Search strategies to identify observational studies in MEDLINE and Embase

Abstract

Six filters from the first study showed sensitivity of 99.5% to 100% and precision of 16.7% to 21.1%. One type of filter was evaluated by two additional systematic reviews (i.e. externally validated) and found that this retrieved 85.2% to 100% of the articles in the reference standard. Twelve filters from the second study had lower sensitivity (48% to 100%) and much lower precision (0.09% to 4.47%).

A typology of systematic reviews for synthesising evidence on health care

“…However, the credibility of the analysis towards evaluating specific interventions or establishing broad generalisations depends on the homogeneity of data from the primary trials or studies, and on the efforts of reviewers to conduct a comprehensive, transparent, unbiased, appropriately weighted and reproducible literature review . Observational studies associated with side effects or rare events have also been reviewed systematically as complementary to evidence from randomised controlled trials . Overall, despite some limitations, meta‐analyses provide useful generalisations for evidence‐based decisions, and with statistical modelling techniques that continue to expand …”

“…Umbrella reviews of quantitative, qualitative or mixed evidence usually consider the effects of exposure to diseases or effectiveness of interventions. They have been used constructively to identify sources of heterogeneity or diversity in observational studies of populations, diagnostic tests, and rare or adverse clinical advents . The term “meta‐narrative review” was coined for a similar umbrella approach to mapping and synthesising information from different sources, perspectives and research traditions .…”

“…The “Population, Intervention, Comparator and Outcome” (PICO) and the “Population, Exposure, Comparator and Outcome” (PECO) checklists help focus the search or evidence within clinical investigations and trials, and a substitution of terms provides the “Setting, Perspective, Intervention, Comparison, and Evaluation” (SPICE) guide when a social context to the search is required . The “Peer Review of Electronic Search Strategies” (PRESS) emphasise the comprehensiveness of a search by attending to MeSH terms and search filters; although with difficulty when terms and filters are poorly developed or misleading, as for reviews of qualitative and observational studies . Comprehensiveness and reproducibility are considered essential search objectives for reducing bias and enhancing confidence in search results, although they may be impractical objectives for most review methods beyond the rigorous controls of a meta‐analysis .…”

“…2,8,51 Observational studies associated with side effects or rare events have also been reviewed systematically as complementary to evidence from randomised controlled trials. 52 Overall, despite some limitations, meta-analyses provide useful generalisations for evidence-based decisions, and with statistical modelling techniques that continue to expand. 2…”

“…They have been used constructively to identify sources of heterogeneity or diversity in observational studies of populations, diagnostic tests, and rare or adverse clinical advents. 52 The term "meta-narrative review" was coined for a similar umbrella approach to mapping and synthesising information from different sources, perspectives and research traditions. 63 When used to search, interpret and synthesise theories, umbrella reviews use an inductive process to locate similar propositions within multiple theories or to construct a metatheory.…”