Systematic reviews of prognostic studies II: 'Risk of bias' assessment in systematic reviews of prognostic studies

Article type
Authors
Moons K1, Wolff R2, Whiting P3, Hooft L4, Damen J4, Riley R5, Mallett S6, Cochrane Prognosis Methods Group 7, Hayden J8
1Julius Center for Health Sciences and Primary Care
2Kleijnen Systematic Reviews Ltd
3University of Bristol
4Cochrane Netherlands
5Prognosis Methods Group
6University of Birmingham
7Cochrane
8Department of Community Health & Epidemiology, Dalhousie University
Abstract
Background:
Prognosis is a description of the probable course of individuals with a health condition. Review and synthesis of prognostic factor and prognostic model studies is a relatively new and evolving area. Critical appraisal of prognostic studies is challenging, but essential to assess and identify biases that are sufficiently large to distort study results.

The QUIPS, CHARMS and PROBAST tools are useful and reliable for the assessment of risk of bias in studies of prognostic factors and models. All three tools are domain-driven with signalling items developed to guide review authors to set up a data extraction list, or to assess the risk of bias in a study.

Objectives:
This workshop will familiarise participants with sources of bias in prognostic factor and prognostic model studies, and introduce participants to the QUIPS, CHARMS and PROBAST tools.

Description:
This interactive workshop will inform and train participants on systematic reviews of prognostic factors and models. The workshop will be split into two sessions. The first session will give an overview of the QUIPS, CHARMS and PROBAST tools and guidance on how to use these. In the second half of the workshop, participants will have the opportunity to experience these new tools and to discuss issues with the creators of the tools.