Become a Patron


Lynn M. Paltrow and Jeanne Flavin

Lynn M. Paltrow and Jeanne Flavin, Arrests of and Forced Interventions on Pregnant Women in the United States, 1973-2005: Implications for Women's Legal Status and Public Health, 38 Journal of Health Politics, Policy & Law 299 (April, 2013)(118 Footnotes)

Lynn M PaltrowOn November 8, 2011, Mississippians voted down Proposition 26, a “personhood” measure that would have changed the state constitution by redefining the word person to include “every human being from the moment of fertilization, cloning, or the functional equivalent thereof” (Mississippi Secretary of State 2011a). The measure's defeat was attributed to the recognition that such a law could have an impact beyond recriminalizing abortion, including outlawing some forms of contraception as well as in vitro fertilization (Parents against Personhood 2012). In addition, it was argued that such measures would create legal grounds for forcing medical interventions on pregnant women and punishing those who, for instance, suffered miscarriages and stillbirths. Proponents of Proposition 26 dismissed the latter concerns in particular as “scare tactics” (Yes on 26 2011). The research findings reported here call this characterization into question.

Subsets of arrests and forced interventions on pregnant women who miscarried or were perceived as riskingJeanne Flavin harm to fertilized eggs, embryos, or fetuses have been identified and discussed in a variety of venues (Kolder, Gallagher, and Parsons 1987; Gallagher 1987; Paltrow et al. 1992; Gomez 1997; Ikemoto 1998; Nelson and Marshall 1998; Adams, Mahowald, and Gallagher 2003; Cherry 2007; Samuels et al. 2007; Fentiman 2006, 2009; Cantor 2012). For example, Paltrow et al.'s 1992 report collected information about 167 cases in which pregnant women who sought to go to term in spite of a drug problem were arrested. Since then, however, there has been no similar documentation, nor has there ever been a comprehensive collection or examination of cases involving the arrest and equivalent deprivations of pregnant women's liberty. As a result, there is a strong possibility that the number of such actions, and their shared legal and public health implications, has been underestimated. Lack of documentation also makes it difficult to evaluate what the likely implications of such things as personhood measures are and whether they pose threats beyond recriminalizing abortion.

A need remains, then, to document the cases, identify which women have been targeted, and determine the legal and public health implications of these arrests, detentions, and forced interventions. We report on more than four hundred such cases that have taken place in forty-four states, the District of Columbia, and federal jurisdictions from 1973 to 2005. We begin by describing the methods by which we identified cases for inclusion in this study and discuss the limitations of our research, leading to the conclusion that our findings represent a substantial undercount of cases. Next, we provide five illustrative cases from among the hundreds that were included in this study. We then report the findings of three separate analyses. First, we describe characteristics of the women and the cases, finding that low-income women and women of color, especially African American women, are overrepresented among those who have been arrested or subjected to equivalent deprivations of liberty. In this section we also describe the circumstances under which arrests, detentions, and forced medical interventions were made and identify leading criminal charges and other actions taken to deprive pregnant women of their liberty. Second, we investigate the legal claims made to justify the arrests, detentions, and forced interventions and their implications. Third, we explore the role played by health care professionals and discuss how arrests and other interventions were carried out in health care settings. We conclude by considering the implications of these cases for the legal status of pregnant women and for maternal, fetal, and child health.

* * *

Five Illustrative Cases

We briefly summarize five cases documented in this study that illustrate some of the varied circumstances in which pregnant women have been deprived of their liberty, the different legal mechanisms used to do that, and some of the consequences of those deprivations. These summaries also bring attention to constitutional issues apart from the right to liberty. For example, they raise questions about whether pregnant women who have been subject to arrests, detentions, and forced interventions have been deprived of their right to procedural due. process, including the right to effective assistance of counsel at critical stages of the proceedings against them.

Regina McKnight

In South Carolina, Regina McKnight, a twenty-one-year-old African American woman, unexpectedly suffered a stillbirth. Although it would later be shown that the stillbirth was the result of an infection, McKnight was arrested and charged with homicide by child abuse. The state alleged that McKnight caused the stillbirth as a result of her cocaine use. A jury found her guilty after fifteen minutes of deliberation. McKnight was sentenced to twelve years in prison. In 2008, as a result of postconviction relief proceedings, the South Carolina Supreme Court unanimously overturned her conviction, concluding that she had received ineffective assistance of counsel at her trial. The court described the research that the state had relied on as “outdated” and found that McKnight's trial counsel had failed to call experts who would have testified about “recent studies showing that cocaine is no more harmful to a fetus than nicotine use, poor nutrition, lack of prenatal care, or other conditions commonly associated with the urban To avoid being retried and possibly sentenced to an even longer term, McKnight pleaded guilty to manslaughter and was released from prison. She had already served eight years of her original sentence.

Laura Pemberton

Laura Pemberton, a white woman, was in active labor at her home in Florida. Doctors, aware of this, believed that she was posing a risk to the life of her unborn child by attempting to have a vaginal birth after having had a previous cesarean surgery (VBAC). The doctors sought a court order to force her to undergo another cesarean. A sheriff went to Pemberton's home, took her into custody, strapped her legs together, and forced her to go to a hospital, where an emergency hearing was under way to determine the state's interest in protecting the fetus still inside her. While lawyers argued on behalf of the fetus, Pemberton and her husband, who were not afforded the opportunity to be represented by counsel, “were allowed to express their views” as she was being prepared for surgery. The judge presiding over the case compelled Pemberton to undergo the operation, which she had refused and believed to be unnecessary. When she later sued for violation of her civil rights, a trial-level federal district court ruled that the state's interest in preserving the life of the fetus outweighed Pemberton's rights under the First, Fourth, and Fourteenth Amendments. Pemberton subsequently gave birth vaginally to three more children, calling into question the medical predictions of harm from a VBAC on which the court had relied.

