Should The Topics Of Cochrane Reviews Be Prioritised?

Jefferson T. Foot soldier, five Collaborative Review Groups and one Field

Introduction: the number of reviews and protocols has steadily increased since the first issue of the Cochrane Database of Systematic Reviews. Given that the Collaboration is still largely voluntary and has little intramural support, the prioritisation of topics for reviews would seem a sensible approach to make best use of scarce research resources. I describe some factors necessary to address the issues of prioritisation and outline the consequences of their adoption.
Objectives:
Methods: I followed epidemiological and economic principles and my experience as a contributor to the Collaboration.
Results: Issues that should be addressed include:

• Should we prioritise? Points in favour are increased efficiency and a clear procedure (if explicit and agreed upon). Points against are the possible loss of the current bottom-up approach and formation of a necessary bureaucracy to assign priorities.
• What should we prioritise? Reviews of single interventions (e.g. beta-Mockers), of whole topics (ischaemic heart disease) or whole areas (e.g. cardiovascular disease). Should we prioritise only reviews or other activities (e.g. handsearching. fund raising etc)?
• What criteria should we use for prioritisation? Whose viewpoint should we assume? (e.g. funder, WHO, society, countries, steering group, reviewers). What model should we use? (see table with examples using a worid-wide viewpoint).

Prioritisation model	Criteria (grade according to:)	Possible top priorities
Epidemiological	Burden of illness	Respiratory diseases
Economic	Where world society gains most benefits at the margin	Childhood immunisations/ Tobacco cessation/Vitamin A supplements
Demand led	User's perceived problems	Prevention and treatment of the common cold/ Gastroenteritis/Circulatory problems
Political	Contribution to getting re-elected	Slow viruses (prion disease) - this week, next week who knows?
Availability of evidence	Number ofRCTs/CCTs/ good Comparative studies	Aspirin for miocardial infarction
Quality of evidence	Number of good quality studies	Maternal & child health
Mixed	Mixture of all methods	Common cold

Discussion: A debate in this issue is necessary and is part of the evolution of the Collaboration.

Rome 1999 O61