Missing data

The problem of missing data is well known, especially in cases where drug companies conceal evidence. However pharmaceutical industry misconduct is not the only cause, and a cluster of papers in this week's BMJ show how aspects of the culture of medical science contribute to the problem.

Elizabeth Loder, BMJ's clinical editor, talks to Harlan Krumholz (Harold H Hines Jr professor of medicine at Yale University) and Joseph Ross (assistant professor of medicine, also at Yale) about missing data from US publicly funded trials. Lisa Bero (professor at the Department of Clinical Pharmacy, University of California) describes how adding missing data to meta-analyses of drug trials can change the results, and Richard Riley (senior lecturer in medical statistics at Birmingham University) explains why individual participant meta-analyses aren't as balanced as we may think.

See also:
Publication of NIH funded trials registered in ClinicalTrials.gov: cross sectional analysis
Effect of reporting bias on meta-analyses of drug trials: reanalysis of meta-analyses
Assessment of publication bias, selection bias, and unavailable data in meta-analyses using individual participant data: a database survey
Missing clinical trial data

Demo video

Audio chapters:

Missing data from US publicly funded trials: 1:30

Effect of reporting bias on meta-analyses of drug trials: 8:1

Bias in individual participant data meta-analyses: 15:38