The lines between law enforcement and health care are blurring. Police increasingly lean on doctors to provide them with genetic samples, prescription histories, and toxicology results that they could not obtain on their own. This often occurs without a warrant or the patient’s consent. At the same time, legislatures are using physicians as regulatory levers to police pregnant and transgender bodies. And due to chronic underfunding of social services, many Americans now receive pseudo-mental health treatment through the courts rather than clinics. Together, these things paint a sinister picture of law enforcement being thrust into medicine in ways that are deeply troubling and vastly underexplored.
In this Article, I reveal how the blurring of law enforcement and health care undermines core principles of medical ethics—such as privacy, accountability, efficacy, honesty, and autonomy—in the service of diametrically opposed prosecutorial goals. I walk through various examples of law enforcement’s co-option of medicine. These examples expose fundamental conflicts between the two professions that support keeping them far apart.
Scholars have done excellent work documenting the overpolicing of emergency rooms. This Article expands the analysis to a wider set of clinical domains and ratchets up the urgency by establishing police involvement in health care as a social determinant of health. When doctors become cops, this does violence to the physician-patient relationship and the (disappearing) trust on which it relies. This in turn causes demonstrable negative health outcomes. Not coincidentally, medical mistrust is highest in communities that are poor, racially diverse, and overpoliced. Thus, to improve the physician-patient relationship and promote health, I provide a few suggestions for how the line between law enforcement and health care may be better enforced.
INTRODUCTION
A. Abortion Restrictions Intensify Police Presence into Health Care
In the wake of the Supreme Court’s decision in Dobbs v. Jackson Women’s Health, 597 U.S. 215 (2022), legislatures in conservative states have significantly restricted access to abortion. One such law is Utah’s SB 174, which was written to go into effect as soon as Roe v. Wade, 410 U.S. 113 (1973), was overturned. This law criminalizes the provision of abortion from implantation in all but three narrow circumstances. One of the three circumstances is if the woman was sexually assaulted.
However, it is not enough that the patient confide in her doctor that she was raped. Before terminating a pregnancy, physicians are required to verify that the rape has been reported to law enforcement. This same provision exists in other states. There is no legislative guidance on the procedure to be followed to protect the confidentiality of the patient’s medical records or the patient-provider privilege. This sort of conspicuous and ill-conceived police intrusion into medical decision-making is both exceptionally troubling and on the rise.
Such laws interfere with patient care by thrusting police into decisions that should be exclusively between patients and their doctors. Rape survivors frequently choose not to report their assault out of concerns that it will revictimize them without leading to a conviction. Under the law of several states, this rational, personal decision will now block these patients from accessing safe medical interventions. Complying with these laws puts physicians in a bind—do they fulfill their ethical obligations, or their legal ones? More importantly, such laws force patients to have their medical needs subordinated to larger prosecutorial goals. Prioritizing the state’s interest in prosecution in this way violates long-standing norms of patient autonomy and medical ethics.
The criminalization of abortion brings law enforcement’s conscription of physicians into agonizingly sharp relief. But this conscription goes well beyond reproductive care and operates on two axes. The first axis involves legislators, acting through the police, leaning on nurses and physicians to enforce moral codes that are only tangentially related to medicine. Examples include not just bans on abortion or gender-affirming care, but the implementation of prescription drug monitoring programs and the provision of addiction or mental health treatments through the carceral system. The second axis of police encroachment into health care is more historic and familiar. It involves cops asking hospital staff to provide them with biological samples that they then use to investigate drug prosecutions, car accidents, murders, and sexual assaults. Everywhere you look, it seems, cops are operating on these two axes to blur together the provision of health care with the prosecution of crime.
While these police encroachments target diverse kinds of data and endanger different aspects of health care, they each do violence to the physician-patient relationship and the trust on which it relies. They take what makes health care so unique and special—the confidentiality and ethical obligations the physicians have toward their patients—and exploit this for prosecutorial ends. The physician-patient relationship is not just some precious ideal that we need to protect out of respect for old-fashioned values. Substantial research now demonstrates that when we undermine the trust patients have in physicians, we do significant damage to population and individual health.
In this Article, I explore how the blurring of health care and law enforcement violates fundamental principles of medical ethics. But I will also reveal how shaky and aspirational these core ethical principles are, and how far we are from realizing them. I establish medical mistrust, which is exacerbated by police intrusion, as a social determinant of health (“SDOH”) that has been massively underexplored. Not coincidentally, medical mistrust leads to poorer health outcomes and is much higher in communities that are poor, racially diverse, and overpoliced. This exacerbates health inequities, as Latinx and Black people have a long history of receiving inferior care, despite having comparable insurance and access. To repair mistrust to improve patient outcomes and respect for autonomy, physicians cannot be perceived to be aligned with policing—an institution that is oppressive and surveillance-oriented.
To justify my thesis—that law enforcement and health care need to be more completely divorced from one another—the Article will proceed in the following way. In the first Part, I will reveal how abortion restrictions are just the tip of the iceberg. In the second Part, I will describe how law enforcement easily gains access to confidential medical data. In the third Part, I establish medical mistrust as a SDOH. Unfortunately, when the police lean on physicians to do their investigative work, this has the potential to annihilate the struggling physician-patient relationship and to lead to poorer health outcomes. In the fourth Part, I provide several illustrations of how law enforcement is meddling inappropriately in health care. These examples reveal not just how pervasive the meddling has become, but also how fundamental the differences are between the two cultures’ ethics. While medicine aspires toward self-regulation, privacy, accountability, efficacy, honesty, and respect for autonomy, law enforcement resists these values. This matters because the co-option of health care by the police undermines principles of medical ethics in the service of diametrically opposed prosecutorial norms. To respect patient autonomy, repair medical mistrust, and promote individual and public health, I will conclude by suggesting a few concrete legal reforms that may better wall health care off from law enforcement.
B. Ubiquitous Police Involvement in Health Care
The post-Dobbs abortion restrictions give us a unique sense of urgency regarding the co-option of health care by law enforcement. However, abortion is just the tip of the iceberg. In many other domains, police are reaching into medical databases and spaces to gain access to information that they cannot obtain on their own.
The reason for the high quantity of police meddling in health care is two-fold. First, physicians are risk-averse and already highly regulated. Attaching threats of criminal penalties to their noncompliance with state laws will deter many of them, and it is easy to add new conditions to the renewal of their medical licenses. This is the “easy deterrence” rationale. Second, cops are constantly looking for ways to improve their crime clearance rates, or the percentage of crimes that are solved. The value of medical and biological evidence to this pursuit is enormous, and this is something hospitals have in spades. This is the “easy access” rationale. Hospitals and physicians are thus both a convenient regulatory lever and a reservoir of evidence.
Astute observers have written about how the police have forced their way into emergency rooms in poor, urban neighborhoods. In one study, close to 80% of patients who had cops transport them to or visit them in the emergency department were Black. Police omnipresence has been shown to compromise clinical care by, for example, prioritizing the interrogation of trauma patients when they need life-saving interventions. Additionally, their swiping of confidential patient data in emergency departments—sometimes without a warrant or consent—appears to be rampant, and increasing, with very little oversight. Overpolicing in urban emergency rooms is a harrowing example of cops injecting themselves into health care in ways that frustrate patient autonomy and clinical care.
But it goes beyond this phenomenon too. To the surprise of many, police have been quietly conducting warrantless searches of various genetic databases. Physicians have also been asked to take on the mantle of law enforcement to limit patient access to not just abortion, but also opioid medications and gender-affirming treatments. Police administer fatal doses of ketamine to patients they diagnose as having “excited delirium” despite this diagnosis being highly contested in the medical community. Counties are expanding their “treatment courts” to provide substandard mental health and addiction treatment, with threats of incarceration for noncompliance.
In many domains, law enforcement is co-opting the practice of medicine. I will go into greater detail for each example below, as they each illustrate different violations of medical ethics norms. However, in the following Section, I will introduce the problems with community distrust of police. I will then compare this to medical mistrust and explore how the deep roots of medical mistrust are an important SDOH.
1. Police Mistrust Is at Record Highs
The public’s trust in the police is at an all-time low. This is related, no doubt, to the fact that in the last year alone, The Washington Post reported that over 1,100 people have been shot and killed by the police. Shockingly, one in twenty homicides in the U.S. are committed by law enforcement. In 2022, 11% of police killings involved no alleged offense by the deceased, 9% involved mental health or welfare checks, and 8% involved traffic violations. One need not linger on these data to understand why the police in the United States spark fear of violence. People should obviously not be killed for being mentally ill or for running a red light.
Now that there is more widespread video recording of officers “as they engage in violence against Black citizens . . . the world has begun to see the human rights violations” committed against Black people by police. In the wake of the police murders of George Floyd and Breonna Taylor, almost ninety percent (88%) of Black Americans say major changes are needed to make policing more accountable. This is compared with 63% of Hispanic Americans and 51% of White Americans who likewise desire greater police accountability. And since the widespread coverage of these incidents, even more examples of police brutality have impaired community trust and reduced incentives to cooperate with the police. Awareness of police brutality has been shown to result in fewer calls to 911 and less information-sharing with the police.
2. Structural Racism in Policing
A large body of scientific research reveals structural racism in “virtually all aspects of the criminal legal system,” with Black Americans “experiencing harsher outcomes in relation to police encounters, bail setting, sentence length, and capital punishment than White people.” While people’s trust in the police varies greatly based on their race, in general, the trust people have in police has taken a huge hit in the last few years.
Researchers have begun to make the connection between police brutality and unmet medical needs. A proposed mechanism for this is impaired medical trust. This Article proposes a direct link between policing and inequitable health outcomes, because patients perceive medicine as aligned with law enforcement. When doctors work too closely with cops, health care is not just seen as affected by the police—it is the police. This takes medicine in the opposite direction from where it needs to go: to correct health inequities. Because trust is so vital and diminishing in health care, we must do more to ensure that the violent and patronizing cop culture does not infect the clinic.
I. MEDICAL MISTRUST IS A SOCIAL DETERMINANT OF HEALTH
A. The Physician-Patient Relationship Is Suffering
Trust is a “vital aspect of clinical care” with mistrust being “a major barrier to a strong patient-clinician relationship.” Unfortunately, the physician-patient relationship is not what it once was. Today, the ideal of the trustworthy, compassionate, attentive physician seems almost quaint. Most of us do not have a long-term, trusting relationship with our primary care doctor. There are many reasons for this.
