With anti-abortion groups energized by the election of President Donald Trump, and abortion supporters worried about the president's pick for the Supreme Court, abortion rights are again at the forefront of the national conversation.
But people who talk about abortion, whether on national news shows or at the dining-room table, are missing a critical piece of information: how many abortions were performed in the U.S. last year.
Incredibly, no one knows – not even the National Center for Health Statistics.
The dearth of current data is an anomaly in an age when information is collected and distributed with mind-boggling speed, and people on both sides of the abortion divide decry the lack of a comprehensive reporting system.
Most states collect data on abortions that occur within their borders, but the questions they ask providers and patients vary widely. A few states don't count abortions at all, so their numbers aren't included in the annual Abortion Surveillance Report issued by the Centers for Disease Control and Prevention.
The most reliable count is generally considered the one issued every three years by a nonprofit organization that was once an arm of Planned Parenthood. That now-independent agency, The Guttmacher Institute, says "more is known about abortion than perhaps any other surgical procedure" but acknowledges the need for "a more robust system for abortion reporting."
The contentious nature of the subject, however, contributes to the lack of progress on a solution. Just as there are wildly different opinions on whether abortions should be legal, there are wildly different opinions on how we should count them.
Should abortions be reported to the federal government as deaths, as one University of North Carolina-Charlotte researcher has argued? Or should potentially inflammatory language be avoided in favor of a legally required accounting that contains only the most basic of information, which may or may not include the age of the fetus?
It's a controversy within a controversy, which is why debate about one of the most divisive issues in the country lacks the most fundamental of information: exactly how many abortions there are, and why women have them. A solution may require financial incentives or a law compelling the collection of data, observers say.
If you ask the Centers for Disease Control and Prevention how many abortions are performed in America, you'll learn that 664,435 abortions were reported to the CDC in 2013.
The language is important; that's the number of abortions reported, not the number of abortions performed. Three states are not included: New Hampshire, Maryland and California. And three-year-old data is the most recent available, according to the CDC.
The Guttmacher Institute, however, has data for 2014 that includes all 50 states, bringing the total number of abortions to 926,200. That's because the institute, which has offices in New York City and Washington, D.C., compiles its own information by calling abortion providers directly instead of asking the states for it, according to Rebecca Wind, Guttmacher's associate director of communications.
The gap shows up not only in the total number of abortions, but in the percentage of pregnancies that end in abortion. Guttmacher's most recent figures show 19 percent in 2014; the CDC's, 18 percent in 2010. And beyond the discrepancy in calendar-year reporting, in its disclosure of study limitations, the CDC acknowledges its figures are consistently about 70 percent of what Guttmacher reports.
Both the CDC and the nonprofit, however, agree that abortion rates have fallen to historic lows. Charles "Chuck" Donovan, president of the anti-abortion Charlotte Lozier Institute in Washington, D.C., attributes the decline to a decreasing number of teenagers having sex, the effect of ultrasound on a pregnant woman's decision whether to have the baby, and what he calls "a significant trend toward pro-life convictions."
The CDC attributes the drop to fewer teen pregnancies, more insurance coverage for contraception, and increased use of long-term contraceptives such as intrauterine devices.
In its most recent polling on abortion, The Gallup Organization found that 47 percent of Americans identified as "pro-choice" and 46 percent as "pro-life." But support for abortion declines after the first trimester of fetal development, and a majority of people favor restrictions such as waiting periods, and spousal and parental consent.
Moreover, the number of people who said abortion is morally acceptable dropped from 45 percent in 2015 to 43 percent in 2016, according to Gallup.
The Guttmacher Institute says that 57 percent of American women aged 15 to 44 live in a state "hostile or extremely hostile" to abortion rights, and counts 288 abortion restrictions enacted in the U.S. between 2011 and 2015.
From its Census Bureau to the Center for Health Statistics, the U.S. government is a virtual geyser of data, spewing numbers, charts and graphs on everything from the number of emergency room visits made by seniors to the number of Americans diagnosed with kidney disease.
Its Abortion Surveillance Report, released every December since 1969, is a deep dive into the abortion data the CDC receives from its yearly request to the 50 states, Washington, D.C., and New York City. But the findings are incomplete since reporting to the CDC is voluntary, even if data collection is mandated by a state.
"Moreover, even in states that legally require medical providers to submit a report for all the abortions they perform, enforcement of this requirement varies and as a consequence several other reporting areas tend to provide CDC with incomplete numbers," the 2013 Abortion Surveillance Report says.
Guttmacher concurs, calling information the states provide to the CDC "incomplete and unreliable" and calling for the federal government to provide technical and financial assistance to improve reporting in states that already do it, and begin it in states that do not.
"Comprehensive, timely abortion reporting is an expensive undertaking, and states need resources to ensure the best systems are put in place," the institute said in a written statement given in response to questions posed by the Deseret News.
The CDC did not respond to questions about what pressure, if any, it puts on states that do not supply information for the reports, nor would it grant an interview with a researcher who has helped compile them for the past 10 years.
The health departments of New Hampshire, California and Maryland did not respond to inquiries about why they do not report to the CDC.
While those states are perennial holdouts, in some years, Delaware and Louisiana also failed to submit its numbers to the CDC. In its annual analysis of abortion trends, therefore, the CDC typically uses aggregate data from regions that regularly report, which typically range from 45 to 47 jurisdictions.