Rachael Lowe

Rachael Lowe, a twenty-year-old pregnant woman, voluntarily went to Waukesha Memorial Hospital in Wisconsin to seek help for her addiction to the opiate Oxycontin. Some hospital staff responded by reporting Lowe to state authorities under Wisconsin's “cocaine mom” law, a statute in the Children's Code that allows the state to take a pregnant woman into custody if it believes that the “expectant mother habitually lacks self-control in the use of alcohol beverages, controlled substances or controlled substance As a result, Lowe was forcibly taken to St. Luke's Hospital in Racine, more than an hour away from where she lived with her husband and two-year-old son. At St. Luke's she was held against her will in the psychiatric ward. While there, she received no prenatal care and was prescribed numerous medications, including Xanax. Although a guardian ad litem was appointed for the fetus, Lowe was not appointed counsel until after the first court hearing in her case, approximately twelve days after being taken into custody. At that hearing, no state official could give the court any information about the health of the fetus or the treatment Lowe was receiving. When a subsequent hearing was held to determine the legality of her incarceration, a doctor testified that Lowe's addiction posed no significant risk to the health of the fetus. At the end of the hearing, the court announced that Lowe would be released from her hospital-based incarceration. Nevertheless, she remained at the hospital in state custody for several days, and under state surveillance and supervision for the remainder of her pregnancy. Lowe was required to provide urine samples and to cooperate with law enforcement and health professionals. As a result of the intervention, Lowe's husband had to take a leave of absence from his job, and Lowe was fired from hers.

Martina Greywind

Martina Greywind, a twenty-eight-year-old homeless Native American woman from Fargo, North Dakota, was arrested when she was approximately twelve weeks pregnant. She was charged with reckless endangerment, based on the claim that by inhaling paint fumes she was creating a substantial risk of serious bodily injury or death to her unborn child. After spending approximately two weeks in the Cass County Jail, Greywind was able to obtain release for a medical appointment. At that appointment Greywind obtained an abortion, despite widely publicized efforts by abortion opponents to persuade her to carry the pregnancy to term. Following the abortion, Greywind filed a motion to dismiss the charges. The state agreed to a dismissal: “Defendant has made it known to the State that she has terminated her pregnancy. Consequently, the controversial legal issues presented are no longer ripe for According to news reports, the prosecutor in the case stated that since Greywind had had an abortion, it was “no longer worth the time or expense to prosecute her” (Orlando Sentinel

Michelle Marie Greenup

In Louisiana, Michelle Marie Greenup, a twenty-six-year-old African American woman, went to a hospital complaining of bleeding and stomach pain. Doctors suspected that she had recently given birth and contacted law enforcement authorities. After repeated police interrogations, Greenup ““confessed” that the baby was born alive, and it died because she had failed to provide it with proper care. Greenup was charged with second-degree murder and was incarcerated. Eventually counsel for Greenup obtained her medical records, which revealed that the fetus could not have been older than between eleven to fifteen weeks and that prior to the miscarriage Greenup had been given Depo-Provera, a contraceptive injection that may cause a miscarriage if administered to a woman who is already pregnant. Greenup was finally released, but only after she agreed to plead guilty to a misdemeanor violation of a public health law that regulates disposal of human remains. There is no indication that the human remains law was intended to apply to pregnant women confronted with a miscarriage.

These five case examples represent only a fraction of the state actions taken against women in the United States, but they provide an important sense of the consequences to the women, including incarceration, forced surgery, coerced abortion, and civil commitment, apparently without regard to the health care that would actually be provided.

Demographic and Case Characteristics

In this section we discuss key quantitative findings on geographic distribution of cases, women's age, stage of pregnancy, mental health status, socioeconomic status, and race (see table 1). We also briefly discuss our findings on men and domestic violence in the women's lives.

We identified state actions taken against 413 women in forty-four states, the District of Columbia, and some federal jurisdictions between 1973 and 2005 (see figure 1). The largest percentage of cases originated in the South (56 percent), followed by the Midwest (22 percent), the Pacific and West (15 percent), and the Northeast (7 The cases took place in every state except Delaware, Maine, Minnesota, Rhode Island, Vermont, and West Virginia. Ten states had ten or more cases. Those ten states also accounted for more than two-thirds of the total number of cases. South Carolina had the largest number of cases (n = 93), followed by Florida (n = 56), Missouri (n = 29), Georgia (n = 16), Tennessee (n = 15), Wisconsin (n = 15), Illinois (n = 14), Nevada (n = 11), New York (n = 11), and Texas (n = 10).

Figure 1 Number of Arrests, Detentions, and Forced Interventions of Pregnant Women in the United States (1973-2005) (Omitted)

In individual states, cases tend to cluster in particular counties and sometimes in particular hospitals. For example, in South Carolina thirty-four of the ninety-three cases came from the contiguous counties of Charleston and Berkeley. Staff at one hospital, the Medical University of South Carolina, initiated thirty of these cases. In Florida twenty-five of the fifty-five cases took place in Escambia County. Of these, twentythree came from just two hospitals: Sacred Heart Hospital and Baptist Hospital. In Missouri twenty-six of the twenty-nine cases came from Jackson County. Of these, twenty cases came from a single hospital: Truman Medical Center.

Overwhelmingly, and regardless of race, women in our study were economically disadvantaged, indicated by the fact that 71 percent qualified for indigent defense. Of the 368 women for whom information on race was available, 59 percent were women of color, including African Americans, Hispanic American/Latinas, Native Americans, and Asian/Pacific Islanders; 52 percent were African American. African American women in particular are overrepresented in our study, but this is especially true in the South (see table 2). Nearly three-fourths of cases brought against African Americans originated in the South, compared with only half of the cases involving white women. Racial disparities are even more pronounced in particular states. Between 1973 and 2005 African Americans in Florida made up approximately 15 percent of the state's population and whites composed 81 percent. Yet approximately three-fourths of Florida's cases were brought against African American women, while only 22 percent were brought against white women. In South Carolina, African Americans made up 30 percent of the state's population, and 68 percent of the population base was white. Yet 74 percent of the cases in the state were brought against African American women and only 25 percent against white women.

We were able to determine the age of the women at the time of their arrest, detention, or forced intervention in approximately 91 percent of the cases. Women in the study range from twelve to forty-three years of age; the average (and median) age was approximately twenty-eight years. We identified two cases involving minors.

One out of five women was still pregnant at the time legal action was taken. In some cases action was taken against a woman early in her pregnancy, when the fetus would not have been viable. In twenty-five cases we found explicit references to a mental health diagnosis, a history of mental health problems, or treatment for mental health problems. Although every pregnancy in this study involved a man, the father or the woman's male partner was mentioned in only 23 percent of cases. Information available in approximately one in ten cases (n = 36) mentioned violence against the pregnant woman.

