5 points on a 100-point index) is small. This result is also disproportionately influenced by the single large (n = 3441), lower quality trial (Witt el at 2006) that used a minimalintervention comparison rather than sham acupuncture. Separate analysis of disability outcomes from the shamcontrolled trials of acupuncture (WMD –6, 95% CI –15 to 3) suggest that the small difference seen between acupuncture and minimal medical care relate to the non-specific effects of provision of care. Similarly, while the results for laser therapy were PF-01367338 mouse promising, the results from the eight included trials varied from exceptionally effective

to slightly harmful. This conflict in the findings is difficult to explain. Pooled results demonstrated no between-group difference at the conclusion of treatment, whereas a significant reduction in pain was found at medium-term follow-up. A delayed analgesic effect does not seem plausible. Furthermore, this pattern of delayed onset of benefit did not consistently appear within trials that measured at both time points, and appears to be partly an artefact of the different studies included at the two time points. The included trials of laser therapy Selleckchem BYL719 investigated similar treatment and dosage protocols, although there was considerable diversity in trial quality and outcomes measured. The lack of consistency between trials in the timing of follow-up assessments resulted in different trials being pooled at post-treatment

and medium-term time points, so the clinical course of symptoms should not be inferred from these data. A more focused review of laser therapy might provide further

explanation about the reasons for the inconsistent trial outcomes. Few trials examined other electrophysical agents and those that did were inconclusive. Two trials of pulsed electromagnetic therapy suggest that this intervention is not effective. There was sparse evidence concerning the various forms of TENS therapy with only one small study reporting no significant results. There were no eligible trials that investigated any of the other electrophysical agents commonly used for neck pain. There is increasing evidence for an association between psychological factors and musculoskeletal old pain and disability (Linton 2000), and therefore a strong rationale supports psychological interventions. However, the role of psychological interventions for neck pain has not been well investigated despite the increasing popularity of these therapies. Some of the psychological therapies, such as those that address coping, adjustment, and problem solving, involve generic pain-management principles and have been investigated in broader spinal pain, or chronic musculoskeletal pain populations (Morley et al 1999). The one trial identified in this review that investigated intensive training in relaxation, a therapy often provided with other psychological interventions, showed that this treatment was not effective for decreasing neck pain.