Open AccessSynthesizing study results in a systematic review
Author(s) -
Jos Verbeek,
Jani Ruotsalainen,
Jan L. Hoving
Publication year - 2011
Publication title -
scandinavian journal of work, environment and health
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.621
H-Index - 103
eISSN - 1795-990X
pISSN - 0355-3140
DOI - 10.5271/sjweh.3201
Subject(s) - judgement , systematic review , mechanism (biology) , similarity (geometry) , flow chart , computer science , process (computing) , psychological intervention , management science , intervention (counseling) , psychology , epistemology , medline , artificial intelligence , political science , economics , philosophy , engineering drawing , psychiatry , law , image (mathematics) , engineering , operating system
A single study rarely suffices to underpin treatment or policy decisions. This creates a strong imperativefor systematic reviews. Authors of reviews need a method to synthesize the results of several studies, regardless ofwhether or which statistical method is used. In this article, we provide arguments for combining studies in a review.To combine studies authors should judge the similarity of studies. This judgement should be based on the workingmechanism of the intervention or exposure. It should also be assessed if this mechanism is similar for variouspopulations and follow-up times. The same judgement applies to the control interventions. Similar studies can becombined in either a meta-analysis or narrative synthesis. Other methods such as vote counting, levels of evidencesynthesis, or best evidence synthesis are better avoided because they may produce biased results. We support ourarguments by re-analysing a systematic review. In its original form, the review showed strong evidence of no effect,but our re-analysis concluded there is evidence of an effect. We provide a flow-chart to guide authors through thesynthesis and assessment process.