A pill routinely prescribed for morning sickness may not be an effective nausea remedy, researchers say - although not all doctors agree.

The warning comes from researchers who conducted a new analysis of a previously unpublished trial that was used to win marketing approval for the drug in the U.S. and Canada.

The drug, pyridoxine-doxylamine, is sold as Diclegis in the U.S. and Diclectin in Canada. It has been taken by millions of pregnant women since it was developed in the 1970s, researchers note in PLOS One. An older version of the medicine with an additional ingredient has been used as far back as the 1950s.

For the current study, researchers reviewed data from a decades-old trial and found little evidence that the medicine is effective, said study co-author Dr. Navindra Persaud, a researcher at St. Michael's Hospital and the University of Toronto in Canada.

"This medication is recommended as the first-line treatment for nausea and vomiting during pregnancy," Persaud said by email. "We now have more information about this 1970s study that should make us question whether this medication should have been approved and whether it was ever proven to be effective."

The original trial in the 1970s was designed to determine if the drug could alleviate morning sickness in the first trimester of pregnancy.

To see if those results proved the drug effective, Persaud and co-author Dr. Rujun Zhang of the University of Toronto examined 36,000 pages of documents from the U.S. Food and Drug Administration, including the original study report, the protocol and summary results, and other documents from Health Canada.

About 2,300 pregnant women participated in the original trial at 14 clinics in the U.S., the researchers found. All of the women were experiencing nausea and vomiting in the first 12 weeks of pregnancy.

Women were randomly assigned to eight treatment groups, including one that received a placebo or dummy pill and seven that received a variety of drugs including the combination currently sold as Diclectin.

Each participant was asked to take two tablets a day at bedtime and one additional pill in the afternoon as needed for one week.

Based on data from 1,599 women who completed the study, participants were 14 percent more likely to report treatment was moderate or excellent with Diclectin than with placebo, the new analysis found.

More on this...

But this finding may not be reliable because the final study results aren't available and 37 percent of participants in the placebo group dropped out before the study ended, researchers note in the current analysis.

In the new analysis, researchers also couldn't see how physicians scored symptoms. And, data for 30 patients recruited in the original trial were excluded from analysis by regulators because there were data recorded without a record of patient visits, the new study notes.

Persaud wasn't able to contact any of the original researchers, and many of them may have died.

Still, Persaud said women prescribed this drug should stop taking it and speak to their doctors about other options to treat morning sickness.

But not all doctors agree with this assessment.

"Studies that have been done since this trial was completed in the 1970s show that doxylamine and pyridoxine are an effective treatment for nausea and vomiting in pregnancy," said Dr. Siripanth Nippita, a researcher at Harvard Medical School and Beth Israel Deaconess Medical Center in Boston who wasn't involved in the current analysis.

"Based on research involving over 200,000 women, we can say that doxylamine and pyridoxine are safe to take during pregnancy," Nippita added by email. "They can be an important part of treatment for nausea and vomiting."