In 1989, the Medical University of South Carolina (MUSC) established, in cooperation with local law enforcement officials, a policy (the Policy) intended to reduce cocaine use by pregnant women. As implemented, the Policy involved the testing of patients' urine to gather evidence of cocaine use and the disclosure of evidence so obtained to law enforcement authorities. Appellants are ten former obstetrical patients at MUSC whose urine was tested pursuant to the Policy; they brought this action claiming, inter alia, that the testing of their urine for cocaine constituted a warrantless, unreasonable search in violation of the Fourth Amendment.1

Following trial, the district court submitted to the jury the question of whether Appellants had consented to the urine drug screens. The jury found in favor of Appellees. We now must decide whether the district court erred in denying Appellants' subsequent motion for judgment as a matter of law. For the reasons set forth below, we hold that no rational jury could conclude, from the evidence presented at trial, that Appellants gave their informed consent to the taking and testing of their urine for evidence of criminal activity for law enforcement purposes. Our holding encompasses two determinations: first, that as to most of the Appellants, the record evidence does not support a finding that Appellants knew that their urine was being analyzed for evidence of criminal activity for law enforcement purposes; and second, that the record evidence does not support a finding that Appellants, for Fourth Amendment purposes, voluntarily submitted to the searches. Excluded from our holding is Ellen Knight, who, as explained below, suffered no Fourth Amendment violation. Additionally, we remand for further proceedings as to Darlene Nicholson, who may not have been searched pursuant to the Policy. In sum, we affirm in part and reverse and remand in part.

Because we are reviewing the denial of Appellants' motion for judgment as a matter of law, we must view the evidence in the light most favorable to Appellees and draw all reasonable inferences in their favor without weighing the evidence or assessing the witnesses' credibility. See Sales v. Grant, 158 F.3d 768, 775 (4th Cir.1998). Viewed in light of that standard, the evidence at trial demonstrated the following facts.

In the fall of 1989, MUSC instituted a policy providing for the testing of the urine of pregnant women for cocaine use and for the reporting, under certain circumstances, of test results to law enforcement officials. The Policy was implemented both at the MUSC hospital and at the MUSC clinic.2 The impetus for the Policy came from Nurse Shirley Brown, a case manager in the obstetrics department at the MUSC hospital. In the previous year, personnel in the obstetrics department had noted an increase in cocaine use by pregnant women. Although these women were referred to substance abuse treatment, such referrals were ineffective in stemming the tide of "cocaine babies" being born at the MUSC hospital. After hearing a report regarding a program in place elsewhere in South Carolina, which involved bringing charges of child abuse against women who used cocaine during pregnancy, Brown spoke with an MUSC official, who in turn contacted the Ninth Circuit Solicitor (chief prosecuting attorney) concerning the development of a program for application at MUSC to detect and deter cocaine use by pregnant women.3

Eventually, a task force was formed that included Nurse Brown, the Solicitor, the Chief of the Charleston City Police Department (CCPD), representatives from the Charleston County Substance Abuse Commission (CCSAC), and doctors from various departments at MUSC involved in perinatal care. By mid-October 1989, these meetings had produced the initial version of the Policy at issue in this litigation. According to the terms of the initial Policy, MUSC maternity patients were to be tested if any of nine criteria was met.4 The initial Policy provided that if a patient had not yet delivered her baby, she was to be given a referral to CCSAC and counseling regarding the harmful effects of drug use during pregnancy. Upon a second positive test or a failure to comply with treatment obligations, the patient would be arrested.5 If a patient tested positive for cocaine upon delivering a child, she was to be arrested "as soon as medically possible." J.A. 1448.

In early 1990, the Policy was amended to focus more on patient education. To that end, two of the doctors involved with the Policy drafted a letter to be given to all women who received prenatal care at the MUSC clinic. Known as the "To Our Patients" letter, it warned patients of the dangers of prenatal drug use and further stated:

If you are using drugs, please stop! If you are unable to stop, please let your doctor know. We want to help mothers get off drugs for the benefit of both you and your baby. We will provide you counseling about the harms of drug abuse and will make arrangements for you to be seen at the Substance Abuse Clinic. We realize that drug abuse is a very difficult problem and we will do all that we can to help you.

If, however, we continue to detect evidence of drug abuse or a failure to follow recommended treatment, we will take action to protect your unborn child. The Charleston Police, the solicitor's office, and the Protective Service Division of [the Department of Social Services] are also committed to the protection of unborn and newborn children from the harms of illegal drug abuse.

We hope that you can understand the tragedy which is being caused by the continued use of illegal drugs during pregnancy. This policy of providing warning, counseling and treatment for pregnant women using illegal drugs is the best way for us to help. For those women who fail this treatment, we must ask for help to protect the life and health of our most innocent unborn children.

Id. at 1437.6 Additionally, patients were shown a video regarding the dangers of drug abuse during pregnancy.

In an effort to publicize the Policy, the Solicitor recorded a public service announcement (PSA) that was aired on local television stations for several months beginning in March 1990. The text of the PSA was as follows:

When you're pregnant, just one line of cocaine, a single hit of crack, rushes to your baby's body and brain. Within minutes your body can be jolted into premature labor risking a still developing child to stroke, even death. And not only will you live with the guilt, you could be arrested.

But this is a tragedy you can prevent. If you have a problem with drugs talk to your doctor or call MUSC at 792-6437. Trained counselors will guide you through drug rehabilitation and advise you about good prenatal care. And if you stay with the program you will not be arrested or prosecuted.

Wake up from the nightmare. Think about your baby first.

Supp. J.A. 1.

The Policy as revised in 1990 also included a greater "amnesty" component. Under the revised Policy, all pregnant women who tested positive for cocaine, including those whose first positive test occurred in conjunction with the delivery of their children, were referred to substance abuse treatment. A patient was not arrested unless she tested positive a second time or failed to fulfill treatment requirements. This aspect of the Policy was explained in a letter from the Solicitor ("the Solicitor's letter"), which was given to patients after a positive urine drug screen. In the case of a woman who tested positive during prenatal care, the letter provided:

During your recent examination you tested positive for drugs. You have been counseled about the harmful effects of drugs to you and your baby and referred to Substance Abuse and Pre Natal Care by the Medical University. By these referrals you are being afforded an opportunity to rehabilitate yourself for the good of yourself and your baby. Please understand that by using drugs during pregnancy you are risking death or at least severe long-term harmful effects to your baby. If you fail to attend Substance Abuse and PreNatal Care you will be arrested by Charleston City Police and prosecuted by the Office of Solicitor.

J.A. 1432 (first emphasis added). A similar letter was given to women who tested positive for cocaine use upon the birth of a baby; this letter informed the patient that a referral had been made to the Department of Social Services in addition to the CCSAC. This version of the letter advised women that they would be arrested and prosecuted "[i]f you fail to complete Substance Abuse Counselling [sic], fail to cooperate with the Department of Social Services..., or if you fail to maintain clean urine specimens during your Substance Abuse rehabilitation."7 Id. at 1433. Women were requested to sign the Solicitor's letter as a means of acknowledging receipt, but a signature on the letter did not signify agreement to anything.

In addition to the documents related to the Policy, women receiving perinatal care at MUSC were generally asked to sign one or both of two consent-to-treatment forms. The first form ("the ambulatory consent form") was signed by women who received prenatal care at the MUSC clinic. This form provided, in pertinent part:

The intention hereof is to grant full authority to such physicians and surgeons and the Medical University Clinics ... to administer and perform all and singular any drugs, treatment, tests or diagnostic procedures.... I further consent to the testing for drugs if deemed advisable by or necessary in the professional judgement of the physician or surgeon.

...

Id. at 1443 (emphasis added). Women admitted to the hospital signed a consent form ("the general consent form") that included the following language:

I acknowledge that I am suffering from a condition requiring Medical/Hospital care and thereby voluntarily consent to such Medical/Hospital care encompassing diagnostic procedures and medical treatment by my physician, assistants or designees, as may be necessary in his or her judgment.... I further consent to the testing of drugs if deemed advisable by my physician.