First, patients and physicians have precious little time to build trust. Since the 1990s, managed care insurance organizations have put considerable pressure on physicians to move quickly from patient to patient. Seeing a high volume of patients is financially rewarded in the allocation of relative value units (“RVUs”), which is a performance metric that impacts physicians’ promotions and salaries. Additionally, doctors “face mounting demands on their time” to fill insurance authorization requests, perform utility review, train mentees, and update their recordkeeping. Patients and physicians simply do not have the time to get to know one another in the way they did decades ago.
Research shows the best way to forge a strong therapeutic alliance is for physicians to “engage more with patients by listening to what they have to say, asking questions and showing sensitivity to their emotional concerns.” But unfortunately, this is not occurring. The fee-for-service model reimburses hospitals and physicians for doing things, but not for talking about whether and how to do things. Relatedly, as medicine has become more technical and specialized, there has also been “a growing substitution” of patient self-reports for those made by machines. Why focus on what a patient is telling you, when you can run a test, bill for it, and see “objective” data for yourself? The overreliance on tests and technology may lead patients to feel like their personal accounts of their health histories do not matter.
Telemedicine may also be damaging the physician-patient relationship. Something critical is lost when our regular physicians cannot see us biting our nails, rapidly losing weight, or struggling to walk. In one study of telemedicine, physicians reported they “were very concerned about the loss of personal connections and touch, which they believed diminished expected rituals that typically strengthen physician-patient relationships.” Of course, telemedicine improves access for many rural Americans. It is not a universally bad thing. But when it comes to initiating strong, trusting bonds between physicians and patients, it is a poor substitute for in-person visits.
Given these systemic pressures, it is impressive that some physicians still take the time to get to know their patients. Though of course, not all physicians do. And data suggests that patients of color who are overweight, who have intellectual disabilities, or who use illicit drugs are more likely to feel rushed by their doctors and to experience poor physician communication.
Most patients are familiar with the feeling of being rushed by their doctor. One famous study found that on average, physicians interrupt their patients after less than twenty seconds of the patient speaking. While interruptions sometimes facilitate deeper sharing, physicians often interrupt patients in ways that discourage patients from offering additional relevant concerns. It is difficult to trust people who do not appear to listen to us.
Together, these factors contribute to a perfect storm that is weakening the physician-patient relationship. Repairing this relationship is not just a good insofar as it promotes health; it is a good on its own because it fosters informed, compassionate, and individualized care. Against this backdrop where doctors are not trusted and police are trusted even less, we need to do more to repair patient trust in health care. Allowing cops to take advantage of the trust that remains in health care to access patient data and prosecute crime does exactly the opposite. In this next Section, I will describe how medical mistrust impairs individual and public health.
B. Medical Mistrust Hurts Health
The trust between clinicians and patients does not exist in the context of an equal relationship. There is a sharp power imbalance, where the “doctor typically has knowledge that the patient lacks, and the power to order investigations and treatments.” But the asymmetry flows in the other direction as well—the patient may hold private knowledge about their symptoms, their values, and their medical history that really ought to be shared.
In theory, patients can adopt the treatment plan or not, they can terminate the relationship, and they can seek treatment elsewhere. But this power is severely limited by practical considerations. For example, many patients live in rural areas, have complicated diagnoses that require specialized care, or do not have access to affordable insurance and a range of providers. Any of these can mean that they have no real choice and are “stuck” seeing particular physicians. Because patients lack the ability to see another doctor, earning the patient’s trust is essential to making sure patients receive quality care and do not feel forced to continue a disrespectful relationship.
Researchers have documented unacceptable levels of medical mistrust, which captures the “overall suspicion of the health care system and beliefs that health care providers and organizations may act contrary to patients’ best interests.” Medical mistrust has been shown to lead to poorer health outcomes, and to “trigger nothing short of a public health crisis.” However, while understood to be “ubiquitous,” precise measurement of medical mistrust has been difficult because it encompasses many different things.
Even so, mistrust has been shown to cause patients to feel stressed about seeing their physician and report less satisfaction with their care. Medical mistrust also leads patients to refuse prescribed medications, to miss cancer screenings, to not see their doctor for regular visits, to discourage others from seeking treatment, to not share sensitive medical information with their providers, and to be less likely to comply with the prescribed treatment or health care plan. There is even some evidence that trusting your health care provider is associated with better self-reported health status. Put simply, when trust suffers, patients suffer.
Medical mistrust is exaggerated in already marginalized communities. It has been shown to discourage Native Americans, Black people, and Latinx people from receiving all kinds of care. In groups with intersectionalities, the effects of mistrust might be even more exaggerated.
For example, researchers have linked vaccine hesitancy to medical mistrust. During the COVID-19 (“COVID”) pandemic, public health researchers observed that Black and Latinx people had more “vaccine hesitancy” despite being disproportionately impacted by COVID deaths. But rather than focusing on “why people of color distrust medicine,” scholars have recently encouraged asking “how has power been abused in medicine?” and “what can institutions do to stop it?” That is, why might people of color rationally fear the COVID vaccine, based on how they have been exploited and abused in the past? The burden of rectifying mistrust cannot be borne exclusively by the disadvantaged communities themselves. This is the great insight of the SDOH framework. It allays the propensity to engage in victim blaming or assuming that the alarming state of health inequity is just.
Unfortunately, the nature of medical mistrust makes it difficult to combat. In a poll by the Association of American Medical Colleges, “six in 10 adults have concerns about sharing their social and online activity data in a healthcare setting” in part due to mistrust. While researchers often need access to race, income, and other sensitive data to measure health inequities, “when patients distrust the system, they are less likely to share self-identifying data.” This creates an unfortunate feedback loop; the very causes of medical mistrust make patients reluctant to provide the data needed to correct it. Of course, if patients know that their health data can be shared with law enforcement, they are even less likely to share their demographic data with their doctors. This will make it even harder to run racially sensitive epidemiological studies to correct health inequities.
As a key factor in racial health disparities, distrust in health care institutions is a massively underexplored SDOH. It needs to be explored in the same way as other well-known factors, such as stress, exposure to high levels of industrial chemicals, air pollution, or living in food deserts. Only by viewing the co-option of health care by law enforcement as a SDOH can we appropriately evaluate the health impacts on communities of color who already experience disproportionate health disparities.
1. Medicine’s History of Injustice
Medicine has a troubled history of policing bodies and reinforcing power structures. For too long, physicians from upper-class, white households have dismissed the stories of patients who were considered “disobedient” or “difficult” simply because they had different backgrounds or experiences. This dynamic continues today. The disparity in class, education, physical ability, and race creates a form of epistemic injustice that prioritizes the voices of patients who tell the physicians familiar stories and what they expect to hear. Add to this the fact that physicians generally have poor listening skills, and we can see why we need to “dethrone” physicians from their presumed superiority to respond to the SDOH.
Hubris and histories of racism in health care have led to justified mistrust of medical providers, especially among people with disabilities, women, Indigenous, Black, and Latinx populations. When reflecting on this, people often call to mind the unethical Tuskegee experiments, in which poor Black sharecroppers were deliberately infected with syphilis and denied available treatments. But by focusing on Tuskegee, we risk minimizing the geographic and temporal breadth of medical racism.
Before the Civil War, physicians “attributed poor health among slaves to their biologic inferiority.” They ignored the fact that servitude was toxic to health while simultaneously exploiting individuals for medical research. Historically, hospitals were intentionally racially segregated, and they continue to be underfunded in racially diverse communities. Up until the 1970s, Black and Indigenous women were formally and involuntarily sterilized by physicians. Women in detention facilities continue to be sterilized through deception and violations of informed consent. To this day, Black women are much more likely to die from childbirth because their health concerns are dismissed. In one study, half of the white physicians surveyed were found to hold false beliefs about Black people’s biological differences—such as thinking they have thicker skin or their blood coagulates more quickly. Infant mortality for Black babies is higher now than it was during the antebellum period. To this day, physicians are more than twice as likely to call the cops when treating a Black patient compared with when they are treating white patients. Reports of these disparities naturally make their way to the affected communities.
2. To Repair Medical Mistrust, Law Enforcement Must Be Walled Off from Medical Spaces and Data
At the population level, physicians are generally trusted as an institution. However, trust in the medical establishment varies significantly by class and race. People of color report lower trust in doctors and the larger health care system than do white people, likely due to the medical racism described above. Members of the LGBTQ community have also experienced discrimination and inferior health care treatment.
Compared with white participants, a multivariate analysis revealed that Black and Hispanic participants had 73% and 49% higher odds, respectively, of reporting mistrust in health care professionals. Additionally, low-income families in the U.S. report being significantly less trusting of physicians and less satisfied with their care than adults who are not from low-income families. Given the volume of data that demonstrates how mistrust impacts health outcomes, it also contributes to significant health care inequalities. By including ratings of trust from marginalized groups, we can see that for many, the health care system is already failing to behave in ways that are worthy of trust.
These datapoints tell us that structural racism and discrimination persist in health care. However, the quantitative data risks masking the lived experiences of people of color, many of whom have a personal story of being mistreated by a medical professional.
As one physician put it, “to improve relationships between doctors and disenfranchised patients[,]” physicians must “recognize that racial preconceptions continue to affect clinical practice, even if it’s not intentional.” Unfortunately, little has changed since the recognition over twenty years ago that “black Americans are less likely than whites to receive a wide range of medical services, including potentially lifesaving surgical procedures.” Black people are “dying not of exotic, incurable, poorly understood illnesses nor of genetic diseases that target only them, but rather from common ailments that are more often prevented and treated among whites than among blacks.”
Medical schools in the U.S. are working to correct this. For example, during the initial COVID lockdowns, doctors protested in solidarity with Black Lives Matter activists in grassroots events called “White Coats for Black Lives.” The American Medical Association (“AMA”) has also worked to incorporate anti-racism and anti-ableist education throughout the medical curriculum. However, they must still do more. Medicine must continue to shift its gaze to the SDOH, rather than “record[ing] politics as biological misfortune.” To trust health care systems, patients must feel heard by their doctor, trust that their doctor is being honest and doing what is in their best interest, and feel the doctor will respect their confidentiality.
Most of the metrics on trust in medicine focus on the physician-patient relationship, because it is the most important aspect of patient trust. As Ji Seon Song recognized, “[w]hen medical and police actors act in concert, there is the potential for their actions to amplify bias and discrimination.” In light of the rampant violence and pervasive lack of trust in the police, repairing medical mistrust cannot be done if law enforcement intrudes on this struggling relationship by invading medical spaces and data.
