Non-publication of clinical trials and other stories . . .

The non-publication of clinical trials is unethical—it breaches the trust of those who consented to take part in the experiment because it would produce useful knowledge. But just under a third of trials remain unpublished four years after completion, according to the latest study based on a survey of 400 trials registered with ClinicalTrials.gov (PLoS One 2014, doi:10.1371/journal.pone.0101826). This is only a modest improvement on the results of Joe Ross and colleagues’ landmark study of 2009 (PLoS Medicine 2009, doi:10.1371/journal.pmed.1000144). Trial participants and all who use knowledge deserve better.

Minerva was surprised to read in a recent Cochrane review of thiazide diuretics for hypertension (Cochrane Database Systematic Reviews 2014;5:CD003824, doi:10.1002/14651858.CD003824.pub2) that these drugs may not cause diabetes. “Chlortalidone increased serum glucose but the evidence was unclear for other thiazides.” Chlortalidone is, in fact, not a true thiazide, unlike hydrochlorothiazide, indapamide, and trichlormethiazide, which were the agents used in the Japanese DIME trial (BMJ Open 2014;4:e004576 doi:10.1136/bmjopen-2013-004576). This was designed to …