Article type
Year
Abstract
Objective:
The workshop aims to train review authors to assess the risk of bias due to missing data when including non-randomized studies (NRS) in systematic reviews of interventions, by using an extended risk of bias tool (ACROBAT-NRS).
Description:
Cochrane recommends that review authors consider and justify whether or not to include NRS for all research questions. The risk of missing data may be higher when including NRS compared to randomized controlled trials (RCT), given that many NRS do not use custom, prospective data collection methods and may rely on routine data sources. Unlike RCTs, NRS have missing data for outcomes, important baseline confounders or classification by intervention; four signalling questions cover these different types of missing data and imbalance across the groups being compared. This workshop aims to give review authors and others intending to include NRS in Cochrane systematic reviews experience in applying ACROBAT-NRS. Participants will work in small groups to answer signalling questions, and judge whether a study is at high or low risk of material bias in each domain for specified outcomes. Responses are made on four-point scales and include a ‘no information’ option; these features will be contrasted with the existing Cochrane risk of bias tool. Check the programme for two other workshops on aspects of ACROBAT-NRS.
The workshop aims to train review authors to assess the risk of bias due to missing data when including non-randomized studies (NRS) in systematic reviews of interventions, by using an extended risk of bias tool (ACROBAT-NRS).
Description:
Cochrane recommends that review authors consider and justify whether or not to include NRS for all research questions. The risk of missing data may be higher when including NRS compared to randomized controlled trials (RCT), given that many NRS do not use custom, prospective data collection methods and may rely on routine data sources. Unlike RCTs, NRS have missing data for outcomes, important baseline confounders or classification by intervention; four signalling questions cover these different types of missing data and imbalance across the groups being compared. This workshop aims to give review authors and others intending to include NRS in Cochrane systematic reviews experience in applying ACROBAT-NRS. Participants will work in small groups to answer signalling questions, and judge whether a study is at high or low risk of material bias in each domain for specified outcomes. Responses are made on four-point scales and include a ‘no information’ option; these features will be contrasted with the existing Cochrane risk of bias tool. Check the programme for two other workshops on aspects of ACROBAT-NRS.