Circumstances of Arrests and Other State Actions

In this section we describe the circumstances in which the arrests and other state actions took place. These circumstances often defy simple categorization. Research into cases that were widely reported in the news media as involving a pregnant woman and her use of an illegal drug or alcohol often revealed that other actions, inactions, or circumstances, in addition to pregnancy, were the primary reason for the state action. These include a pregnant woman who had been in a location while pregnant that exposed her unborn child to dangerous “fumes that permeate in the and another case in which the woman did not follow her doctor's medical advice to rest during her pregnancy and did not get to the hospital quickly enough on the day of delivery.

In several cases a woman's efforts to seek help after having been physically abused resulted in her arrest, although factors such as drinking alco hol or using an illegal drug while pregnant were cited as grounds for those arrests. In South Carolina a twenty-three-year-old African American woman was charged with homicide by child abuse after she experienced a stillbirth. The charging documents, including the arrest warrant and incident report, alleged that her use of drugs and alcohol caused the stillbirth. Further research into her case revealed that she had used a small amount of powder cocaine, consumed alcohol, and taken eight Tylenol in an effort to commit suicide on her twenty-third birthday.

Another case provides a particularly good example of one that defies simple categorization and characterization. Deborah Zimmerman, a thirty-four-year-old white woman from Franksville, Wisconsin, had been drinking alcohol and was allegedly intoxicated when she was brought to St. Luke's Hospital two days before she was scheduled to deliver her baby. Declining a “biophysical profile” at a prenatal care appointment a week earlier, as well as drinking alcohol and smoking cigarettes while pregnant, all legal activities, were mentioned in the criminal complaint describing the grounds for her arrest on charges of attempted first-degree intentional homicide and first-degree reckless.injury. The case received widespread national attention, focusing on Zimmerman's alcohol use and the claim that she wanted to “kill” her unborn child through her use of alcohol. A review of the case reveals something unreported in the media: medical staff decided to contact the police and characterize her as a criminal only after she refused to consent to fetal monitoring and cesarean surgery.

According to the criminal complaint, “Once at St. Luke's Hospital, Deborah Zimmerman was combative and refused monitoring and Although Zimmerman “kept talking about a gentleman and how he was abusing her,” neither the nurses nor the doctors apparently saw this information as bearing on why Zimmerman might object to being touched by the strangers who made up the medical staff (Terry Eventually, however, staff performed an ultrasound on Zimmerman. Based on their interpretation of the results, medical staff believed that cesarean sur gery was necessary because of “fetal intolerance to labor and suspected intra-uterine growth retardation.”According to testimony from a surgical technician in the labor and delivery unit, Zimmerman refused to consent to the surgery:

Q. What did you and the hospital personnel do as a result of her refusal to consent to the C-section?

A. Well, I was assisting the RN ... and as I recall when we said we, we told her she needed a C-section, she said no one is doing this f-ing thing to me and I don't want to be here. Like I said, she did threaten to leave quite a bit, got up out of bed a few times. We then realized we had to do something, so we consulted the physician again and our nurse supervisor, who then decided to call in the police after [Zimmerman] had made a statement to me.
Q. What sort of statement did she make?
A. ... I was in the room alone with her, trying to talk to her, explained to her the situation, that she needed a lot of help here, that she had to cooperate, it wasn't just for her health, it was for the baby's health, and she had said if--at this time there was talk about that she might not be staying and, I recall her saying to me, if you don't keep me here, I'm going to go home and keep drinking and drink myself to death and I'm going to kill this thing because I don't want it anyways.
The first half of this hearsay statement has been interpreted by some as an explicit suicide threat made in the presence of doctors and nurses--one that generated no psychiatric consultation, evaluation, or treatment (Armstrong 2003: 2). The second half of the statement became the excuse for the arrest and the subject of national news. The fact that her refusal of cesarean surgery prompted the idea to call the police did not make the news at all.

The difficulty of categorization notwithstanding, we found that the majority of cases identified in this study focused on women who became pregnant, sought to continue to term, and were believed to have used one or more illegal drugs, with cocaine most often identified as one of them. Eighty-four percent (n = 348) of cases involved an allegation that the woman, in addition to continuing her pregnancy, had used an illegal drug. Two hundred and eighty-two cases identified cocaine as one of the drugs being used, 51 identified methamphetamine or amphetamines, 23 mentioned heroin or another opiate, and 43 identified marijuana. In 6 cases marijuana was the only illegal drug mentioned.

More than half the 348 cases (n = 177) in which a woman was identified as having used an illegal drug also specifically referred to other factors, in addition to the pregnancy, as part of the rationale or circumstances justifying the arrest or detention. Regardless of whether there was a drug-related allegation, refusal to follow treatment orders was identified as part of the justification for the arrest, detention, or forced medical intervention in nearly one in five cases. In 41 cases alcohol was mentioned. Lack of prenatal care was identified as a factor in 68 cases. The fact that the woman smoked cigarettes was mentioned in 12 cases.

Other factors explicitly described in arrest warrants and other legal documents justifying state intervention in cases that also involved an allegation of drug use included the fact that the pregnant woman had a sexually transmitted infection, was HIV positive, or gave birth at home or in another setting outside a hospital. In one case the state indicated that it would use the fact that the woman had refused offers of voluntary sterilization in support of its prosecution. In numerous cases the fact that a pregnant woman had other children, some of whom were identified as having been exposed to alcohol or another drug, was referenced as part of law enforcement officials' explanation for the arrest (Rizzo 2002; Murphy

Sixteen percent of the cases (n = 65) involved no allegation that the woman had used an illegal, criminalized drug. These include cases in which women were deprived of their liberty based on claims that they had not obtained prenatal care, had mental illness, or had gestational diabetes, or because they had suffered a pregnancy loss. In fifteen of these cases alcohol was the only drug mentioned. Thirty of these cases involved efforts to force women to submit without consent to medical interventions. These forced interventions included pregnant women who had diabetes or sought to have a vaginal birth and refused to undergo cesarean surgery or other surgical intervention, those who refused to submit to a blood transfusion, and one who refused to allow a public health nurse who had been appointed as a guardian ad litem for the fetus to monitor the pregnancy, “check on the welfare of the unborn and provide any medical services that the nurse deemed necessary (Sealey

In eight cases pregnant women were alleged to have self-induced an abortion that the state claimed violated the state's abortion laws. In two cases state action was used to detain women who expressed an intention to have an abortion, and in one of those the woman's incarceration prevented her from having an abortion.

