Mifepristone, MMR, and the Perils of Washing Coal

Federal law protects the well-being of study participants. How about the rest of us?

In this newsletter I'll be discussing two lines of research currently being planned by Health and Human Services (HHS) agencies.

The research concerns unrelated drugs ­– mifepristone and the MMR vaccine – but there are common themes.

For instance, in each case, data collection won't harm the participants themselves, but the findings may jeopardize public health.

(Ironically, safety concerns are what motivate the research in the first place.)

More than two decades of evidence attest to the safety of these drugs. Unfortunately, ideological bias rather than science is fueling the quest for new data.

The new data will either be useless (because it confirms what we already know) or dangerous (because it falsely impugns the safety of the drugs and curtails access).

In Chinese, this is called "washing coal". It's a pointless activity, and it might create a mess.

A key distinction

Wait, you may be thinking. Isn't this the worst possible time to advocate against health research?

As the New York Times reported yesterday, nearly 2,500 NIH grants have been terminated or delayed. The Trump administration's 2026 budget would cut funding for health research across a variety of agencies. At Harvard alone, more than $100 million in funding has been withheld for ongoing studies of cancer, pediatric HIV, and Parkinson's disease. The health sciences are currently under siege.

I'm not suggesting that any study be stopped. Rather, I'll be calling out two lines of health research that shouldn't be (re)started. This research is unethical.

Some context

Up until the 1970s, research that's clearly unethical wasn't necessarily illegal. In the U.S., the most infamous and influential example was the Tuskegee Syphilis Study.

In this study, 399 African-American men with syphilis were neither treated for the disease nor informed they had it. Beginning in 1932, the U.S. Public Health Service and the Tuskegee Institute simply recorded the progression of symptoms.

The Tuskegee study didn't harm the American public, but the consequences for participants and their families were horrific. More than 100 of the men died from syphilis or its complications, 40 wives were infected, and 19 children were born with congenital syphilis.

Public exposure of the Tuskegee study in 1972 was a major impetus for the National Research Act of 1974 and other legislation meant to protect the well-being of study participants.

Nowadays, federal law requires that most research on humans be approved in advance by an Institutional Review Board (IRB).

For nine years I served as chair of the IRB for my university. Safety is never 100% guaranteed (in one of the studies we approved, a PC monitor fell and broke a participant's toe), but I know from personal experience that the IRB process is rigorous and genuinely helps minimize risk.

So, if research participants are protected from harm, how could the results of a study imperil the rest of us?

Plans for reassessing the safety of mifepristone and the MMR vaccine illustrate how this might happen.

1. Mifepristone.

Last month, HHS secretary RFK Jr. directed the FDA to review regulations on the abortion medication mifepristone. On Tuesday, FDA commissioner Marty Makary pledged to closely monitor new safety data.

Makary's promise seems innocuous. The FDA attempts to monitor the post-marketing safety of all the drugs it approves.

In this case though, monitoring is dangerous. Not to the women whose data are "monitored", but, as we'll see, to any woman who might need an abortion.

Some background

Mifepristone was developed in the early 1990s by the French scientist Dr. Étienne-Émile Baulieu, who passed away last week.

In 2000, the FDA authorized the use of mifepristone and misoprostol to bring about an abortion. Typically, mifepristone is taken to end the pregnancy, and, one to two days later, misoprostol is used to induce a miscarriage.

Hundreds of studies have shown that mifepristone and misoprostol are both effective and safe. Adverse effects do occur, as with any medication, but the rates of serious complications are lower than for medications such as penicillin and Viagra.

Why the sudden doubts about mifepristone's safety?

Junk science. In particular, a pair of deeply flawed studies sponsored by the Charlotte Lozier Institute, a pro-life advocacy group. (I discuss these studies at length here, in what I call a "case study on how researchers can lie with statistics".)

The Lozier-sponsored studies have no standing in the scientific community. Sage, the publisher, retracted them in 2024 owing to methodological errors and undisclosed conflicts of interest. RFK Jr. and Marty Makary continue to cite them nonetheless.

In addition, RFK Jr.'s directive to Makary, as well as Makary's promise on Tuesday, were spurred by a new "study" that appeared on April 28.

The new "study"

This "study" is neither peer-reviewed nor published in a scientific journal. Rather, it's simply posted on the Ethics and Public Policy Center website, an advocacy group that prioritizes

"pushing back against the extreme progressive agenda while building a consensus for conservatives."

The authors of the "study" are not health experts and appear to have no background publishing peer-reviewed research.

The authors examined 865,726 insurance claims from 2017-2023 and concluded that 10.93% of women who took mifepristone experienced a serious adverse event.

Prior studies indicate rates of about 0.3 to 0.5%. Why did the new "study" yield a much higher estimate?

In some respects it's hard to address the question. The authors didn't identify the database they used, as routinely done in this kind of research. Nor did they explain how they defined "other abortion-specific complications", which represent 51.96% of the adverse events they recorded.

These lapses in transparency make it impossible for other researchers to directly evaluate their claims.

All the same, there's clear evidence that adverse events were overestimated. Two examples:

• Diagnostic coding did not reliably distinguish between severe hemorrhaging and the routine bleeding that occurs after an abortion. It's likely that some of the adverse events represent the latter.

• 4.7% of women in the study visited an emergency room within 45 days of their abortion. This is consistent with the 4.6% figure currently listed on the FDA-approved label for the drug (sold as Mifeprex). But ER visits constituted 43.27% of the adverse events the authors recorded. In other words, nearly half of their data is consistent with prior work on mifepristone safety.

