Quality Assessment for Epistaxis RCTs

Cochrane Risk of Bias Tool

Study, Year (Reference) Adequate Sequence Generation? Allocation Concealment? Blinding? Incomplete Outcome Data Addressed? Free of Selective Outcome Reporting? Free of Other Bias? Source of Funding Other Comments ClinicalTrials. gov Quality Summary:
Cochrane Risk of Bias Tool
Boyer et al. 2015 (15) U U N Y Y U American Rhinologic Society New Investigator Research Grant (HB) Unblinded NCT01408732 Unclear if adequate sequence generation or allocation concealment Unblinded.
Dupuis-Girod et al. 2016 (27) Y Y Y Y Y Y Hospices Civils de Lyon grant supported by PHRC 2013 and Association Pour la Maladie de Rendu-Osler NCT02106520 Low risk of bias.
Gaillard et al. 2014 (135) Y Y Y Y Y Y Hospices Civils de Lyon and the French Ministry of Health (2005 Hospital Clinical Research Programme (PHRC) grant. Even though the recruitment target (213) was not met at the end of the planned enrolment period, recruiting stopped because not enough patients met eligibility criteria, and the treatment units reached their expiration date. NCT00355108 Low risk of bias.
Geisthoff et al. 2014 (11) Y Y Y Y Y Y Pharmacia GmbH, (now part of Pfizer) & Baxter Deutschland GmbH.  After completion MEDA Pharma GmbH & Co. KG, paid 3500 Euros to UWG for access to the data to gain government approval of TA for epistaxis in HHT in Germany. "No funding source was involved in study design, collection, analysis and interpretation of the data nor in the decision to submit the manuscript" Results presented mainly derived from ITT analysis including 20 patients. Per protocol analyses including 18 patients, were also performed but did not reveal any relevant changes to the results (data not shown). NCT01031992 Funding provided by Pharmacia GmbH, (now part of Pfizer) and Baxter Deutschland GmbH.

After study completion, MEDA Pharma GmbH & Co. KG, paid 3500 Euros for access to the data.

Riss et al. 2015 (28) Y Y Y Y Y Y An unrestricted grant by Roche Austria to two of the authors (M.B. and D.R.). "The company was not involved in any other part of the study (eg, protocol design, data acquisition, statistical analysis, or manuscript editing)." A random block size of 6 was used, but because of the 2 stratification criteria and the low number of patients, 9 patients were randomized by chance to the bevacizumab group and 6 patients to the placebo group. NCT01314274 Two of the authors received an unrestricted grant by Roche Austria.
Whitehead et al. 2016 (9) Y Y Y Y Y Y Cure HHT NCT01408030 Low risk of bias.
Yaniv et al. 2009 (64) U Y U Y U U Not stated No description of blinding or of sequence generation. Pre-specified outcomes not clearly described.   QoL outcome added after study began. NCT00375622 No description of blinding or of sequence generation. Pre-specified outcomes not clearly described.   QoL outcome added after study began.

Source of funding not stated.