In October 2020, John Carlisle reported that after scrutinizing more than 500 randomized controlled trial (RCT) manuscripts over three years and obtaining anonymized individual participant data for over 150 trials, 44% of those trials contained at least some flawed data and 26% were judged to have problems so widespread that the trials were impossible to trust.
October 01, 2020
high
temporal
Findings from John Carlisle's analysis of RCT manuscripts and available individual participant data.
In October 2020, John Carlisle reported that when only aggregated summary tables (and not individual participant data) were available for randomized controlled trials, 2% of trials were judged to have flawed data and 1% were judged to be trials with problems so widespread that they were impossible to trust.
October 01, 2020
high
temporal
Comparison of detectability of flawed data with and without access to individual participant data in Carlisle's analysis.
A randomized controlled trial (RCT) is widely regarded as the gold standard of medical research for evaluating interventions.
high
descriptive
General methodological characterization of RCTs in clinical research.