In the next Section, I will pivot to explaining how it is that police gain access to confidential health information. Given that health data is considered especially sensitive, federal statutes, state statutes, and evidence rules recognize it as privileged and confidential. However, ironically, the Privacy Rule of the Health Insurance Portability and Accountability Act (“HIPAA”), which was enacted to shield medical data from unauthorized disclosure, incorporates a number of exceptions for the police. These exceptions oddly give the state access to this highly private information.
II. HOW POLICE HAVE EASY ACCESS TO MEDICAL DATA AND SPACES
Many of the articles and cases that have analyzed whether police can co-opt health data have viewed this query exclusively through the lens of a criminal defendant’s Fourth Amendment constitutional right to be free from unreasonable searches and seizures. Fourth Amendment cases nearly always find a way to permit the unconsented-to blood draw, search of a private database, or forced involvement of physicians in policing.
The Fourth Amendment framing makes sense as an initial matter, because it provides the constitutional floor for privacy protection. However, it does not provide the ceiling. Governments can do much more to protect medical privacy than what is required by the Fourth Amendment. And by myopically focusing on constitutionality when assessing medical privacy, courts and legislatures do not take seriously the social good that comes from walling health care off from law enforcement.
This is not to say that Fourth Amendment doctrine is irrelevant; it provides a critical piece of the tapestry. I will therefore explain how it has operated to generally grant police access to private medical data. For starters, if a valid warrant is issued on a showing of probable cause, the medical search would not violate the Fourth Amendment’s protection against unreasonable searches and seizures. But even without a warrant, police access to medical records is often permitted as either not being a search, being a reasonable search, or being part of a “special need” that is incidental to law enforcement.
Even so, depending on the circumstances, police can be found to have violated a patient’s Fourth Amendment rights if they obtained private medical data without a warrant and the search was deemed unreasonable. To determine reasonableness, courts balance the government interest in obtaining the data, which is often compelling, against the “manner in which the information was gathered and maintained, and whether confidentiality is protected.” In practice, this does not limit police access too much. This is especially true when police obtain toxicology results from the emergency department to prosecute drunk driving. As Song points out, this is because emergency departments are often treated as though they are public spaces, where patients forfeit their expectations of privacy. Thus, even if the access to medical data is deemed a “search,” it will often be considered a reasonable one in light of the patients’ reduced expectations of privacy.
Given the unpredictability of the Fourth Amendment balancing test, many are concerned about the protection medical data will be afforded in the context of reproductive or gender-affirming care. At present, the case law is inconsistent in terms of whether the involuntary disclosure of private, non-emergency medical data runs afoul of the Fourth Amendment, especially when the evidence might be permitted under federal law, but not under state law. And if the courts come out differently on the reasonableness of searches in less politicized contexts of drunk driving and gunshot wounds, they will certainly be variable when dealing with prosecutions of physicians who provide highly-politicized abortion and trans care. And no matter how or why the private medical data are disclosed, once the state has access to this sensitive health data, the damage will be done. This will almost certainly chill patient access to many types of treatment.
However, to be clear, the police rarely need warrants to search in health care settings. Instead of obtaining a warrant by demonstrating probable cause, they can simply rely on subpoenas, state statutes, and HIPAA exceptions to gain warrantless access to patient medical data. Health care data has become the warrantless, backdoor, investigatory jackpot. So, how does this happen?
First, hospitals can play a quite active role in sharing patient data. The coercive prenatal drug testing program deemed unconstitutional in Ferguson v. City of Charleston, 532 U.S. 67 (2001), was a hybrid program run by both the hospital and the local police. It threatened women with criminal punishment if they did not “agree” to drug treatment. Hospitals and law enforcement conspired together to violate patient trust in what was something like a medical “sting operation.”
This program exploited vulnerable patients while generating negative public health impacts. That is, when pregnant women worry that they can be prosecuted for drug use, they are discouraged from seeking prenatal and addiction care, which then harms them and their newborns. To be clear, routine drug testing of pregnant individuals does not violate the Fourth Amendment so long as it is done initially for medical purposes and with patient consent. However, the legal consequences of the consent to the blood draw or urinalysis might not be explicit, and the pervasiveness of drug testing has led one physician to ask whether it is a “policing practice masquerading as clinical care.”
Cases like Ferguson are extreme, and hopefully rare. But the police do not need to formally run sting operations within prenatal clinics to obtain confidential medical data. Instead, they can do something much more banal. So long as the data are either incidental to law enforcement or relevant to an ongoing investigation, they can just ask nurses and physicians to provide them with it. And often, health care workers comply.
A. Cooperation by Health Care
One of the most common ways police access patient data without a warrant is to rely on nurse or physician complicity. Because the Fourth Amendment case law is messy, nurses and physicians report not being sure what the police are constitutionally allowed to do. And even when they know disclosure is neither required nor permitted, health care workers have admitted to turning over confidential patient data simply because law enforcement asked. One nurse reported how a police officer started asking investigative questions of a patient while he was being resuscitated. She knew this was not allowed. But she noted that it can be awkward to tell a loitering cop, who is armed, to leave the radiology suites, or to stop observing lifesaving treatments.
Qualitative research has documented how physicians and nurses may divulge test results, patient treatments, and patient locations to cops because they feel pressured or “cornered” by armed officers. This is especially apparent in bustling emergency departments where providers may feel safer having police around. A symbiotic relationship sometimes develops—in which the police protect staff from seemingly unruly patients and in turn the providers give the police access to confidential medical data.
The pressure that nurses report feeling to comply with police requests is warranted. In 2017, nurse Alex Wubbels was arrested after explaining to a University of Utah cop that she could not draw blood from an unconscious patient. She told the officer that because the patient was not under arrest, there was no warrant, and the patient could not consent, she could not provide a blood sample according to the institution’s policy and interpretation of the Fourth Amendment. After confirming she would not comply with his request, the cop lunged at Wubbels, forcing her into handcuffs and taking her outside while she screamed in protest. She ultimately settled a suit against the university for $500,000 for battery and false imprisonment. However, the media coverage of this event will likely deter other nurses from similarly pushing back against illegal police overreach. Police have come to feel entitled to emergency room patients, as if the hospital is just an extension of what happens out on the public street.
There is great inconsistency in how cases are treated when patients allege Fourth Amendment violations against the police having access to their medical data. However, the trend seems to be to allow law enforcement to use medical data and even obtain blood samples without a warrant. And there is sadly little accountability for the health care workers who are complicit. When providers cross the line and violate patient privacy, as Ji Seon Song points out, “[o]nly in the rare case has a doctor been taken to task by the courts for assisting police officers by performing procedures,” such as when a doctor “participated in a number of forced rectal examinations.”
We do not have good data on how often nurses and physicians turn over private medical data without a warrant. However, qualitative studies suggest that health care staff often comply with warrantless police requests to obtain biological samples or test results of patients without probable cause. In cases where patients are later prosecuted, they may challenge the constitutionality of the search of their medical records or the acquisition of specimens for law enforcement. But sometimes the medical data is just used to identify individuals and is never introduced at trial. In these cases, the legal remedies are quite unsatisfying.
B. Permissive State Laws
Another way cops gain access to patient medical data is by broadly interpreting permissive state laws. State statutes explicitly allow law enforcement to access patient data—such as toxicology or urine results in the event of a vehicle crash, prescription history to regulate controlled substances, paramedic data on opioid overdoses for public health reporting, or positive test results for sexually transmitted disease. But often, police use these statutes to obtain medical data for very different reasons than those contemplated by the statute.
Permissive state laws can provide a tremendous amount of wiggle room. In State v. Russo, 790 A.2d 1132 (Conn. 2002), the defendant challenged a warrantless search of his prescription history as violating the Fourth Amendment. The state responded that the search was reasonable because a Connecticut statute granted law enforcement access to prescription records in order to enforce controlled substance laws. It did not matter that the police inspected the patient records for a very different purpose—to prosecute fraud. The court still upheld the search under a broad and deferential interpretation of the statute.
For another example, a state attorney general was investigating whether abortion clinics were performing abortions after fetal viability in violation of Kansas law. The Kansas statute permitted disclosure of patient information “to comply with the obligation to disclose the medical basis and specific medical diagnosis relied upon [for the abortion.]” When the attorney general was later being investigated himself for misconduct and harassing the clinics, the court permitted thirty complete patient medical records to be produced—from an abortion clinic, no less—despite neither the provider nor the patients being under investigation. Broad interpretation of disclosure statutes with narrow statutory purposes provides another avenue for police access to confidential medical data.
Even when permissive statutes are deemed unconstitutional under the Fourth Amendment, police reliance on them will not always trigger exclusion at trial. That is, even when statutes specifically exclude use of the medical data, or the statute itself has been found to violate the Fourth Amendment, the prosecution might still be allowed to rely on the data in a criminal trial if the police can show they were confused, and relied on the statute in “good faith.” The latitude shown to law enforcement in these contexts is astounding.
C. HIPAA Exceptions
In addition to state confidentiality provisions, most readers will be familiar with the Privacy Rule of the Health Insurance Portability and Accountability Act (“HIPAA”). This federal law requires that covered entities like health care providers and insurance companies obtain the patient’s authorization before sharing their protected health information (“PHI”). Protected health information is anything identifying an individual, created or received by a health care provider that relates “to the past, present, or future physical or mental health or condition of an individual, the provision of health care to an individual, or the past, present, or future payment for the provision of health care to an individual.”
HIPAA is often invoked as the universal health privacy protector. But it poses almost no obstacles to the police use of medical data. For starters, the police themselves are not a “covered entity,” so once medical data resides with them, they are not subjected to HIPAA privacy or security rules. And even when the police request PHI from a hospital or other covered entity, the Privacy Rule permits disclosure to them for a broad variety of reasons, and without judicial involvement. For example, disclosure is permitted (1) if a state law permits it (such as the statutes above mandating disclosure of opioid overdoses), or (2) to respond to a court order, warrant, subpoena, or summons issued by a judge or grand jury. Perhaps most broadly, HIPAA allows covered entities to share PHI (3) to respond to an administrative subpoena or written statement by police that specifies how the request is “relevant and material” to an investigation.
The only limit on these administrative requests is that they must be limited in scope “to the extent reasonably practicable in light of the purpose for which the information is sought” and “[d]e-identified information could not reasonably be used.” This squishy language from the Privacy Rule seemingly does not employ any balancing test. If law enforcement has made a request that asks for information that is relevant and material (a low bar) and the request fits within one of the stated uses, then HIPAA will allow it.
