Developing efficient search strategies to identify reports of adverse effects in medline and embase

Abstract

Objective: This study aimed to assess the performance, in terms of sensitivity and precision, of different approaches to searching  and  to identify studies of adverse effects. Methods: Five approaches to searching for adverse effects evidence were identified: approach 1, using specified adverse effects; approach 2, using subheadings/qualifiers; approach 3, using text words; approach 4, using indexing terms; approach 5, searching for specific study designs. The sensitivity and precision of these five approaches, and combinations of these approaches, were compared in a case study using a systematic review of the adverse effects of seven anti-epileptic drugs. Results: The most sensitive search strategy in  (97.0%) required a combination of terms for specified adverse effects, floating subheadings, and text words for 'adverse effects'. In , a combination of terms for specified adverse effects and text words for 'adverse effects' provided the most sensitive search strategy (98.6%). Both these search strategies yielded low precision (2.8%). Conclusions: A highly sensitive search in either database requires a combination of approaches, and has low precision. This suggests that better reporting and indexing of adverse effects is required and that an effective generic search filter may not yet be feasible.

Poor reporting and inadequate searches were apparent in systematic reviews of adverse effects

“…For reviews that search for all adverse effects on a given intervention, a search using floating subheadings (such as ''adverse effects,'' or ''complications'') is recommended [4,7,29]. In this study, only one review reported using floating subheadings for adverse effects.…”

“…Free access to particular sources (such as MEDLINE) may influence some of the decisions on which sources to search. Yet studies have indicated that MEDLINE is not the most useful source of adverse drug effects information [7,27,28]. The absence of more detailed research may also prevent researchers from selecting an optimal core set of databases.…”

“…Unlike the assessment of clinical effectiveness, the difficulties with identifying adverse effects information are compounded by poor reporting and indexing of adverse effects terms [4e7]. The adverse effects of interest may also include new, previously unrecognized effects, so it may not be known which particular adverse effects the searches should be designed to retrieve [1,4,7,8]. Moreover, there are a wide variety of study designs that are potentially useful in evaluating adverse effects [9], and this complicates the search further [10].…”

Appropriateness and Adequacy of the Keywords Listed in Papers Published in Eating Disorders Journals Indexed Using the MEDLINE Database

“…Many studies stress the importance of the correct use of MeSH in comparison with free text when recovering scientific literature (Golder et al, 2006;Chang et al, 2006). The suitability of search equations (themed filters or documentary languages) is highlighted by using Descriptors to recover specific articles or a specific type of document with a high degree of sensitivity (Haynes et al, 2005).…”

Identification of Evidence for Key Parameters in Decision-Analytic Models of Cost Effectiveness: A Description of Sources and a Recommended Minimum Search Requirement

“…It should be noted that whilst Medline is recommended here due to its availability, EMBASE and the Science Citation Index (SCI) might produce higher yields of adverse effects evidence [20]. A search filter for adverse effects has been developed and validated [21]. The filter focusses on drug adverse events and a different approach might be required for non-drug interventions [22].…”