Id. at 1442 (emphasis added). These forms were the same forms signed by every patient of the MUSC clinic and hospital; they were not altered in any way to account for the Policy.

During the initial stages of her pregnancy, Griffin received prenatal care at a clinic in McClellanville, South Carolina; that clinic referred her to the MUSC clinic. In connection with her prenatal care at the MUSC clinic, Griffin signed an ambulatory consent form on June 28, 1989. The following day, her urine tested positive for cocaine. During her testimony at trial, Griffin denied being informed that she had tested positive.

On October 7, Griffin was admitted to the MUSC hospital in pre-mature labor. She did not sign a general consent form in connection with this admission. See id. at 1523. During this admission, Griffin's urine again tested positive for cocaine. Griffin was arrested on October 10, just as she was released from the hospital. She remained incarcerated for the next three weeks, until she delivered her baby on October 26.

In January 1990, Griffin was indicted on one count of possession of cocaine and one count of distribution of cocaine to a person under the age of 18. She was not prosecuted.

Powell initially received prenatal care from a physician in McClellanville, who referred her to the MUSC clinic because of her high blood pressure. Powell signed an ambulatory consent form on May 24, 1989. Powell attended her prenatal appointments until late September, when Hurricane Hugo struck the Charleston area. On October 13, Powell went into labor and was transported to the MUSC hospital by ambulance. A general consent form was signed by Powell's mother. Powell delivered her baby shortly after arriving at the hospital; both Powell's and the infant's urine tested positive for cocaine. Powell was subsequently arrested, but she was not prosecuted.

During her pregnancy, Knight received prenatal care at the MUSC clinic. In connection with this care, Knight signed an ambulatory consent form on October 11, 1989. On the morning of November 6, 1989, Knight was admitted to the MUSC hospital in labor and subsequently delivered her baby. The record does not contain a general consent form signed in connection with this admission.

Knight's baby's urine tested positive for cocaine; the record does not indicate that Knight's urine was tested. Based on the positive test from her infant, Knight was arrested on November 8. The charges were dismissed after she successfully completed a drug treatment program.

Singleton delivered two children at MUSC during the existence of the Policy. During the first of these pregnancies, in 1989, Singleton received no prenatal care. She delivered a son on November 9 at the MUSC hospital after signing a general consent form upon her admission to the hospital. Singleton was in active labor when she signed the form and delivered her baby shortly thereafter. At that time, her urine tested positive for cocaine. Singleton told Nurse Brown that she consented to a urine drug screen "because she thought she was clean." Id. at 1862 (internal quotation marks omitted). Singleton was arrested on November 10. In July 1990, she was charged with child neglect; she was not prosecuted.

In November 1990, Singleton delivered another baby, this time in the ambulance en route to the MUSC hospital. She later signed a general consent form. In connection with her post-partum care, Singleton again tested positive for cocaine. As an alternative to arrest, Singleton voluntarily admitted herself to inpatient treatment at the MUSC Institute of Psychiatry (IOP).

In the summer of 1990, when she was six or seven months pregnant, Pear was referred to the MUSC clinic because she was at high risk for preterm labor. Pear signed an ambulatory consent form for care at the clinic on July 13, 1990. At that visit, her urine tested positive for cocaine. Following the positive test, Pear reviewed and signed the Solicitor's letter, reviewed the "To Our Patients" letter,9 and watched the video on substance abuse during pregnancy. She was provided with an appointment for substance abuse counseling. In late August, Pear was admitted to the MUSC hospital in preterm labor and again tested positive for cocaine. Pear signed a general consent form in connection with this admission. Pear was arrested upon her discharge on September 1 and spent the remainder of her pregnancy in jail. She delivered her child on September 14. Although Pear was charged with distributing cocaine to a person under the age of 18, the charge was subsequently nolle prossed.

During her trial testimony, Pear acknowledged that she had been aware, during her pregnancy, that MUSC was testing women for drug use and that those who tested positive were arrested.

Hale delivered a baby at the MUSC hospital on December 13, 1990, having received no prenatal care during her pregnancy. The record does not contain a general consent form signed in connection with this admission. Both Hale and the baby tested positive for cocaine. Hale was informed of the hazards of drug use and signed the Solicitor's letter. She was not arrested at that time, but a warrant for Hale's arrest was issued in March 1991, and Hale was arrested in June of that year. Given the terms of the Policy, we assume that Hale's arrest was triggered by her failure to comply with treatment obligations. As with the other Appellants, however, Hale was not prosecuted.

Ferguson began receiving prenatal care at the North Charleston Health Department in early or mid 1991. At some point prior to or during her pregnancy, Ferguson saw the PSA. At her request, she was referred to the MUSC clinic after admitting to Health Department personnel that she was using cocaine. Ferguson signed an ambulatory consent form in connection with her first visit to the clinic on June 19, 1991. When she signed the form, Ferguson was experiencing pre-term labor. At that visit, Ferguson tested positive for cocaine. She thereafter viewed the video on substance abuse and signed the Solicitor's letter. Ferguson was encouraged to enter inpatient substance abuse treatment but refused because she lacked adequate child care.

On August 4, Ferguson underwent an emergency cesarian section based on her doctor's suspicion, later confirmed, that the placenta was separating from the uterine wall. Ferguson signed a consent form specific to the surgery, but the record does not include a general consent form related to Ferguson's admission to the hospital. Ferguson again tested positive for cocaine. Although the terms of the Policy provided that Ferguson was to be arrested at that time, she was allowed to go home to attend to the details of her mother's funeral. Ferguson was arrested on August 19 and charged with distributing drugs to a person under the age of 18; this charge was based on the positive urine drug screen on June 20. The charge was dismissed in 1993.

Joseph saw the PSA while she was pregnant in 1991. Joseph did not seek prenatal care because she was using cocaine and heroin; in view of the PSA, she was afraid of being arrested. On June 6, when she was approximately 15 weeks pregnant, Joseph was admitted to the MUSC hospital for treatment of a severe infection in her right foot. Joseph was in dire need of medical treatment; hospital employees warned her that failure to complete treatment could result in amputation of her foot or even death. The record does not include a general consent form signed by Joseph in connection with this admission. Joseph's urine tested positive for cocaine. Nursing notes from Joseph's medical records indicate that a transfer to the IOP was contemplated and that Joseph was to be shown the video and the Solicitor's letter prior to this transfer. However, Joseph left the hospital against medical advice on June 7.

Joseph returned to the hospital on June 14 because of excruciating pain in her foot; at that time, she signed a general consent form. She again tested positive for cocaine. Joseph was shown the video, given the Solicitor's letter, and advised that she had appointments for prenatal care and substance abuse counseling. Joseph was informed that failure to attend either of these appointments could result in her arrest.

When Joseph was 32 weeks pregnant, she was admitted to the MUSC hospital for performance of an emergency cesarian section, and she signed a general consent form. She again tested positive for cocaine and was arrested on October 23, the day she was discharged from the hospital. As with the other Appellants, Joseph was not prosecuted.

Williams began receiving prenatal care at the North Charleston Health Department in December 1991, when she was five or six months pregnant. At her request, she was transferred to the MUSC clinic for assistance with her cocaine problem. Williams requested the transfer because she had heard of the Policy and understood that MUSC was trying to help pregnant women with their drug problems. Williams signed an ambulatory consent form at her first visit to the MUSC clinic on January 15, 1992. She tested positive for cocaine at that time, whereupon she signed the Solicitor's letter and watched the video. She also was referred to substance abuse counseling. Williams tested positive for cocaine several more times during her pregnancy but was not arrested (the record does not indicate why). She delivered her baby at the MUSC hospital on March 10, and again tested positive for cocaine. The record does not contain a general consent form signed by Williams in connection with this admission. She was arrested on March 12 and was charged with distributing cocaine to a person under the age of 18 based on one incident of prenatal cocaine use. Williams was not prosecuted, apparently because she successfully completed an inpatient drug treatment program.

