Dr. Jen Gunter is a board-certified OB/GYN in both Canada and the US and a pain medicine physician. She blogs on a range of topics, often with great insight.
On May 1, 2015, Dr. Gunter posted a piece titled “I took Zofran in pregnancy. My son has a heart defect. I’m not suing. Here’s why.” on her blog. You can read the post here.
In her piece, Dr. Gunter used a photo of Michael Monheit, Esq., the lead sponsor of ZofranLegal.com, to illustrate the many ads and articles about Zofran lawsuits that have cropped up in the last few months:
Obviously, Dr. Gunter’s opinion on the Zofran litigation differs from our own. That’s fine; we welcome spirited debate. And we have no problem with her using Michael’s likeness. But we did feel compelled to respond to her arguments.
We also noticed several factual inaccuracies in Dr. Gunter’s post, and corrected them in our response. We provided screenshots from her article where we thought context was needed.
As a medical professional, your work is vital. Anti-vaxxers, politicians abusive (or simply ignorant) of current scientific knowledge, celebrities promoting snake-oil treatments: you take them all down with gusto, rigor and evidence. We respect your experience, and your views on the use of scientific research to justify legal action. But we’d like to clarify some aspects of the Zofran litigation that we think you left out.
We agree that determining “what causes what” is difficult, but accepted legal doctrine embraces a notion of causality premised on increased risks. For many years, federal courts have employed a two-criteria test for studies on a drug’s possible effects:
- the study has to find a more than two-fold increase in the relative risk
- the study’s authors must ascribe statistical significance to their findings
In our own efforts representing clients in Zofran lawsuits, the studies we describe satisfy those conditions.
The study published in Reproductive Toxicology isn’t the Andersen study. It’s the Danielsson study, and it analyzed records from the Swedish Medical Birth Register. Out of more than 1.5 million pregnancies, the researchers identified 1,349 women who had been prescribed Zofran or its active ingredient. As you note, 17 of these exposed women gave birth to children with cardiac septal defects. Another 2 had children with other congenital heart defects. Their conclusion: “the risks for a cardiovascular defect and notably a cardiac sept[al] defect were increased and statistically significant,” with a 2.05 increased relative risk for septal defects.
Danielsson et al write that “a final answer to the question of moderate teratogenicity is seldom obtained from one study but that repeated studies are necessary.”
The “other group” to which you link, Pasternak et al, looked at birth records logged in the Danish Medical Birth Registry, not the same Swedish registry used by Danielsson’s group. The researchers concluded that no significant increase in the risk for birth defects was associated with Zofran exposure.
The real Andersen et al reanalyzed the Danish birth records used in Pasternak, but made two notable changes. They widened the study period to include about 300,000 more pregnancies and limited their exposure group to women who had been prescribed Zofran during the first trimester. Of the 1,368 women who redeemed a prescription during early pregnancy, 61 (4.7%) gave birth to children with major birth defects. 3.5% of the unexposed pregnancies resulted in a major malformation. An increase in the prevalence of heart defects was pronounced and statistically significant, with odds ratios of 2.1 for atrial septal defect, 2.3 for ventricular septal defect and 4.8 for atrioventricular septal defect.
When we refer to a study that found less than a doubling in risk for a specific defect, or in which the authors couldn’t ascribe statistical significance, we say so. Like when we concluded a discussion of the Pasternak study by saying, “However, the report, published in the New England Journal of Medicine, concluded that Ondansetron is not associated with increased malformation rates when used for morning sickness.” Or when, in reference to an Australian study, we wrote that its “sample size was too small to draw statistically significant conclusions.”
We have no intention of misleading people, or twisting scientific results to serve some Machiavellian purpose.
Beyond the subject of specific study results, you make an important point: risks become relative for individual patients.
We agree: prescribing decisions should emerge from an “informed, collaborative decision making” process.
Your use of the epilepsy medication example is salient. There are many drugs with well well-documented risks, but in some cases their use is justified in light of the benefits. Our conviction is not that any risk is too high, but that patients should be thoroughly informed, and allowed to participate in risk-benefit analyses along with their medical professionals.
Many Zofran lawsuits claim that GlaxoSmithKline actively concealed evidence of Zofran’s potential risks. This is an allegation we’re currently investigating. If it’s true, if the company failed to submit adverse event reports linking Zofran to birth defects as plaintiffs have alleged, that’s not just an issue for the FDA. We believe the public in general, and patients specifically, have a right to know the full story behind the medications they’re prescribed.
Again, we’re investigating, not presupposing, these claims. For adverse event reports, Monheit Law submitted a Freedom of Information Act (FOIA) request and obtained every report sent to the FDA involving Zofran or its active ingredient. Then we combed through to find the reports in which Zofran was a suspected cause of birth defects. We made all of this information, the files sent to us by the FDA and the results of our analysis, public here. And we presented those results without bias.
But what may be the most serious allegation raised against GlaxoSmithKline involves the company’s marketing of Zofran. In 2012, the federal government filed a lawsuit against GSK, charging the company for promoting a number of drugs “off label,” for indications not approved by the FDA. One of those drugs was Zofran.
In the eventual settlement agreement, the Department of Justice claimed GSK had “promoted the sale and use of Zofran for a variety of conditions other than those for which its use was approved as safe and effective by the FDA (including hyperemesis or pregnancy-related nausea),” “made and/or disseminated unsubstantiated and/or false representations or statements about the safety and efficacy of Zofran concerning [its use as a treatment for pregnancy-related nausea” and “offered and paid illegal remuneration to health care professionals to induce them to promote and prescribe Zofran.”
If those allegations are true, and the DOJ has never backed down from its position, then GSK deprived physicians of the information needed to make a well-informed risk benefit decision, too.
A survey of more than 25,000 women taken by Boston University found that between 2003 and 2008, Zofran was prescribed twice as often as any other common off-label morning sickness treatment. And while we don’t know whether or not the drug’s being used as a “first line” treatment, it looks like Zofran is being favored over other pharmaceuticals.
As you note in your article, pharmaceutical companies are extremely unwilling to study a drug’s effects during pregnancy. And as you know, off-label prescription is extremely common. For some conditions and certain drugs, it’s the industry standard. But with a new antihistamine-based treatment (however expensive) finally approved as safe and effective, we believe patients should be apprised of Zofran’s potential association to birth defects. That just seems right to us.