Because hospitals need not publicly report when they disclose PHI to law enforcement, we have no idea how often the police use this backdoor route. The exception is a gigantic loophole. If nurses and physicians incorrectly interpret this exception as requiring rather than permitting disclosure of PHI, it could result in much less privacy protection against government disclosure than the drafters envisioned. And even if they interpret it correctly, the exception permits law enforcement to have access to confidential medical information any time it might be useful to their criminal investigations. They do not need a warrant. They do not even need a subpoena.
Unfortunately, in addition to allowing quite a lot of unconsented-to disclosures, trial courts interpreting these HIPAA exceptions sometimes read them to suggest that they preempt common law or constitutional privacy requirements. Thus, if something is permitted under HIPAA, some trial courts treat it as permissible under various tort theories and the Fourth Amendment. This is an error, as the Privacy Rule applies to a very narrow context of covered entities and was never meant to be coterminous with preexisting privacy rights. However, confusion about HIPAA’s application is rampant. This may lead cops, health care workers, and patients to think that it replaces common law privacy protections and gives the final say on what may legally be disclosed. It is counterintuitive and troubling that a statute meant to expand confidentiality over medical data is likely having the opposite effect.
Additionally, if the disclosure of a suspect’s medical data was unauthorized under HIPAA, that does not automatically lead to suppression in a related criminal trial. Conversely, if the disclosure was permitted under HIPAA, it might still violate a defendant’s Fourth Amendment rights. If the medical data does need to be introduced at trial, it could be suppressed if the police are found to have conducted an unreasonable search under the Fourth Amendment. This risk might be why the police are increasingly relying on nondefendants’ medical data. Nondefendant’s medical data poses no risks of suppression or Fourth Amendment violations.
When seeking genetic data to identify suspects, police can rely on “close enough” genetic matches by requesting the medical data of a suspect’s children, siblings, or even cousins. They can then use the biological sample from the suspect’s relative to connect the suspect to the crime scene through familial matching. When done this way, the investigative use of a relative’s genetic material might never see the light of day at trial and may never need to be introduced. Because the remedy for a Fourth Amendment violation is suppression at trial, if the relative whose specimen was tested by police is never criminally charged, the relative would need to seek redress by filing a civil rights violation under Section 1983 of Title 42 of the U.S. Code. In addition to being time-consuming and expensive, the evidentiary hurdles are quite high. And these civil proceedings would have no impact on the related conviction. This renders the legal guardrails against police overreach in this context insufficient.
In the previous Section, I explained how there are few legal hurdles to law enforcement accessing private patient data. Specifically, the way the Fourth Amendment has been interpreted, provider compliance, and broad HIPAA exceptions permit a great deal of police access to confidential medical data. Additionally, even when police overreach, the remedies that are available are insufficient to fully protect patient privacy. In the next Section, I describe why police access to medical data has the potential to do so much violence to the physician-patient relationship and core principles of medical ethics.
III. THE GUN AND THE STETHOSCOPE: TWO VERY DIFFERENT CULTURES
Conscripting clinicians to do the bidding of law enforcement is lamentable for a number of reasons. This is because the goals of law enforcement are diametrically opposed to the goals of health care. Consider the most common symbols of each profession. For physicians, it is the stethoscope, which amplifies the patient’s body and allows its rhythms to be heard. Now, consider the symbol of policing in the U.S.—the gun, which is violent and silencing. While these two professions embody more than can be encapsulated by these two symbols, the stethoscope and the gun reflect deep concrete and metaphorical differences between the two institutions. These differences are reflected in their values, ethics, and in the community’s expectations of them.
But let us start with how they are similar, and then we can highlight their significant differences. First, doctors and cops are both considered part of the helping professions. Each owes fiduciary obligations to the people they serve, which in theory, are built on trust. They both wield tremendous power over vulnerable populations, given their status as gatekeepers to important social goods. However, mounting awareness of the disparities experienced at their hands has led to reduced trust in both professions.
In practically every other way, there could not be two more different institutions. Health care is about healing and promoting evidence-based measures, while protecting patient autonomy, privacy, and health care accountability. Law enforcement is not about healing, does not promulgate evidence-based measures, and frequently disrespects individual autonomy and privacy, while dodging most forms of legal accountability. Some have argued for police to adopt a “culture of health” to improve their policing, and perhaps they should. But it cannot operate the other way. That is, the culture of medicine cannot be infected with the culture of policing. I will now explore various case studies of police intruding into health care. Each example reveals the diametrically opposed norms and ethics of each profession, and why this creates a strong argument for their being kept separate.
A. Different Self-Regulation Norms
In the last century, fueled in part by complicity in the Nazi concentration camps, physicians have developed and internalized sweeping reforms to their codes of ethics. Emerging from this grisly past, physicians have documented many instances of their abuses of power. Principles of respect for autonomy, nonmaleficence, privacy, accountability, and justice have since been incorporated not only into the medical school curriculum, but also into the fabric of how hospitals and clinics should be run.
These ideals are not consistently realized, and sometimes they are pit against one another. However, because the principles of autonomy, privacy, accountability, and nonmaleficence are near-absolute, when physicians fail to realize them, these norms are likely to be regulated by the physicians and hospitals themselves—even in the absence of any legal liability. As compared with law enforcement, physicians extensively self-regulate through governing bodies and professional associations. And of course, on top of their self-regulation, if physicians breach confidentiality, fail to obtain patient consent, or cause harm through their negligence, they are frequently civilly sued and held accountable for malpractice.
When it comes to honesty, almost seventy percent (69%) of Americans now rate the honesty and ethical standards of physicians as “very high” or “high.” Most, but certainly not all, Americans feel safe entrusting our physicians with sensitive health information about ourselves or our family members that we would never willingly give to the cops. This did not happen by accident. This is the result of decades of developing cultures of honesty and disclosure. Overall, trust in clinicians in the U.S. remains relatively high, though it could and should be much higher, particularly among the most socially disadvantaged groups.
By contrast, police rarely hold themselves accountable for the violence that they perpetrate, which is often not just careless, but intentional. If anything, police culture “tolerate[s] and encourages” misconduct by framing things in terms of “us versus them”—where a certain amount of brutality is considered necessary to protect law enforcement from the communities they serve.
And when it comes to honesty, rather than regularly exposing mistakes, as is done during medical peer review meetings, cops are discouraged from disclosing misconduct or perjury by fellow officers. This practice of solidarity has been referred to as the “blue wall of silence.” This code of silence and resulting lack of transparency makes it difficult to identify problem officers and widespread corruption. Experts who have examined root causes of police misconduct have concluded that “the roots of police misconduct rest primarily within the organizational culture of policing.”
To the extent police officers support ethical codes of conduct, they appear to do so instrumentally to prosecute crime. And perhaps this makes sense because they are accountable to the public, and not to individuals. That is, respect for autonomy does not appear to be an end in itself, but a means of having a better rapport with the community and increasing crime clearance rates. Conversely, in medicine, there are precious few actions a clinician can take that violate a competent individual’s autonomy, because autonomy is a deontological obligation that clinicians are bound to uphold. These different institutional values and goals lead to very different patient expectations of clinicians and cops. If cops intrude on clinical spaces in ways that violate autonomy, patients may worry that physicians will do the same—and indeed they might at police behest.
B. Different Privacy Norms
Cops and clinicians also have very different norms around privacy and confidentiality. Patient privacy is considered by the American Medical Association (“AMA”) to be “fundamental, as an expression of respect for patient autonomy and a prerequisite for trust.” The AMA Code of Ethics even spells out the different zones of privacy that physicians must respect: physical space, personal data, decisional privacy, and associational privacy. Ethically, physicians should inform patients when their privacy has been breached, separate from any legal obligations.
Physicians also have common law fiduciary obligations for “confidences communicated by a patient [to] be held as a trust.” Texas law stresses that due to the “highly personal nature of [medical] information[,] [it] places a heavy responsibility on the trial court to prevent any disclosure that is broader than necessary.” And every state has now created a statutory physician-patient privilege of some form that prohibits disclosure of statements and medical records in court without the patient’s consent. The privilege promotes “free and full communication between a patient and his doctor so that the doctor will have the information necessary to competently diagnose and treat the patient.” Indeed, courts have warned against jeopardizing the “sanctity” of the physician-patient relationship by divulging patient information to third-parties. Others have stated that “the patient [has] the power to reveal the private information to the persons the patient chooses, reinforcing the [physician-patient] privilege’s policy of patient autonomy and privacy.” It is difficult to square these unequivocal statements with granting cops broad access to sensitive medical information, just because they ask.
If patients do not trust that their communications will be private, they will keep secrets from their doctors that will frustrate their care. Even worse, they may refuse to seek important prenatal or emergency care altogether. Additionally, confidentiality fosters a feeling of safety and sanctuary between patients and physicians, which signals that, as it relates to the physician’s goals, the needs of the patient are paramount even to those of society.
People (sometimes reluctantly) must also share sensitive and humiliating information with cops. Even so, there are no norms of confidentiality in law enforcement. There are also no common law, statutory, or evidentiary requirements that information shared with police be kept confidential. If anything, in law enforcement, the norms generally run in the opposite direction.
Witnesses largely lose control over their testimony when it becomes part of an investigative record. For example, Chanel Miller recounts in her beautiful memoir, Know My Name, how painful it was to realize that intimate photos of her naked, assaulted body would be shared in a crowded courtroom during the prosecution of Brock Turner. Once the police are involved, the victim’s images are no longer private. This is true regardless of how sensitive or embarrassing those images or statements might be. This is not to say that the police are unethical—this is just to say that their code of ethics do not include honoring confidentiality.
1. Police Search Clinical Laboratories and Biobanks
Genetic information is one of the most sensitive types of information connected to us: it reveals predictive information about our future health and our relatives, it does not change over time, and we are only scratching the surface of what it all means. Recognizing that genetic material is highly sensitive, but might be used nefariously, Congress passed a federal statute in 2008 to restrict its use in the provision or pricing of health insurance. Despite its extensive protection in health care, genetic information receives meager confidentiality safeguards when used by cops. The Fourth Amendment jurisprudence has shockingly analogized discarded genetic samples to trash, permitting its warrantless seizure.