During her pregnancy in the latter part of 1993, Nicholson received prenatal care at the MUSC clinic. On October 20, Nicholson signed an ambulatory consent form in connection with her prenatal care; at this visit, her urine tested positive for cocaine. During a subsequent prenatal visit on December 17, she was informed that she was dehydrated and needed to be admitted to the hospital. Nicholson signed a general consent form in connection with this admission. During that admission, Nicholson's urine again tested positive for cocaine. At that time, she was given a choice between voluntary and involuntary commitment to the IOP. Nicholson voluntarily admitted herself to the IOP for inpatient drug treatment and remained there until January 7, 1994.

After her release, Nicholson continued to use drugs. On February 7, she was involuntarily committed to the IOP by the CCSAC. She was released two days later, on February 9. Less than a week later, Nicholson readmitted herself to the IOP for another 30 day course of substance abuse treatment. On February 21, Nicholson was temporarily released from the IOP and admitted to the MUSC hospital for the birth of her child; after two days of recovery, she was readmitted to the IOP and completed her course of treatment.

Appellants filed this action on October 5, 1993, claiming, inter alia, that the performance of the urine drug screens violated their Fourth Amendment right to be free from unreasonable searches and seizures. In response, Appellees contended that the searches were justified by the special needs doctrine, see Nat'l Treasury Employees Union v. Von Raab, 489 U.S. 656, 665-66, 109 S. Ct. 1384, 103 L. Ed. 2d 685 (1989), or, alternatively, that Appellants had consented to the searches.

The district court rejected the special needs theory, concluding that the involvement of law enforcement personnel in the formulation of the Policy precluded the application of the special needs doctrine. Following trial, the district court submitted the Fourth Amendment claim to the jury. The court instructed the jury that it should find for Appellees if it concluded, after considering the totality of the circumstances, that Appellants had consented to the searches. However, the court also instructed the jury that neither the general consent form nor the ambulatory consent form was adequate to establish voluntary consent to the searches.

The jury found in favor of Appellees. Appellants moved for judgment as a matter of law, maintaining that a rational jury could not find, from the evidence presented, that any of the Appellants had consented to the searches. The district court denied the motion, stating that "[t]here is evidence in the record from which the jury could reasonably conclude that the plaintiff[s] consented to these searches." Id. at 2728. However, the district court did not specify what evidence supported the verdict.

On appeal, a majority of this panel affirmed on the basis that the searches were sustainable under the special needs doctrine. See Ferguson v. City of Charleston, 186 F.3d 469, 476-79 (4th Cir.1999) (Ferguson I); cf. id. at 486-89 (Blake, J., dissenting). This ruling was subsequently reversed by the Supreme Court, which concluded that because "the immediate objective of the searches was to generate evidence for law enforcement purposes," the special needs doctrine did not apply. Ferguson v. City of Charleston, 532 U.S. 67, 83, 121 S. Ct. 1281, 149 L. Ed. 2d 205 (2001) (Ferguson II). The Court therefore vacated our decision and remanded for consideration of whether there was sufficient evidence to support the finding of consent by the jury. See id. at 86, 121 S. Ct. 1281. Specifically, the Court instructed us to determine whether Appellants gave their "informed consent" to the searches. Id. at 76, 121 S. Ct. 1281.

We review the denial of judgment as a matter of law de novo. See Konkel v. Bob Evans Farms Inc., 165 F.3d 275, 279 (4th Cir.1999). In so doing, we must view the evidence in the light most favorable to Appellees and draw all reasonable inferences in their favor without weighing the evidence or assessing the witnesses' credibility. See Sales, 158 F.3d at 775. "The question is whether a jury, viewing the evidence in the light most favorable to [Appellees], could have properly reached the conclusion reached by this jury." Benesh v. Amphenol Corp. (In re Wildewood Litig.), 52 F.3d 499, 502 (4th Cir.1995). We may reverse only if "there can be but one reasonable conclusion as to the verdict," Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986); if reasonable minds could differ, we must affirm, see Sales, 158 F.3d at 775.

Before considering the merits of the ruling of the district court, it is necessary to consider some preliminary matters.

At the outset, we must define with some specificity the Fourth Amendment event with which we are concerned. One could argue that the reach of the Fourth Amendment extends only to the process by which the urine sample was obtained and that any subsequent handling or testing of that sample does not have Fourth Amendment ramifications. This is the position taken by Justice Scalia in his dissent in Ferguson II. See Ferguson II, 532 U.S. at 92, 121 S. Ct. 1281 (Scalia, J., dissenting) ("There is only one act that could conceivably be regarded as a search of petitioners in the present case: the taking of the urine sample.").10 At the other end of the spectrum, one may consider all law enforcement activities related to the urine sample — i.e., the taking of the sample, the testing of it for evidence of drug use, and the reporting of the results to law enforcement officials — as covered by the Fourth Amendment. This seems to be the view taken by the Supreme Court, see id. at 77, 121 S. Ct. 1281, and it is the way the complaint was framed by Appellants and presented to the jury. In the middle is the position that the Fourth Amendment event at issue comprises the taking and testing of the urine but not the reporting of results to law enforcement officers.

For purposes of resolving this appeal, it is sufficient to reject the first possibility — that the only Fourth Amendment event was the obtaining of the sample. In the special needs context, where the Court has most often addressed the constitutional implications of urine drug screens, the Court has uniformly addressed the constitutionality of the taking of a urine sample and the testing of that sample for evidence of drug use.11 See, e.g., Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 652, 115 S. Ct. 2386, 132 L. Ed. 2d 564 (1995); Skinner v. Ry. Labor Execs.' Ass'n, 489 U.S. 602, 617, 109 S. Ct. 1402, 103 L. Ed. 2d 639 (1989). Because there was, as we explain below, insufficient evidence from which a rational jury could conclude that any Appellant (excluding Ellen Knight and Darlene Nicholson, discussed infra Part II.A.2) validly consented to the taking and testing of her urine for law enforcement, as opposed to medical, purposes, we do not consider whether the disclosure of test results to law enforcement authorities also implicated the Fourth Amendment.

Additionally, we must address Appellees' arguments that two of the Appellants do not have any claim under the Fourth Amendment. These Appellants are Ellen Knight, who was arrested based on a search of her newborn's urine, and Darlene Nicholson, whose urine was tested, according to Appellees, after the Policy had been suspended.

As noted in the statement of facts, Knight herself was never searched pursuant to the Policy; rather, a search was conducted on the urine of her newborn child.12 Appellees assert that because Knight did not have a reasonable expectation of privacy in her child's urine, she lacks standing to claim a violation of the Fourth Amendment. See Rakas v. Illinois, 439 U.S. 128, 139-40, 99 S. Ct. 421, 58 L. Ed. 2d 387 (1978) (explaining that term "standing" refers, in Fourth Amendment context, to question of whether party's own expectation of privacy has been infringed).

We agree with Appellees that Knight suffered no violation of her Fourth Amendment rights. "Fourth Amendment rights are personal rights which, like some other constitutional rights, may not be vicariously asserted."13 Alderman v. United States, 394 U.S. 165, 174, 89 S. Ct. 961, 22 L. Ed. 2d 176 (1969); see United States v. Taketa, 923 F.2d 665, 670 (9th Cir.1991) (noting that "the Supreme Court [has] explicitly rejected concepts of `vicarious' or `target' standing to assert fourth amendment rights"). We are aware of no decision holding, or even suggesting, that a mother has a reasonable expectation of privacy in her newborn child's bodily fluids. Indeed, such a holding would conflict with the general rule that an expectation of privacy does not arise from one's relationship to the person searched. See 5 Wayne R. LaFave, Search and Seizure § 11.3(i), at 223-24 (3d ed.1996). Accordingly, we affirm as to Knight.

