PAGE 3

Page 3

Details of the Medline search strategy are available in the Appendix. Additionally, the reference lists of the located articles and key SRs of OMT were manually searched for further relevant literature. Hard copies of all retrieved articles were read in full. All titles and abstracts identified in the electronic database search were screened for relevance. Articles appearing to meet the inclusion criteria were retrieved in full for further evaluation and validation according to predefined criteria. The data screening and selection process were carried out in-dependently by 2 reviewers (Drs Posadzki and Lee). In case of disagreement, a third independent reviewer (Dr Ernst) was asked to decide. The Cochrane tool was used to assess the risk of bias (ROB) of the RCTs.12 This tool consists of 7 domains: adequate sequence generation, allocation concealment, patient blinding, assessor blinding, addressing of incomplete data, selective outcome reporting, and other sources of bias. Each domain can be scored as follows: H, high ROB; L, low ROB; and U, unclear ROB. Quality assessment process was conducted by 2 independent reviewers (Drs Posadzki and Lee) and subsequently validated by the third reviewer (Dr Ernst). Dis-agreements about whether a study was of low or high quality were settled through joint discussions. Data extraction was conducted by 2 re-viewers (Drs Posadzki and Lee) by using a predefined form and subsequently validated by another reviewer (Dr Ernst).

1 2 3 4 5 6