A Cochrane review [Abstract] 1 included 24 studies with a total of 2 334 subjects. The quality of the trials was poor: 5 trials were quasi-randomised, 5 showed differences at baseline and 17 lacked follow up data. The outcome was better after hypnosis than imipramine in one trial (RR for failure or relapse after stopping treatment 0.42, 95% CI 0.23 to 0.78). Psychotherapy appeared to be better in terms of fewer children failing or relapsing than both alarm (RR 0.28, 95% CI 0.09 to 0.85) and rewards (0.29, 95% 0.09 to 0.90) but this depended on data from only one trial. Medicinal herbs had better results than desmopressin in one trial (RR for failure or relapse after stopping treatment 0.35, 95% CI 0.14 to 0.85). Acupuncture had better results than sham control acupuncture (RR for failure or relapse after stopping treatment 0.67, 95% CI 0.48 to 0.94) in a further trial. Active chiropractic adjustment had better results than sham adjustment (RR for failure or relapse after stopping treatment 0.76, 95% CI 0.60 to 0.95). However, each of these findings came from small single trials, and need to be verified in further trials. The findings for diet and faradization were unreliable, and there were no trials including homeopathy or surgery.
Comment:The quality of evidence is downgraded by study limitations (inadequate or unclear allocation concealment), and by imprecise results (few patients for each comparisons).
Primary/Secondary Keywords