without beneficial clinical outcome" is a bonus for those who work in pain management units. Comparative films demonstrate to the patient that there is little or no disease progression and that they haven't done their back any harm. No amount of "reassurance" has the same power, and the challenge to radiology and the editor is to come up with something betternot just to talk about it.

Ed Charlton, consultant in pain management and anesthesia. 
Royal Victoria Infirmary, Newcastle upon Tyne, UK jecharlton{at}compuserve.com

Authors' reply

IN REPLYIn response to the letter from Stephen Brealey, the GPs in our study did not have open access to MRI or CT scans through the National Health Service. All investigations ordered by hospital consultants were recorded and included in the health economic analysis, which we will be presenting separately. In addition, details of private referrals and private investigations have been recorded, costed, and included in the analysis.

We were unable to show any difference in outpatient referrals (the mechanism through which further investigation would be accessed) between the treatment groups. Thus, we consider it unlikely that the provision of alternative imaging techniques may have confounded our results.

Dr Charlton's argument that radiographic results will reassure and lead to improved function in patients with low back pain is not supported by our findings. Our study demonstrates that in a primary care population with low back pain of a median duration of 10 weeks, a lumbar spine x ray image does not improve function, nor does it get people back to work more quickly. If anything, those who underwent radiography in our study reported slightly worse function.

Radiographic imaging is not an innocuous test. We need to find ways of reassuring the large number of patients with low back pain who are extremely unlikely to have serious pathology without exposing them to radiation. Unfortunately we do not seem to be very good at doing this at present. As suggested in our paper, further work in this area is required.

Denise Kendrick, senior lecturer in general practice. 
University of Nottingham, UK denise.kendrick{at}nottingham.ac.uk