In 2017, researchers working at the University of Toronto in Canada and the Keenan Research Centre of the Li Ka Shing Knowledge Institute of St. Michael's Hospital, also in Toronto, Canada, reanalyzed a pivotal doxylamine and pyridoxine clinical trial from the 1970s and made a fairly surprising finding: the data had substantial flaws.
Digging even deeper, Dr. Navindra Persaud — from the Department of Family and Community Medicine at St. Michael's Hospital as well as the Department of Family and Community Medicine at the University of Toronto — and colleagues reanalyzed another clinical trial using an updated version of the drug, this one from 2010.
The recent trial was part of the reason that the Food and Drug Administration (FDA) approved the popular morning sickness drug known as Diclegis.
Commenting on his findings, Prof. Persaud told me, "The medication seems to be ineffective based on the results of this trial. I was shocked to learn this about a commonly prescribed medication."
The results of Prof. Persaud's analysis are now published in the journal PLOS ONE.
'Only FDA-approved' morning sickness drug
Diclegis is the only drug approved by the FDA for the treatment of morning sickness in pregnancy. According to its manufacturer Duchesnay, it has been prescribed to 33 million women worldwide.
In Canada, where the drug is known as Diclectin, it is prescribed at least once for every two births.
I asked Prof. Persaud why he decided to look into the drug. "I used to prescribe this medication," he explained. "I was taught to prescribe it. The medication was recommended as the first line medication for nausea and vomiting during pregnancy."
"When I looked carefully at the clinical practice guidelines that recommended this medication, they did not cite supporting studies. So I tried to find the basis for the recommendations. It was surprisingly difficult to obtain information about this commonly prescribed drug."
Prof. Navindra Persaud
When the results of the clinical trial were initially published in the American Journal of Obstetrics & Gynecology in 2010, the study authors concluded, "Diclectin [...] is effective and well-tolerated in treating nausea and vomiting of pregnancy."
Drug 'significantly better than placebo'
The study involved 261 pregnant women, all of whom completed a 2-week course of daily Dicletin or placebo treatment. As the authors explained in the paper, "Diclectin led to significantly greater improvement in NVP [nausea and vomiting of pregnancy] symptoms as compared with placebo."
This was based on a drop in what is known as the PUQE score. PUQE stands for pregnancy-unique quantification of emesis/nausea, with a score of 3 meaning no symptoms and a score of 15 being the most severe.
Pregnant mothers in the placebo group saw a 3.9 drop in their PUQE score from 8.8 at the start of the study, while those in the treatment group saw a 4.8 point drop down from 9.0.
In addition to the PUQE score, the team also saw a greater improvement in the global assessment of well-being score, less time taken off work, and fewer women seeking alternative treatments when they received the drug.
More pregnant mothers asked to keep taking the drug after the study finished than the placebo.
The key finding remains the PUQE score.
But while the difference seen in the clinical study may be statistically significant, Prof. Persaud uncovered that the results were not in line with what the clinical trial set out to achieve.
'Expected difference in PUQE score of 3'
Citing the original clinical study report and the FDA review of the study, Prof. Persaud explains that the difference in PUQE score between the drug and the placebo groups was expected to be 3 points — a far cry from the difference reported in the clinical trial.
He told me that "[he] was [...] surprised that important information about the trial was hidden until now. Although some results were published in 2010, the earlier reports did not mention the fact that a difference of 3 points on the 15-point symptom scale was prespecified as the minimal important difference (or the smallest difference that a patient would deem as important)."
I asked him why he thought the FDA decided to license the drug in light of the results not meeting the expected differences.
"While the review by the FDA was very thorough, the review does not address the fact that the 3-point difference between groups was not found in the trial."
Prof. Navindra Persaud
In his paper, he explains, "The FDA summary review indicated a 'small, but statistically significant improvement'" and noted that "although the treatment effect is small, there are no other FDA-approved treatments for nausea and vomiting of pregnancy."
Placebo just as good as drug?
Commenting on the results of the trial and his reanalysis, Prof. Persaud explained, "In this trial, women given a placebo had large improvements in symptoms over 2 weeks. By the end of the 2-week trial, women given a placebo had symptom scores around 4 and the lowest possible score on the symptom scale is 3."
"So the results of this study indicate that no studied treatment could have had a substantial benefit over the placebo," he added.
With millions of women worldwide having taken the drug over the years, could it be the placebo effect that causes the improvement in symptoms?
Explaining his take on the situation, Prof. Persaud said, "We may find that nausea and vomiting during pregnancy may be like the common cold: it is common, it causes substantial suffering, it can occasionally cause serious complications, and it does not have a highly effective treatment."
So, is it likely that the drug will be withdrawn from the market based on Prof. Persaud's findings? He doesn't think so.
"It is very unusual for medications to be removed from the market because of ineffectiveness," said Prof. Persaud. "Medications are withdrawn when they are found to be harmful after approval but even this is quite rare. So it is unlikely that this medication will be withdrawn."
What to do when morning sickness strikes
I am no stranger to morning sickness, having endured a seemingly never-ending bout last summer during my second pregnancy. Here is what Prof. Persaud told me about other treatment options.
"[...] [R]ecommended treatments include P6 acupressure, antihistamines such as diphenhydramine, and other nausea treatments such as metoclopramide." However, he did add the following caveat: "None are proven to be highly effective."
Prof. Persaud also pointed me in the direction of a 2015 systematic review examining treatments for nausea and vomiting during pregnancy. While some — such as ginger, chamomile, vitamin B-6, and lemon and mint oil — were effective for some women, the "[...] review found a lack of high-quality evidence to back up any advice on which interventions to use."
"[...] All of those results [...] should be taken with a grain of salt," Prof. Persaud said, echoing the review's findings.
If you are looking for more information about coping with morning sickness, check out our handy guide "Morning sickness: 10 tips to relieve it."