In the November issue of the BJGP, Somerville et al report an interesting systematic review of usual care in low back pain primary care trials.1 They highlight the variability that currently exists in the usual care packages, which function as the comparator treatment in many trials, and point out that the content of treatment in this arm is often inadequately described. They draw our attention to the difficulties associated with the interpretation of results when this comparator treatment is poorly described. We hope their message does not fall on deaf ears.
Notwithstanding this, we feel that their interpretation that the speed at which the Roland Morris Disability Questionnaire (RMDQ) score changes over time is dependent on the duration of low back pain, may be inaccurate. The clinical trajectories in several acute/sub-acute trial arms to trajectories from two small chronic low back pain trials were compared and Somerville et al note a more rapidly improving trajectory in the acute/sub-acute arms.2,3 Meng et al's report2 on a trial of acupuncture (n = 24) for chronic low back pain in older patients describes follow-up results at 2, 6, and 9 weeks; rather than the 30 week follow-up period indicated by Somerville et al. Licciardone's report of a trial of osteopathic manipulation (n = 20)3 did not provide point estimates for follow-up RMDQ scores in their usual care group; rather, it was stated ‘there was no significant decrease’. This was represented as Figure 1 in Somerville et al's report as a stable trajectory.1
We have done a new analysis of data from the UK BEAM trial4 (Brealey et al5 in the original paper), extracting data on outcomes for those in the usual care group of this trial with pain lasting for more than 3 months (n = 205). Furthermore, we obtained data from the usual care arm of Licciardone et al's trial of osteopathic manipulation (thanks to John Licciardone). For completeness, we present a figure including these data (Figure 1).
The usual care arm in the UK BEAM trial was nearly eight times the size of the two previously included studies combined. Data from patients with chronic pain show a markedly different trajectory from those described in Somerville et al's report (Figures 1 and 2). It more closely mimics the acute/sub-acute trajectories they presented. In contrast to their conclusion, we conclude that patients in usual care arms are likely to improve in trials of both acute and chronic back pain.
- © British Journal of General Practice, 2009.