The Guttmacher Institute, which says its mission is to advance sexual and reproductive health and rights in the U.S. and globally, wants more accurate reporting because it says the incidence of abortion is an important public health indicator.
"On its own, the understanding of abortion levels, rates and trends is key to documenting the success of efforts to help women avoid unintended pregnancy, the precursor to most abortions. As a component of other statistics, abortion incidence data are essential to calculating levels and rates of pregnancy overall, teen pregnancy and unintended pregnancy," Joerg Dreweke wrote in a position paper for the institute.
The CDC, likewise, argues that accurate reporting matters, saying national abortion surveillance is needed "to guide and evaluate the success of programs aimed at preventing unintended pregnancies," help determine "the relative safety of abortion practices," and accurately estimate the number of pregnancies in the nation.
Donovan, the co-author of several papers critical of the nation's patchwork reporting system, said more timely reporting could help lower the rates of abortion if it reveals strategies that seem to make a difference.
"In Arizona a few years ago, a pro-life group conducted a campaign to promote pregnancy help centers and their services, and the state's relatively rapid reporting allowed them to analyze whether they had impacted the abortion rate," he said.
Abortion rates could be affected by a recession or an economic boom, a new abortion procedure or a new contraceptive. "We don't want to wait five years to find out if something has had a significant impact," Donovan said.
The CDC has helped to speed the dissemination of other health data, most recently in New Hampshire, where a newly developed app allows rapid sharing of causes of death on death certificates.
Stephen Wurtz, the state's registrar of vital records, told the Concord Monitor, "In almost real time, we can share this cause-of-death information with New Hampshire's health department or the CDC, so if they're doing surveillance, looking for any (outbreak), they can say, 'It looks like we've got a pocket of something'... and do more." The CDC paid about $200,000 to develop the app, the Monitor's article said.
Since abortion tracking lacks similar technology, Donovan said Guttmacher's numbers are "the best we've got, but a long way from ideal."
"Everybody needs to have access to data that is neutrally gathered. It can't just be from the clinics," he said, saying that state health departments should be involved.
"The CDC national data is not faulty or untrustworthy; it is just incomplete. You want a source that has little interest in broad public policy concerns about a phenomenon, one that just pays attention to integrity and scientific utility of the data," Donovan said.
But abortion rights supporters and abortion opponents don't agree on what that data should comprise. The Lozier Institute, the education and research arm of the anti-abortion nonprofit Susan B. Anthony List, applauds Oklahoma for the extensive list of information it requires, which includes method of payment for the abortion, number of previous pregnancies, whether the woman saw an ultrasound of the fetus, and the reason for the abortion.
The Guttmacher Institute says asking the reason for the abortion is wrong and unduly invasive, and cites Oklahoma's reporting as a "particularly egregious example of a state that has crossed the line into politicizing abortion reporting."
"Abortion is a constitutionally protected right, and requiring that women be queried about their reasons for obtaining an abortion comes dangerously close to making them justify why they are availing themselves of this right," Guttmacher says.
The institute also criticizes South Dakota and Texas for using the term "unborn child" in questions posed to women.
In 1998, Minnesota passed a law that required abortion providers to file a report on each procedure with the state health department. The health department, in turn, must submit a report on the previous year's abortions by July of the following year. The reports are available online, and Minnesota's data on 2015 is some of the most recent — and comprehensive — available within the U.S.
Minnesota reports on the woman's age, marital status and education, among other demographics. The Lozier Institute, which ranks states' abortion reporting on a scale of 0 to 100 points, gives it a 76. Only Oklahoma is higher. (Utah got a 65.)
Guttmacher lauded Minnesota for "comprehensive and timely abortion reporting" and said the state "could in many ways serve as a model for other states." But the institute cites "problematic" questions on the state's abortion reporting form, including questions about parental consent, the reason for the abortion, and the method of disposal of the fetal remains.
In 2015, the Minnesota Legislature added new questions: whether the abortion resulted in the birth of a live infant, and if so, what steps were taken to save the life of the child.
While those questions and the one about the reason for the abortion are unpopular with abortion-rights supporters, the report generally stirs little controversy, in part because it's been around for so long, said Carol Hajicek of the Minnesota Department of Health.
The state has been collecting the data with paper forms, but is switching to an electronic system this year, she said.
And Hajicek said Minnesota is careful to protect the privacy of both women having abortions and the doctors who perform them, going so far as to withhold potential identifiers in the report if there's a chance that someone could deduce who the patient was. (For example, if the woman obtaining an abortion was an Asian-American living in a small community that was predominantly Caucasion.)
Donovan, an abortion opponent, agrees that any improvement of the system must ensure that a woman's privacy is not compromised.
"The risk (of compromising privacy) goes up the more granular you get with any type of data. We have to be aware of that concern on the pro-life side," he said.
Noting that some states have full reports on the previous year's abortions by summer, Donovan said "there's really no reason" for national reporting to be behind by several years.
"Abortion data sounds like a lot, but compared to other things doctors have to capture, it's not that much," he said.
Both Indiana and Ohio are looking at strengthening their reporting, and the Lozier Institute hopes other states will join them. Donovan said the institute plans to put forth some ideas for reform, possibly later this year, but is proceeding carefully since previous attempts at change on the federal level have failed.
For its part, Guttmacher said it supports "good-faith efforts" to improve existing state and federal government data collection systems, but "would have concerns about any punitive measures that single out abortion reporting for unique penalties."
The next report from the CDC is expected in late November and will contain data from 2014.