Dr. Mitchell Creinin never expected to be in the position of investigating a treatment he doesn't think works.
And yet, Creinin will be spending the next year or so using a research grant from the Society of Family Planning to put to the test a treatment he sees as dubious — one that recently has gained traction, mostly via the Internet, among groups that oppose abortion. They call it "abortion pill reversal."
The technique — a series of oral or injected doses of the hormone progesterone given over the course of several days — arose outside the usual avenues of scientific testing, says Creinin, a medical researcher and professor at the University of California, Davis.
Creinin, an OB-GYN, has spent the bulk of his career in family planning research. He has studied topics ranging from different treatments for miscarriage to how women choose birth control methods.
Performing abortions, he says, has always been a part of his practice and philosophy. "I need to provide these services to help women," Creinin says.
Proponents of "abortion pill reversal" say it can stop a medication-based abortion in the first trimester, if the progesterone is administered in time.
But Creinin says the progesterone treatments are ineffective at best in halting an abortion that has already begun. And, Creinin says, promotion of the treatment can be potentially harmful by giving pregnant women misleading information that an abortion can be undone.
Though critics of abortion pill reversal say the term is an unproven misnomer, it has been such a compelling phrase that it's already been written into the laws of a number of states.
Legislators in Arkansas, Idaho, South Dakota and Utah have made it a legal requirement in recent years that doctors who provide medical abortions must tell their patients that "reversal" is an option, although they are not prevented from also telling patients if they think the treatment doesn't work.
Medical researchers such as Creinin and the American College of Obstetrics and Gynecology are concerned by that trend.
"You create a law based on no science — absolutely zero science," Creinin says.
Proponents of the technique say they do have evidence. But it's anecdotal, Creinin says, or comes from studies that lack rigorous controls. It's time, Creinin says, for a formal study that can be definitive.
"I want to own that," he says.
In the first 10 weeks of a pregnancy, women who are seeking abortions generally have two options: a surgical procedure or a medication-based abortion (after that, only surgical abortions are performed).
Mifepristone pills work by blocking progesterone, a hormone that helps maintain a pregnancy. The second medicine, misoprostol, makes the uterus contract, to complete the abortion. Studies suggest that 95 percent to 98 percent of women who take both drugs in the prescribed regimen will end the pregnancy without harm to the woman. Surgical evacuation can complete the abortion, if necessary.
So what happens if a woman takes mifepristone, then changes her mind and wants to continue with the pregnancy?
If the change of heart comes in the first hour after she's swallowed that initial medicine, her doctor might help her induce vomiting. If she hasn't yet absorbed the first drug, the process may be stopped before it starts.
The bigger question, and one for which the data are murkier, is: What happens if a woman takes the first medicine but never goes on to take misoprostol, the second drug in the regimen?
According to the American College of Obstetrics and Gynecology, "as many as half of women who take only mifepristone continue their pregnancies." (If the pregnancy does continue, mifepristone isn't known to cause birth defects, ACOG notes.)
In 2012, a San Diego physician named George Delgado said he had hit upon a chemical way of stopping the abortion process with more certainty — a way to give more control to a woman who changed her mind. He called his protocol "abortion pill reversal."
A family medicine physician, Delgado calls himself "pro-life," not anti-abortion. He says about a decade ago he got interested in the 24-hour window after a woman takes mifepristone but before she takes misoprostol.
He'd received a call from a local activist who said a woman needed Delgado's help. She had swallowed the first pill in the abortion regimen but had reconsidered and no longer wanted to end her pregnancy.
"People do change their minds all the time," Delgado says.
Hoping to help the woman, Delgado gave her progesterone — a medication that has many uses, including as a treatment for irregular vaginal bleeding and as part of hormone replacement therapy during menopause. If progesterone is useful in these other ways, Delgado figured, it might stop the action of the progesterone-blocker mifepristone, and halt an abortion.
Delgado says the pregnancy of that first patient continued uneventfully, which he credits to the progesterone.
He then started giving the progesterone treatment to more patients who came to him. He went on to develop a network of clinicians around the country willing to give progesterone to patients who no longer want to go through with their abortions, although he wouldn't say how many of those clinicians took part in his research.
These days, Delgado says, most women who come to him for the progesterone treatment are self-referrals. While searching online, many find the website for the Abortion Pill Rescue Network, a nationwide group of clinicians who provide the treatment.
The network is backed by Heartbeat International, an anti-abortion rights group, and, according to spokesperson Andrea Trudden, includes more than 500 clinicians willing to prescribe progesterone to patients who have initiated the medication abortion process.
In support of their claims about abortion pill reversal, Delgado and colleagues have published their research in medical journals.
In 2012, Delgado co-authored a report in the Annals of Pharmacotherapy on the experiences of six pregnant women who received mifepristone and then injections of progesterone. Four of the women, the paper said, were able to carry their pregnancies to term.
In a statement released in August 2017, ACOG said the results of the study, a type known as a case series that didn't include a comparison group, "is not scientific evidence that progesterone resulted in the continuation of those pregnancies." ACOG's statement also said: "Case series with no control group are among the weakest forms of medical evidence."
