Article type
Year
Abstract
Background: These years the quality of meta-analysis/systematic review in screening test has not been evaluated. So we want to use PRISMA checklist to assess the overall review quality.
Objective: To explore the quality of meta-analysis/systematic review in screening test.
Methods: Eight databases (included MEDLINE and EMBASE) were searched from the beginning to December 2010. Data were extracted into Excel spreadsheets. The PRISMA checklist was used to assess reporting characteristics and methodological quality.
Results: The searches reported 4726 relevant articles. After application of inclusion criteria, 65 studies were included. Compliance with PRISMA checklist items ranged from 0 to 100% (Table 1). Most reviews were compliant with the following checklist items: reported a title (100%), rationale (100%), objectives (100%), eligibility criteria (100%), information sources (98.5%), search (98.5%), study selection (96.9%), data collection process (96.9%), data items (100%), summary measures (96.9%), synthesis of results (95.4%), study selection (95.4%), study characteristics (96.9%), conclusions (100%), structured summary (81.5%), risk of bias across studies (80%); More than half of the reviews were compliant with the following checklist items: just as risk of bias within studies (41.5%), results of individual studies (72.3%), synthesis of results (72.3%), summary of evidence (76.9%), limitations (72.3%); Less than half of reviews were compliant with the following checklist items: described protocol and registration (6.15%), risk of bias in individual studies (35.4%), additional analyses (35.4%), risk of bias across studies (33.8%), funding (41.5%).
Conclusions: Our purpose was to provide readers with a broad overview of the reporting and methodological characteristics of SRs in screening test. Although quality assessment inmeta-analyses has improved with the introduction of QUADAS in 2004, the reporting quality is still troubling. Especially in items like risk of bias, protocol and registration and funding reported. So, the most urgent strategy is to focus on increasing the value of SRs in screening test, rather than continuing to publish them in great quantity.
Objective: To explore the quality of meta-analysis/systematic review in screening test.
Methods: Eight databases (included MEDLINE and EMBASE) were searched from the beginning to December 2010. Data were extracted into Excel spreadsheets. The PRISMA checklist was used to assess reporting characteristics and methodological quality.
Results: The searches reported 4726 relevant articles. After application of inclusion criteria, 65 studies were included. Compliance with PRISMA checklist items ranged from 0 to 100% (Table 1). Most reviews were compliant with the following checklist items: reported a title (100%), rationale (100%), objectives (100%), eligibility criteria (100%), information sources (98.5%), search (98.5%), study selection (96.9%), data collection process (96.9%), data items (100%), summary measures (96.9%), synthesis of results (95.4%), study selection (95.4%), study characteristics (96.9%), conclusions (100%), structured summary (81.5%), risk of bias across studies (80%); More than half of the reviews were compliant with the following checklist items: just as risk of bias within studies (41.5%), results of individual studies (72.3%), synthesis of results (72.3%), summary of evidence (76.9%), limitations (72.3%); Less than half of reviews were compliant with the following checklist items: described protocol and registration (6.15%), risk of bias in individual studies (35.4%), additional analyses (35.4%), risk of bias across studies (33.8%), funding (41.5%).
Conclusions: Our purpose was to provide readers with a broad overview of the reporting and methodological characteristics of SRs in screening test. Although quality assessment inmeta-analyses has improved with the introduction of QUADAS in 2004, the reporting quality is still troubling. Especially in items like risk of bias, protocol and registration and funding reported. So, the most urgent strategy is to focus on increasing the value of SRs in screening test, rather than continuing to publish them in great quantity.
Images