Here's a broad takeaway: 95.13% of the adverse events the authors reported were either inadequately defined (abortion-specific complications) or comparable in prevalence to what has been previously documented (ER visits).

I could go on, as others have, but you get the idea. Almost all of the flaws in the measurement of ER visits would tend to overestimate rather than underestimate adverse events attributable to mifepristone.

Why additional research might be dangerous

Further analysis of mifepristone's safety record, beyond the FDA's routine reviews, might simply confirm that the drug has a low rate of adverse effects.

Alternatively, the research might create false concerns about safety.

How could that happen? Any or all of the following might yield overestimates of risk:

• ideological bias

• flawed methodology

• statistical flukes

Ideological bias is my greatest concern. RFK Jr. has not acknowledged the hundreds of studies attesting to the safety of mifepristone. Instead, he keeps mentioning that handful of retracted or non-peer-reviewed studies I've discussed.

Ideological bias leads methodological choices that yield ideologically-favorable outcomes. It's not hard to create data to fit a narrative. This has happened before – as in the April 28 "study" – and may happen again unless the FDA examines new mifepristone safety data competently and fairly.

2. The Measles-Mumps-Rubella (MMR) vaccine.

On March 7, under orders from RFK Jr., the CDC announced plans for a large-scale study aimed at determining whether the MMR vaccine causes autism (i.e., Autism Spectrum Disorder, or ASD).

In fact, more than two decades of research, and a systematic review of 338 studies including over 1.2 million children, have shown that MMR vaccination does not increase the risk of ASD.

Why would anyone ever think they were related?

Mainly, three kinds of fallacious or false evidence:

(a) Anecdotally, signs of ASD are often noticed in the months following an MMR vaccine.

As I mentioned last week, the fact that X comes before Y doesn't prove that X causes Y. Infants often start walking shortly before hints of ASD are observed, but walking doesn't cause ASD.

(b) A 1998 Lancet study claimed to show an MMR-autism link as well as gastrointestinal causes.

This study is the major source of false MMR-ASD claims.

The study only had 12 participants, so it never stood a chance of providing strong evidence. In any case, it has been discredited in almost every possible way.

For instance, the lead author, Andrew Wakefield, was found to have fabricated data, altered numbers, and concealed financial conflicts of interest (e.g., a patent for a measles vaccine). The study was eventually retracted and Wakefield was stripped of his medical license.

(c) Thimerosal, a preservative used in vaccines, has been blamed for causing autism.

As the FDA itself notes, more than two decades of research has shown no connection between thimerosal and ASD, nor did rates of ASD decline after this preservative was removed from the MMR vaccine in 2001.

Mifepristone clearly has adverse effects, on rare occasions, and it's clear why those effects occur. As for the MMR vaccine, there's neither a connection to ASD nor a plausible hypothesis for why there would be.

Back in March, Jerome Adams, the U.S. Surgeon General during Donald Trump's first term, noted that

“This initiative [the CDC's planned research] if—and I want to emphasize the "if" here—done thoughtfully and without any agendas, could help reaffirm the safety of vaccines, could help reassure patients, and could ultimately strengthen public trust and immunization programs.”

I'm doubtful that new evidence would persuade skeptics. So much data already fails to show an MMR-autism link – nor is there reason to expect one, based on what we know about the causes of ASD, which reflect a complex, poorly understood interplay of genetic and environmental influences.

(Perhaps it's comforting for non-scientists like RFK Jr. to fabricate a villain, but not all health conditions have a single cause.)

As with mifepristone, the best-case scenario for the MMR vaccine is that the CDC research confirms what we already know. Ideological bias, bad methodology, and/or statistical flukes may falsely impugn its safety but, once again, my main concern is the prospect of bias. RFK Jr. has already hired David Geier, a discredited vaccine skeptic, to run the CDC study.

If new findings do indicate an MMR-ASD link, vaccination rates will surely drop, and people will suffer. The graph below, created from CDC data, offers a snapshot life before and after the vaccine:

What can we do?

Our tax dollars are being spent on FDA and CDC research that will either be useless or endanger the well-being of women and children who need access to mifepristone and the MMR vaccine.

This money would be better spent on the health research currently being delayed or terminated, the health data no longer being gathered, and health programs such as USAID and PEPFAR.

It's easier to call out corruption than to fix it, but calling it out is a good starting point. Beyond that, here are a few things we might do:

• Check out the Stand up for Science website for guidance on contacting your congressional representatives, the opportunity to sign an open letter, and information on local events and volunteer opportunities. (If nothing else, there's solace in finding communities of people who share your concerns.)

• Consider donating to organizations that promote women's reproductive rights, including those that support access to abortions, as well as organizations devoted to countering vaccine misinformation, such as the Vaccine Integrity Project.

• Finally – I'm dreaming now – let's push for a constitutional amendment requiring subject-area expertise for cabinet-level positions and other leadership roles in the federal government. This wouldn't protect us from the likes of Marty Makary or Dr. Oz, but RFK Jr. at least would be excluded, because has no background in science or medicine. (What he does have, among other things, is a talent for washing coal when his ideological biases call for it.)

Thanks for reading!

---

洗煤: Washing coal – i.e., a pointless activity. In Mandarin, it's pronounced "she may", with a rising tone on the second word, as if asking a question.