Skip to Main Content

In overturning the Trump administration’s attempt to expand the so-called conscience rule for health care workers this week, a federal judge has brought renewed attention to a long-simmering debate in medicine over when doctors can decline to provide treatment to patients without abdicating their professional responsibilities.

The revised rule, issued last spring by the Department of Health and Human Services, was aimed at protecting doctors, nurses, and others from, in the words of HHS, being “bullied out of the health care field” for refusing to participate in abortions, gender reassignment surgery, or other medical procedures based on religious beliefs or conscience. Critics of the rule charge that it would enable discrimination by allowing providers to deny health care to certain patients, particularly women and LGBTQ+ individuals.


U.S. District Judge Paul Engelmayer ruled that HHS overstepped its authority, though the rule sought to “recognize and protect undeniably important rights.” But what are those rights, and in what circumstances can physicians ethically withhold treatment that a patient wants?

There are three general contexts in which it is permissible and sometimes obligatory to refuse care: when doctors are subjected to abusive treatment, when the treatment requested is outside a doctor’s scope of practice, or when providing the requested treatment would otherwise violate one’s duties as a physician, such as the Hippocratic mandate to “first do no harm.” But none of these rationales can justify physicians denying care based on their personal beliefs.

When patients are abusive

If a patient walks into my office using threatening language or behaving violently toward me or my staff and fails to improve his behavior despite good-faith attempts at redirection, I can ask him to leave without receiving care. Of course, there may be extenuating circumstances. A patient in the midst of a mental health crisis who is abusive clearly requires immediate attention. And a critically ill patient who comes to the emergency room engaging in violent behavior but desperately in need of care cannot be dismissed, as this would cause her immediate harm, though security personnel may be required to assist in the delivery of care. Still, in the absence of urgent care needs, I am within my rights to not provide treatment to an abusive patient rather than allow him or her to continue with behavior that disrupts the care of other patients or threatens my safety or that of other health care workers.


Scope of practice limitations

Doctors should not provide treatment outside their scope of practice. As a cardiologist, I have expertise in treating cardiovascular disease and its risk factors, but I do not manage non-cardiac conditions. If a patient of mine with heart disease asks me for pain medication for a lower back strain or antibiotics for an ear infection, I should decline to provide this treatment because it is outside my area of practice or expertise. I should, however, advise him on how best to proceed by referring him back to his primary care physician.

While that may be an inconvenience to my patient, my providing non-cardiac treatment without being up to date on current guidelines and practice standards presents a real potential for harm. My prescribing the wrong antibiotic, for example, might delay him from getting the right treatment and put him at higher risk for infectious complications, which would violate my duty as a physician to do no harm.

Upholding physician duties

The third context in which doctors can refuse to provide certain treatments deserves a closer look. Patients seek care from physicians not only to treat illness but also to promote wellness and flourishing, and physicians have duties to provide this care to the best of their abilities. These include the imperatives to respect patient autonomy, to improve quality of life and longevity when possible, to alleviate suffering, to promote fair allocation of medical resources, and, perhaps most importantly, to avoid doing harm.

When a patient’s request comes into conflict with these duties, a doctor may need to refuse it — though he or she is obligated to do so with kindness and an appropriate explanation of the rationale.

Consider antibiotics again as an example. If a patient comes to her primary care physician seeking treatment for ear pain and requests antibiotics, but the exam points to a viral rather than bacterial process, her doctor can and should refuse to prescribe antibiotics.

First off, antibiotics are not effective against viral infection and thus provide no benefit. In addition, all medications carry the potential to cause harmful side effects. Prescribing antibiotics in this situation would place the patient at an admittedly small risk of harm with zero chance of benefit.

Second, inappropriate antibiotic prescriptions contribute to the growing problem of antibiotic resistance, which causes harm to society and thus violates a physician’s duty to act as a steward of medical resources.

Opioids offer another example. These medications can provide powerful pain relief, but their use may expose patients to a significant risk of abuse and addiction. As such, they require judicious prescribing. Not all pain warrants their use, and they should not be prescribed to placate patients if they are not indicated, no matter how strongly they are requested.