Although deprivations of women's liberty are often justified as mechanisms for protecting children from harm, we found that in a majority of cases the arrest or other action taken was not dependent on evidence of actual harm to the fetus or newborn. As noted earlier, in two out of three cases no adverse pregnancy outcome was reported. In many cases criminal charges rested on the claim that there was a risk of harm or a positive drug test but no actual evidence of harm. Similarly, in numerous cases where court orders were sought to force medical interventions, a risk of harm was identified that did not materialize.

In cases where a harm was alleged (e.g., a stillbirth), we found numerous instances in which cases proceeded without any evidence, much less scientific evidence, establishing a causal link between the harm and the pregnant woman's alleged action or inaction. In other cases we found that courts failed to act as judicial gatekeepers to ensure, as they are required to do, that medical and scientific claims are in fact supported by expert testimony based on valid and reliable scientific evidence (Neufeld 2005; Paltrow and Jack 2010).

The lack of scientific evidence was especially clear in the Geralyn Susan Grubbs case. Grubbs, a twenty-three-year-old white woman, gave birth to a son in Alaska. Two weeks after birth, the baby died unexpectedly. The state asserted that Grubbs's use of cocaine while pregnant caused her son's death and charged her with manslaughter as well as two drug-related offenses. Facing a potential thirty-year sentence, Grubbs accepted a plea bargain to the lesser charge of criminally negligent homicide. Grubbs's conviction and sentence remained in full force even though, in response to a separate civil suit, the state admitted that it had since discovered that the autopsy, which had formed the basis of Grubbs's conviction, was erroneous and that cocaine was not the cause of the infant's death.

In re Unborn Child of Starks provides a clear example of a judicial proceeding in which witnesses were allowed to express opinions about medical and scientific facts even though they were not qualified to do so. Julie Starks, a twenty-five-year-old white pregnant woman in Oklahoma, was arrested in a trailer that was allegedly being used, or that had once been used, to manufacture methamphetamine. In addition to arresting Starks and charging her with manufacturing methamphetamine, the state began proceedings in the Rogers County juvenile court to declare her unborn child “deprived” (in danger due to parental neglect, abuse, cruelty, or depravity). The juvenile court took emergency custody of Starks's fetus and also raised her bond from the $25,000 set by the criminal court to $200,000, with the added condition that if Starks posted bond she would be placed in a foster home until she gave birth. While incarcerated in the county jail, Starks experienced dehydration and premature labor, developed urinary tract infections and sinus problems, and lost twelve pounds. She spent more than a month in jail before the Oklahoma Supreme Court ruled that the juvenile court judge's order raising Starks's bond to $200,000 was “an unauthorized application of judicial

The lower court, however, continued its emergency order, giving custody of Starks's fetus to the Oklahoma Department of Human Services. A jury trial in the juvenile court went forward to determine if the fetus was “deprived” under the state's Children's Code. The state alleged that Starks had placed “the unborn child at risk of injury, serious bodily injury, with defects or death.”Because there was no evidence to support the state's claim that Starks was using any illegal drugs while pregnant, the case focused on the argument that while pregnant, she had “inhaled” dangerous chemicals allegedly used in the manufacture of methamphetamine. The state was allowed to rely on testimony from local law enforcement officials to support this claim. For example, a police sergeant agreed with the prosecutor that he did not “need a medical degree” to testify that a pregnant woman should not have been in the environment in which they found her. The prosecutor argued that it did “not take a rocket scientist, so to speak, to figure out that these kinds of chemicals would be harmful to not only the mother but the unborn child,” and was allowed to make this claim without any scientific experts at all. The jury reached a verdict, later overturned, that the fetus, while still inside Starks, had been “deprived.”

Criminal Charges and Other Efforts to Deprive Pregnant Women of Their Liberty

Overwhelmingly, the deprivations of liberty described here occurred in spite of a lack of legislative authority, in defiance of numerous and significant appellate court decisions dismissing or overturning such actions, and contrary to the extraordinary consensus by public health organizations, medical groups, and experts that such actions undermine rather than further maternal, fetal, and child health (American College of Obstetricians and Gynecologists 1987, 2005, 2011; National Perinatal Association 2011; American Psychiatric Association 2001; American Nurses Association 1991; American Academy of Pediatrics 1990; Cole 1990; March of Dimes 1990; National Council on Alcoholism and Drug Dependence 1990). The American Medical Association, the American Academy of Pediatrics, and the American College of Obstetricians and Gynecologists, for example, have concluded that threats of arrest and punishment deter women from care and from speaking openly with their doctors (Cole 1990; American Academy of Pediatrics Committee on Substance Abuse 1990). The American Medical Association statement also notes that such threats could pressure some women to have unwanted abortions rather than risk being subject to criminal penalties.

Due in part, no doubt, to the strong public health opposition to such measures, no state legislature has ever passed a law making it a crime for a woman to go to term in spite of a drug problem, nor has any state passed a law that would make women liable for the outcome of their pregnancies (Paltrow, Cohen, and Carey 2000; National Abandoned Infants Assistance Resource Center 2008; Guttmacher Institute 2012a). Similarly, no state legislature has amended its criminal laws to make its child abuse laws applicable to pregnant women in relationship to the eggs, embryos, or fetuses that women carry, nurture, and sustain. No state has rewritten its drug delivery or distribution laws to apply to the transfer of drugs through the umbilical cord. To date no state has adopted a personhood measure, and no law exists at the state or federal level that generally exempts pregnant women from the full protection afforded by federal and state constitutions. In 1997, as a result of a judicial ruling (not legislation), South Carolina became the only state during the time period covered by our study (1973-2005) to authorize the prosecution of pregnant women.

Nevertheless, our study documents hundreds of arrests or equivalent deprivations of liberty, with the majority relying on interpretations and applications of criminal laws that were never intended to be used to punish women in relationship to their own pregnancies. In 86 percent of the cases (n = 354), the efforts to deprive pregnant women of their liberty occurred through the use of existing criminal statutes intended for other purposes (see table 1). In those cases the charges most frequently filed were child abuse or child endangerment (n = 204).