Given the value of genetic samples to solving crime, it is no wonder police use of genetic material from health care databases is so common. In 2005, police in Kansas subpoenaed a DNA sample from a woman’s pap smear to connect her father to a string of murders. Through the use of her medical sample (which had been held in a Kansas health clinic), without her notice or consent, the police confirmed that her father, Dennis Rader, was likely the BTK serial killer. This connection gave them probable cause to arrest the suspect. Rader was later convicted of ten counts of first-degree murder.
2. Police Search Public Health Newborn Screening Genetic Databases
Recently in New Jersey, a child’s genetic sample—also obtained in a health care setting—was used to implicate her father in a sexual assault. The child, who is now nine years old, had her heel stuck by hospital staff when she would have been just a couple of days old. The heel stick was part of New Jersey’s mandatory newborn screening program, which blends together individualized patient care with public health goals. Because the database is maintained for public health reasons, the consent process is not as robust as it is in regular clinical decisions. In the absence of a formal refusal, hospital staff collect the blood samples on filter paper that they use to test the baby for dozens of serious, hidden diseases. The residual samples are often maintained in public health facilities.
In the New Jersey case, law enforcement suspected that the child’s father had committed a sexual assault. When they realized he had a child in New Jersey, they accessed the child’s state newborn screening database without a warrant. A comparison of the crime scene DNA and the blood on the screening card revealed the newborn to be a first-degree relative of the perpetrator. This gave them the probable cause they needed to arrest the father for the assault.
News of this prompted a lawsuit by the New Jersey Office of the Public Defender against the Department of Health (which maintains the screening cards). The complaint asserted that parents are not told that their child’s DNA will be placed in this database and potentially used by law enforcement. The litigation resulted in the state admitting to using the newborn bloodspots, without parental consent, in at least five other cases that resulted in prosecutions. Legislators in New Jersey proposed a bill in 2022 to prohibit the use of newborn blood spots for criminal investigations without parents’ affirmative consent, but the bill has stalled in committee.
For Fourth Amendment purposes, the lack of any consent for secondary law enforcement use makes this case very different from the use of a public genetic database, called GEDMatch, to apprehend the Golden State Killer in 2018. So-called forensic genetic genealogy relies on public genetic databases, in which users voluntarily upload their genetic information from Ancestry.com or 23andMe to facilitate their personal genealogical research. This was also how police identified the man who murdered four college students at the University of Idaho.
In this case, however, parents were not uploading their child’s genetic profiles to a public website for recreational purposes. Rather, the genetic newborn screening bloodspots were obtained in a hospital, exclusively for medical purposes. This draws hospital staff into the prosecutorial process in a very different way.
Because the newborn screening program is not completely voluntary, and parents usually must affirmatively opt out rather than opt in, there should be no unconsented-to secondary uses of this data. It only takes one highly publicized instance of misuse, such as in the New Jersey case, to deter parents from participating in the screening program. If parents opt out, newborns will lose the huge benefit of detecting rare diseases that can be fatal if untreated. Given that overpolicing and distrust is already more rampant in communities of color, there is reason to worry that blurring the public health goals of newborn screening with the prosecutorial goals of criminal investigations will cause greater health disparities.
Some may argue that the use of newborn screening bloodspots is justified, on an instrumentalist account. On this view, the data is just sitting there in a warehouse, waiting to serve the public and identify rapists, murderers, and other “bad” people. In response to this, I offer two thoughts. First, we ought not to concede that the inquiry is a consequentialist one, as opposed to holding firm in the obligations to keep medical data private. And second, even on the instrumentalist account, it is not clear that providing cops access to genetic newborn screening databases will yield net benefits to society. Are dozens of children dying from preventable diseases obviously worth less than a murder prosecution? The life of a child who dies because his parents opted out of the screening program is just as valuable as the life of a murder victim for whom the police seek justice. The public should be able to debate the competing values of these institutions, rather than the prosecutorial goals always trumping those of public health.
Additionally, the police may not be able to handle these sensitive specimens in a way that adequately maintains confidentiality. The storage, labeling, and transfer of genomic materials is highly regulated in health care settings and laboratories. Because data stored by law enforcement is not subject to health privacy or clinical laboratory regulations, medical data could find its way into an investigative police file, and be treated far too cavalierly.
This fear is not exaggerated. In San Francisco in 2016, a woman provided her DNA to police as part of a sexual assault rape kit. Without her knowledge or consent, in 2021, police used her genetic sample to charge her with retail theft. This horrific use of her genetic information reveals the lack of respect for privacy in law enforcement. Once they have your data for one purpose, they may feel entitled to use it however they please. And just as we worry that parents will opt out of newborn screenings if they think their samples are accessible to law enforcement, sexual assault victims are now given yet another reason not to report their assault to police.
All of this blurring of genetic data with law enforcement occurs amidst considerable police secrecy. We do not know how often police access medical or public health databases; law enforcement does not report this. This is an additional difference between cops and clinicians. Clinicians are much more likely to report on their own internal practices, either due to accreditation requirements or because of their civil commitment to transparency. Cops, on the other hand, tend to be very secretive about their own internal processes—using privacy as a shield and sword when it comes to their own institutional data.
C. Different Accountability Norms
One reason for the different privacy norms has to do with who each profession sees as their “client.” Medical institutions must address social injustice at the policy level. However, individual physicians do not owe duties to society generally when making treatment decisions. Instead, their duty is to the individual patient. This has been enshrined in some states’ common law. For example, physicians in Arizona owe a fiduciary duty to their patients “to exercise the utmost good faith” and to “act in the best interests of his patient so as to protect the sanctity of the physician-patient relationship.”
This focus on the individual can create conflict between medical and public health ethics. For example, during pandemics, governments might ask physicians to allocate scarce resources like ventilators or antivirals to serve population health. Because physicians see themselves as agents for individual patient care and not as instruments of public health, having to do this caused some physicians moral distress.
1. Police Are Increasingly Unaccountable to Private Individuals and the Public They Serve
To law enforcement, conversely, the client is the public, not any one individual. The first sentence of the International Association of Chiefs of Police Code of Ethics states that the police officer’s “fundamental duty is to serve mankind.” Police success is measured in terms of public safety statistics measured at the population level. The utilitarian framework contributes to an ideology of police being unaccountable to individuals for the harms they cause.
The unaccountability of law enforcement plays out directly in negligence law, where police are often immunized from liability under the “public duty doctrine.” This common law doctrine varies a bit from state to state. However, it essentially holds that cops cannot be held liable for an injury if they fail to perform a duty, and the duty is “owed to the general public rather than to an individual plaintiff.” For police to owe a duty to affirmatively act to protect an individual, some special, pre-injury relationship must exist between the cop and the injured party, the injured party must be in custody, or the cop must have created detrimental reliance. Even grossly negligent failures by a cop may not be considered a basis for tort liability, as cops are protected by the discretionary nature of their decision, the public duty doctrine, and other government immunity provisions.
The common law doctrine of qualified immunity further shields the police from individual liability for constitutional rights claims unless the officials violated “clearly established law.” The specter of immunity impacts the likelihood that individuals will even file suit. Qualified immunity for civil rights violations has received well-deserved criticism in recent years due to media coverage of appallingly high levels of police killings. Shielding police from civil lawsuits in all but a tiny fraction of cases, the doctrine “serves as a barrier to incentivizing police officers to do better when dealing with the public.”
This patchwork of immunities reveals a picture of police that are increasingly unaccountable to the public they serve. This led one judge to write that “liability for municipal employees is effectively dead.” Meanwhile, the duties of medical professionals are expanding to protect nonpatients and unnamed third-parties the physicians have never met.
These opposite trajectories bear emphasizing. In the last fifty years, physicians, nurses, and therapists have seen sharp increases in their liability through expanding notions of negligence obligations. Physicians in Pennsylvania have duties to protect or warn a patient’s neighbors, even when they have never met them or know their names. Somewhat paradoxically, physicians have been asked to pay for accidents that may even be principally caused by law enforcement. I will explain how this came to be, below.
2. Physician Accountability to the Public Is Expanding
The landmark case of Tarasoff v. University of California, 551 P.2d 334 (Cal. 1976), is an important illustration of the asymmetrical obligations owed to the public by health care professionals and law enforcement. Because it was a key factor in the erosion between health care and law enforcement, I will go into a bit of detailing describing the case here. In Tarasoff, a young man named Prosenjit Poddar brutally killed fellow Berkeley student Tatiana Tarasoff. In Poddar’s criminal trial, the family discovered that Poddar’s therapist had breached confidentiality—risking tort liability—to report Poddar to local law enforcement. The therapist was concerned for the unnamed woman (Tatiana) whom his client had admitted he wanted to harm. Discovering this, Tatiana’s parents then sued the police, the university, and the individual therapist who had treated Poddar in negligence—for failing to warn their daughter of Poddar’s threats.
The facts of this case are quite remarkable. When the therapist called the campus police, the police detained Poddar and then let him go because he “appeared rational.” The police were sued but were given statutory immunity because their decision to release Poddar was part of their public duties and was discretionary. The court reasoned this was appropriate because courts should not be second-guessing the executive branch.
The therapist, on the other hand, was found to have a duty to protect Tatiana. The California Supreme Court suggested that the therapist was careless for not figuring out that Poddar’s target was Tatiana, and then personally reaching out to warn her. How the therapist was reasonably supposed to do this in the 1970s, without Google or Tatiana’s contact information, much less any training in risk assessment or violence prevention, remains a mystery.
An oft-cited section from the opinion states that patient-psychotherapist confidentiality “must yield to the extent to which disclosure is essential to avert danger to others. The protective privilege ends where the public peril begins.” Many follies in the co-opting of medical privacy to serve law enforcement goals may be traced to this opinion, and this singular sentence. Even the HIPAA exception that permits unauthorized disclosure if someone threatens themselves or others is referred to as the “Tarasoff exception.”
After Tarasoff, judges extended affirmative duties to warn nonpatients to social workers and physicians. Clinicians may be liable for negligence if they do not violate patient confidentiality to prevent imminent, serious bodily harm to an identifiable person. While physicians already had duties to report infectious diseases, Tarasoff can be viewed as creating a sea change in creating broad obligations for therapists, and then other clinicians, to protect the public health and safety. Unsurprisingly, therapists really dislike being asked to violate patient confidentiality to disclose imminent risks to third parties, as they worry it chills therapeutic speech and can stymie treatment. It also runs counter to principles of beneficence—the ethical duty to prioritize the interests of their individual patients.
