Efficiency and contribution of strategies for finding randomized controlled trials: a case study from a systematic review on therapeutic interventions of chronic depression.

backgroundIdentifying all existing evidence is a crucial aspect in conducting systematic reviews. Since the retrieval of electronic database searches alone is limited, guidelines recommend the use of additional search strategies. The aim of this investigation was to assess the efficiency and contribution of additional search strategies for identifying randomized controlled trials in conducting a systematic review on interventions after performing a sensitive electronic database search. DESIGN AND

methodsSeven electronic databases, 3 journals and 11 systematic reviews were searched. All first authors of the included studies were contacted; citation tracking and a search in clinical trial registers were performed. A priori defined evaluation criteria were calculated for each search strategy.

resultsA total of 358 full-text articles were identified; 50 studies were included in the systematic review, wherefrom 84.0% (42) were acquired by the sensitive electronic database search and 16.0% (8) through additional search strategies. Screening reference lists of related systematic reviews was the most beneficial additional search strategy, with an efficiency of 31.3% (5) and a contribution of 10.0% (5/50), whereas hand-searching and author contacts contributed two and one additional studies, respectively. Citation tracking and searching clinical trial registers did not lead to any further inclusion of primary studies.

conclusionsBased on our findings, hand-searching contents of relevant journals and screening reference lists of related systematic reviews may be helpful additional strategies to identify an extensive body of evidence. In case of limited resources, a sensitive electronic database search may constitute an appropriate alternative for identifying relevant trials. Significance for public healthSystematic reviews provide the strongest form of evidence synthesis for therapeutic interventions and are of high relevance for decision makers in public health. Preparing high quality systematic reviews can be very time-consuming since all existing evidence should be identified, but the retrieval of electronic database searches is limited and therefore additional search strategies are recommended. However, the time needed for conducting full systematic reviews does often not address the need for urgent evidence. Thus, priorities of summarizing all available evidence and providing fast evidence-based recommendations may conflict. So far, no consensus exists regarding which additional search strategies are beneficial and sparsely time-consuming for conducting systematic reviews. We examined the efficiency and contribution of additional search strategies performed after a sensitive electronic database search. Our results provide highly relevant information for researchers conducting systematic reviews in various fields of public health research and for establishing guidelines for conducting rapid reviews.