Sixty-eight cases involved women who experienced miscarriage, stillbirth, or infant death. In all but six cases, prosecutors attributed the loss entirely to actions or inactions that occurred during the woman's pregnancy. In forty-eight of those cases, women were charged under variations of the state's homicide laws, including such crimes as feticide, manslaughter, reckless homicide, homicide by child abuse, and firstdegree murder. In four cases in which a woman's actions were described as inducing a self-abortion, she was also charged under murder or manslaughter statutes.

Some of those statutes did not require any intent to end the pregnancy. For example, Regina McKnight, the African American woman from South Carolina discussed above, was convicted of homicide by child abuse even though all parties in the action, including the state, agreed that she had no intention of ending the pregnancy.

The vast majority of women (n = 295) were charged with felonies, which are offenses punishable by more than one year of incarceration. African American women were significantly more likely than white women to be charged with felonies (see table 2). Eighty-five percent of African American women were charged with felonies, compared with 71 percent of white women.

Identifying the Underlying Legal Theory

As discussed above, appellate courts have overwhelmingly rejected efforts to use existing criminal and civil laws intended for other purposes (e.g., to protect children) as the basis for arresting, detaining, or forcing interventions on pregnant women (Fentiman 2006). Given the lack of specific legislative authority, we sought to determine what legal theory was offered. In virtually every case in which we could identify the underlying legal theory, we found it to be the same as that asserted by proponents of personhood measures: namely, that the fertilized egg, embryo, or fetus should be treated as if it were completely legally separate from the pregnant woman herself. Prosecutors, judges, and hospital counsel argued that the legal authority for their actions came directly or indirectly from feticide statutes that treat the unborn as legally separate from pregnant women, state abortion laws that include language similar to personhood measures, and Roe v. Wade, misrepresented as holding that fetuses, after viability, may be treated as separate persons.

Today, thirty-eight states and the federal government have passed feticide or unborn victims of violence acts or amended their murder statutes to include the unborn (National Conference of State Legislators Such laws make it a crime to cause harm to a “child in utero” and recognize everything from a zygote to a fetus as an independent “victim,” with legal rights distinct from the woman who has been harmed. These laws are generally passed in the wake of a violent attack on a pregnant woman and, as in Texas, are described as creating “a wall of protection for pregnant women and their unborn children” (Hupp 2003; emphasis added). These laws, however, have also been used to provide the purported authority for arresting pregnant women themselves.

As cases documented in this study demonstrate, women in California, Georgia, Tennessee, South Carolina, and Utah who suffered stillbirths or delivered babies who died shortly after birth have been charged directly under state feticide laws. In Utah a feticide law was used as the basis for arresting and charging Melissa Rowland. Rowland gave birth to twins, one of whom was stillborn. Rowland was arrested on charges of criminal homicide, a first-degree felony, based on the claim that she had caused the stillbirth by refusing to have cesarean surgery two weeks earlier. A spokesman for the Salt Lake County district attorney's office explained the homicide charge this way: “The decision came down to whether the dead child--a viable, if unborn, being as defined by Utah law--died as a result of another person's action or failure to take action. That judgment ... is required by Utah's feticide law, which was amended in 2002 to protect the fetus from the moment of conception” (Johnson 2004).

Even when women are not charged directly under feticide laws, such laws are used to support the argument that generally worded murder statutes, child endangerment laws, drug delivery laws, and other laws should be interpreted to permit the arrest and prosecution of pregnant women in relationship to the embryos or fetuses they carry.

Texas's feticide law (SB 319), enacted as the Prenatal Protection Act, was used in precisely this way. As the Austin Chronicle reported, “The bill passed, was signed into law by Gov. Rick Perry, and took effect on Sept. 1, 2003. A mere three weeks later, 47th District Attorney Rebecca King (prosecuting in Potter and Armstrong counties) penned a letter to ‘All Physicians Practicing in Potter County'--Amarillo--informing them that under SB 319 ‘it is now a legal requirement for anyone to report a pregnant woman who is using or has used illegal narcotics during her pregnancy”’ (Smith 2004).

Rather than refuse this demand from the district attorney, health care providers complied. As a result, more than fifty Potter County women were reported, charged with crimes, and in many cases incarcerated (Thomas 2006). Some of these arrests were challenged. In 2006, a Texas Court of Appeals finally held that the Prenatal Protection Act did not authorize the arrests. In spite of this decision, however, some of the women were incarcerated for years while their cases worked their way through the court system.

Antiabortion statutes that include statements of separate rights for the unborn, similar to those asserted by personhood measures, are also routinely used to justify arrests, detentions, and forced surgeries on women who had no intention of ending a pregnancy. For example, the 1986 Missouri Abortion Act includes a preamble stating that life begins at conception and that “the laws of this state shall be interpreted and construed to acknowledge on behalf of the unborn child at every stage of development, all the rights, privileges, and immunities available to other persons, citizens, and residents of this Although the statute contains an explicit provision protecting pregnant women from punishment, Missouri prosecutors have used the law to justify the arrests of scores of pregnant women, including one who admitted to using marijuana once while she was pregnant and another who drank alcohol. An Illinois abortion law stating that “an unborn child is a human being from the time of conception and is, therefore, a legal person for the purposes of the unborn child's right to life” was cited as authority for forcibly restraining, overpowering, and sedating a pregnant woman in order to carry out a blood transfusion she had refused.

In Roe v. Wade, the US Supreme Court explicitly rejected the claim that fetuses, even after attaining viability, are separate legal persons with rights independent of the pregnant women who carry, nurture, and sustain them. Still, consistent with the goals of personhood measures, prosecutors, hospital attorneys, and judges frequently misrepresent the decision to stand for the opposite meaning (Gallagher 1987). They claim that Roe instead establishes that viable fetuses must be treated as legal persons fully separate from the pregnant woman. This misstatement of Roe's actual holding has been used in numerous cases as authority for depriving pregnant women of their liberty.

