Tsai 2016.

Study characteristics
Methods	Study design Parallel RCT Study dates Duration of follow‐up: 4 weeks Time frame: March 2014 to January 2017
Participants	Study characteristics Setting: single centre (units of the nephrology department affiliated with the Kaohsiung Chang Gung Memorial Hospital in Taiwan) Country: Taiwan Inclusion criteria: patients met the KDOQI guidelines for intradialytic hypotension; pre‐HD SBP of 100 mm Hg and less or a decrease in SBP > 20 mm Hg, accompanied by at least one of the following: diaphoresis, nausea, vomiting, cramps, headache or dizziness; aged 20 to 75 years; were on maintenance HD for at least 3 months; had suffered intradialytic hypotension in at least 15% of their dialysis sessions during the past 2 months; were willing to sign the consent form were included Exclusion criteria: severe disorders of the heart, brain, liver, or haematopoietic system; active malignancy; mental disorders; pregnancy or lactation; or had experienced hypersensitivity skin reactions to herbal acupoint therapy Baseline characteristics Number (analysed/randomised): intervention group (14/18); control group (13/14) Mean age ± SD (years): intervention group (62.29 ± 4.80); control group (59.46 ± 9.38) Sex (M/F): intervention group (6/8); control group (2/11) Dialysis type: HD Mean dialysis vintage ± SD (years): intervention group (15.00 ± 6.88); control group (10.31 ± 6.96,) Comorbidities CVD: intervention group (3/14); control group (2/13) Diabetes: intervention group (3/14); control group (4/13) Hypertension: not reported Depression (clinician diagnosis): not reported
Interventions	Intervention classification Non‐pharmacological intervention Indication: study targeting fatigue Intervention group Herbal acupoint therapy Control group Sham herbal acupoint treatment Co‐interventions The same dialyser was used for each patient during the entire study period
Outcomes	Outcomes reported Fatigue outcome measures used: validation data available Frequency of intradialytic hypotension: assessed at 0 and 4 weeks Episodes and number of nursing interventions: assessed at 0 and 4 weeks Pre, nadir and post‐dialysis BP: assessed at 0 and 4 weeks Change in fatigue: assessed at the 0 and 4 weeks 10‐point VAS: scores of 1 to 3 represent mild levels, scores of 4 to 6 represent moderate levels, and scores of 7 to 10 represent severe levels Recovery time from fatigue after dialysis Rated as within minutes (0), when arriving home (1), at bedtime (2), the next morning (3), and by next HD (4): assessed at 0 and 4 weeks Blood chemistry: assessed at the 0th and 4th week Hb White blood cell BUN Potassium Calcium Potassium Albumin HCT Treatment failure: assessed at the 0th and 4th week Dry weight: assessed at the 0th and 4th week Dialysis adequacy: assessed at the 0th and 4th week Ultrafiltration goal decrease: assessed at the 0th and 4th week Volume of study fluid used: assessed at the 0th and 4th week Early discontinuation of dialysis: assessed at the 0th and 4th week Adverse events: assessed at the 0th and 4th week Case report form
Notes	Additional information Funding: Chang Gung Memorial Hospital with grant number CMRPG 8D0341 and CMU under the Aim for Top University Plan of the Taiwan Ministry of Education Conflicts of interest/disclosures: none Trial registration identification number: NCT02210377 A priori published protocol: research protocol was published and approved by the Institutional Review Board of Chang Gung Memorial Hospital (CGMH) (IRB no. 102‐4749A3 and 104‐3156C)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from Tsai 2016: "Participants were randomly and equally allocated to either the herbal acupoint therapy (HAT) or placebo group by computer‐generated randomisation." Quote from Tsai 2016 protocol: "Randomisation will be generated by a computerised random number function in Microsoft Excel, and the patients, programme assessors and statisticians will be unaware of the group to which they have been assigned. A block randomisation procedure (based on age, comorbidities such as cardiovascular disease and diabetes mellitus) will be employed to ensure that group allocation is equal and that the characteristics of the trial participants are similar." Comment: A computer‐generated sequence with random numbers is considered as low risk of bias. No imbalance between intervention groups was apparent
Allocation concealment (selection bias)	Unclear risk	Method of allocation concealment was not reported in sufficient detail to permit judgement.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote from Tsai 2016: "All patients, program assessors, outcome assessors, and statisticians were blind to the group allocations until the end of the clinical trial." Comment: A double blind study is considered as low risk of bias
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote from Tsai 2016: "Patient subjective assessments of the degree of fatigue and recovery time from fatigue after dialysis in both groups. [...] All patients, program assessors, outcome assessors, and statisticians were blind to the group allocations until the end of the clinical trial." Comment: The outcomes were assessed with an appropriate measure, without differences between groups. However, subjective measures were used, it was stated that outcomes were assessed without knowledge of treatment allocation, and knowledge of treatment assignment that may have influenced reporting. However, objective and subjective outcomes were assessed. Overall the outcome assessment was blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote from Tsai 2016: "In all, 27 patients (84%) completed the entire study. [...] These patients were randomly divided into a group receiving HAT therapy (18 patients) and a group receiving sham‐HAT therapy (14 patients), and 5 patients (15.6%) dropped out before week 2. The remainder of the patients provided complete data at follow‐up." Comment: As reported in Figure 1, 14/18 participants in the intervention group and 13/14 participants in the control group participants completed the study (> 5% lost to follow‐up, with difference between groups). Reasons for discontinuations seemed to be not related to the treatment allocation (discontinuation for disease progression and withdrawal in the intervention group and withdrawal in the control group
Selective reporting (reporting bias)	High risk	Protocol was published and approved by the Institutional Review Board of Chang Gung Memorial Hospital. Fatigue was reported in accordance with a pre‐specified analysis plan, using multiple eligible outcome measurements (scales, time points). Fatigue was reported in a format that was extractable for meta‐analysis. All outcomes that should be addressed (fatigue, cardiovascular disease, and death) were not reported
Other bias	Low risk	There was no evidence of different baseline characteristics, or different non‐randomised co‐interventions between groups. Funding was unlikely to influence the data analysis and authors did not have conflicts of interest. The study seemed to be free from other sources of bias