words returned the following results: – cohort AND acupuncture (208) – case control AND acupuncture (377) – case series AND acupuncture (150). The Acupuncture Evidence Project © AACMA 7 2 CONDITIONS NOW RATED AS ‘EVIDENCE OF POSITIVE EFFECT’ IN THIS REVIEW 2.1 Migraine prophylaxis [Positive effect] For migraine prophylaxis, acupuncture was rated as ‘effective’ in the Australian DVA review (2010) and ‘evidence of positive effect’ in the USVA Evidence map of acupuncture (2014) (5, 6). Since March 2013 a narrative review of high quality randomised controlled trials and two systematic reviews including a Cochrane systematic review update, have confirmed that acupuncture is superior to sham acupuncture and seems to be at least as effective as conventional preventative medication in reducing migraine frequency (40-42). Moreover, acupuncture is described as ‘safe, long-lasting and cost effective’ (40). Subgroup analysis in the Cochrane systematic review found that 16 or more treatment sessions showed a larger effect size (Z=4.06) than 12 treatments or fewer (Z=2.32). Evidence levels in these three reviews was moderate to high quality. 2.2 Headache (chronic tension-type and chronic episodic) [Positive effect] Chronic tension-type headaches and chronic episodic headaches were not reviewed in the Australian DVA review (2010) and rated as ‘evidence of positive effect’ in the USVA Evidence map of acupuncture (2014) (5, 6). The most recent Cochrane systematic review update confirmed that acupuncture is effective for frequent episodic and chronic tension-type headaches with moderate to low quality evidence (43). A brief review of systematic reviews and meta-analyses described acupuncture as having a ‘potentially important role as part of a treatment plan for migraine, tension-type headache, and several different types of chronic headache disorders’ (44). Studies in Germany and the UK found acupuncture for chronic headaches to be cost-effective (44). 2.3 Low back pain [Chronic – positive effect; acute – potential positive effect] For low back pain, acupuncture was rated as ‘effective (possibly)’ in the Australian DVA review (2010) and ‘unclear’ in the USVA Evidence map of acupuncture (2014) (5, 6). The main reason given for the ‘unclear’ rating in the USVA Evidence map was that ‘sham acupuncture controlled trials tended towards statistically nonsignificant results’ and a Cochrane systematic review in 2005 which drew no firm conclusions on the effectiveness of acupuncture for acute low back pain (6). The problems associated with the interpretation of the effects of sham acupuncture have already been discussed. When reviews separate the evidence on chronic low back pain from acute low back pain, as was done in the recent review by the US Agency for Healthcare Research and Quality, it was found that there is moderate quality evidence for the effectiveness of acupuncture in chronic low back pain for both pain intensity and function, but only low quality evidence for pain intensity and function in acute low back pain (45). In a systematic review of 32 randomised controlled trials for acupuncture treatment of chronic non-specific low back pain, acupuncture was superior (both statistically and clinically significant) to sham acupuncture in pain reduction and improved function immediately after treatment (46). Acupuncture was equivalent but not superior to usual care in pain and function, but acupuncture plus usual care was superior to usual care alone (46). The reviewers rated the evidence level of their review as Level of Evidence I (46). A systematic review of 11 randomised controlled trials on acute low back pain, acupuncture was superior to NSAIDS for improving symptoms (small effect), and superior to sham for pain but not function (47). A narrative review of non-invasive and alternative treatments for chronic low back pain rated the evidence for the effectiveness of acupuncture as high and for acupressure as moderate (48). The Acupuncture Evidence Project © AACMA 8 A review of 16 systematic reviews found that acupuncture alone, or when added to usual care, provided short-term improvement in pain and function for chronic low back pain (medium to large clinical effects) and hence ‘should be advocated in routine clinical practice’ (49). For acute low back pain, the reviewers could not make firm conclusions about the effectiveness of acupuncture due the inclusion of only two systematic reviews (49). Two studies found that acupuncture is likely to be cost-effective for low back pain or chronic non-specific low back pain, respectively (50, 51). 2.4 Knee osteoarthritis pain [Positive effect] Knee osteoarthritis pain was not reviewed in the Australian DVA review (2010) and rated as ‘evidence of potential positive effect’ in the USVA Evidence map of acupuncture (2014) (5, 6). In a network meta-analysis comparing 22 interventions in 152 studies, acupuncture was found to be equal to balneotherapy and superior to sham acupuncture, muscle-strengthening exercise, Tai Chi, weight loss, standard care and aerobic exercise (in ranked order) (52). Acupuncture was also superior to standard care and muscle-strengthening exercises in a sub-analysis of moderate to high quality studies (52). In a systematic review of 12 randomised controlled trials, acupuncture was found to