The Tarasoff majority created new duties for therapists by exploiting the special relationship between the therapist and patient and using it to generate duties to people outside of that relationship. Put differently, the court took the very things that made therapeutic relationships special and sacrosanct—confidentiality and trust—and exploited them for law enforcement goals. Many take this “special relationship” argument at face value and assume that because the therapist had a trusting relationship with the murderer, this naturally led to his owing a duty to the murderer’s victim. But this move was and is extremely counterintuitive.
Imagine extolling the cherished relationship between mother and son, and then requiring the mother to divulge her son’s secrets to prioritize some other kid’s needs. The obligations to third parties is a weak link in the reasoning of Tarasoff that is assumed rather than defended. As a landmark example of the co-option of health care by law enforcement, the Tarasoff decision has likely done a great deal of violence to the therapeutic relationship.
3. The Tarasoff Ruling Frustrates Psychiatric Care
Indeed, shortly after the case was decided, a whopping eighty percent of therapists “observe[d] greater patient reluctance to discuss violent thoughts,” and sixty percent felt “that patients were at least somewhat more reluctant to discuss sensitive information.” In addition to chilling patient willingness to discuss sensitive matters, therapists also erred on the side of restricting patients’ freedoms. A third of therapists reported being “more likely after Tarasoff to commit patients involuntarily to the hospital.” This might not be terrible policy if it helped patients or prevented violent crime. But that appears not to be the case. One study using a fixed-effects model found that mandatory duty-to-warn laws actually resulted in an increase in homicides of five percent.
Of course, recognizing the vulnerability of patients and the potential for abuse, physicians and nurses are accountable to the public in many other ways. If they abuse patient trust or fail to execute their duties, they can lose their professional licenses, be denied hospital admitting privileges, and may also be liable for fines for violating the Health Insurance Portability and Accountability Act. They also may be prosecuted for many health-care-specific types of self-dealing, fraud, and misrepresentation. And of course, unless they work for a public, state-run hospital, they can be, and frequently are, sued for ordinary common law malpractice claims.
D. Different Efficacy Norms
Therapists worried (correctly) that the Tarasoff ruling would pierce a huge hole in patient confidentiality, demand clairvoyance, and not deter much crime. They were right. Without gathering any data whatsoever, Tarasoff-type expansions of duties have knocked down the preexisting wall between health care and law enforcement. This is concerning, but unsurprising. Judges do not conduct research on the efficacy of the policies they propose, and neither do the police.
Even law enforcement interventions that are passed by statute and capable of being carefully studied first by experts are almost never evidence based. Of the scant few police interventions that have undergone validity or reliability testing after the fact, the large majority have not been shown to work at all. This means, in effect, that law enforcement is allowed to experiment on the public. This makes it all the more appalling that they are then immunized from negligence when they get it wrong.
This is diametrically opposed to ethical principles in clinical medicine. Virtually no treatments should be offered to patients without testing their safety and efficacy first. An untested intervention might actually cause harm as opposed to just not work, which would lead to violations of the ethical principle of nonmaleficence (colloquially, “do no harm”). While all health care involves some uncertainty, to experiment on patients, physicians must first demonstrate that no safer options have worked. To deviate from an accepted standard of care, physicians must have data that the new path is equally safe and potentially more effective. This is not an ethical norm held by law enforcement. In the following Section, I explore two examples of law enforcement dabbling in health care in ways that resemble raw experimentation rather than evidence-based care.
1. Examples of Law Enforcement Providing Experimental “Treatment”
i. Excessive Ketamine Administration
In August 2019, a twenty-three-year-old Black man named Elijah McClain was stopped by police on his way home. Someone reported he was “acting suspicious” though he was not suspected of any crimes. Three cops restrained Elijah in a chokehold and then handcuffed him while he stated he was an introvert who never even hurt flies. The police claimed after the fact that McClain went for a gun, but there is no evidence of this. When the paramedics arrived, at law enforcement’s request, they administered five milliliters of ketamine without knowledge of Elijah’s vital signs, asthma, or mental health history. Elijah went into a coma and later died. The final autopsy report indicated a high dose of ketamine contributed to his death. The police officer who put McClain in a neck hold after he was administered ketamine was acquitted of negligent homicide and manslaughter.
While law enforcement may need to restrain individuals who are physically threatening them, they seem to have no internal criteria for how the restraint will be done ethically and according to best medical practices. In Elijah’s case, there was no evidence that he needed the ketamine to be restrained. Indeed, he was already in handcuffs.
Unfortunately, police often rely on the controversial diagnosis of “excited delirium” to administer ketamine to stressed individuals. As Osagie Obasogie has written, “there is little scientific evidence to support claims that excited delirium exists as a legitimate psychiatric condition.” Even so, it is increasingly being used to explain “suspicious deaths that occur in police custody.” Reviews of police custody deaths validate that excessive use of ketamine is often the primary cause of death, as law enforcement are administering doses of sensitive medication without adequate medical training or supervision. This is yet another troubling example of the medicalization of criminal investigations, using nonevidence-based “treatments.” Police have inadequate training on how to assess the health status of individuals. And yet, they, along with paramedics, routinely experiment on a case-by-case basis, injecting powerful medications into people experiencing mental health crises.
ii. Drug Courts Provide Nonevidence-Based Treatment
Drug courts have proliferated in the last two decades as a result of the failed War on Drugs and mass incarceration. Qualifications for drug court vary, but state and county programs typically allow people charged with nonviolent crimes to take a plea in abeyance, which will disappear so long as they graduate from a diversion program such as drug court. Drug courts often involve a county judge overseeing someone’s treatment program with regular check-ins with parole officers, treatment providers, and counselors. Participants must appear before their judge and explain any relapses they have had. Judges will often place requirements on graduation from drug court, such as obtaining and keeping a job.
Drug courts provide an on-ramp to treatment. Compared with probation-as-usual, some studies suggest they reduce recidivism rates. However, because addiction is a disease, the primary comparison should not be criminal metrics, but clinical ones. We do not know how drug courts compare to routine, affordable evidence-based medical care for addiction because access to this treatment does not exist in the United States. Instead, we funnel addiction “treatment” through the prison system or through highly unregulated addiction clinics that often do not have even one licensed medical provider.
The blending of punishment and treatment goals problematically blurs the lines between health care and law enforcement. People in recovery are expected to check in with their parole officer and attend drug court regularly with the threat of incarceration looming in the background. These check-ins make it difficult for poorer people living on the margins and without transportation to sustain employment. The result is that the carceral state is engaged in a great deal of surveillance and treatment that should be provided not in the courtroom, but in the clinic.
While some of the treatment programs mandated through drug court are evidence-based, many are not. As a result, participants in drug courts often get mixed messages about the value of medication to their recovery. For example, despite being highly effective and the “gold standard” for opioid use disorder, some drug court programs prohibit the use of buprenorphine to curb opioid cravings. For decades, many drug courts have considered medication for opioid use disorder (such as methadone) to be taboo because it is “just another addiction.” This perspective ignores the reliable empirical data that when properly dosed, opioid agonists like buprenorphine can keep people employed and able to break the cycle from craving to binging.
There are unfortunately great disparities between courts in the level and quality of addiction care that participants receive. Even some of the evidence-based drug courts that contract with licensed addiction providers require participants to attend group therapy through Alcoholics Anonymous meetings. Group therapy can be a very useful form of social support to individuals in recovery. However, in some counties it is the primary method of “treatment” for individuals in drug court. This is a problem because its faith-based message is “not a substitute for scientifically valid addiction treatment and should not constitute the primary form of medical assistance received by drug court participants.”
2. Police Are the Default Providers for Patients in Crisis
Because the U.S. has failed to fund many types of social services and behavioral health care, the police are often called on when people are in crisis. When police are the hammer and the only tool we have, every social service—from “traffic stops and noise complaints to evictions, overdoses, and psychiatric emergencies”—becomes their nail. We have culturally come to depend on the police as the key point-of-access for the delivery of addiction treatment, mental health care, and even the reporting of child abuse, despite their inability to meet vulnerable patients where they are. To be clear, police departments may not desire being frontline medics. However, due to the chronic underfunding of behavioral health services, police are much more likely to be the first and last resort for people in crisis.
While crisis-trained police can reduce escalation and help divert people to treatment, these programs are only successful relative to the status quo of carceral, law enforcement methods. And while they may be “associated with reducing recidivism and lowering costs, . . . there is little association between program participation and improved behavioral health.” That is, they might reduce crime, but they do not appear to promote health. And yet, state governors and mayors are expanding nonevidence-based programs to funnel “treatments” for mental illnesses through law enforcement and the criminal justice system.
In California, the “CARE Court” will “connect[] people in crisis with a court-ordered treatment plan for up to two years, while diverting them from possible incarceration, homelessness or restrictive court-ordered conservatorship.” Programs like this reveal how we have completely given up on providing actual, evidence-based mental health treatment and instead criminalize poverty and mental illness. The vast majority of unhoused people with addiction or other mental illnesses receive no treatment. And when this is the reality—where we have given up on providing universal mental health treatment—the illnesses do not simply disappear. They show up in ways that are funneled by default through law enforcement and the criminal justice system.
CARE Court and similar programs only seem progressive if they are compared to incarceration. When compared with the receipt of top-notch, evidence-based treatment, they fall woefully short. And yet, we lack the collective imagination or will to treat people in the clinic rather than the courtroom. It would be far better to provide mental health treatment wholly divorced from the threat of punishment. The looming stick of incarceration distracts from the provision of quality, evidence-based treatment.
Providing “treatment” through criminal courts confuses the roles of law enforcement and clinicians. Participants may come to view clinicians as being inherently aligned with the police state. If you are used to your treatment provider talking about your mental health struggles with your parole officer, counselor, or judge, you might come to think these disclosures are normal outside of this context. You might also not tell your court-appointed provider everything that they need to know to adequately treat you; if you say the wrong thing, it might delay graduation from drug court or any other diversion program. But most importantly for our purposes, the treatment that is provided by these diversion programs is often out-of-date and not evidence-based. This will impair patient trust in health care. In the next Section, I will explain how another ethical principle, honesty, is also critical for trust. I will discuss how it is a cornerstone of medicine, but not of law enforcement, and why this matters.
E. Different Honesty Norms
There used to be norms of “benevolent deception” in medicine, where physicians did not disclose terrible prognoses to patients to give them hope. This is no longer ethically or legally permitted. The AMA Code of Ethics requires physicians to maintain “open communication between physician and patient” as “essential for trust in the relationship.”
