Kwak 2012

Methods	RCT Number analysed/randomised: 40/40 Intention‐to‐treat: reported Power analysis: NR Funding source: University of Tsukuba, Japan
Participants	WAD symptoms 3 months or longer Participant recruitment: through advertisements in local newspapers and on the homepage of the Kyung Hee Medical Center
Interventions	INDEX TREATMENT Acupuncture Acupuncture points from the gallbladder (GB), small intestine (SI), bladder (BL), triple energiser (TE) and large intestine (LI) meridian systems located on shoulder, neck, head and upper limbs. Needles (length 40 mm; diameter 0.16 mm, SEIRIN Co. Ltd) were inserted perpendicularly to a depth of 1.0 to 2.0 cm in place for 15 minutes COMPARISON Wait‐list CO‐INTERVENTION Both groups maintained their usual care, including physiotherapy, exercise and sufficient rest Treatment schedule: Treatment consisted of 6 sessions, 3×/wk for 2 weeks No follow‐up
Outcomes	PAIN INTENSITY (0 to 10 VAS scale) Baseline mean: acupuncture 4.59, wait‐list 4.88 End of study mean: acupuncture 2.74, wait‐list 4.47 immediate post treatment Absolute benefit: acupuncture 1.85, wait‐list 0.41 Reported results: statistically significant favouring acupuncture Drop‐outs: none Adverse effects: 3 acupuncture participants reported mild adverse events (2 with bruising, 1 with fatigue) No serious side effects reported Costs of care: NR
Notes	Unable to reach study author to obtain necessary data for abstraction
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computerised number table generated random sequences
Allocation concealment (selection bias)	Unclear risk	Envelopes sealed but not sequentially numbered (numbers inside)
Blinding (performance bias and detection bias) All outcomes ‐ patient?	High risk	Not possible because of design
Blinding (performance bias and detection bias) All outcomes ‐ care provider?	High risk	Not possible because of design
Blinding (performance bias and detection bias) All outcomes ‐ outcome assessor?	Unclear risk	Participant was the outcome assessor with self reports
Incomplete outcome data (attrition bias) All outcomes ‐ Drop out rate acceptable?	Low risk	No drop‐outs
Incomplete outcome data (attrition bias) All outcomes ‐ Analyzed in the group to which they were allocated?	Low risk	Acceptable
Selective reporting (reporting bias)	Low risk	See registration
Similarity of baseline characteristics	Low risk	Reported as similar in Tables 2, 3, 5
Co‐interventions avoided or similar?	Unclear risk	Not reported
Compliance acceptable?	Unclear risk	Not reported
Similar timing of outcome assessment?	Low risk	Baseline and 2 weeks
Fatal Flaw	High risk	Flawed; wrong analysis was used; Figure 2 should use independent t‐test, not paired t‐test