Most observers of the abortion debate would likely agree that it is primarily a matter of ethics, a contest between the Hippocratic standard of respect for human life and various forms of humanistic pragmatism that allow abortion as a way to address personal and social problems. Resolving that debate has engaged philosophers of medicine, religious leaders, and the few political figures who seriously address it, for decades, with shifts in public policy and law back and forth. The debate also necessarily involves the matter of real-world consequences, making data collection and analysis a crucial part of the discussion, even if they are not ultimately the primary drivers of convictions on so profound a set of questions.

This fact has led to a range of skirmishes and debates all their own. One example is the decision of the Food and Drug Administration (FDA) during the Obama administration to drop the requirement that abortion pill (mifepristone) manufacturers report non-fatal events to the FDA involving the drug. Only deaths were, and still are under the Trump administration, subject to mandatory reporting.

As fresh controversy swirls around the FDA’s promise to review the safety of mifepristone, the status of future injury reporting remains in doubt, although it would be a welcome restoration of a policy with virtually no downside, especially as other FDA policy changes have increased risks to women who are taking the abortion drug regimen later in pregnancy, without a doctor’s visit, and no tests to rule out ectopic pregnancy and other serious health concerns.

November brought another challenge regarding abortion data. For many years, on the day after Thanksgiving, a media-sensitive U.S. Centers for Disease Control brought out its annual abortion surveillance report, a much-criticized and certainly incomplete account based on voluntary state reports. The reports failed to include states like California and New Jersey that refused to submit data despite being high-abortion-rate states, and the snail-like pace of the CDC typically meant the report arrived years after the reporting period closed, rendering them less useful for detecting and analyzing trends, including the rate at which surgical abortions were giving way to pills by prescription. The new Trump administration’s personnel cuts at various agencies were the prime reason the CDC report is not available. Sources say a new team will produce a fresh report in the spring of 2026, and one can only hope for a better and more thorough product, even in the face of stronger opposition to abortion reporting in some blue states.

But there is another level of concern about the dearth of data, and it is perhaps the most serious of all. These are the deliberate strategies of certain groups and publishers to suppress any study that identifies harms or potential harms from abortion. The first prong of these strategies is one that treats a researcher’s views on the ethics of abortion as inherently disqualifying — even a conflict of interest, even if the researcher states plainly, as a matter of course, their affiliations and sources of funding.[1] This is one of the nubs of pending litigation between scholars working with the Charlotte Lozier Institute (CLI) and Sage Publishing, owner of several prominent academic journals. The entire story is set forth at the CLI website “Assault on Science,” and it represents a deplorable abandonment of professional standards by the prominent publisher. While several studies were ultimately entangled in Sage’s retraction rampage, the primary object of its disaffection was a pathbreaking study of a massive Medicaid dataset that tracked the occurrence, via billing records, of various health complications for women having abortions in the one-third of states that finance them for the poor.

The lead author of the contested study, which was released in 2021, is James Studnicki, one of the most distinguished data and public health experts in the United States. Studnicki is currently vice president and director of data analytics at CLI, the latest phase of a career that has spanned five decades at leading institutions. Studnicki holds both Doctor of Science (Sc.D.) and Master of Public Health (M.P.H.) degrees from Johns Hopkins University and a Master of Business Administration (M.B.A.) degree from the George Washington University. He has contributed over 100 peer-reviewed articles to the health services research and public health systems and services research literature. His articles have appeared in the New England Journal of Medicine, the Journal of the American Medical Association, the American Journal of Preventive Medicine, and the American Journal of Public Health, among many others.

