Article type
Year
Abstract
Background: The overall aim of this work is to develop a rigorous search filter for health promotion and public health (HP & PH) literature in PubMed. This is part of ongoing work to update the Health Promotion and Public Health Field s specialised database used by Cochrane Collaboration HP & PH Field members. It contains reviews and trials of effectiveness specific to HP & PH.
Objectives: To develop a HP & PH-specific filter, two research questions were examined: First, what is the sensitivity (the number of retrieved references) and precision (the number of relevant references) of current search terms for reviews/meta-analyses specific to HP & PH? This question examines the use of both HP & PH terms and review/meta-analysis terms. Second, how does this result compare to the sensitivity and precision of searching PubMed using the new meta-analysis and review one-click limit features? This question examines the use of current review/meta-analysis search terms developed by our research team versus those developed for PubMed.
Methods: Searches for reviews evaluating HP & PH effectiveness in PubMed were undertaken from January 2001 to December 2003. Two searches were run in PubMed: one located reviews using our research unit s developed reviews strategy; the other was run to locate reviews using PubMed s meta-analysis and review limits. Three HP & PH journals indexed in PubMed for the same period were hand searched and served as the gold standard: American Journal of Health Promotion, Canadian Journal of Public Health, and Health Education and Behavior. Two independent reviewers screened both PubMed reference sets for relevant studies using previously established criteria.
Sensitivity and precision of current searches to locate HP & PH reviews were established. The yield of HP & PH reviews located using PubMed s reviews/ meta-analysis limit was also obtained. Both results were compared against each other and then against hand searching to determine if (1) current searches were locating all reviews located by hand searching; (2) all reviews located using current searches were all HP & PH topics; and (3) PubMed s one-click feature for reviews/meta-analyses were locating all HP & PH reviews found by hand searching.
Results: Conducting this study will provide evidence on the sensitivity and precision of current HP & PH search terms already in use, as compared with the gold standard of hand searching. It will also provide information on whether PubMed s one-click limit feature for meta-analyses and reviews gets all of the references that would be located with our present search terms.
Conclusions: This study will provide the evidence to make an informed decision about whether current HP & PH search strategies need to be improved to either search wider (improved sensitivity) or to search more narrowly (improved precision). It will also indicate whether PubMed s one-click reviews/ meta-analysis limit features locate all HP & PH reviews/meta-analyses.
Objectives: To develop a HP & PH-specific filter, two research questions were examined: First, what is the sensitivity (the number of retrieved references) and precision (the number of relevant references) of current search terms for reviews/meta-analyses specific to HP & PH? This question examines the use of both HP & PH terms and review/meta-analysis terms. Second, how does this result compare to the sensitivity and precision of searching PubMed using the new meta-analysis and review one-click limit features? This question examines the use of current review/meta-analysis search terms developed by our research team versus those developed for PubMed.
Methods: Searches for reviews evaluating HP & PH effectiveness in PubMed were undertaken from January 2001 to December 2003. Two searches were run in PubMed: one located reviews using our research unit s developed reviews strategy; the other was run to locate reviews using PubMed s meta-analysis and review limits. Three HP & PH journals indexed in PubMed for the same period were hand searched and served as the gold standard: American Journal of Health Promotion, Canadian Journal of Public Health, and Health Education and Behavior. Two independent reviewers screened both PubMed reference sets for relevant studies using previously established criteria.
Sensitivity and precision of current searches to locate HP & PH reviews were established. The yield of HP & PH reviews located using PubMed s reviews/ meta-analysis limit was also obtained. Both results were compared against each other and then against hand searching to determine if (1) current searches were locating all reviews located by hand searching; (2) all reviews located using current searches were all HP & PH topics; and (3) PubMed s one-click feature for reviews/meta-analyses were locating all HP & PH reviews found by hand searching.
Results: Conducting this study will provide evidence on the sensitivity and precision of current HP & PH search terms already in use, as compared with the gold standard of hand searching. It will also provide information on whether PubMed s one-click limit feature for meta-analyses and reviews gets all of the references that would be located with our present search terms.
Conclusions: This study will provide the evidence to make an informed decision about whether current HP & PH search strategies need to be improved to either search wider (improved sensitivity) or to search more narrowly (improved precision). It will also indicate whether PubMed s one-click reviews/ meta-analysis limit features locate all HP & PH reviews/meta-analyses.