A Massachusetts trial-level court relied on this distortion of Roe when it ordered Rebecca Corneau, a thirty-two-year-old white woman, imprisoned so the state could force her to undergo medical examinations over her religious objections. In Pennsylvania a hospital sought a court order to force Amber Marlowe, a twenty-five-year-old white woman, to undergo cesarean surgery. Counsel for the hospital cited Roe for the proposition that “Baby Doe, a full term viable fetus, has certain rights, including the right to have decisions made for it, independent of its parents, regarding its health and The court granted the order, awarding the hospital custody of a fetus before, during, and after delivery and giving the hospital the right to force Marlowe to undergo cesarean surgery without her consent. In Florida Roe was misused as authority for taking Pemberton, the Florida woman discussed above who attempted a VBAC, into police custody and forcing her to undergo cesarean surgery. As a trial-level federal court asserted, “Whatever the scope of Pemberton's personal constitutional rights in this situation, they clearly did not outweigh the interests of the State of Florida in preserving the life of the unborn child .... This is confirmed by Roe v.

In other words, where prosecutors, judges, and other state actors have articulated legal arguments for depriving pregnant women of their liberty, they are the same as those made in support of personhood measures; both rely on the idea that state actors should be empowered to treat fertilized eggs, embryos, and fetuses as completely, legally separate from the pregnant women.

Interventions in Health Care Settings and the Role of Medical Professionals

In this section we discuss findings indicating that some medical and public health professionals have worked with law enforcement and other state officials to deprive pregnant women of their liberty. Although it is often presumed that medical information is confidential and rigorously protected by constitutional and statutory privacy protections as well as principles of medical ethics, cases we have identified challenge that assumption. Similarly, the results of those disclosures, including bedside interrogations by police and other state authorities, likely contradict most medical patients' expectations of privacy and humane treatment.

We note that state and federal law is extremely variable in terms of when and whether health care providers may be required to report information to civil child welfare authorities that would reveal evidence of a pregnant woman's drug or alcohol use or abuse (Paltrow, Cohen, and Carey 2000; Ondersma, Malcoe, and Simpson 2001). These laws also sometimes fail to define what must be reported (i.e., the term “drug-affected” newborn in the federal law addressing this issue is not defined) (Weber 2007). Mandated reporting and civil child welfare responses deserve more attention than can be provided here. Instead, we focus on our findings indicating a wide variety of disclosures, some of which are clearly prohibited by law and all of which challenge the idea that medical and public health approaches are distinct from law enforcement approaches addressing drug use and maternal, fetal, and child health issues (Gomez 1997).

In two-thirds of the cases (n = 276), we were able to identify the mechanism by which the case came to the attention of police, prosecutors, and courts. In 112 cases, the disclosure of information that led to the arrest, detention, or forced intervention was made by health care, drug treatment, or social work professionals, including doctors, nurses, midwives, hospital social workers, hospital administrators, and drug treatment counselors (Dube 1998). In at least 47 cases, health care and hospital-based social work professionals disclosed confidential information about pregnant women to child welfare or social service authorities, who in turn reported the case to the police.

Hospital-based health care providers and social workers appear more likely to disclose information about patients of color (see table 2). In 240 cases, both race and reporting mechanism were known. Nearly half (48 percent) of African American women were reported to the police by health care providers, compared to less than one-third (27 percent) of white women. White women, by contrast, were far more likely (45 percent) to have their cases come to the attention of the police through other mechanisms, such as reports by a probation or parole officer, an arrest unrelated to pregnancy, or a report from a boyfriend or family member.

Far from being a bulwark against outside intrusion and protecting patient privacy and confidentiality, we find that health care and other “helping” professionals are sometimes the people gathering information from pregnant women and new mothers and disclosing it to police, prosecutors, and court officials. In some cases hospital medical staff have specifically collaborated with police and prosecutors to develop a coordinated system of searching pregnant women for evidence of illegal drug use, reporting women who test positive to the police, and helping the police carry out arrests of the hospitalized women. In Ferguson v. City of Charleston, the US Supreme Court held that such collaboration violated a patient's Fourth Amendment constitutional rights to privacy. Ferguson also held that medical staff who collect and disclose patient information in order to advance law enforcement purposes may be held liable for damages. Nevertheless, as our earlier discussion of cases from Amarillo, Texas, demonstrates, collection of patient information for law enforcement purposes has occurred since Ferguson.

Our research also revealed that in some cases making a report to child welfare authorities was no different than making a report directly to law enforcement officials. For example, as part of a long-standing partner ship among social workers, local police, and the Maryland state attorney's office, medical personnel at Easton Memorial Hospital reported positive drug test results of new mothers or their newborns to the Talbot County Department of Social Services, which in turn, and by agreement, passed that information on to the police. In Tennessee, Anita Gail Watkins, a forty-three-year-old African American woman, was reported to the Department of Human Services (DHS) after she confided in her doctor that she had used cocaine before the birth of her son. A doctor at the hospital explained that “our goal from a medical standpoint is the best outcome for the infant. When there is evidence of drug use, we notify DHS. Where the trail goes from there is not up to us.”The disclosure to DHS led to a Clarksville Police Department detective, who arrested Watkins and charged her with the crime of reckless endangerment (Crosby

Disclosures of patient information to law enforcement authorities, whether directly from health care providers or conveyed through child welfare agencies, have resulted in bedside interrogations that are reminiscent of the days before Roe when women suspected of having illegal abortions were subjected to humiliating police questioning about intimate details of their lives while lying, and sometimes dying, in their hospital beds (Reagan 1998). For example, Sally Hughes DeJesus, a twenty-eight-year-old white woman from North Carolina, experienced a relapse and used cocaine after eleven months of abstinence. She told her midwife what had happened, reporting that “I told her I needed help .... I was afraid for my baby” (Beiser 2000). According to a news story, the midwife told the hospital where DeJesus was having the baby about her drug use. When the doctors there performed a drug test on the healthy newborn and found that it had been exposed prenatally to cocaine, they called the police. Following this report, “As DeJesus lay recuperating in her hospital room in Henderson County, North Carolina, sheriffs marched in to interrogate her” (ibid.). She was then charged with felonious child abuse. Cases in this study reveal that women who had recently given birth, suffered a stillbirth, or were believed to have self-induced an abortion were subjected to bedside interrogations. Women have been interrogated while still experiencing the effects of sedatives given during cesarean surgery. In one case, police were called so quickly that they were present when the woman was informed she had lost the pregnancy. The detective who interrogated the bereaved woman in that case asked, among other things, “Did you do everything in your power to ensure that you'd have a healthy

In many cases, hospital staff disclosed information to police and prosecutors despite principles of patient confidentiality and apparently without any court order or other legal authority requiring them to do so. Such disclosures were clear in the Melissa Rowland case discussed above. The probable cause statement (describing the grounds for the fetal homicide charge) relied extensively on statements made by doctors and nurses who had examined Rowland. The fact that Rowland signed a form acknowledging that she was leaving the hospital against medical advice was used against her. While health care providers at LDS (Latter Day Saints) Hospital freely discussed Rowland's case with the police, the hospital's official spokesperson nevertheless cited “medical privacy” as one of the reasons for declining to comment on the case to the press (Sage 2004).

