Bye 2011.

Methods	Quasi‐randomised controlled trial No blinding reported Unclear if ethics approval and informed consent obtained Randomisation occurred at the level of wrists, with no constraint that each participants' wrist be allocated to different treatments
Participants	Total N randomised = 12 (20 wrists) randomised Intervention group N = Not reported Control group N = Not reported 0 males, 12 females Mean ± SD (range) age: Group 1: 43.50 ± 10.9 years Group 2: 35.60 ± 5.79 years Inclusion criteria: 1. Housewives aged 30 to 65 years 2. Chronic hand or wrist pain with paraesthesia or numbness in at least one finger innervated by Median nerve 3. Nocturnal symptoms for at least 3 months 4. Mild to moderate CTS based on electrodiagnostic findings Exclusion criteria: 1. History of corticosteroid injection in wrist during the past three months 2. History of carpal tunnel surgery in the affected hand 3. History of wrist fracture in the affected hand 4. Pregnancy 5. Upper motor neuron problems
Interventions	Intervention group: MANU hand brace (keeps third and forth finger in extension, and which was developed by Manente 2001) at nights for 4 weeks Control group: Short 'Cock‐up' splint (with natural wrist angle) at nights for 4 weeks
Outcomes	Outcomes assessed at baseline, after two weeks, and at the end of four weeks of treatment: 1. Pain VAS was filled by the participant at the beginning of the study. The participant was instructed to mark a point for her diurnal pain and one for her nocturnal pain for every day during the study. The VAS was a 100 mm line in which 0 meant no pain and 100 meant the most possible pain, so each point corresponded to a number indicating severity of the pain. For each participant the numbers were added and their mean was considered as her mean pain. 2. Symptoms assessed using the Levine carpal tunnel questionnaire, with 11 questions, each with five answers rating from 1 (least) to 5 (most), so the highest possible score is 55. 3. Function assessed using the Levine carpal tunnel questionnaire, with 11 questions, each with five answers rating from 1 (least) to 5 (most), so the highest score possible score is 40.
Notes	Study written in Turkish and was translated into English by a translator recruited by the Neuromuscular Disease Review Group. Some participants in each group had bilateral CTS, however, the number of participants and wrists allocated to each group was not reported. Some bilateral CTS participants received the same intervention for both wrists while others received different interventions for each wrist. It is not clear if analysis was undertaken at the participant‐ or wrist‐level for outcomes, . Attempts to clarify this information from the trialists were unsuccessful. Therefore, it is not clear whether a unit of analysis error occurred. As the sample size for each outcome is unclear, we could not include this outcome data in the Data and analyses section.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Comment: Alternation was used, in that the first participant with mild symptoms was placed in MANU hand brace group and the next participant in the other group. Also, the first participant with moderate symptoms was placed in MANU hand brace group and the next participant in the other group. This was done to promote severity of the disease equality between groups.
Allocation concealment (selection bias)	High risk	Comment: As alternation was used, the personnel responsible for recruiting participants would be aware of what the next group participants would be allocated to, thus the allocation was not adequately concealed.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Comment: It is not clear whether participants and personnel were blind to treatment.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Comment: The authors did not mention if they had assessed how often and how the participants used the splints. Not enough information was provided in the paper to determine whether outcome assessors (who are the participants only as only self‐reported outcomes were used) were blind to treatment.
Incomplete outcome data (attrition bias) 3 months or less	Unclear risk	Comment: In this study, there was no explanation about excluded patients or percentage of participants appearing for follow‐up. It seems that all the patients who entered the study finished the study. However, it was not clear how many participants were allocated to each group.
Selective reporting (reporting bias)	High risk	Comment: The means and SDs of all outcomes were reported, but as the number of participants allocated to each group was not clear, it was not possible to enter these data into RevMan.
Other bias	Low risk	Comment: No other sources of bias identified.