Nicholson's urine was tested for cocaine in October and December 1993. Appellees assert that the Policy had been discontinued by this time, and thus any such test was not a search because it was not performed by MUSC personnel acting in conjunction with law enforcement authorities. It is not clear to us whether Appellees made this assertion before the district court; if they did not, the argument is forfeited. See Trandes Corp. v. Guy F. Atkinson Co., 996 F.2d 655, 665 (4th Cir.1993). If, however, the question was raised by Appellees but not resolved by the district court, the court must address the issue before allowing Nicholson to seek damages. Therefore, we remand for the district court to determine (1) whether Appellees' claim that Nicholson was not searched under the Policy was forfeited; (2) if not, whether Appellees' assertion is correct; and (3) if not, whether a rational jury could find that Nicholson gave her informed consent to the search. If it reaches the last question, the district court should be guided in its determination by the reasoning of this opinion. Finally, if the district court determines that Nicholson does have a Fourth Amendment claim and that the finding of consent by the jury is not supported by the evidence presented at trial, a jury should be allowed to consider Nicholson's claim for damages.

Also before proceeding to the merits, it will be helpful to note two issues with which we are not concerned. First, this appeal does not turn on whether any Appellant was prosecuted or on whether Appellants were arrested only after a second (or subsequent) positive test. While these questions might be relevant to Appellants' damages, the consent question addresses only the search — i.e., the taking and testing of Appellants' urine for law enforcement purposes — not any events that followed from the search.

This case is also not about the motives of those involved. There can be little doubt from the record that the creation and implementation of the Policy were motivated by the sincere desire of all involved to help pregnant women and their babies. Our ruling is in no way an indication that the MUSC personnel responsible for the Policy engaged in purposefully wrongful conduct. Rather, our ruling concerns only a question of constitutional procedure: whether Appellants gave their informed consent to be searched, i.e., whether Appellants gave their informed consent to the taking and testing of their urine for law enforcement purposes.

It is to that question that we now turn. The Fourth Amendment prohibits unreasonable searches, see U.S. Const. amend. IV, and searches conducted without a warrant are per se unreasonable unless a valid exception to the warrant requirement applies, see Schneckloth v. Bustamonte, 412 U.S. 218, 219, 93 S. Ct. 2041, 36 L. Ed. 2d 854 (1973). Voluntary consent to a search is such an exception. See id. In determining whether consent to search was freely and voluntarily given, the factfinder must examine the totality of the circumstances surrounding the consent. See id. at 227, 93 S. Ct. 2041. "[T]wo competing concerns must be accommodated in determining the meaning of a `voluntary' consent — the legitimate need for such searches and the equally important requirement of assuring the absence of coercion." Id.

In considering the totality of the circumstances, the factfinder must take into account the characteristics of the person from whom consent is sought, such as age, maturity, education, and experience. See United States v. Lattimore, 87 F.3d 647, 650 (4th Cir.1996) (en banc). Of particular relevance here, the factfinder must consider "the possibly vulnerable subjective state of the person who consents." Schneckloth, 412 U.S. at 229, 93 S. Ct. 2041.

In addition to the characteristics of the consenter, the factfinder must also look to the circumstances of the request for consent. See Lattimore, 87 F.3d at 650. Indeed, the remand by the Supreme Court in Ferguson II, requiring us to determine whether Appellants gave their informed consent to the searches, rests upon the particular factual circumstances of this case. Simply put, the circumstances of the encounter between Appellants and those conducting the searches were unusual. In the vast majority of cases, consent to search is requested by one known to the suspect to be a law enforcement officer. Here, in contrast, any request for consent was made by medical personnel acting (unknown to Appellants) as agents of law enforcement.

In remanding for further consideration of the consent issue, the Supreme Court set specific parameters tailored to these unique circumstances, as indicated by its decision to use the term "informed consent," language that does not appear in the Court's other consent-to-search cases. Under these parameters, Appellants' knowledge of law enforcement involvement in the Policy is critical to the existence of informed consent. The Court admonished that "when [medical personnel] undertake to obtain ... evidence from their patients for the specific purpose of incriminating those patients, they have a special obligation to make sure that the patients are fully informed about their constitutional rights." Id. at 85, 121 S. Ct. 1281 (citing Miranda v. Arizona, 384 U.S. 436, 86 S. Ct. 1602, 16 L. Ed. 2d 694 (1966), by comparison); see id. at 78, 121 S. Ct. 1281 (noting that in special needs cases involving urine drug screens, "there was no misunderstanding about the purpose of the test"); id. at 78 n. 13, 121 S. Ct. 1281 (stating that nothing in mandatory reporting laws "would ... lead a patient to anticipate that hospital staff would intentionally set out to obtain incriminating evidence ... for law enforcement purposes"). Thus, the Court made abundantly clear in Ferguson II that any finding of informed consent must rest on a determination that Appellants had knowledge, from some source, that no medical purpose supported the testing of their urine for cocaine; further, Appellants must have understood that the tests were being conducted for the law enforcement purpose of obtaining incriminating evidence. Phrased somewhat differently, critical to the question of whether Appellants voluntarily consented to the searches is the antecedent question of whether they understood that the request was not being made by medical personnel for medical purposes, but rather by agents of law enforcement for purposes of crime detection.14 See Gouled v. United States, 255 U.S. 298, 305-06, 41 S. Ct. 261, 65 L. Ed. 647 (1921) (holding that suspect did not consent to search of his office by granting admission to friend who claimed to be making social call but in fact, unbeknownst to suspect, was a government agent).

Appellees assert that the tests were, in fact, done for medical purposes. Insofar as the intent of the Policy was to provide medical assistance to the women and their unborn and newly born children, this may be an accurate statement. In the context of Appellants' Fourth Amendment claim, however, to say that the searches were motivated by medical purposes is to say that the collection and testing of the urine was done independently of the Policy. We emphasize that there was no evidence introduced into this record to support the conclusion that any of the searches here would have been conducted in the absence of the Policy; for that reason, we are compelled to proceed on the basis that the primary purpose of the urine drug screens was crime detection, not medical treatment. Cf. Ferguson II, 532 U.S. at 78 n. 13, 121 S. Ct. 1281 (noting difference between reporting of information "acquired in the course of routine treatment" and "intentionally set[ting] out to obtain incriminating evidence ... for law enforcement purposes").

We note also that the parties agree that Appellees bore the burden of proving, by a preponderance of the evidence, that Appellants had consented to each of the searches.15 There is some dispute among the courts of appeals regarding which party bears the burden of proving consent, or the absence thereof, in a civil suit alleging a violation of the Fourth Amendment. See Trulock v. Freeh, 275 F.3d 391, 401 n. 4 (4th Cir.2001) (citing cases). Because this case was litigated by all parties on the premise that Appellees had voluntarily assumed the burden of proof, we will assume for purposes of this decision that the burden of proof rested on Appellees.

With these principles in mind, we proceed to an examination of the evidence pertaining to each Appellant to determine whether a rational jury could have found that that Appellant consented to the taking and testing of her urine by agents of law enforcement for the purpose of obtaining evidence of criminal activity. Initially, we consider whether either the ambulatory consent form or the general consent form was adequate to establish Appellants' informed consent.16 As noted above, the relevant language of the forms provided that the patient consented "to ... testing for drugs if deemed advisable by or necessary in the professional judgement of the physician or surgeon," J.A. 1443 (ambulatory consent form), or that the patient consented "to the testing of drugs if deemed advisable by my physician," id. at 1442 (general consent form). Neither this language, nor anything else in either form, advised or even suggested to Appellants that their urine might be searched for evidence of criminal activity for law enforcement purposes. Rather, to the extent the forms alerted Appellants to the possibility that their urine would be tested for drugs, Appellants were led to believe that such tests would be conducted only if an Appellant's treating physician deemed such a test advisable in the particular circumstances of that Appellant's medical care. As discussed above, there is no evidence that any of the urine drug screens were conducted as a result of a doctor's independent medical judgment; to the contrary, it appears that all of the tests were performed pursuant to the strictures of the Policy. We therefore conclude that, as a matter of law, neither the ambulatory consent form nor the general consent form could serve as sufficient evidence of Appellants' informed consent to the searches.17 Cf. Stoner v. California, 376 U.S. 483, 489-90, 84 S. Ct. 889, 11 L. Ed. 2d 856 (1964) (holding that express or implied consent of hotel guest for entry by maids, janitors, or repairmen did not extend to entry by police officers to search for evidence of crime).

