Effectiveness of acupuncture across indications: overview of Cochrane Systematic Reviews

Article type
Authors
Piso B1, Reinsperger I1, Breuer J1
1Ludwig Boltzmann Institute for Health Technology Assessment, Austria
Abstract
Background: Acupuncture is a traditional Chinese medicine therapy method that is often used in addition or exclusively for the treatment of diseases.
Objectives: Our aim was to provide social insurance funds with an overview of the effectiveness of acupuncture for different indications based on Cochrane Systematic Reviews.
Methods: In May 2014, we conducted a systematic literature search in the Cochrane Database of Systematic Reviews that yielded 132 hits. Two authors independently screened 68 full-text articles for fulfillment of the inclusion criteria. Cochrane Reviews comparing needle, laser or electro-acupuncture with no/sham-acupuncture or other interventions were included. Cochrane Reviews were excluded, if acupuncture was only used as comparator. We extracted review details (e.g. number of included studies and patients) and finally, authors´ conclusions as ‘primary endpoint’.
Results: We included 55 Cochrane Reviews. Authors of nine Cochrane Reviews found evidence that acupuncture (mostly for indications that aimed to reduce pain) was more effective than the control intervention to a certain extent (at least for some major endpoints and comparisons). For four indications authors concluded that acupuncture seemed to be ineffective (e.g. for epilepsy or irritable bowel syndrome). In the remaining, and majority, of reviews no firm conclusions on the effectiveness of acupuncture could be drawn due to missing or inconclusive evidence (42 acupuncture indications).
Conclusions: The considerable number of published Cochrane Reviews and protocols (we additionally found 37 protocols for planned Cochrane Reviews on acupuncture) shows that questioning the effectiveness of acupuncture is of high relevance. Nevertheless, data in only a few Cochrane Reviews allow convincing answers. High quality evidence from primary studies is lacking for most of the therapeutic indications. This evidence gap should be targeted by randomised controlled trials of high methodological quality and with sufficient sample sizes.