In 2018, Delgado and colleagues in his network of health providers published a larger case series, this one involving 754 patients, in the journal Issues in Law and Medicine. The paper concluded that the reversal of mifepristone's effects with progesterone "is safe and effective."
The researchers acknowledged that the study didn't randomly assign women to receive a placebo or mifepristone. A study like that, called a randomized placebo-controlled trial, would provide strong evidence. But Delgado and his colleagues wrote that doing this kind of trial "in women who regret their abortion and want to save the pregnancy would be unethical."
"There's no alternative treatment," he says. "You can't always wait for the [randomized, controlled trials]. If it's lifesaving, there's no alternative."
State legislatures consider "abortion reversal" bills
One of Delgado's most outspoken critics, Dr. Daniel Grossman, an OB-GYN at the University of California, San Francisco, says all of the published studies supporting this use of progesterone have been marred by methodological flaws that inflate the "success rate" of the reversal treatment.
Last October, Grossman and Kari White, a sociologist at the University of Alabama, Birmingham who studies family planning issues, wrote an editorial in the New England Journal of Medicine criticizing Delgado's research methodology, saying he used flawed statistics and didn't set rigorous criteria for the characteristics patients had to fulfill to be included in the study.
"A systematic review we coauthored in 2015 found no evidence that pregnancy continuation was more likely after treatment with progesterone as compared with expectant management among women who had taken mifepristone," they wrote.
"I think there's a big bias against abortion pill reversal," Delgado says in response. "ACOG typifies that bias by coming out with strong statements. ... This is a new science, but we have a substantial amount of data, and it's been proven to be safe."
The critics haven't slowed Delgado's supporters.
Already in 2019, legislators in several states — Kansas, Kentucky, North Dakota and Nebraska — have been considering bills that would require abortion providers to tell their patients about abortion reversal. Back in 2017, Delgado testified in support of similar legislation in Colorado, although the proposal never made it into law.
Grossman says he's furious that states are forcing abortion providers to give their patients inaccurate information related to abortion care.
What's more, Grossman says, "these laws take an extra step ... and essentially are encouraging patients to be a part of clinical research that isn't really being appropriately monitored. ... This is really an experimental treatment."
Progesterone hasn't been evaluated by the Food and Drug Administration for reversing a medication abortion. Doctors are permitted to prescribe drugs for uses not approved by the FDA as part of the practice of medicine. It's known as off-label use.
Until Delgado published his 2018 paper, Delgado told his patients they were receiving a "novel treatment." He says he believes there is now enough research to support the routine off-label prescription of progesterone for women who don't want to complete a medication-based abortion.
"Now we have a substantial amount of data. There is no alternative. And it's been proven to be safe," Delgado says. "Why not give it a chance?"
Although Creinin disagrees that the evidence supports this use of progesterone, he is sympathetic to the idea that women who seek an abortion may not be certain about the decision at their first appointment. Creinin says he supports policies that allow women as much control as possible over the decision about whether or not to terminate a pregnancy.
"There are people who change their minds," Creinin says. "That's a normal part of human nature."
UCSF's Grossman agrees.
He encourages abortion providers, when possible, to send the mifepristone and misoprostol home with the patient, if she requests it. That way, she can start the protocol only if and when she's ready, rather than make the decision in a clinic where she might feel rushed. (FDA rules about mifepristone say the pill can only be dispensed in certain types of clinics — usually clinics that provide abortions. And some states have additional restrictions on how and where the drugs can be prescribed and taken.)
Putting abortion reversal to the test
Creinin's study, approved by the UC Davis institutional review board in December, has been registered with ClinicalTrials.gov, which tracks medical research.
The study is slated to involve 40 women who are between 44 and 63 days of pregnancy and are seeking to have a surgical abortion. As a condition of the research, the women would have to be willing to take mifepristone, the initial pill that would normally trigger a medical abortion, and then a placebo or progesterone.
Two weeks later, researchers will see if there's any difference in the rates of continued pregnancy. If progesterone can prevent the effects of mifepristone, Creinin says, he'll find that more women in the group that got progesterone are still pregnant, with a pregnancy that's progressing.
The key ethical point, the researchers say, is that all the women in this study want to have an abortion and will get one by the study's end. The study isn't enrolling women who are seeking a "reversal." They will be told upfront that if the mifepristone doesn't prompt an abortion, they will be offered a surgical abortion.
Creinin says the study participants will be compensated for their time in the study, but won't be paid for having an abortion. And patients will still be responsible for the cost of the surgical procedure — either through their insurance or out-of-pocket.
Creinin is skeptical that progesterone will have any effect, since it is thought that mifepristone irreversibly blocks progesterone in the body.
But if it does have a clinically significant effect, he says, "I want to know that."
Creinin hopes that his work will help medical researchers better understand if this kind of treatment can actually help women who change their minds after taking mifepristone for a medication abortion.
If the results show the progesterone doesn't work, Creinin hopes that it will discourage state legislators from mandating that doctors tell their patients about an ineffective treatment.
Creinin started enrolling patients in the study in February. He isn't sure how long the study will take, but says he probably won't have preliminary results for at least a year.
Dr. Mara Gordon is the NPR Health and Media Fellow from the Department of Family Medicine at Georgetown University School of Medicine.