Popular morning sickness drug is not effective, new research finds
A new report on Diclectin is based on clinical trial records kept secret by the drug’s manufacturer and Health Canada.
|Report an Error|
Share via Email
Drug trial records kept secret by the Canadian government and a pharmaceutical company show a popular morning sickness medication is not effective, a new published report says.
The findings about Diclectin, published Wednesday in the online journal PLOS ONE, are the culmination of a seven-year transparency battle by Toronto doctor Nav Persaud to see the evidence behind a go-to treatment for pregnant women across the country.
Persaud recommends doctors stop prescribing the drug.
“This information that’s been hidden until now indicates that (Diclectin) is not effective, and therefore women should not be taking it,” said Persaud, a researcher, lecturer and family physician at St. Michael’s hospital.
This article is about transparency and the drug’s efficacy, not its safety. The maker of Diclectin, Duchesnay, said the drug is the most studied medication approved for use by pregnant women, and its use is supported by multiple studies.
“(Diclectin) has been prescribed for more than 40 years in Canada and has been repeatedly proven safe and effective for use throughout pregnancy,” Duchesnay spokeswoman Fiona Story said in a statement.
As part of Persaud’s protracted fight to learn what he can about the popular pill, he has in recent years authored papers exposing flaws in the research purporting the drug’s effectiveness.
The 2009 clinical trial that is the focus of Persaud’s new analysis was funded by the company that makes the drug.
Health Canada only released the 2009 clinical trial records to Persaud and his co-authors to analyze after he signed a confidentiality agreement promising not to share the original trial data that underpinned his findings.
Before this, the regulator had told him the information was confidential and belonged to Duchesnay. The drug company would also not allow the information to publicly shared, Persaud said.
“It’s disturbing that the company that sponsored the study has hidden the information. It’s also disturbing that the federal government that’s charged with protecting the health of Canadians has decided to hide this information,” Persaud said. Duchesnay did not answer questions about why it would not release the trial records.
Diclectin is a combination of an antihistamine and vitamin B6. Research has shown one prescription is filled for every two live births in Canada.
The clinical trial found that women who took Diclectin reported a reduction in symptoms that were 0.7 points greater than women using a placebo, based on a 13-point scale.
That difference, however, fell far short of the company’s own threshold for proving the drug was effective, Persaud’s re-analysis of the trial records found.
Those records, given to Health Canada by the drug’s manufacturer, specified that the 2009 trial findings would be clinically important only if there was a three-point reduction in symptoms.
“It’s not a minor detail of the study you or may not want to consider. It goes to the core of the main conclusion of the study. Either the medication is effective or it’s not,” Persaud said.
The 0.7 difference in symptom reduction is statistically significant, Persaud said, but not large enough to be noticed by women taking the drug. Most of the women taking the placebo sugar pill had little to no nausea or vomiting symptoms by the end of the two-week trial.
The findings of the new analysis, authored by Persaud and four other Canadian researchers, are based on 9,174 pages of previously secret clinical trial records obtained from Health Canada.
The regulator would not allow Persaud to show the journal or the public any of the underlying data in his re-analysis to support his findings.
In a statement, Health Canada said the “available evidence continues to support Diclectin in the treatment of nausea and vomiting during pregnancy.”
Health Canada reviewed Diclectin’s safety and efficacy in 2016, during which time Persaud presented his findings about the 2009 clinical trial. An expert panel “noted methodological issues in the study” but determined that they did not require any changes to Diclectin’s labeling, a Health Canada spokesperson said.
The 2009 clinical trial led to the drug being approved for use in the United States, where it’s marketed as Diclegis. The company said the trial was designed “in collaboration with” the U.S. drug regulator and found a “statistically significant” difference between the drug and placebo treatments.
Like Health Canada, the U.S. Food and Drug Administration said its determination that the drug is safe and effective remains unchanged.
The U.S. regulator is reviewing the findings from Persaud’s re-analysis.
The results of the clinical trial were first reported in a 2010 medical journal article co-authored by Dr. Gideon Koren, who in 2015 left his prominent post at Sick Kids hospital in disgrace following a scandal at the Motherisk lab.
That 2010 journal article cherry picked certain findings (known in clinical trials as “secondary endpoints”) to make the medication appear more effective, Persaud said. The article noted that Diclectin led to a reduction in missed work time due to morning sickness, but did not mention that there was no difference compared to the placebo for things such as time lost from household tasks or the number of visits to health-care providers.
Koren, now living and working in Israel, did not respond to an email request for comment.
Persaud said even the reported 0.7 difference in symptom reductions between the trial’s drug and placebo groups is misleading because of how the 2010 study analyzed participants who dropped out of the clinical trial.
The original study recorded trial dropouts by assuming the severity of their nausea and vomiting remained the same as the last day it was recorded, even though symptoms tapered off over time for most of those using the real medication and the sugar pill.
“Because more participants in the placebo group dropped out than the Diclectin group, that tended to make the symptom scores in the placebo group look worse,” he said.
A 2015 Star investigation found Diclectin’s maker, Duchesnay, has had financial relationships with Koren and his Motherisk program, as well as the Society of Obstetricians and Gynaecologists of Canada.
The Society’s 2016 clinical guidelines on nausea and vomiting in pregnancy recommend Diclectin or vitamin B6 as the first-line drug treatment when dietary and lifestyle changes haven’t worked. The guidelines cite the Koren-led study about the 2009 clinical trial as evidence to support its recommendation.
The Society’s clinical committee does “reviews of all relevant scientific evidence and will adjust its recommendations if a change is indicated. At this point we stand behind our clinical expert’s recommendations,” Society President Dr. Jennifer Blake said in a statement.
The Society said the financial support it receives from Duchesnay and other companies is arms-length for educational purposes and follows strict ethical standards.
In January of last year, the Canadian Family Physician medical journal said in a commentary that it no longer recommends the popular medication as the first-line treatment. The commentary renounced the journal’s previous publication of clinical guidelines advocating the drug’s use, which the journal says were developed by Motherisk. At the time, the journal said it ended its longstanding relationship with Motherisk in part because of concerns about its independence from Duchesnay.
As for why PLOS ONE decided to publish Persaud’s article without being able to see the underlying data that Health Canada insisted he keep secret, an editor of Persaud’s re-analysis said the doctor’s methods and approach were “rigorous and well described.”
“The study report underwent a thorough review, with extensive queries and clarification back and forth on methods and results,” said Barbara Mintzes, a senior lecturer at the University of Sydney in Australia. She also set up the peer-review process so the research could be properly vetted.
Mintzes said it is important for doctors and pregnant women to have as much information on how well the popular drug works as possible.
“The evidence of effectiveness was marginal at best in this trial,” she said. “In addition to informing clinicians and patients, (Persaud’s research) also has important implications for pharmaceutical policy, highlighting the need for full public access to clinical trial results, including individual level data, to allow independent researchers to re-analyze these results.”
Health Canada has proposed updating regulations that would make publicly available reams of previously confidential clinical trial documents about prescription drugs and medical devices.