Because neither the general nor the ambulatory consent form establishes express consent to a search for evidence of criminal activity for law enforcement purposes, we must consider whether Appellants gave implied consent, i.e., whether Appellants had knowledge of the Policy from one or more sources and implicitly consented to the searches by thereafter going to MUSC for treatment. Courts generally are reluctant to infer consent from mere notice followed by voluntary conduct. See McGann v. Northeast Ill. Reg'l Commuter R.R. Corp., 8 F.3d 1174, 1180-81 (7th Cir.1993) (stating that notice and voluntary conduct are "necessary but not sufficient conditions" for implied consent). For the reasons set forth below, we conclude that the evidence presented at trial was insufficient for a rational jury to find that Appellants implicitly consented to the searches. First, as to many of the Appellants, there simply is not enough evidence to justify a finding of knowledge of the critical portions of the Policy—namely, that the Policy involved searches of patients' urine for evidence of criminal activity for law enforcement purposes. Second, even if all of the Appellants could be found to have had such knowledge, no rational jury could conclude that Appellants voluntarily submitted, for purposes of the Fourth Amendment, to treatment at MUSC.

Griffin was searched under the Policy during a hospital admission for preterm labor in early October 1989.18 When Griffin was subjected to the Policy, neither the Solicitor's letter nor the "To Our Patients" letter was in use, and the PSA had not yet begun to air. And, no testimony was elicited from Griffin regarding any knowledge she had of the Policy. There was therefore no evidence from which a rational jury could infer that Griffin had knowledge of the terms of the Policy.

Powell's urine tested positive for cocaine when she delivered her son in October 1989.19 As with Griffin, Powell was not provided with any documents related to the Policy and was searched under the Policy before the PSA began to air. Furthermore, there is no evidence that Powell was aware of the Policy prior to the search of her urine. We accordingly conclude that the evidence was insufficient for a rational jury to find that Powell had knowledge of the Policy.

Singleton first tested positive for cocaine in connection with the birth of a child in November 1989. When confronted with the positive test, Singleton informed Nurse Brown that she had given permission for the urine screen "because she thought she was clean." J.A. 1862 (internal quotation marks omitted). This statement clearly indicates a before-the-fact awareness by Singleton that her urine would be tested for cocaine. However, the question for the jury was not simply whether Singleton knew that her urine would be tested; rather, the question was whether Singleton knew that her urine would be searched for evidence of criminal conduct for law enforcement purposes. Appellees note that Singleton "was arrested after another Appellant's well-publicized arrest." Supplemental Br. of Appellees at 25. Even assuming that Singleton's knowledge of this arrest would be sufficient for a rational jury to infer her knowledge of the critical provisions of the Policy, there is no evidence that Singleton was aware of the arrest. The mere fact that an arrest was "well-publicized" provides no indication that a particular individual was aware of the publicity. We therefore conclude that there is no evidence from which a rational jury could infer Singleton's knowledge of the critical portions of the Policy.

Singleton's urine was searched again in November 1990, after she delivered a baby in an ambulance en route to MUSC. Although it is a close question, we conclude that a rational jury could infer that Singleton had knowledge of the critical aspects of the Policy when she gave this sample of her urine. A year earlier, Singleton had been arrested by law enforcement officers after her urine tested positive for cocaine. The connection between the positive urine drug screen and the arrest could not have escaped Singleton's notice, and a rational jury could determine that the connection between the urine drug screen and the arrest was sufficient to apprise Singleton that the November 1990 urine drug screen was going to be performed for law enforcement purposes.

Pear's urine was searched under the Policy during a prenatal appointment at the MUSC clinic in July 1990 and during an admission to the MUSC hospital for preterm labor in August 1990. At trial, Pear testified that she was aware, at the time of these searches, that women were being arrested based on positive urine drug screens. Specifically, Pear agreed that "before going to MUSC for ... prenatal care [she was] aware that MUSC was testing pregnant women for illegal drugs" and that "there was a policy regarding arresting pregnant women [who] used illegal drugs during . . . pregnancy." J.A. 444. This evidence is sufficient to support a finding of knowledge by a rational jury.

Hale delivered a baby at MUSC in December 1990, having received no prenatal care during her pregnancy. Hale's urine was searched and tested positive for cocaine. There is no evidence in the record from which a rational jury could infer that Hale had knowledge of the Policy.

Ferguson requested a transfer to the MUSC clinic after viewing the PSA. Accordingly, we must consider whether Ferguson's viewing of the PSA is evidence from which a rational jury could infer knowledge of the critical aspects of the Policy. We conclude that it is not. The PSA simply informed viewers that cocaine was harmful to fetuses and that MUSC would provide help to pregnant women who used cocaine. Nothing in the PSA provided any notice that the "help" provided by MUSC involved searches of patients' urine for evidence of criminal activity for law enforcement purposes.

According to Nurse Brown's trial testimony, see supra note 6, Ferguson would have received the "To Our Patients" letter before her urine was searched. We determine, however, that no rational jury could infer that Ferguson gained knowledge of the critical aspects of the Policy from reading this document. As discussed previously, in order to be considered informed of the critical aspects of the Policy, Appellants had to understand that MUSC personnel were working with and for law enforcement in testing Appellants' urine for cocaine. Not only does the "To Our Patients" letter not impart such information, its text — which informed readers that law enforcement authorities would become involved only after continued drug-use problems — actively negates any such reading. In any event, there is no evidence that Ferguson read the "To Our Patients" letter or, having read it, that she came away with the necessary understanding of the Policy. Any such finding by a jury would be speculative.

After Ferguson's urine tested positive for cocaine on June 19, 1991, she viewed the video on substance abuse and signed the Solicitor's letter. In August, when Ferguson delivered her child, she was again searched pursuant to the Policy, raising the question of whether the video or the Solicitor's letter imparted to Ferguson knowledge of the critical aspects of the Policy. Clearly, no such knowledge was provided by the video, which was simply a general production regarding the dangers of prenatal drug use and which included no discussion of the Policy. And, similar to the "To Our Patients" letter, the Solicitor's letter provided no information regarding the coordinated efforts of MUSC and law enforcement personnel. Certainly, a reader could infer from the Solicitor's letter that positive urine drug screens would be reported to law enforcement authorities. Critically, however, the Solicitor's letter does not inform the reader that the urine drug screen which prompted the letter had been a search conducted in conjunction with law enforcement personnel or, even more importantly, that any subsequent urine drug screen would be done for law enforcement purposes. Rather, the only reference to the involvement of law enforcement personnel concerns the consequences of a patient's failure to attend substance abuse counseling or prenatal care appointments. Accordingly, neither of these items provided evidence from which a rational jury could infer Ferguson's knowledge of the Policy.

Like Ferguson, Joseph also saw the PSA during her pregnancy. For the same reasons that apply to Ferguson, we conclude that Joseph's viewing of the PSA did not provide her with knowledge of the critical aspects of the Policy. The same is true of Joseph's later review of the Solicitor's letter. And, there is no other evidence from which a rational jury could infer that Joseph had knowledge of the critical aspects of the Policy.20

Like Ferguson, Williams requested prenatal care at the MUSC clinic after learning of the Policy. According to her testimony at trial, Williams' understanding of the Policy did not include any awareness that MUSC was searching patients' urine for evidence of criminal activity for law enforcement purposes. And, there is no other evidence from which a rational jury could infer that Williams had knowledge of the critical aspects of the Policy.

For the reasons set forth above, we conclude that a rational jury could have found the requisite knowledge of the Policy only as to Singleton and Pear. As to the other Appellants, the evidence does not support a finding of knowledge.

