Chen 2007.

Methods	Design: randomised controlled trial, parallel design Setting: outpatients and hospitalised patients, single centre Country: Republic of China
Participants	Sample size/available for analysis: 70/70 Gender: 30 females/40 males Age (years, mean and SD): 42.47 (± 14.07) years Duration of disease (months, mean and SD): 18.51 (± 8.43) months Inclusion criteria: "cervicogenic headache in accordance with the diagnostic criteria of Sjaastad; and at the same time one of the following conditions: course ≥ 6 months; nearly 3 months without any medication; cervical spine X‐ray inspection shows that the cervical spine has no degenerative changes." Exclusion criteria: "neurological diagnosis of other types of headache; cervical spine fracture, dislocation or discectomy and severe osteoporosis are not suitable for manipulation or nerve stimulation therapy."
Interventions	Group 1: TENS (n = 34) Group 2: manipulation treatment (n = 36) Treatment duration: 4 weeks Follow‐up duration: 6 weeks Schemes of TENS: two 75 mm × 115 mm electrodes were placed in the upper cervical spine on both sides, symmetrical two‐way square wave output, frequency 100 Hz, pulse width 250 μs, 20 min each time during the course of treatment (10 sessions)
Outcomes	Pain (numerical pain scale) Range of motion (in degrees)
Notes	Funding sources: not reported Conflict of interest: not reported Full text language: Chinese
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "...with cervicogenic headache were randomly allocated to receive manipulation treatment and TENS treatment". Comment: information provided is not sufficient.
Allocation concealment (selection bias)	Unclear risk	Quote: "...with cervicogenic headache were randomly allocated to receive manipulation treatment and TENS treatment". Comment: information provided is not sufficient.
Blinding of participants	High risk	Comment: different interventions (TENS and manipulation therapy) were compared and blinding was not possible.
Blinding of personnel/providers	High risk	Comment: due to the nature of the intervention, it was not possible to blind the physician who provided the intervention.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Comment: no available information
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: there were no losses of participants.
ITT analysis	Low risk	No losses
Selective reporting (reporting bias)	Unclear risk	Comment: the study protocol was not available.
Group similarity at baseline	Low risk	Comment: the three groups seem to be similar at baseline.
Cointerventions	Low risk	No cointerventions
Compliance	Unclear risk	Comment: compliance with the intervention was not reported.
Timing of outcome assessments	Low risk	Comment: timing of outcome assessment is identical for all intervention groups and for all primary outcome measures.
Other bias	Low risk	Comment: no other biases were identified.