If a clinician is dishonest in any way, this will negatively impact the care the patient receives, the willingness of the patient to seek care, and the likelihood that the patient will see the clinician as protecting their best interests. Deception is universally condemned in medicine. It impairs shared decision-making and “fundamentally undermines patients’ autonomy in nearly all situations.” Clinicians who lie to patients can be sued and investigated for professional ethics violations. They could also potentially lose their license.
Police, on the other hand, routinely lie to witnesses to get them to cooperate. For example, they may tell someone their statement is not being recorded when it is, or say they have video footage of the defendant committing the crime, even if this is not the case. Courts do not find this violates defendants’ due process rights. It is well-recognized that “[t]he police are permitted to lie or use some deceptive methods in their questioning as long as the deception was not . . . so extensive as to induce a false confession . . . .” In this regard, health care and law enforcement are quite distinct. Physicians should not assume that patients are lying, because the physician and the patient should theoretically be on the same page. However, for the police, the need to investigate a crime means everyone is presumed to be hiding something. They have therefore “come to believe that lying is a necessary and justifiable component of their jobs.”
Police are allowed to use deceptive methods to obtain evidence. For example, DNA that a suspect unwittingly leaves on an armchair may later be analyzed without a warrant and without violating the Fourth Amendment. If one technically agrees to provide a DNA sample, but did not feel they were free to refuse, this is presumed to be voluntary consent in the police context.
Some have argued that police surreptitiously obtaining genetic samples for investigative purposes is unethical because they “bypass[] the codes of informed consent.” In the medical context, the phrase “informed consent” is a term of art. It “requires that a patient understands what is being done to her before she agrees to be touched by a physician or researcher.”
A physician’s failure to provide informed consent can result in battery or medical malpractice liability. It is a well-developed concept and obligatory. But this is not true for law enforcement, which is not an institution built on honesty, transparency, or trust. The different disclosure and transparency norms between health care and law enforcement stem from the different weight each puts on the value of autonomy. In the next Section, I will explore the principle of autonomy, and how it steers medical ethics but not police practice.
F. Different Autonomy Norms
Despite physicians and nurses having a great deal more clinical education than most of their patients, they should not tell the patient what to do with their bodies. In medicine, respecting autonomy means that patients have the freedom to make decisions for themselves and determine the course of their care. Autonomy does not mean patients get to demand specific things. Rather, it means they should be presented with the options—their risks and benefits—and be given the chance to ask questions.
Patients cannot make autonomous decisions if physicians do not share information that would be material to their decision. Respect for autonomy means that patients might make choices that appear irrational to physicians (like refusing chemotherapy), but stem from personal goals of care. Thus, respecting autonomy means respecting the patient’s idiosyncratic values and right of self-determination.
There is no corollary respect for autonomy in law enforcement. There is not a police code of ethics that ensures that autonomy is respected. Indeed, manipulative or sneaky tactics might be instrumentally encouraged to detain or interrogate individuals. This provides yet another critical difference between the two institutions and another basis for keeping them separate. To trust physicians, patients must not come to expect their doctors to treat them like cops do.
The next two examples I will explore demonstrate how the legislature disrespects autonomy by refusing to allow patients to make medical decisions for themselves. If the legislature has decided that a type of health care—such as abortion or gender-affirming care—is morally suspect, they regulate it heavily, using health care licenses and professionals as the enforcement mechanism. And yet, states only have the ostensible moral authority to regulate these practices in the way they do because at base, they do not consider them to be legitimate health care. If they did honor them as valid treatments, they would have no authority to stick their noses in a conversation that should be between patients and their doctors.
1. Limiting Access to Lifesaving Abortion Care
Governments are presently using their police power to block access to critical health care—showing perhaps the greatest disrespect possible for the autonomy of people who can become pregnant. Statutes that have restricted access to abortions since Roe v. Wade was overturned demonstrate a remarkable lack of recognition of how they require physicians to violate medical ethics.
Because these power grabs by law enforcement are not evidence-based, the way the exceptions are crafted reveals complete naivete about how medicine is practiced. For example, the meaning of terms like “medical emergency” have already been tested in hospitals across the country. In Missouri, a woman named Mylissa Farmer went into labor at 18 weeks with a desired pregnancy. Her water broke and she had vaginal bleeding and cramping. When emergency physicians assessed her, they said the baby girl she was carrying would not survive. However, because it was technically still alive, Missouri’s post-Dobbs abortion ban would not permit them to perform an abortion, even to prevent serious complications to Mylissa. Unfortunately, her physicians did not think she was close enough to death to justify the termination under Missouri’s law. Mylissa’s doctors told her she would just need to wait until her “vitals plummeted or infection set in, or the fetus’ cardiac activity stopped” before they could intervene. This put her at great risk, as she attempted to travel to a neighboring state while in a precarious and unstable situation.
The Missouri statute, and others like it, tie physicians’ hands and do not let them do what they know is best for the birth mother and her baby, which is to terminate the pregnancy compassionately and safely. There is already mounting evidence that in other states, like Texas, patients are suffering from preventable complications that are caused not by medicine, but by criminal laws. These abortion bans are obviously not meant to respect the autonomy of women and people who can become pregnant. In fact, they are written in such medically naive ways, they might even be designed to do the opposite.
In an effort to advance a pro-life, anti-choice agenda, abortion bans forget that they are conscripting physicians to violate their code of ethics in the service of prosecutorial goals. They also forget that medical decisions are not black-and-white. Several of the statutes are written as if a siren goes off when someone’s life is at risk. But that is not how critical situations arise. A pregnant person’s vital signs can change rapidly, often without advance notice. Someone can seem stable, and then two minutes later they are crashing.
Finally, “medical emergencies” are not declared without patient input. Respect for autonomy means that physicians do not unilaterally tell patients when heroic life-saving interventions will be used. Whether a patient wants chest compressions or a ventilator depends on their values and personal goals of care. The exceptions that only permit abortions when the pregnant person is near death ignore the very purpose of medicine. Medical ethics requires that physicians do what they can to promote healing and to help people thrive—by promoting autonomy and beneficence—and not just to prevent death.
Other abortion laws exhibit additional “black-and-white” thinking about medicine. Utah’s current eighteen-week ban permits abortions if the fetus has a lethal defect or severe brain abnormality that is “uniformly diagnosable.” However, this fundamentally misunderstands the way bodies present themselves and the uncertainty inherent in many clinical situations. While some fetal defects are expressed with near certainty, whether a particular defect will be fatal and within a particular timeframe is rarely something that can be uniformly diagnosed. Medically speaking, the phrase “uniformly diagnosable” is nonsense. These are just some of the terrifying real-world problems that arise when legislators try to commandeer the practice of medicine, with the threat of criminal consequences to physicians for noncompliance.
Physicians in Utah and elsewhere have expressed concern over how to thread the needle in these situations—that is, to do what is best for their patients while also not exposing themselves to criminal fines or jail time. Every physician’s risk preference will vary, but these laws will certainly have a chilling effect and put patients’ lives at risk. They are also part of a long-term, disturbing trend of requiring physicians to violate autonomy by shoehorning politics into health care. Below is yet another example of this phenomenon, which is tragically on the rise.
2. Punishing the Provision of Gender-Affirming Treatment
State laws banning gender-affirming care also require physicians to disrespect patient autonomy. In 2023, the Governor of Utah signed a ban on gender-affirming surgeries for minors into law. The Utah law prohibits a health care provider from “providing a hormonal transgender treatment” or “providing sex characteristic surgical procedures” to new patients who were not diagnosed with gender dysphoria before a certain date.” The penalty for violating this law is either losing one’s license or being sued for malpractice.
On April 6, 2021, the Arkansas State Legislature overrode the governor’s veto to prohibit the provision of gender-affirming treatment to minor patients—or to even refer them to other physicians for this treatment. The Arkansas law states that “[t]he risks of gender transition procedures far outweigh any benefit at this stage of clinical study on these procedures.” In Arkansas, unlawful provision of gender-affirming care is likewise considered unprofessional conduct and can provide a basis for a malpractice lawsuit. Because physicians who provide gender-affirming care could lose their license in Arkansas, this will effectively remove this care for many youths in the state and will almost certainly result in increased mental distress and suicide.
Similar legislation is pending in approximately fifteen other states. Most of these bills propose banning all health care professionals from prescribing or administering puberty blockers, hormone therapy, or gender-affirming surgeries to anyone under the age of eighteen unless the patient can verify that they are genetically intersex. Penalties can be extreme; in Idaho, a proposed bill would punish physicians with a felony punishable by a life sentence. Fortunately, the Idaho bill has not yet passed their senate.
In the aforementioned laws, the legislature is deciding what counts as the medical standard of care—disrespecting patient autonomy by going against the great weight of medical evidence and expertise and imposing fines and professional penalties. The American Academy of Pediatrics issued a statement saying “[p]olitics has no place here. These are individual conversations between clinicians, patients and families about what’s best.” In testimony regarding the harms the legislation will cause, pediatricians said it is based on “myths and misinformation” and a “misunderstanding about medical and surgical aspects of gender-affirmative care.”
While legislators claim these bills protect children, physicians argue they do the opposite. Long-term data shows that “access to gender-affirming care in childhood and adolescence can have profoundly important mental health benefits,” including reducing the risk of suicide, decreasing depression, and decreasing anxiety. At some point, these statutes will need to be enforced, and when they are, law enforcement will trump important goals of clinical treatment simply because the government holds itself to a lower standard of care, permits experimentation on the public, and violates patient autonomy.
Once again, we see the state invading private medical domains through the apparatus of law enforcement. This is concerning, because as I have laid out above, law enforcement fails to promote the key tenets of privacy, honesty, evidence-bases, respect for autonomy, and accountability, which are key ingredients to stimulate trust. Once in the hands of law enforcement, vulnerable patients’ data may be used for nefarious, punitive purposes, which might discourage patients from receiving necessary treatments. The disrespect for individuals in the law enforcement context cannot infect patients’ perspectives in the medical space; otherwise, patient trust in medicine will suffer. In the next Section, I will explain why patient trust is critical to the delivery of quality care.
G. Different Trust Norms
Compared with other countries, public trust in physicians in the U.S. is low—unacceptably low. It is therefore imperative that health care policies do more to earn the trust of the patient populations they serve, and not less. Promoting patient trust cannot be accomplished if people perceive physicians to be prosecutors, or as working in tandem with law enforcement.
