I welcome the comments and additional data provided by Froud and Underwood. I quite agree with them that the endpoint for data collection in Meng et al's report1 was at 9 weeks rather than 30 weeks as represented in the published version of in Figure 2b of our paper.2 The version of our paper that was accepted for publication in fact shows the endpoint correctly at 9 weeks. The error appears to have occurred during the production phase at the journal and was not spotted by me in the final proof. I apologise and take full responsibility for this. However, this does not alter the fact that the Roland Morris Disability Questionnaire (RMDQ) scores in Meng et al's study of people with chronic back pain did not improve significantly during the course of their study.
For Froud and Underwood to obtain the data that they present in Figure 1 overleaf they have had to re-interpret results from their own study3 and contact the author of another4 to present previously unpublished information. The main thrust of our article is that usual care should be adequately described in any published paper so that the average reader can understand how the study participants and their own patients compare. It should not be necessary to have to make further enquiries.
Given this, I am grateful to Froud and Underwood for the data that they have already supplied to us. This new information adds to our understanding of the course of chronic back pain. While agreeing that their new data show a trend for chronic back pain sufferers in their study to improve over time, I would beg to differ, however, with their conclusion that it ‘more closely mimics the acute/sub-acute trajectories’. The information about the UK BEAM study3 that they sent to me after the publication of our article gave RMDQ scores for patients with acute, mixed, and chronic duration of back pain. From this I calculate the following falls in RMDQ scores at 3, 6, and 12 months respectively: acute (1.66, 1.3, 0.26), mixed (1.45, 0.94, 0.66), and chronic (1.31, 0.7, 0.91). This would seem to back up our conclusion that ‘the longer the duration of the pain, the slower the rate of improvement,’ certainly in the first 6 months. The falls in RMDQ score between 6 and 12 months are less than one point which is unlikely to be clinically significant. Data obtained from Licciardone et al by Froud and Underwood show that RMDQ scores of patients with chronic back pain fell by only by 1.1 in 24 weeks, which the original authors described as showing ‘no significant decrease’.4 The overall picture remains that the outlook for patients with chronic back pain is less favourable than those with more acute onset.
Despite these very valid points, the overwhelming message that arose from our review remains true. The content and outcome of usual care of low back pain in primary care is not well described and is quite variable. This has implications for research when usual care is employed as a control, but also when considering how to bring evidence-based medicine into routine primary care.
- © British Journal of General Practice, 2009.