SR (for acupressure) will not occur. If a return to primary studies is required to check or confirm information, we will note any discrepancies or adjustments made. 3.3.3 Requests for data Eligible SR protocols, ongoing trials, studies not published in English, and studies published as conference abstracts will be identified for inclusion. Study authors will be contacted through an open-ended request for data or further information. If no further data are available, the study will be noted as 'Ongoing’ or within the ‘Studies Awaiting Classification’ table and will not be included in the evidence appraisal. No attempts will be made to obtain or clarify data from authors of published peer-reviewed primary studies (shiatsu) or SRs (acupressure). 3.3.4 Data items Shiatsu The following characteristics of included studies will be extracted: study design, year conducted, setting and location, participant characteristics (including demographics, comorbidities, etc.), intervention and comparator characteristics (including number of treatment sessions, program duration, co‐interventions), outcomes (including measurement method, timing, or severity), method of analysis, and funding sources. Acupressure The following characteristics of included SRs will be extracted: review objectives, study design (e.g. qualitative review, meta-analysis), year conducted, databases searched, date (and range) of documented search, SR eligibility criteria for participant characteristics (including demographics, comorbidities, etc.), SR eligibility criteria for intervention and comparator characteristics (including number of treatment sessions, program duration, co-interventions), outcomes reported in the SR (including measurement method, timing or severity), the method of synthesis/analysis employed, characteristics of included primary studies (number, study design features), risk of bias tool used to appraise included primary studies and their rating (noting review authors comments or concerns), funding sources and the overall conclusion of the SR. The data extraction forms will also note whether the SR searched for and included publications in languages other than English. Included primary studies will also be listed by author, date of publication and eligibility for inclusion in this review. When evaluating the effectiveness of acupressure across the same PICO (See Section 3.3.11and 3.3.12), overlaps and omissions between SRs will be noted in a matrix (see Figure 1). Research Protocol HTANALYSTS | NHMRC | EVIDENCE EVALUATION ON THE CLINICAL EFFECTIVENESS OF SHIATSU 27 Where data from a selected SR is augmented with data obtained from another SR this will be documented using footnotes. 3.3.5 Missing data No imputation for missing data will be conducted. Studies with missing data will be included alongside other studies for that condition; either in the narrative (non-quantitative) synthesis of results or on forest plots showing the sample size. Implications of the missing data will be considered when interpreting the evidence and will be discussed under ‘Overall completeness and applicability of evidence’. Investigations into missing data within a study (e.g. a review of the clinical trial protocol) will be noted when assessing the risk of bias for that study (see Section 3.3.6). 3.3.6 Tools to assess risk of bias in individual studies The risk of bias of included studies will be assessed using the most appropriate risk of bias assessment tool according to the type of study as follows: • SRs (acupressure): AMSTAR-2 quality assessment checklist (45) • RCTs (shiatsu): Revised Cochrane risk of bias tool v2.0 (46, 47) • NRSIs (shiatsu): ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions (31, 48) Shiatsu Randomised controlled trials The risk of bias of RCTs will be assessed using the revised Cochrane Risk of Bias tool (46, 47). This tool is made up five domains assessing bias arising from the randomisation process; bias due to deviations from intended interventions; bias due to missing outcome data; bias in measurement of the outcome; and bias in selection of the reported result. Each domain will be assessed for bias, which will be recorded as ‘high’, ‘low’, or ‘some concerns’. An overall risk of bias for each outcome in the RCT will be judged based on the following criteria: • overall low risk of bias – low risk of bias for all key domains • some concerns – at least one domain has some concerns raised, but none are found to be at high risk of bias • overall high risk of bias – high risk of bias for one or more key domains Nonrandomised studies interventional studies (NRSIs) Critical appraisal of NRSIs will be guided by the methods described by Cochrane using the ROBINS-I tool (31, 48). Potential confounders and cointerventions for each population identified for inclusion will be identified and agreed through discussion with the NTWC prior to assessment of the risk of bias. ROBINS-I evaluates the risk of bias observed in the following domains: confounding, selection of participants, classification of interventions, deviations from intended interventions, missing data, measurement of outcomes, and selective reporting. Each domain will be judged, which will be recorded as ‘low’, ‘moderate’, ‘serious’, ‘critical’, or ‘no information provided’.