Acupuncture for low back pain: traditional diagnosis and treatment of 148 patients in a clinical trial.

Aims: To assess patterns of diagnosis, including concordance, and treatment within a clinical trial of traditional acupuncture for low back pain.

Setting: In a pragmatic randomised controlled clinical trial, 148 patients with low back pain, of between 4 weeks and 12 months duration, were randomised to the offer of individualised acupuncture and received up to 10 treatments.

Methods: Standardised diagnosis and treatment records were completed by practitioners for 148 patients. The diagnosis was based on three pre-defined low back pain syndromes. For a subgroup of patients, one of the six practitioners then independently re-examined the patients, blind to the original diagnosis. The diagnostic inter-rater reliability was assessed in terms of percentage congruent classifications and Cohen's Kappa. Structured interviews of practitioners established further details about practice styles.

Results: The most commonly diagnosed syndrome associated with low back pain was Qi and Blood Stagnation (88% of patients), followed by Kidney Deficiency (53%) and Bi Syndrome (28%), with more than one syndrome being identified for 65% of patients. For the subgroup examined twice, practitioner concordance was reasonable: between 47 and 80% of classifications were congruent, while Kappa values lay between 0 ("the same as chance") and 0.67 ("good"). Practitioners provided 1269 treatments in total, using 177 different acupuncture points. Most commonly used channels were Bladder and Gall Bladder, and the commonest points were BL-23 and the two lowest Huatuojiaji points. Auxiliary treatments were utilised by all practitioners to varying degrees.

Conclusions: Diagnostic concordance among practitioners was reasonable, and clear themes emerged for treatment. Further research is required to develop a flexible trial protocol with scope for individualised treatment.