Why the number of Emergency Room visits are a very misleading measure of the safety of chemical abortions

By Randall K. O’Bannon, Ph.D. NRL Director of Education & Research

With chemical abortions [“medication” abortions, in pro-abortion parlance] continually on the rise, you would expect that pro-abortion advocacy groups would go out of their way to assure the public that these abortions are very safe. And you would be correct.

To take just one example, the folks at UCSF (the University of California – San Francisco, sometimes called “America’s Abortion Academy”) cranked out a study claiming that abortion is very safe, based on there being relatively few cases of abortion-related visits to U.S. emergency rooms.

While that might sound initially impressive to sympathetic reporters or even medical professionals who don’t read carefully or know the issue, to those who understand how the abortion industry works and how the reality of abortion plays out in field, the claim is on much shakier ground.

Such a study might capture some, or even most of abortion-related complications if a woman presenting herself at the ER reveals she has had or is having an abortion. But what if she doesn’t, or she can’t, or the attending physician wants to keep it out of the medical records, then the study results are bogus.

And many of those promoting chemical abortions in particular have explicitly counseled women that they don’t have to reveal their abortions to the doctors treating them; they can just claim to be having a miscarriage.

Then, if the woman has serious problems, or even dies, the “miscarriage” gets blamed, while her abortion is counted as another ‘safe’ procedure.

Moreover, as we shall see, serious problems are simply classified otherwise.

Study “Results”

The study, “Abortion-related emergency department visits in the United States: An analysis of a national emergency department sample” was published in the online edition of BMC Medicine.

A UCSF team led by abortion researcher Ushma D. Upadhyay looked at samples from a national database of visits to emergency departments made by women from 2009 to 2013 (at the time, the five most recent available years), examining all records coded for abortion. Doing so, they found 6,342 with an identified abortion diagnosis which after eliminating duplicates or cases clinicians determined were unrelated to the abortion, resulted in 6,239 ER visits in their sample.

When projected on the population as a whole, researchers concluded that there would have been 27,941 emergency department visits somehow connected to abortion. They then whittle these down to 5,673 “major incidents,” discarding over 22,000 incidents they dismiss as “minor” or undeterminable or “no incident at all.”

“Major incidents” were those requiring an overnight inpatient stay, a blood transfusion, or surgery. “Minor incidents” included all others that involved an abortion -related diagnosis or treatment, although numbers show that this included some “repeat abortions” or “uterine reaspirations” that (according to authors’ definitions) did not count as “surgery.” Anything connected to a pre-existing condition (e.g., high blood pressure, diabetes, obesity) were categorized as “no incident.”

Notice from these criteria how many complications fail to get counted as major complications. A woman could come in with severe vomiting and diarrhea. She could have to have an IV to restore her fluids. But if she rallies and makes it home without being admitted for overnight observation, her visit isn’t counted.

If the baby doesn’t completely abort and some doctor performs a suction abortion on the woman, that doesn’t count. She can be in such significant pain that she needs high power opiate pain killers, but it isn’t a major complication if she goes home.

These women may have undergone tremendous ordeals, but according to the study, their ordeals were listed as “minor” or “no incident” at all.

The biggest way the study sweeps numbers of complications under the rug, however, was in its initial coding selection. Upadhyay and colleagues selected only those records, they said, with an ICD-9 diagnosis code for abortion.

But then in listing the ones they examined, they mentioned the codes for legal abortion (635), illegal abortion (636), unspecified abortion (637), and failed attempted abortion (638). Notably missing? The one for ectopic pregnancy (633), whose symptoms often mimic chemical abortion, and the one for miscarriage, or “spontaneous abortion” (634).

Given how promoters of chemical abortions have often advised women to tell doctors treating them for complications that they are simply having a miscarriage, this last is an irresponsible and potentially fatal omission for the study.

But, nevertheless, considering the 5,673 identified “major incidents” out of the 5,282,500 total abortions Guttmacher reports for the five years under examination, Upadhyay and her team come up with an incidence of rate of 0.11%, or 108 “major incidents” per 100,000 abortions.

