Intended for healthcare professionals

News

Flawed invisible trial underpins US morning sickness drug, researchers say

BMJ 2017; 356 doi: https://doi.org/10.1136/bmj.j132 (Published 09 January 2017) Cite this as: BMJ 2017;356:j132
  1. Owen Dyer

A 1975 trial that led to the approval of the US’s only approved morning sickness drug—long buried in the Food and Drug Administration archives but newly published—is marred by “questionable data integrity” and should not be relied on “because of the high risk of bias,” say researchers who obtained the original trial from the FDA under a freedom of information request and published it in the journal PLoS One.1

The 40 year old research has been cited frequently by the FDA, Health Canada, and the drug’s manufacturer, Duchesnay, as the principal evidence for the safety and efficacy of the vitamin B6/doxylamine combination, formerly known as Bendectin but today sold as Diclegis in the US and as Diclectin in Canada.

It was not published, however, until a Toronto family doctor, Nav Persaud, responded to a 2013 call in The BMJ to restore invisible and abandoned trials—the RIAT initiative.2 Researchers planning to publish such trials were invited to state their intentions in the Rapid Response section, which Persaud did.

He sought data from Health Canada but received only heavily redacted documents at the price of signing a confidentiality agreement. Duchesnay did not provide the trial, either, but Persaud told The BMJ that obtaining it from the FDA “was relatively easy; I made a request and they provided a large quantity of information quickly, much more information than Health Canada.”

Among the flaws the researchers identified were that the final results of the study were missing, data from over 30% of recruited subjects were never analyzed, and few details about statistical analyses were provided.

Data from one trial investigator had been excluded because he was found to have been “data recording in absence of patient visits.” Using the Cochrane collaboration risk of bias tool, the researchers reported a high overall risk of bias in the study.

The vitamin B6/doxylamine treatment has been taken by about 33 million pregnant North American women since the 1950s. It is still given in half of all full term pregnancies in Canada and is the only treatment for nausea and vomiting in pregnancy approved in the US.

Originally sold as Bendectin by Merrell Dow, it was taken off the market voluntarily in 1983 after a string of lawsuits alleged that it caused birth defects. A prominent expert witness in these cases was William McBride, then recognized as the physician who first warned of thalidomide’s dangers. McBride was found to have falsified research to make Bendectin seem teratogenic, and he was later struck off in his native Australia.

In 2013 the FDA approved a return of the treatment, manufactured by Duchesnay. The US brand Diclegis made headlines in 2015, when the FDA ordered the correction of an Instagram post by the reality television star Kim Kardashian claiming that the drug had cured her morning sickness. Kardashian’s endorsement omitted any information on contraindications or adverse effects, and Duchesnay later acknowledged that Kardashian had been paid to endorse the drug.

In response to the PLoS One article, covered by several national US newspapers, Duchesnay issued a summary of evidence supporting the use of its product.3 Health Canada and the FDA both said that they were confident in their assessment of the drug’s safety and efficacy but that they would review the new evidence.

References

View Abstract