This is not about trying to keep groups separate for the sake of purity or to fetishize medicine. Rather, to prevent negative impacts on patient health, it is critical that we erect sharp boundaries between law enforcement and treatment. The lack of respect for patient privacy, honesty, and autonomy, and the lack of accountability in law enforcement for causing brutal, individual harms, makes the police in the U.S. less worthy of public trust. In the next Section, I will explore instances when the lines between health care and law enforcement have been blurred, and how these blurred lines have likely caused significant harm to the physician-patient relationship, the trust on which it relies, and the quality of care.
1. Prescription Drug Monitoring Programs Rely on and Exacerbate Mistrust
Prescription drug monitoring programs (“PDMPs”) are databases that track prescriptions and patient requests for controlled substances. They proliferated over the last decade in response to the opioid crisis and now exist in nearly every state. From the outset, they were “an instrument of law enforcement.” PDMPs operate differently in every state, but generally require physicians and pharmacists to enter prescribing data for controlled substances or to check such data before initiating a new controlled substance prescription.
Proponents argue that PDMPs are a helpful investigative tool to assess potential criminal diversion of drugs by patients, physicians, and providers. Critics argue that they are inefficient and poorly designed. PDMPs are often not developed with a “clear orientation towards health promotion” and instead complicate “communication and relationships between prescribers, pharmacists, and patients.” Put simply, they place law enforcement between a patient and their physician and can violate the trust between them.
PDMPs also may trigger unintended consequences that harm patients. Fearing law enforcement oversight, physicians may under-prescribe necessary pain medications. This may lead patients to seek treatment for their pain illegally on the streets, with a drug supply that could be impure and deadly. In addition to the significant public health concerns, there are enormous potential privacy risks of PDMPs when this sensitive data lands in the hands of law enforcement.
For the purposes of this Article, the chief problem with PDMPs is that they destroy patient trust by injecting law enforcement norms and goals into medicine. If patients know that their physician is stepping away for a few minutes to check a police database to see if the patient is telling the truth, this brings a prosecutorial element into the decision-making process that will destroy rapport. When programs like PDMPs start chipping away at the principle of confidentiality in medicine, there are no obvious principled limits on cops’ use of doctors as their prosecutorial instruments. This loss of confidentiality will further hurt the sanctity of the physician-patient relationship, and the disappearing trust on which it relies.
CONCLUSION
For the many reasons laid out above, it is important that the norms of law enforcement not creep into the world of medicine. As the Association of American Physicians and Surgeons so fittingly recognized, “[p]hysicians are not agents of the police power of government, and should not be forced to choose between protecting their patients against prosecution or protecting them against disease.” Unlike cops, physicians must embody ethical norms that respect self-regulation, privacy, accountability, efficacy, honesty, and autonomy. Respecting these ethical norms is critical not only to promote public health, but also to repair the broken trust between physicians and patients. In this Article, I described the rampant mistrust of medical providers and institutions, which is exacerbated by police intrusion into health care. Medical mistrust is a SDOH that disproportionately impacts patients of color. To build health care systems that are more trustworthy and equitable, physicians must be walled off from law enforcement.
A number of efforts can mitigate against the police intrusion into health care. However, there is no simple solution to this complex problem, and more detailed research needs to be completed on each solution. Nevertheless, below are some concrete steps courts and legislatures could consider to better separate health care from law enforcement.
A. HIPAA Should Be Revised to Make It Harder for Cops to Obtain Medical Data
The HIPAA exceptions described above make it too easy for law enforcement to pierce medical privacy and obtain access to confidential health data. In addition to permitting disclosures of patient data in response to subpoenas, HIPAA also permits disclosures on a simple written administrative request; this is what health privacy scholar Leslie Francis has called an “open-ended provision.”
HIPAA has thus had the counterintuitive effect of making it easier, rather than harder, for the government to access health data. HIPAA was not intended to provide cops with greater access to confidential data. However, because its permissive exceptions are sometimes read by law enforcement and health care staff to create entitlements to patient data, the statute has become not a shield, but a sieve. This is unfortunate because the “HIPAA drafters worried as much about personal data being abused by the government as they did about misuse by researchers or the insurance industry.”
Indeed, while I was writing this Article, Vanderbilt Medical Center became embroiled in controversy over its disclosure of transgender patients’ medical records to the Tennessee attorney general “as part of an investigation into medical billing.” The disclosure appears to have been pursuant to a civil investigative request, and it is unclear whether this runs afoul of existing Fourth Amendment precedent. The concern, of course, is that the investigation could be pretext, and a means for harassing clinics that provide gender-affirming care as well to their patients. Regardless, HIPAA does almost nothing to stop this kind of disclosure. As Nashville LGBTQIA+ advocate Lance Preston said: “[d]o we believe that Vanderbilt Medical Center could have fought this and taken a bigger stand? Absolutely. But at the same time, we believe that it would have just prolonged the inevitable because the attorney general unfortunately, has the law on his side.”
This Article calls for revisions to HIPAA to remove the breezy access the police have to medical data. The exceptions have become too easy to overcome, without demonstrating a strong investigative need. However, precisely how this statute ought to be revised is beyond the scope of this Article. There is, at least, a strong prima facie argument for requiring cops to do more than issue a written request asking for medical data. The status quo is too permissive. Ironically, HIPAA currently gives the very entity most people worry about sharing their medical data with—the police—incredible free access.
B. Common Law Courts Should Rethink Tarasoff and Expansive Immunity
The Tarasoff-style duty to warn was ill-conceived from the start. Empirical research has shown that such duty has not resulted in demonstrable improvements in the prevention or prosecution of crime. Instead, it has stifled vulnerable patients’ willingness to obtain mental health treatment. While likely resulting from an intuitive impulse to hold someone accountable for heinous crimes, the Tarasoff-style duty takes the very thing that makes health care special—confidentiality and patient trust—and exploits it in a way that harms not only public health, but also medical ethics. One way to course-correct is to limit the rampant recognition of duties of physicians to warn or protect third parties whom they have never met. Therapists and physicians lack good risk-assessment tools, they are often wrong, and their warnings to third parties are generally ineffective. But requiring them to contact the police to report their patients chills therapeutic trust.
C. Physicians Need Greater Autonomy to Practice Ethical Medicine
In Cruzan v. Director, Missouri Department of Health, 497 U.S. 261 (1990), the Supreme Court recognized that competent patients have a “constitutionally protected liberty interest in refusing unwanted medical treatment” under the Fourteenth Amendment. While this has been interpreted as a negative liberty rather than a positive right, it nonetheless underscores the importance of patient autonomy in our nation’s history and tradition. In another case that refused to recognize a right to physician-assisted suicide, the Court heralded the state’s legitimate interest in “protecting the medical profession’s integrity and ethics and maintaining physicians’ role as their patients’ healers.” Federal courts have long-recognized that the physician-patient relationship is special and “common law and historical American practices have traditionally trusted individual doctors and their patients with almost complete autonomy to evaluate the efficacy of medical treatments.” While a legitimate government interest can justify limiting certain rights and is not itself recognition of a right, the articulated state interest in protecting the physician-patient relationship is worth emphasizing.
The Dobbs decision exposes the current Court’s hostility to substantive due process jurisprudence. However, there was a time, not too long ago, when the Court recognized patients’ liberty interest in making autonomous medical decisions. Indeed, before the Dobbs about-face, privacy, informed consent, and medical autonomy were considered deeply rooted in our nation’s history and tradition. Returning to this orthodox view, and bolstering it in other health care contexts, could make it harder for the police to encroach on health care decisions and disrespect patient autonomy. Of course, this will require focusing strategic appellate efforts on future Courts. In the meantime, advocates should try to develop respect for patient autonomy under state constitutions, and in contexts that are less politicized than abortion or trans care. This can provide helpful counterbalancing to the increasing encroachment of law enforcement and the state into health care.
D. Health Care Providers Need Training on the Difference Between Permissive and Mandatory Disclosures
As discussed above, nurses and physicians comply with police requests for confidential medical data because they are unsure of what they are allowed to do. Consequently, hospitals and clinics therefore must provide better training for their staff on when they can exclude law enforcement from clinical data and spaces. They also must provide real-time security and support if cops threaten staff for not granting them access to this data.
Additionally, nurses and physicians must be trained on the difference between permissible and required disclosures. While HIPAA permits many unauthorized disclosures to law enforcement, it does not require them unless they are accompanied by a judicial order. Institutions should educate providers on the long-term negative health effects of blurring health care and law enforcement, so that they will reject more requests for health data that are not accompanied by a judicial subpoena or warrant.
E. We Need to Reimagine Health Care as Being Off-Limits from Police
I leave the reader with many open questions about how far my proposal should go. When I first began this project, I asked myself whether my desire to divorce law enforcement from health care could justify keeping medical data from the police even in cases of child abuse or neglect. Every state requires physicians to report suspected abuse, and while the efficacy of these laws is debated, they are thought to at least reduce ongoing abuse. Surely, despite the negative impact of police encroachment on the physician-patient relationship and trust, there could be a way to justify it when innocent children are at risk.
Upon reflection, I realized that I had fallen into a common trap. This trap presents complex social problems such as poverty, homelessness, mental illness, and child abuse as being either handled by the police, or not being handled at all. However, this is a false binary. Physicians should continue to report suspected abuse and neglect, but to trained social workers and abuse prevention experts, not to the police. Police should only be involved much later in the process, if at all, after allegations of abuse are validated through an investigation led by trauma-informed social workers. And even then, we should prioritize addressing and treating the root cause of the abuse, rather than reflexively removing children from the home.
Cops are trained to be “authoritative, physical, and commanding”—traits that might be helpful in some policing situations, but are “ineffective” when responding to most health care needs. But because of chronically underfunded social service programs, when people are in crisis, it is the cops who show up. This leads to law enforcement having a de facto monopoly on the provision of many forms of treatment, which are delivered poorly and in ways that violate medical ethics. We simply lack the collective imagination to see many calls to 911 as health problems deserving of compassionate, ethical, and evidence-based treatment by doctors, and not by cops.
On another front, because physicians are easy regulatory levers, states have increasingly been shoehorning politics into medicine. This takes the very thing that makes the physician-patient relationship sacred—trust—and violates it for ulterior government motives. We have seen this recently with bans on gender-affirming care and in the many laws restricting abortion access. Everywhere we look, law enforcement is encroaching on the practice of medicine. To repair the physician-patient relationship, uphold principles of medical ethics, and promote greater trust in health care, we need to keep doctors from becoming cops.