The authors conclude from this data that “Perceptions that abortion is unsafe are not based on evidence,”

Upadhyay went further with the Los Angeles Times, saying that those regulating abortion in the name of safety should know better. “The people who are supporting these laws must know that abortion is safe,” Upadhyay told the Times. “They’ve seen the research – studies like this one – which only adds to the abundant scientific data that abortion is safe.”

Deficiencies in the Data

There are problems with the study, some major, which casts serious doubt on the authors’ conclusions.

The study is premised on the assumption that abortion-related complications will be reported as such. But there is no guarantee that this will happen.

Even when the outcome is as drastic as death (1) physicians may not note the abortion or pregnancy. Why? Past studies suggest it could have been either because they did not know, or felt there were reasons to avoid mentioning in the public death certificate (Journal of Contemporary Health Law & Policy, Spring 2004).

Busy ER doctors may consider the injury irrelevant, or may honor the request of the woman to keep it out of the record. They may even be concerned that this could be used against abortionist colleagues. A hemorrhage will be noted simply as a hemorrhage, an infection as just an infection, regardless of whether it was precipitated by an abortion.

And note that women who received treatment for their complications–at the clinic or by returning to the clinic rather than going to the ER–would not have been captured by UCSF study.

The abortion industry has a history of covering up its mistakes. A June 16, 2011, report by the Chicago Tribune found that mandatory reporting of abortions and complications to the state was often being ignored. More recently, in May of 2017, the director of Missouri’s Department of Health and Senior Services Randall Williams released a statement saying that he had found that abortionists had not been complying with a state law requiring them to report every abortion complication they diagnosed or treated to his department within 45 days.

Upadhyay and her colleagues admit that their method, using billing codes to discern the nature of ER visits could be “imprecise” and “incomplete.” However they want to blame the “stigma” surrounding abortion as a possible reason for error. “The estimates here may be conservative,” says the UCSF team, “if patients did not report having had an abortion due to fear of stigmatization or if relevant diagnosis and procedure codes were systematically misreported.”

While granting further that “this study may miss abortion-related incidents that were inaccurately coded as a miscarriage,” Upadhyay and UCSF team, long connected to the promotion of “medication abortion,” failed to note (as noted above) that prominent promoters of chemical abortion have advised women that they do not need to tell ER physicians they have taken abortion pills.

“Women on Web,” an international service that ships abortion pills all over the world and has garnered a great deal of publicity here in the U.S., explicitly tells women “You do not have to tell the medical staff that you tried to induce an abortion; you can tell them that you had a spontaneous miscarriage. Doctors have the obligation to help in all cases and know how to handle a miscarriage.”

They advise women that concealment is easy to accomplish. “The symptoms of a miscarriage and an abortion with pills are exactly the same and the doctor will not be able to see or test for any evidence of an abortion, as long as the pills have completely dissolved.”

A website by “woman family doctors” from New York City called “earlyabortion.com” tells women that they can “disguise” their abortion as a “bad menstrual period” so that others in the household will not know what they are going through. “In addition,” they note, “a medical professional will not be able to tell that you took the abortion pill. Even in an emergency room, a doctor can’t differentiate between a miscarriage and a medical abortion” (www.earlyabortion.com/abortion-pill-step-7-having-the-miscarriage-at-home, accessed 6/21/18).

Unwarranted Conclusions about Abortion Safety

We already know, from other studies by abortion advocates, that women are not always forthcoming about their abortion histories. Rachel Jones and Katryn Kost found that fewer than half of abortions to women in the U.S. from 1997 to 2001 were reported by women in face-to-face interviews in the 2002 National Survey of Family Growth (Studies in Family Planning).

If there is the stigma that Upadhyay and her team talk about, it would mean that a large percentage of women would not be inclined to reveal their abortions to ER doctors, especially if they felt they didn’t have to.

If just a fraction of the million or so women using mifepristone (one of the two drugs that make up “medication” abortions) during the study period saw the internet advice suggesting they conceal their abortions from medical personnel, and then followed it when they showed up hemorrhaging at the ER – which seems not only possible, but likely – the study’s conclusions of abortion’s safety would be significantly compromised.

(1) Upadhyay’s team did uncover 15 deaths in their study, but somehow missed more than a third of those reported to the U.S. Centers for Disease Control,