Ma 2006.

Methods	A 2‐armed, double‐blinded, parallel‐group randomised controlled trial with 1 week duration of treatment and 1 week duration of follow‐up
Participants	Location: Beijing, China Setting: single‐centre study, Department of Otorhinolaryngology, Third Hospital Beijing, from January 2005 to April 2005 Sample size: Number randomised: 30 in betahistine group and 30 in vitamin B6 group Number completed: 29 in betahistine group and 28 in vitamin B6 group Participant baseline characteristics: Age (mean (SD) years): 47 (18) in betahistine group versus 46 (17) in vitamin B6 group (P value 0.802) Gender (male/female, n): 16/14 in betahistine group versus 15/15 in vitamin B6 group (P value 0.796) Duration of tinnitus (n (%)): < 3 months: 7 (23); 4 months to 1 year: 7 (23); > 1 year: 16 (53) in betahistine group versus < 3 months: 8 (27); 4 months to 1 year: 9 (30); > 1 year: 13 (43) in vitamin B6 group (P value 0.731) Tinnitus loudness (≤ 15 dB, n (%)): 17 (57) in betahistine group versus 20 (69) in placebo group (P value 0.329) Tinnitus laterality (n (%)): bilateral: 16 (53); left‐sided: 9 (30); right‐sided: 4 (13); inside head: 1 (3) in betahistine group versus bilateral: 18 (60); left‐sided: 6 (20); right‐sided: 6 (20) in the vitamin B6 group (P value 0.548) Tinnitus pitch (n (%)): 125 to 1500 Hz: 4 (13); 2000 to 4000 Hz: 9 (30); ≥ 6000 Hz: 17 (57) in the betahistine group versus 125 to 1500 Hz: 5 (17); 2000 to 4000 Hz: 6 (20); ≥ 6000 Hz: 18 (60); pitch could not be matched: 1 (3) in the vitamin B6 group (P value 0.697) Hearing loss (n (%)): 17 (57) in betahistine group versus 20 (67) in vitamin B6 group (P value 0.426) Inclusion criteria: patients aged 18 to 75 years with subjective tinnitus Exclusion criteria: Middle‐ear and outer‐ear tinnitus Objective tinnitus Ménière's disease Pregnancy (History of) allergic reaction to any of the components of betahistine mesilate (History of) peptic ulcer Bronchial asthma Malignancy, including adrenal medullary tumour and brain tumour Uncontrolled acute infection Non‐compliant patients using less than 80% of the prescribed medication Use of concurrent medication
Interventions	Intervention group: betahistine mesilate tablets, 18 mg daily (6 mg 3 times a day), 1 week Comparator group: vitamin B6 tablets, 30 mg daily (10 mg 3 times a day), 1 week Use of additional interventions: flunarizine hydrochloride, 5 mg daily (5 mg once a day), 1 week, both treatment arms
Outcomes	Treatment effect on tinnitus loudness match (5 levels: tinnitus match decreased to 0 dB; reduced by ≥ 15 dB; reduced by ≥ 5 dB and < 15 dB; reduced by < 5 dB or exacerbated by < 5 dB; exacerbated by ≥ 5 dB) at 1 week Efficiency analysis based on a derivative of the tinnitus loudness match data (2 levels: tinnitus match reduced by ≥ 5 dB or tinnitus match reduced by < 5 dB or exacerbated) at 1 week Subjective treatment effect (5 levels: tinnitus disappeared; significantly reduced; slightly reduced; not changed; exacerbated) at 1 week Change in hearing at 1 week Adverse effects at 1 week
Funding sources	No information provided
Declarations of interest	No information provided
Notes	—
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed by workers from the Epidemiological Investigation Room Peking University Third Hospital, using SAS software
Allocation concealment (selection bias)	Unclear risk	Methods of allocation concealment not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	In the abstract it was mentioned that this was a double‐blind trial. However, the authors did not report who was blinded for what.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	In the abstract it was mentioned that this was a double‐blind trial. However, the authors did not report who was blinded for what.
Incomplete outcome data (attrition bias) All outcomes	Low risk	3 participants did not complete the trial: 1 participant from the test group and 2 participants from the control group. An intention‐to‐treat and per‐protocol analysis was performed.
Selective reporting (reporting bias)	Low risk	There is no protocol available. The outcomes listed in the methods section of the article are all reported in the results section of the article.
Other bias	Unclear risk	Conflicts of interest and funding were not reported