Cleland 2005.

Methods	Type of trial: RCT Number analysed/randomly assigned: 36/36 Intention‐to‐treat analysis: not applicable (NA)
Participants	Chronic neck pain
Interventions	INDEX TREATMENT Manipulation (manip): technique: thoracic manip; frequency: 1 session; dose: 1 to 2 manip per segment, average 3.7 manip per participant; route: thoracic spine to each restricted segment COMPARISON TREATMENT Placebo manip (cntl): technique: use of flat open hand on participant naive to manip; frequency: 1 session; dose: average 3 placebo manip per participant; route: thoracic spine COINTERVENTION: NR Duration of therapy period: 1 session Duration of follow‐up: none
Outcomes	PAIN (VAS, 0 to 100) Baseline mean: manip 41.6, placebo 47.7 End of study mean: manip 26.1, cntl 43.5 Absolute benefit: manip 15.5, cntl 4.2 Reported results: significant favouring manipulation SMD: ‐0.93 (95% CI ‐1.62 to ‐0.24); NNTB 2; treatment advantage 29% FUNCTION: NR PARTICIPANT SATISFACTION: NR GPE: NR QoL: NR SIDE EFFECTS: none COST OF CARE: NR
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Page 129, column 1, paragraph 3
Allocation concealment (selection bias)	Low risk	Page 129, column 1, paragraph 3
Blinding of Participants (performance bias)	Low risk	Page 130, column 2, paragraph 2
Blinding of Personal (performance bias)	High risk	Not possible owing to design
Blinding of the Outcome assessor (detection bias)	High risk	Not possible owing to design
Incomplete outcome data (attrition bias)	Low risk	Page 130, column 2, paragraph 2
Randomized Participants analysed were allocated (attrition bias)	High risk	Not reported
Selective outcome (reporting bias)	Low risk	Protocol available
Similar groups at baseline?	Low risk	Table 1, page 131
co‐interventions avoided or similar?	Low risk	Table 1, page 131
Compliance acceptable?	Low risk	Pre‐test/Post‐test design
Similar timing of outcome assessment?	Low risk	Pre‐test/Post‐test design