A Wisconsin obstetrician who was providing twenty-four-year-old Angela M. W. with prenatal care suspected that she was using cocaine or other drugs. When blood tests allegedly confirmed the obstetrician's suspicion, he confronted Angela about her drug use. She then stopped coming in for scheduled appointments, at which point the obstetrician reported her to the Waukesha Department of Health and Human Services (DHHS). Relying on this information, DHHS petitioned the juvenile court for an order directing the Waukesha County Sheriff's Department to take Angela's fetus into protective custody. With the obstetrician's sworn statement against his patient as the sole source of information about the case, the juvenile court appointed a guardian ad litem for Angela's fetus and issued an order requiring that the fetus “be detained ... and transported to Waukesha Memorial Hospital for inpatient treatment and protection.”According to the order, “Such detention will by necessity result in the detention of the unborn child's mother, This 1997 Wisconsin case occurred before the state adopted a law specifically permitting the commitment of a pregnant woman who “habitually lacks self-control in the use of alcohol beverages or controlled substances.”Notably, however, this law does not mandate that health care providers report their pregnant patients to state authorities (Martino 1998; Quirmbach and Montagne 1998).

The Angela M. W. case illustrates that threats of punitive responses discourage some women from continuing medical care. In the Marlowe case discussed earlier, Marlowe fled the hospital while in active labor rather than submit to unnecessary surgery. She found a hospital that respected her decision making and delivered a healthy baby vaginally. In South Carolina, a thirty-three-year-old biracial woman, Theresa Joseph, was in her first trimester of pregnancy when she was admitted to the Medical University of South Carolina for treatment of a severe foot infection. Because Joseph was pregnant and acknowledged having a drug problem, she was threatened with arrest under the hospital's policy. Joseph responded to the threat by leaving the hospital against medical advice and avoiding both prenatal care and drug treatment for the remainder of her pregnancy. Several other women not only avoided prenatal care and hospital births because they feared child removal or arrest but also delayed seeking, or failed altogether to obtain, medical care for themselves or their newborn babies for the same reasons.

Alma Baker, a thirty-four-year-old white woman in Texas, was arrested on charges of delivering a controlled substance to a minor when her twins were born and tested positive for THC, a chemical compound found in marijuana. Baker squarely addressed how fear of reporting and punishment may have a deterrent effect when she said, “If I would have known that I'd get in trouble for telling my doctor the truth [that she was using cannabis to calm her nausea] I would have either lied or not gone to the doctor” (Gorman 2004).

Individual health care providers and social workers have in some instances arguably violated ethical standards by breaching privacy and confidentiality, overriding patient decision making, and facilitating the arrest or other punitive detention of a patient (Jos, Marshall, and Perlmutter 1995). To be sure, professional medical, public health, and social work organizations and individuals have also played a vital role in challenging such actions. Our research found that more than 250 professional and advocacy organizations and individual experts have joined one or more amicus curiae (friend of the court) briefs in cases documented in this study. These briefs bring courts' attention to the dangerous impact that arrests, detentions, and forced interventions have on maternal, fetal, and child health (e.g., Abrahamson et al.


The hundreds of cases this study documents raise numerous concerns about the health and dignity afforded to pregnant women in the United States. Pregnancy and childbirth continue to carry significant life and health risks (Centers for Disease Control and Prevention [CDC] 2000, 2008; Amnesty International 2010; Save the Children 2010; Raymond and Grimes 2012). In many of the cases, women experienced those risks (often voluntarily undergoing cesarean surgery to bring forth life) only to find that doing so provided the basis for being charged with a crime. Some affidavits in support of the arrest describe giving birth as part of the alleged crime. For example, one affidavit explained that the woman “did willfully and unlawfully give birth to a male In some cases the criminal charges filed and comments made by arresting officers, prosecutors, and judges were explicit in denying dignity to both women and their children. Accordingly, the woman did not give birth to a child but rather to a “victim,” a “bastard,” or a ““delinquent.”

Our findings challenge the notion that arrests and detentions promote maternal, fetal, and child health or provide a path to appropriate treatment. Significantly, detention in health and correctional facilities has not meant that the pregnant women (and their fetuses) received prompt or appropriate prenatal care. Our research into cases claiming that arrests and detentions would ensure that pregnant women were provided with appropriate drug treatment or that only women who had refused treatment would be arrested or prosecuted overwhelmingly found that such claims were untrue. In some cases women were arrested despite the fact that they were voluntarily participating in drug treatment. Our findings also lend support to the medical and public health consensus that punitive approaches undermine maternal, fetal, and child health by deterring women from care and from communicating openly with people who might be able to help them (Roberts and Pies 2011; Roberts and Nuru-Jeter 2010; Jessup et al. 2003; Poland et al. 1993; Gehshan 1993; US General Accounting Office 1990). Cases documenting pregnant women's unwillingness to seek help for themselves, and in some cases for their newborns, provide compelling anecdotal evidence that punitive measures and the legal arguments supporting them will undermine rather than advance state interests in public health.

Our study also challenges the idea that arrests, detentions, and forced interventions of pregnant women are extremely rare and occur only in isolated, exceptional circumstances against a narrowly definable group of women. Quite to the contrary, cases documented in this study make clear that arrests, detentions, and forced interventions have not been limited to pregnant women who use a certain drug or engage in a particular behavior. Our research shows that these state interventions are happening in every region of the country and affect women of all races.

