| Ref ID | 89 |
| First Author | M. Kicinski |
| Journal | STATISTICS IN MEDICINE |
| Year Of Publishing | 2015 |
| URL | https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/sim.6525?download=true |
| Keywords |
• General medical • Publication bias • Cochrane |
| Problem(s) |
• Poor consideration of publication bias • Perpetuates citation of poor quality primary study data • Weaknesses identified in some Cochrane reviews |
| Article Type | Empirical |
| Article Subtype | Cross-sectional survey/Methodological systematic review |
| First Author Country | Belgium |
| Aim | To assess the presence of publication bias of all 1106 meta-analyses from the Cochrane Database of Systematic Reviews comparing treatment with placebo or no treatment up to August 2012 using a Bayesian hierarchical selection model. |
| Level of Investigation | Analytical |
| Summary of Findings | Data from 1106 meta-analyses from 329 Cochrane reviews comparing treatment with placebo or no treatment were included. In the meta-analyses of efficacy, outcomes favouring treatment had on average a 27% (95% Credible Interval (CI): 18% to 36%) higher probability to be included than other outcomes. In the meta-analyses of safety, results showing no evidence of adverse effects were on average 78% (95% CI: 51% to 113%) more likely to be included than results demonstrating that adverse effects existed. In general, the amount of over-representation of findings favourable to treatment was larger in meta-analyses including older studies. The number of statistically significant findings in meta-analyses was larger than expected from the power of the individual studies. |
| Number of systematic reviews included | 329 |
| Number of eligible systematic reviews assessed | 3845 |
| Treatment impacted | No |
| Treatment impacted description | |
| Interpretation impacted | Not Applicable |
| Interpretation impacted description |