In addition to knowledge, Appellees also bore the burden of proving voluntary conduct — as earlier stated, a burden they assumed throughout trial and on appeal. The conduct of Singleton and Pear particularly, and the remaining Appellants generally, in presenting themselves to MUSC for treatment cannot be considered "voluntary" in a constitutional sense.

As noted previously, one of the critical factors in assessing voluntariness under the totality of the circumstances is "the possibly vulnerable subjective state of the person who consents." Schneckloth, 412 U.S. at 229, 93 S. Ct. 2041. Singleton was searched for the second time immediately after she had given birth in an ambulance. The second search of Pear was performed when she was in preterm labor. Beyond these specific searches, we note that of the 13 searches discussed in the previous section, six (including the two searches of Singleton) were performed on patients who were in active labor or who had given birth immediately prior to the search (in two cases, by emergency cesarian section). Another three searches (including the second search of Pear) were performed on patients experiencing preterm labor, and two searches — the June 1991 searches of Theresa Joseph — were performed on a patient who was seriously ill and in excruciating pain.

Medical distress may create a vulnerable subjective state that is inimical to voluntary consent in two ways. First, a patient who is in dire need of medical treatment will feel less free to question or refuse certain portions of that treatment, even if she is physically capable of doing so. Second, the physical strain of labor, birth, or serious illness will have a deleterious effect on the patient's mental process, limiting her ability to rationally consider whatever choices she has. While it is true that physical distress does not invariably vitiate consent, see United States v. Mason, 966 F.2d 1488, 1494 (D.C.Cir.1992), Appellees assumed and thus bore the burden of proving that despite their physical condition, Appellants possessed sufficient mental faculties and awareness to allow them to give informed consent. No evidence bearing on this issue was presented by Appellees.

More generally, we note that Appellants were all insured by Medicaid and that MUSC was the only medical system in the Charleston area that accepted Medicaid as a form of payment. Furthermore, the MUSC hospital and clinic were responsible for treating high-risk pregnancies, into which category most or all of Appellants fell. Thus, if Appellants wished to receive prenatal care and assistance at delivery, the record conclusively demonstrates that they had no choice but to seek such services at MUSC. And, Appellees presented no evidence that Appellants would have been treated if they had refused to provide a urine sample. While we cannot imagine that such treatment would have been denied, especially for those patients in medical distress, this is a matter on which Appellees voluntarily assumed the burden of proof. Appellees' failure to present evidence on this point requires us to assume that Appellants did not have the option of refusing to be searched and still obtaining medical treatment. The choice to be searched or forego necessary medical treatment "is the antithesis of free choice" to consent or refuse.21 Garrity v. New Jersey, 385 U.S. 493, 87 S. Ct. 616, 17 L. Ed. 2d 562 (1967); cf. id. at 496-97, 87 S. Ct. 616 (holding that choice between forfeiture of Fifth Amendment right to remain silent and loss of employment rendered confession involuntary).

This voluntariness inquiry is particularly important as to Singleton and Pear, the only two Appellants as to whom a rational jury could find knowledge of the critical aspects of the Policy. Singleton was in significant physical distress at the time of the November 1990 search (at which time she had knowledge of the critical aspects of the policy), which was performed very shortly after she had given birth. The second search of Pear occurred when she was in preterm labor. There is no evidence in the record to indicate that Singleton and Pear were in a condition to voluntarily consent to a search. And, even though Pear was not in any physical distress the first time her urine was searched, Appellees failed to present evidence that would have allowed a jury to find that she could have received prenatal care without submitting to a search of her urine, or that she could have foregone such care altogether. We therefore conclude that even though Singleton (at the time of the November 1990 search) and Pear possessed the requisite knowledge of the policy, a rational jury could not find voluntariness.

It may be the case, as Justice Scalia suggested, that the outcome of this appeal is evidence "that no good deed goes unpunished." Ferguson II, 532 U.S. at 104, 121 S. Ct. 1281 (Scalia, J., dissenting). It is certainly disheartening that, as a result of our holding today, damages may be imposed on those who acted with the best interest of Appellants and their children at heart. But, however noble Appellees' intentions, the regrettable and inescapable conclusion remains that Appellees did not bear their burden of proving that they obtained constitutionally valid consent.22 We therefore must reverse the denial of the motion for judgment as a matter of law as to all Appellants except Ellen Knight, and remand for further proceedings. Because Knight suffered no constitutional violation, we affirm as to her. On remand, the district court should conduct proceedings regarding Darlene Nicholson's standing and Appellants' damages.

AFFIRMED IN PART, REVERSED AND REMANDED IN PART

NIEMEYER, Circuit Judge, concurring in the judgment in part and dissenting in part:

Under the limited review that we are now directed to conduct by the remand order of the Supreme Court and that is required by the deferential standard applicable to that review — taking the facts in a light most favorable to the Medical University of South Carolina — I would affirm the jury's verdict. Even if I were to apply the new legal standard for consent articulated in dictum by the Supreme Court in its remand order, I would affirm with respect to nine of the appellants, either because the facts support a finding that the search under Policy M 7 was consented to with full knowledge or because the appellant's claim did not implicate the Policy. Under this new standard, I would reverse the judgment only as to Laverne Singleton because of an insufficiency of evidence to demonstrate her knowledge of the consequences of her voluntarily supplying a urine sample. My reasons follow.

* The ten women who are plaintiffs in this action contend that Policy M-7 of the Medical University of South Carolina ("MUSC") — providing that urine samples, given by certain pregnant women, be screened for the presence of illicit drugs and that the results be forwarded to law-enforcement authorities for prosecution — violated their right against unreasonable searches, as protected by the Fourth Amendment. The district court rejected MUSC's claim that the purported searches were justified by the "special needs" exception to the Fourth Amendment, but, following a trial, a jury found that each of the ten plaintiffs consented to the searches, and therefore the Fourth Amendment was not implicated.

On appeal, we affirmed, but not on the issue of consent. We concluded that the searches themselves were reasonable under the "special needs" doctrine, which justifies certain searches designed to serve non-law-enforcement ends — in this case the medical interests of the mothers and the babies — even though law-enforcement means were employed. Ferguson v. City of Charleston, 186 F.3d 469, 477 n. 7, 479 (4th Cir.1999); see also Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 652-53, 115 S. Ct. 2386, 132 L. Ed. 2d 564 (1995); Mich. Dep't of State Police v. Sitz, 496 U.S. 444, 451-55, 110 S. Ct. 2481, 110 L. Ed. 2d 412 (1990); Griffin v. Wisconsin, 483 U.S. 868, 873, 107 S. Ct. 3164, 97 L. Ed. 2d 709 (1987). We grounded our holding on the conclusion that the urine drug screens conducted under the Policy were, in fact, for medical purposes wholly independent of the incidental law-enforcement efforts and that the law-enforcement efforts were intended to reinforce the medical purposes. Ferguson, 186 F.3d at 477.

On review by certiorari, the Supreme Court reversed our judgment and remanded the case for further proceedings. Ferguson v. City of Charleston, 532 U.S. 67, 86, 121 S. Ct. 1281, 149 L. Ed. 2d 205 (2001). The only issue decided by the Supreme Court was whether the MUSC policy was a search justified by the "special needs" doctrine. On this issue, the Court concluded:

While the ultimate goal of the program may well have been to get the women in question into substance abuse treatment and off of drugs, the immediate objective of the searches was to generate evidence for law enforcement purposes in order to reach that goal. The threat of law enforcement may ultimately have been intended as a means to an end, but the direct and primary purpose of MUSC's policy was to ensure the use of those means. In our opinion, this distinction is critical.... Given the primary purpose of the Charleston program, which was to use the threat of arrest and prosecution in order to force women into treatment, and given the extensive involvement of law enforcement officials at every stage of the policy, this case simply does not fit within the closely guarded category of "special needs."