At the same time, disturbing patterns emerge from our data, which show that the majority of cases have included an allegation relating to the use of an illegal drug (overwhelmingly cocaine), that low-income women, especially in some southern states, are particularly vulnerable to these state actions, and that pregnant African American women are significantly more likely to be arrested, reported by hospital staff, and subjected to felony charges.

These findings are consistent with investigative news articles reporting that African Americans are more likely to be subjected to drug testing and reporting (Rotzoll 2001; Anderson 2008); studies finding racial disparities in drug testing and reporting of African American women (Chasnoff, Landress, and Barrett 1990; Ellsworth, Stevens, and D'Angio 2010; Roberts and Nuru-Jeter 2011), and previous research concerning court-ordered interventions (Kolder, Gallagher, and Parsons 1987). They are also consistent with well-documented racially disproportionate application of criminal laws to African American communities in general and to pregnant African American women in particular (Roberts 1997; Flavin 2009; Alexander 2010; Tonry 2011).

A full discussion of the implications of our research with regard to race, gender, and the war on drugs is beyond the scope of this article. It is important to note, however, that the clear racial disparities identified cannot be explained as the consequences of “color-blind” decisions to exercise state control over pregnant women who use drugs or more specifically those who use cocaine. Although which substances are most likely to be used may vary with population subgroups and geography, rates of drug use and dependency are similar across races (Mathias 1995; Hans 1999; National Institute on Drug Abuse 2003; Substance Abuse and Mental Health Services Administration 2009, 2011; Roberts and Nuru-Jeter 2011).

Moreover, the risks of harm from prenatal exposure to cocaine are not qualitatively different from risks posed by other factors (legal and illegal), and the harms that have been associated with prenatal exposure to cocaine are not easily distinguishable from other contributing and often correlated factors (Zuckerman et al. 1989; Mayes et al. 1992; Little, Wilson, and Jackson 1996; Slotnick 1998; Addis et al. 2001; Chavkin 2001; Lewis et al. 2004; Ackerman, Riggins, and Black 2010). In 2001 the Journal of the American Medical Association published a comprehensive analysis of the developmental consequences of prenatal exposure to cocaine that concluded:

Among children aged 6 years or younger, there is no convincing evidence that prenatal cocaine exposure is associated with developmental toxic effects that are different in severity, scope, or kind from the sequelae of multiple other risk factors. Many findings once thought to be specific effects of in utero cocaine exposure are correlated with other factors, including prenatal exposure to tobacco, marijuana, or alcohol, and the quality of the child's environment. (Frank et al. 2001: 1613-14)

The authors of the study condemned as “irrational” policies that selectively ““demonize” in utero cocaine exposure (ibid.: 1620). Indeed, the US Sentencing Commission (2007), in adjusting the penalties associated with crack-related offenses, did so in part because it concluded that “the negative effects from prenatal exposure to cocaine, in fact, are significantly less severe than previously believed” and that those negative effects are similarly correlated with the effects of prenatal exposure to other drugs, both legal and illegal.

Finally, as has been compellingly argued by historians, sociologists, legal scholars, and others, the willingness to believe that cocaine, and especially crack cocaine, required uniquely punitive responses was derived in large measure from racist assumptions about African Americans in general and African American mothers in particular (Gomez 1997; Morgan and Zimmer 1997; Reinarman and Levine 1997; Roberts 1997; Humphries 1998, 1999; Collins 2000: 69-96; Zerai and Banks 2002; Hart 2012). The harsh treatment imposed on the pregnant women in our study, including being taken straight from their hospital beds and arrested shortly after delivery, being taken in handcuffs, sometimes shackled around the waist, and at least one woman being shackled during labor, is consistent with a long and disturbing history of devaluing African American mothers (Roberts 1997; Ocen 2011; Roth 2012).

Our review of the legal authority articulated in support of the actions taken against the pregnant women identified in this study found that it rested on the claim that state authorities should have the power to arrest, detain, and forcibly intervene on pregnant women in order to protect the fertilized eggs, embryos, and fetuses inside them. We believe the implications are clear: if feticide statutes that purport to protect pregnant women and fetuses from third-party attacks and existing laws that declare separate rights for eggs, embryos, and fetuses are already being used as the basis for justifying depriving pregnant women of their liberty, we must expect that personhood measures will be used this way, too. Thus, far from being a scare tactic, our findings confirm that if passed, personhood measures not only would provide a basis for recriminalizing abortion, they would also provide grounds for depriving all pregnant women of their liberty.

Our findings also make clear that far more than the right to decide to have an abortion is at stake if such laws pass. All pregnant women, not just those who try to end a pregnancy, will face the possibility of arrest, detention, and forced intervention as well as threats to and actual loss of a wide range of rights associated with constitutional personhood (Gallagher 1987; Johnson 1989; Roberts 1991; Daniels 1996; Boyd 1999; Campbell 2000; Solinger 2002; Roth 2003; Fentiman 2006; Cherry 2007). Indeed, we have identified more than two hundred cases initiated against pregnant women since 2005 that also overwhelmingly rest on the claim of separate rights for fertilized eggs, embryos, and fetuses (see, e.g., James 2010; Pilkington 2011; Robinson 2012; Calhoun 2012; ABC News 2 2012).

While voters in Colorado and Mississippi defeated personhood ballot measures three times (Colorado Secretary of State 2008, 2010; Mississippi Secretary of State 2011b), Personhood USA, the organization sponsoring these measures, has promised to continue its efforts to get them passed (Pesta 2012; Vanderveen 2012). Similar bills, including the so-called Sanctity of Human Life Act (H.R. 212, 112th Cong. [2011]), have been introduced in Congress. In light of these continued efforts and our findings, we challenge health care providers, law enforcement and child welfare officials, social workers, judges, and policy makers to examine the role they play in the arrests and detentions of and forced interventions on pregnant women. We call on these same people to develop and support only those policies that are grounded in empirical evidence, that in practice will actually advance the health, rights, and dignity of pregnant women and their children, and that will not perpetuate or exacerbate America's long and continuing history of institutionalized racism. Finally, our study provides compelling reasons for people who value pregnant women, whether they support or oppose abortion, to work together against personhood and related measures so women can be assured that on becoming pregnant they will retain their civil and human rights.