We critically appraised articles using the Scottish Intercollegiate Guidelines Network (SIGN) criteria for randomized controlled trials [56]. The SIGN criteria were selected by the steering committee for ease of use and relevance, and adapted for the purpose of our review by adding the following questions to the generic checklists:

“The definition of the non-musculoskeletal condition is clear?” (Yes/No)

“The participants are free from the non-musculoskeletal condition studied at the beginning of the study?” (Yes/No/Can’t say) (only for studies investigating primary prevention)

“The spinal manipulative therapy (spinal manipulation, spinal mobilization, and spinal traction) intervention is described in sufficient detail?” (Yes/No)

“The control intervention (if any) is described in sufficient detail?” (Yes/No)

“The follow-up period is sufficient (long enough for the outcome to occur) to answer the research question?” (Yes/No/Can’t say)

In addition, we edited the following item (in the primary prevention form) to ensure that the measurement properties of the method used to identify the non-musculoskeletal condition were clearly captured. The item “Are all outcomes measured in a standard, valid and reliable way?” was split into two questions, “The non-musculoskeletal condition is measured in a reliable way” (Yes/No/Can’t say) and “The non-musculoskeletal condition is measured in a valid way” (Yes/No/Can’t say). Detailed notes accompanied the SIGN generic checklists, and these were also edited to match the purpose of this review.

The risk of bias assessment was informed by the items from the SIGN checklists that focused on methodological quality. All risk of bias assessments were conducted by two independent investigators who were unaware of each other’s ratings. The risk of bias items included: clarity of the research question, definition of the non-musculoskeletal condition, randomization procedure, blinding of participants, clinicians and investigators, description of manipulation and control interventions, outcome measurements, drop-outs, co-interventions, intention-to-treat analysis and follow-up period.

The risk of bias assessment was conducted in four sequential steps. Prior to the Global Summit, independent pairs of reviewers (within each of the primary, secondary and tertiary prevention review groups) critically appraised relevant RCTs to determine their methodological quality. At the Global Summit, the quality was discussed and agreed upon by the respective group. After the summit, all RCTs were critically appraised a third time by two independent experienced methodologists (CLY, JW, IA, SM, JH, PC) to ensure that the SIGN criteria were interpreted and applied in a similar manner across reviewers and review groups. Two participants (SF, EP) then performed quality assurance by reviewing all SIGN forms and risk of bias tables developed from the third round of reviews to ensure their accuracy and standard application.

A study was rated as low risk of bias (high or acceptable quality according to the SIGN methodology) if reviewers judged that potential sources of selection bias, information bias and confounding were minimal or acceptable [56]. In particular, reviewers focused on potential biases related to the randomization procedure, concealment of treatment allocation, blinding, administration of sham intervention, and attrition [5761]. The presence of a validated sham procedure was considered particularly important.

Finally, all researchers involved in the systematic review of the literature reviewed the risk of bias tables and were asked to vote on the outcome of the critical appraisal through an electronic survey. We used 75% agreement (38/50 participants) as the threshold for consensus. One researcher (CLY) was not involved in the systematic review at the Global Summit but was, in case of problems, available as referee, to thereafter participate in the validation process and therefore also in the two voting sessions.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.