Id. at 82-84, 121 S. Ct. 1281. Distinguishing its earlier cases in which it justified drug testing, the Court said that "[i]n each of those earlier cases, the `special need' that was advanced as a justification for the absence of a warrant or individualized suspicion was one divorced from the State's general interest in law enforcement." Id. at 79, 121 S. Ct. 1281. The Court then reiterated its basis for concluding differently in this case, stating that "the central and indispensable feature of the policy from its inception was the use of law enforcement to coerce the patients into substance abuse treatment." Id. at 80, 121 S. Ct. 1281.

The Court explicitly did not review the question of whether the patients gave their consent to the searches, as the jury had found. Rather, on this point, the Court said, "[W]e necessarily assume for purposes of our decision — as did the Court of Appeals — that the searches were conducted without the informed consent of the patients." Id. at 76, 121 S. Ct. 1281; see also id. at 77, 121 S. Ct. 1281. After reversing on the one issue that it did decide, the Court remanded the case for our review on the previously unreviewed question of whether the jury had evidence to support its finding of consent. As dictum, the Court explained that hospital employees have "a special obligation to make sure that the patients are fully informed about their constitutional rights, as standards of knowing waiver require." Id. at 85, 121 S. Ct. 1281. In making that statement, however, the Court did not review its jurisprudence of consent, nor did it appear to be adopting a new standard for giving consent and thereby overruling the pre-existing law. Under the pre-existing law, consent to waive a Fourth Amendment right did not depend on a knowing and intelligent decision or on full or accurate information but rather whether the defendant "voluntarily" provided the information, regardless of whether the act was knowing or intelligent or whether the inducement was complete, true, or accurate. See, e.g., Schneckloth v. Bustamonte, 412 U.S. 218, 235-46, 93 S. Ct. 2041, 36 L. Ed. 2d 854 (1973); Hoffa v. United States, 385 U.S. 293, 300-02, 87 S. Ct. 408, 17 L. Ed. 2d 374 (1966). Were it otherwise, the procedures of law enforcement would have been dramatically changed, and doubt would be cast upon virtually every consent that is routinely given today without full knowledge of the consequences. See Schneckloth, 412 U.S. at 243, 93 S. Ct. 2041 (noting that it would be "next to impossible" to apply such a standard).

If mere "voluntariness" remains the correct standard for waiving a governmental search, then it is beyond dispute that the jury had, in each instance in this case, sufficient evidence to support its finding of voluntariness. There is no evidence that any patient was compelled to provide a urine sample or that any patient objected to providing one when requested to do so by medical authorities. And the majority agrees with this observation. See ante at 390 n. 8. Indeed, even before adoption of Policy M-7, it was established medical protocol to obtain urine samples for medical purposes when the patient presented signs that she may have been on drugs during her pregnancy. As testified to by Nurse Shirley Brown, the Obstetrics Case Manager at MUSC, MUSC had been following such a medical protocol before adoption of Policy M-7, and the adoption of Policy M-7 in October and November of 1989 was only intended to put teeth into the pre-existing protocol by systematically subjecting the patient to the possibility of prosecution for the abuse of drugs revealed by the drug screen tests.

But even if the standard suggested by the Supreme Court by dictum turns out unwittingly to be a new standard, the evidence, in my judgment, supports the jury's finding of consent in favor of MUSC in at least five of the instances presented. Under this "new" standard, the inquiry stated by the majority is the appropriate one: (1) did the patient have knowledge that the urine screens could be used as a basis for arresting the patient and (2) did the patient, having that knowledge, provide the urine samples voluntarily. Because there is no evidence that the urine samples were coerced, the only factual question is whether the patients provided those samples with knowledge that the test results might be used for prosecutorial purposes.

In reviewing the jury's findings in this case, it is important first to understand the law that the jury was applying. The district court, after telling the jury that there were no search warrants in this case, instructed the jury that any finding of a constitutional search depended on the patient's consent. The court said:

There not being a warrant issued, [the searches] are unreasonable and in violation of the Constitution of the United States, unless the defendants have shown by the greater weight or preponderance of the evidence that the plaintiffs consented to those searches.

* * *

[I]t is conceded by the parties that if the searches were consented to then they are constitutional. And again you may reach a different conclusion as to consent from plaintiff to plaintiff. That doesn't have to be so, that's dictated by the facts. But it doesn't mean that if there is consent in one case that there is necessarily consent in the other or vice versa. Again, you have to look at the facts that relate to each plaintiff and reach a decision on the question of consent.

After telling the jury that the State carried the burden of proving consent, the court then proceeded to explain how the jury was to conclude whether each defendant consented:

The matter of consent more often than not is discussed in terms of voluntariness and willingness to give the consent. But before there can be any voluntariness, before there can be any consent, there must be knowledge. The person giving the consent must have knowledge of what she is doing. It's basic that before you can consent to something, you have got to know what it is. You have to have enough information to know what is being done before you can consent to it. And the scope of the consent must be broad enough to include the search in question. Consent for one purpose may be broad enough to cover one actor, but may not be broad enough to cover another actor.

The written consent in this case may be sufficient to cover the taking of a urine sample and the testing thereof by Medical University of South Carolina officials alone, but it is not sufficient to cover the sharing of that information with law enforcement officials. That written consent is not sufficient consent to warrant a search where the search information is furnished to law enforcement officers. There must be something in addition to that written consent before you can say that these were consent searches.

* * *

And therefore to show that there is a valid consent in this case there must be something else that convinces you that the plaintiffs consented not only to the taking of this urine for the use by the medical people, but also the sharing of it with law enforcement people for the possibility of prosecution for drug offenses.

(Emphasis added).

Thus the district court instructed the jury to decide both whether the plaintiff—patients had knowledge that their urine screens could be used to prosecute them and whether their giving urine samples, in light of this knowledge, was voluntary. And applying these instructions to the facts, the jury found that each of the appellants knowingly and voluntarily consented to the searches.

Recognizing that the jury was not misinstructed, even under the "new" standard, we need only concern ourselves with the question of whether there was evidence from which the jury could reasonably have reached the conclusion that consent was in fact given. Benesh v. Amphenol Corp. (In re Wildewood Litig.), 52 F.3d 499, 502 (4th Cir.1995). In applying this standard of review, we take the evidence in a light most favorable to MUSC. Id.

The record shows that before Policy M-7 was adopted in mid-October or November 1989, MUSC utilized a medical protocol solely for medical reasons to test urine samples of pregnant women who presented themselves with certain medical conditions. Nurse Brown testified that, under this pre-Policy medical protocol, "if people met certain criteria that were listed that were published by the faculty, you know, to go out into the clinics that the following are indicators and would need to be screened, you know, for medical evaluation, then they would have [been screened]." She explained that the urine screen for drugs would be prescribed under the protocol if the patient had "no prenatal care, late prenatal care, abruptio placentae, intrauterine fetal death, preterm labor, and there was one other one that was supposed to be screened prenatally or at delivery." This protocol was adhered to by all physicians and was communicated to medical residents by memorandum. Thus, when a patient was abrupting, "the placenta was separating, [and] it was putting her life and [the] baby's life in jeopardy, ... the physician order[ed] a [urine drug] test as part of many tests that they order." Under the protocol, if the drug screen proved positive for drug abuse, the woman was counseled about the harmful effects of drugs on her and her child and was referred to substance abuse treatment. No one has argued that this pre-Policy medical protocol involved an unconstitutional search.

Policy M-7, the subject of this case, was agreed upon by MUSC, the police, and the solicitor (chief prosecuting attorney) on or about October 12, 1989, when the police department sent Nurse Brown a proposed form of the Policy to which the various entities had agreed orally. This communication was followed by a memorandum dated October 17 actually announcing the "policies concerning drug abusing pregnant women." And, thereafter, Policy M-7 was formally adopted as part of the MUSC medical center policy manual on November 27, 1989. As announced and informally adopted, Policy M-7 included a form solicitor's letter dated October 18, 1989, addressed to patients and candidly warning them of their potential arrest: "Please understand that by using drugs during pregnancy, you are risking death or at least severe long-term harmful effects to your baby. If you fail to obtain Substance Abuse and Pre Natal Care, you will be arrested by Charleston City Police and prosecuted by the Office of Solicitor." The date of the solicitor's letter, which was to be employed as part of the Policy, confirms when Policy M-7 was informally implemented, as does the testimony of Nurse Brown. Summarizing her testimony, the district judge stated, "She has testified, as I understand it, to two things. She said late October of '89 or the first part of November of '89 the policy was first adopted," and the plaintiffs' attorney agreed with that summary. (J.A. 587.) In short, on or about October 17 or 18, 1989, Policy M-7 was first adopted, albeit informally, and beginning then, it authorized drug screening, the results of which could be used to support arrests of pregnant women.

