Article type
Year
Abstract
Background: Searching electronic databases is an important part of conducting systematic reviews (SR). In order to avoid publication bias, the aim is to maximize sensitivity. However, if sensitivity is maintained, optimizing precision can reduce the workload of reviewers and the cost of retrieving full text articles. Recently, the HEDGES team proposed a set of search strategies1 more sensitive than the Cochrane Highly Sensitive Search Strategy (HSSS)2,3. The relevance of this difference is not clear.
Objectives: To compare Cochrane HSSS and the HEDGES strategy for searching four electronic databases.
Methods: We used the search strategies for MEDLINE, EMBASE, CINAHL and PsycINFO which we developed for a Cochrane SR on the effects of non hormonal interventions for vasomotor symptoms in women with a history of breast cancer. We compared the results in terms of number of total hits, potentially eligible articles (based on a review of the title), selected for full text retrieval (based on title and abstract) and included studies, with the HEDGES strategies ('higher sensitivity'). All the searches were run at the same time, were identical except for the 'publication type phases' of the strategy and were evaluated independently by two reviewers. Results: For MEDLINE, EMBASE, CINAHL and PsycINFO, the HSSS yielded 85, 192, 20 and 0 total citations, and the HEDGES 109, 192, 23 and 8, which means 11.78% more citations for the latter. The potentially eligible articles were 36, 93, 18 and 0 respectively for the HSSS, and 51, 98, 18 and 3 for the HEDGES, 15.64% greater for HEDGES. The articles with full text retrieved and the included studies were identical in the four databases for both strategies (full text 10, 9, 7 and 0; Included 9, 8, 6 and 0).
Conclusions: In the development of this SR, performing searches with the HEDGES strategies did not report any benefit and increased workload for reviewers.
References
1. Haynes RB, McKibbon KA, Wilczynski NL, Walter SD, Werre SR. Optimal search strategies for retrieving scientifically strong studies of treatment from Medline: analytical survey. BMJ 2005; 330:1179-82.
2. Dickersin K, Scherer R, Lefebvre C. Identifying relevant studies for systematic reviews. BMJ 1994; 309:1286-91.
3. Robinson KA, Dickersin K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. International Journal of Epidemiology 2002; 31(1):150-3.
Objectives: To compare Cochrane HSSS and the HEDGES strategy for searching four electronic databases.
Methods: We used the search strategies for MEDLINE, EMBASE, CINAHL and PsycINFO which we developed for a Cochrane SR on the effects of non hormonal interventions for vasomotor symptoms in women with a history of breast cancer. We compared the results in terms of number of total hits, potentially eligible articles (based on a review of the title), selected for full text retrieval (based on title and abstract) and included studies, with the HEDGES strategies ('higher sensitivity'). All the searches were run at the same time, were identical except for the 'publication type phases' of the strategy and were evaluated independently by two reviewers. Results: For MEDLINE, EMBASE, CINAHL and PsycINFO, the HSSS yielded 85, 192, 20 and 0 total citations, and the HEDGES 109, 192, 23 and 8, which means 11.78% more citations for the latter. The potentially eligible articles were 36, 93, 18 and 0 respectively for the HSSS, and 51, 98, 18 and 3 for the HEDGES, 15.64% greater for HEDGES. The articles with full text retrieved and the included studies were identical in the four databases for both strategies (full text 10, 9, 7 and 0; Included 9, 8, 6 and 0).
Conclusions: In the development of this SR, performing searches with the HEDGES strategies did not report any benefit and increased workload for reviewers.
References
1. Haynes RB, McKibbon KA, Wilczynski NL, Walter SD, Werre SR. Optimal search strategies for retrieving scientifically strong studies of treatment from Medline: analytical survey. BMJ 2005; 330:1179-82.
2. Dickersin K, Scherer R, Lefebvre C. Identifying relevant studies for systematic reviews. BMJ 1994; 309:1286-91.
3. Robinson KA, Dickersin K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. International Journal of Epidemiology 2002; 31(1):150-3.