The clash with Sage, which is now in arbitration, is not isolated. Disputes over the authority and quality of peer-reviewed articles can be helpful if carried on without bias and in good faith. It is disturbing, however, when the disputes arise because of a partisan objection to a finding in an article or to the assumed policy views of a study’s authors, especially when this invalid objection is lodged against only parties with a particular view. This holds whether the topic under examination is climate change, gender transition, COVID vaccines, or abortion, each of them frequent subjects of controversy. Because institutional forces in American society often embrace a strict creed of alignment with their belief systems, these topics should spur even more care to avoid the kinds of injustice at issue here. If this level of mistreatment can reach someone of Studnicki’s caliber and integrity, no researcher is immune. It can quickly devolve into an across-the-board rejection of sound science as babies are thrown out, one by one, with the bathwater.

These background factors should deepen appreciation for studies that attempt to gauge the factors that contribute to pregnancy decisions and that influence pregnancy outcomes. These factors can be familial, social, creedal, economic, and medical. Multiple factors are typically in play. One area that has long presented itself as a challenge to quality research is the matter of sequence — to what extent does the order and frequency of a woman’s pregnancy experiences influence subsequent pregnancies?

Now Dr. Studnicki and a fresh team of scholars have combined to produce a first-of-its-kind study that attempts to answer at a statistical level whether a prior abortion or abortions is associated with subsequent natural pregnancy loss (miscarriage). It has just appeared in the journal JMIR Formative Research and has the mouthful title, “Patterns of Prior Induced Abortions and the Likelihood of Subsequent Natural Pregnancy Loss: Exploratory Application of Pregnancy Outcome Sequencing.” The study asked the same question regarding natural loss with respect to prior natural losses and natural loss with any other prior combination of pregnancy outcomes.

The size of this statistical challenge illustrates why nothing of this complexity has been previously attempted. Numerous studies have looked at discrete questions about induced abortion and impacts on subsequent pregnancies, but most such studies, Studnicki’s team noted, “are limited by the data available to a single pregnancy outcome as the focus of the analysis, such as the woman’s first pregnancy or another identified ‘index’ outcome.” To achieve a more accurate result, researchers needed data for a massive variety of POSs (pregnancy outcome sequences) and access to long-term data for each woman whose history would be examined, data that was hard to come by, especially so as the POS grew more complex and the cohort of women with each type of POS became smaller and statistically uninformative. Specifically, Studnicki and his team “identified a study population of 508 unique POSs, representing 5455 women, each of whom had 1 to 16 pregnancy outcomes, for a total of 14,198 pregnancies.” Suffice it to say, a Ph.D. in statistics would barely suffice to describe the mathematics involved in parsing each of these POSs to identify impacts on natural loss.

The whole study is worth a look as an example of how research on a very difficult subject can be done without resort to predetermined conclusions. The study’s strongest conclusion is that such an analysis can be done and a sequencing approach has value in identifying potential conclusions that were previously undetectable. In this instance, the conclusion is that prior abortion is associated with subsequent natural loss and that there is a “dose effect” wherein multiple abortions have a stronger association with such losses.

The authors also conclude that the risk of a subsequent natural loss is statistically significant after two natural losses and between the first and second natural loss. The authors supply these conclusions based on the data, which itself was obtained from the Center for Medicare and Medicaid Services Chronic Conditions Data Warehouse, and then spend the bulk of the remainder of their paper discussing the limitations of this new approach to the topic and why these limitations, as the authors elegantly frame it, “restrain assurance in the results.”

In many ways, here and elsewhere Studnicki and CLI have demonstrated what science, including statistical science, is all about. It involves ideas thoroughly explained and tested, modeled and applied, while at the same time recognizing limitations. Here those limits include population sizes for various sequences, the lack of data about intervals between pregnancies, patient characteristics, the economics and other demographics of the study population, and more besides.

The point is: good research moves the ball down the field and invites more debate. It does not presume outcomes and seek results to win a media fight. It provides more clarity, invites response and, most important of all, welcomes further debate. In short, it seeks the truth about things that truly matter.

[1] The author of this article was president of Charlotte Lozier Institute from 2011 to 2024; no current staff member of the Institute reviewed this article prior to publication in The Washington Stand.