A modification of Policy M-7, providing a second chance to patients testing positive for drugs if they successfully completed treatment, was adopted in January 1990. And during this period, a public service announcement was developed and broadcast publicly and privately to patients. That announcement stated:

When you're pregnant, just one line of cocaine, a single hit of crack, rushes to your baby's body and brain. Within minutes your body can be jolted into premature labor risking a still developing child to stroke, even death. And not only will you live with the guilt, you could be arrested.

But this is a tragedy you can prevent. If you have a problem with drugs talk to your doctor or call MUSC at 792-6437. Trained counselors will guide you through drug rehabilitation and advise you about good prenatal care. And if you stay with the program, you will not be arrested or prosecuted.

Wake up from the nightmare. Think about your baby first.

It is against these facts about when Policy M 7 was adopted and implemented that we must consider the personal circumstances of the appellants in this case.

The record evidence shows that Lori Griffin and Sandra Powell each gave urine samples to MUSC under MUSC's medical treatment protocol as it existed before adoption of Policy M-7 and thus the screens were not searches implicating the Fourth Amendment. Griffin signed an ambulatory consent form on June 28, 1989, then gave a urine sample, and tested positive for cocaine. On October 7, 1989, again before Policy M-7 was adopted, Griffin went to MUSC for pre-term labor. During this admission, she gave another urine sample that tested positive for cocaine. As she was released from the hospital, she was arrested on the basis of the positive showing and remained in custody until she delivered her baby on October 26. In January 1990, Griffin was indicted for possession of cocaine and distribution of cocaine to a minor on October 6 and 7, 1989, as revealed by medical testing conducted before the implementation of Policy M-7.

Powell similarly signed an ambulatory consent form on May 24, 1989, when she came to MUSC for prenatal care before Policy M-7 came into effect. On October 13, 1989, she was taken to MUSC in an ambulance when she went into labor, and at the time of that admission, a urine sample was provided. It tested positive for cocaine, and the next day, Powell was arrested on the basis of that medical record for her cocaine abuse. Again, all of this occurred before adoption of Policy M-7.

With respect to each of these patients, I submit that, and certainly the jury could have found that, the urine tests were conducted pursuant to the medical protocol in existence before Policy M-7 and that the urine samples were voluntarily provided for medical reasons. No appellant has contended that the medical protocol as it existed before the adoption of Policy M-7 in mid-October and early November 1989 was conducted for prosecutorial purposes and thereby violated the Fourth Amendment.

With respect to the urine samples provided by Pamela Pear, Paula Hale, Crystal Ferguson, Theresa Joseph, and Patricia Williams, Policy M-7 was fully in effect. But in each case, when a sample was taken and tested, the prosecutorial potential was explained to the patient. Indeed, the majority opinion agrees that Pear was aware that MUSC was testing pregnant women for illegal drugs and that she knew that MUSC had a policy regarding arresting women who used illegal drugs during pregnancy. Hale similarly had full knowledge of the Policy. She came to MUSC on December 13, 1990, and delivered a baby. Urine samples provided at her delivery tested positive for cocaine. She signed a copy of the solicitor's letter specifically warning of arrest and has admitted to knowing at the time that drug screens at MUSC were being used to support arrests of pregnant women. And Ferguson likewise had full knowledge. She came to MUSC because she was using drugs and needed help. She signed the consent form, viewed the public service announcement warning of dangers of using drugs and warning that she could be arrested, and signed the solicitor's letter, which specifically warned patients that they could be arrested. Again, Joseph also saw the public service announcement and, on presenting herself to MUSC, signed a consent form. In addition, she signed the solicitor's letter specifically warning her of the potential for arrest. Finally, Williams specifically asked to be transferred from North Charleston Health Department to MUSC for prenatal care because of her drug problem. She had heard about the MUSC policy and admitted knowing that MUSC was testing women for the presence of drugs. When she presented herself at MUSC, she signed the consent form and tested positive. She was then shown the video on substance abuse, given a copy of the solicitor's letter and scheduled for an appointment at the Charleston County Substance Abuse Center. With this information, Williams gave urine samples on two occasions thereafter, as well as an additional time at the delivery of her baby. On each of these three occasions, she tested positive for drug abuse. Thus, with respect to each of these five appellants, the record amply supports the jury's conclusion that the patients were fully informed of prosecutorial purposes when they provided their urine samples for drug testing.

The facts relating to Laverne Singleton are distinguishable from these five and do not support a finding that she consented to use of test results for prosecutorial purposes. Singleton delivered two children through the medical assistance of MUSC. In connection with her first child, delivered on November 9, 1989, a postpartum urine screen tested positive for cocaine. She had received no prenatal services at the obstetrics clinic. She admitted at the time that she consented to the urine drug screen because she "thought she was clean." Based on the November 9 test, she was arrested. During the course of a second pregnancy in 1990, she had two positive drug screens. When she was admitted for delivery in November 1990, she again tested positive for cocaine. As an alternative to arrest for drug abuse, Singleton voluntarily admitted herself to inpatient treatment at the MUSC Institute of Psychiatry. While her arrest in 1989 shows her knowledge of the Policy when she submitted samples with respect to her second pregnancy and delivery in 1990, the record fails to support a finding that she had knowledge of the Policy or of prosecutorial purposes when she was tested in 1989. Accordingly, under the "new" standard under which we are now considering Singleton's claim, I would conclude that a reasonable jury could not find that Singleton consented to the search in 1989 that led to her arrest.

With respect to Ellen Knight, the majority opinion concludes that she does not have standing to assert a claim because only her child's urine was tested. I agree. And with respect to Darlene Nicholson, the majority opinion concludes that she was tested after the Policy had been discontinued. While the majority opinion would remand Nicholson's claim to determine whether in fact she was searched under the Policy, I respectfully submit that the burden is on Nicholson to demonstrate that she was searched pursuant to the Policy, and she has failed to carry that burden.

In adopting, modifying, and implementing Policy M-7, MUSC has made a serious effort to address a serious problem — a problem that exists not only in South Carolina but also nationwide. Drug abuse by pregnant women harms the woman as well as the fetus that she carries, permanently injuring the fetus and imposing on the baby a life-time of problems from birth. Far beyond the problems of the mother and child are the enormous social problems borne by the community. Even as well-intended as Policy M-7 may be, it must nonetheless comport with constitutional constraints. The Supreme Court has concluded that coupling the medically necessary drug testing of women who present evidence of drug abuse with the concomitant purpose of prosecuting them based on the results of the drug tests is not justified by a "special needs" exception to the Fourth Amendment. The Court held that only if such testing were, as a factual matter, consented to could such a search be made, and on the consent question, the Court committed the issue to us to resolve.

I respectfully submit that the record amply supports the jury's finding that all of the women voluntarily supplied urine samples for testing. And, if we must require an informed consent, the record amply supports the jury's finding that five of the women — Pamela Pear, Paula Hale, Crystal Ferguson, Theresa Joseph, and Patricia Williams — knowingly consented to the searches and that four other women do not have claims because three — Lori Griffin, Sandra Powell, and Darlene Nicholson — submitted urine samples under a purely medical protocol, and Ellen Knight was never personally tested under the Policy. Only as to Laverne Singleton does the evidence fail to support the district court's judgment under this standard.

Accordingly, on remand from the Supreme Court, I would affirm, or, alternately, I would reverse the judgment of the district court as to Singleton and affirm as to the other appellants.