Workshop type: Discussion.

Objective: To present the current ‘state of the art’ for designing and performing systematic reviews of prognosis and prognostic factors. We will discuss the benefits, limitations and challenges of prognosis systematic review.

Summary:
Similar to systematic reviews of interventions, systematic reviews of prognosis are important to inform consumers, health professionals, service providers and policy makers of evidence-based clinical management. Prognosis plays a central role in medical decisions, as the diagnostic classification of a patient is often based on information about the prognosis of their risk factors or condition. Moreover, in diagnostic testing, clinical course is regularly used as reference standard. Treatment effectiveness may be related to the “baseline” event rate, which is obtained from prognostic studies, and identification of prognostic factors may lead to new opport unities for development of interventions that focus on these factors. 

The results of single studies of prognosis are rarely conclusive; systematic review can provide a transparent way to synthesize study findings. Observations across studies from different populations, different exposures, and outcome measures can inform generalizability, or identify important sources of heterogeneity of results. Furthermore, approaching a systematic review in an exploratory way can point to areas in need of research and it can highlight methodological limitations and assess the potential impact of these.

An international prognosis review network was formed four years ago, and an application to become a Methods Group for Cochrane Collaboration is currently being progressed. Simultaneously many groups are undertaking systematic reviews and working on systematic review methods. However, there is limited empirical evidence regarding the most appropriate methods for conducting prognosis systematic reviews, leading to discrepancies in the methods used. Although basic principles to reduce bias and random error are similar to those used for intervention reviews, there are several challenges unique to systematic reviews of prognosis.

In this workshop we will present current methods and future directions for five key steps of prognosis systematic reviews:
• Defining the review question
• Identifying and selecting studies
• Assessing study quality
• Collecting data
• Synthesizing and interpreting results. 

Intended audience: Review authors and methodologists with an interest in systematic reviews of prognosis.

Technical level expected of participants: Intermediate.