Birch 1998

Methods	RCT Number analysed/randomised: 36/46 Intention‐to‐treat: NR Power analysis: NR Funding source: intramural grant from the Anesthesia Department of Brigham and Women's Hospital, Boston, MA
Participants	Myofascial neck pain lasting ≥ 6 months Participant recruitment: hospital‐based pain management centre, neurology clinic, public announcement
Interventions	INDEX TREATMENT Group 1 ‐ relevant acupuncture (believed to be effective for the condition treated) Pre‐determined relevant points; shallow needling (2 to 3 mm) used with copper wire containing silicone diode applied for stage 1 of session (10 minutes); infrared lamp also used over points during stage 2 of treatment (10 minutes). Treatment done without Deqi by a licensed acupuncturist with 13 years of experience COMPARISON TREATMENT 1 Group 2 ‐ irrelevant acupuncture Same treatment dosage and duration as Group 1, using pre‐determined irrelevant points with copper wire attached but connections severed (stage 1) and placebo light (no heat during stage 2) COMPARISON TREATMENT 2 Group 3 ‐ medical control NSAID 500 mg OD Trilisate, no acupuncture CO‐INTERVENTION Avoided Treatment schedule: 14 sessions over 12 weeks, with each session averaging 30 minutes in duration Duration of follow‐up: immediate post treatment
Outcomes	PAIN INTENSITY (VAS 0 to 10 scale) Baseline mean: relevant 4.8, irrelevant 4.7, control 4.9 End of study mean: relevant 1.87, irrelevant 3.37, control 4.73 Absolute benefit: relevant 2.93, irrelevant 1.33, control 0.17 Reported results: Relevant acupuncture group had significantly lower pain scores, including hourly ratings SMD ‐2.52 (95% CI random ‐3.49 to ‐1.54) immediate post treatment Reasons for drop‐out: 2 moved, 6 not specified, 2 lost to contact Adverse effects: NR Costs of care: NR
Notes	‐‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding (performance bias and detection bias) All outcomes ‐ patient?	Low risk	Adequately described
Blinding (performance bias and detection bias) All outcomes ‐ care provider?	High risk	Not possible
Blinding (performance bias and detection bias) All outcomes ‐ outcome assessor?	Low risk	Participant blinded as outcome assessor of self reports
Incomplete outcome data (attrition bias) All outcomes ‐ Drop out rate acceptable?	Low risk	Described and acceptable
Incomplete outcome data (attrition bias) All outcomes ‐ Analyzed in the group to which they were allocated?	High risk	Did not analyse; all randomised
Selective reporting (reporting bias)	Unclear risk	No protocol
Similarity of baseline characteristics	Low risk	Reported to be similar
Co‐interventions avoided or similar?	Unclear risk	Not reported
Compliance acceptable?	Low risk	Each participant attended 14 sessions, twice weekly
Similar timing of outcome assessment?	Low risk	All assessed at 12 weeks
Fatal Flaw	Low risk	Acceptable