Fig 1.
(A) Prevalence of reporting of randomisation, blinded assessment of outcome, sample size calculation, and conflict of interest in 146 publications describing in vivo research identified through random sampling from PubMed; change in prevalence of (B) randomisation, (C) blinded assessment of outcome, and (D) conflict of interest reporting in quintiles of year of publication.
Vertical error bars represent the 95% confidence intervals of the estimates (S1 Data).
Fig 2.
Prevalence of reporting of (A) randomisation, (B) blinded assessment of outcome, (C) sample size calculations, and (D) conflict of interest reporting in 2,671 publications describing the efficacy of interventions in animal models of Alzheimer’s disease (AD, n = 324 publications), focal cerebral ischaemia (FCI, 704), glioma (175), Huntington’s disease (HD, 113), intracerebral haemorrhage (ICH, 72), experimental autoimmune encephalomyelitis (EAE, 1029), myocardial infarction (MI, 69), and spinal cord injury (SCI, 185) identified in the context of systematic reviews.
Vertical error bars represent the 95% confidence intervals, and the horizontal grey bar represents the 95% confidence interval of the overall estimate (S2 Data).
Fig 3.
Change in prevalence of reporting of (A) randomisation, (B) blinded assessment of outcome, (C) sample size calculations, and (D) conflict of interest reporting in quintiles of year of publication for 2,671 publications describing the efficacy of interventions in animal models of eight different diseases identified in the context of systematic reviews.
Vertical error bars represent the 95% confidence intervals of the estimates (S3 Data).
Fig 4.
Prevalence of reporting of (A) randomisation, (B) blinded assessment of outcome, (C) sample size calculations, and (D) conflict of interest reporting by decile of journal impact factor in 2,671 publications describing the efficacy of interventions in animal models of eight different diseases identified in the context of systematic reviews.
Black lines indicate the median value in that decile, and grey lines indicate the 95% confidence limits derived from nonparametric median regression (S4 Data).
Fig 5.
Prevalence of reporting of randomisation, blinded assessment of outcome, inclusion or exclusion criteria, and sample size calculation in 1,173 publications describing in vivo research published from five leading UK institutions (labelled A through E).
For each institution, the vertical error bars represent the 95% confidence intervals, and the horizontal grey bar represents the 95% confidence interval of the overall estimate for that risk-of-bias item (S5 Data).
Table 1.
Reporting of measures to reduce the risk of bias in publications from 2009–2010 that were randomly selected, identified in the context of systematic reviews or from leading UK institutions.