While a physician’s refusal to prescribe antibiotics or opioids may disappoint a patient and potentially result in negative patient satisfaction reviews, physicians are obligated to do no harm and promote wellness over the dubious metric of satisfaction surveys. The customer may always be right, but the patient is not a customer or a client.

We have seen the pendulum of medical ethics swing from a focus on beneficent paternalism (the doctor knows best) toward a focus on autonomy (the patient knows best). I think the right path lies in between. In a typical patient encounter, after I explain my diagnostic and treatment plan to a patient, I ask if it makes sense and if he is on board. The response is often, “Doc, you’re the boss!” to which I invariably reply, “I am the expert, but you are the boss.”

In other words, the patient’s goals and values should dictate treatment, while it is the doctor’s duty to propose potential approaches that are in line with those values and review options to determine the best path toward achieving those goals. Doctors should not try to force treatments upon patients that conflict with their values, and patients should not try to coerce doctors into providing treatments that are medically inappropriate.

Conflicting physician duties

There are some situations in which professional duties inevitably come into conflict with each other. Several states have legalized physician-assisted suicide, though typically with strict criteria such as the need for multiple physicians to confirm the presence of terminal disease and psychiatric evaluation to exclude treatable mental illness. The ethics of physician-assisted suicide are controversial, with compelling moral arguments on both sides of this debate.

Those in favor cite the imperative to respect patient autonomy or right to self-determination, as well as doctors’ duty to relieve suffering. Those opposed argue that helping a patient take her own life profoundly violates the principle of non-maleficence or avoiding harm. This is a situation in which conscientious objection may be ethically invoked. Doctors may ethically decline to participate in physician-assisted suicide if they believe that doing so would violate their professional duties. That said, they should make a good-faith effort to refer the patient to another physician who might be more inclined to consider such a request.

It is not, however, ethical to refuse a patient’s request for treatment simply on the basis of personal beliefs, including religion. Much like our country’s founding principles that enshrine the separation of church and state, medical ethics must recognize the boundaries between church and medicine.

American moral and legal theory have traditionally embraced the Rawlsian conception of liberty — the idea that individual liberty must be respected and protected until one individual’s action encroaches upon another’s liberty. For example, a person does not have the right to act violently toward another because this action robs the second individual of his right to freedom from violence. Through this lens, the term “religious liberty” is disingenuous in that it actually limits the liberty of patients to receive medical care free from the constraints of a clinician’s religion that his or her patients may or may not embrace.

Here is a secular example to illustrate this point. I am a pesco-vegetarian who has chosen to follow a predominantly plant-based diet for health and environmental reasons, and also because I object to factory farming practices involving the slaughter of animals to produce meat. As a cardiologist, my duty is to provide the best evidence-based heart care for my patients. This, of course, includes counseling them on the significant cardiovascular benefits of a plant-based diet in addition to prescribing medications as needed. But I have no business trying to coerce them into adopting my position on food by trying to morally shame them out of their current habits or by refusing to prescribe a cholesterol-lowering medication because that would enable or encourage their consumption of meat.

I cannot imagine anyone would argue that it would be ethically permissible for me to refuse to treat patients who eat meat after having had a heart attack because I object to their diets. This would be morally (and legally) unacceptable. In the same vein, it is no more permissible for physicians to refuse or alter their care of patients based on religious convictions.

It is unethical for a physician to deny care to LGBTQ+ patients because of personal objections about whom his or her patients choose to love in their private lives. It is unethical to refuse to prescribe contraception to single individuals because of personal or religious objections to premarital or nonprocreative sex.

Abortion is a thornier issue because a legitimate metaphysical argument can be made that life begins at conception and, similar to physician-assisted suicide, performing an abortion could be seen as violating a physician’s duty to preserve life and avoid doing harm. Yet forcing women to carry unwanted pregnancies fundamentally violates their autonomy, and thus their personhood.

Abortion is an essential part of health care in that it must sometimes be performed to preserve the health or life of the mother, and in other cases it is necessary to ensure a woman’s right to self-determination as an autonomous adult. While physicians should be allowed some discretion if they truly believe performing an abortion in certain cases would violate their duties as a medical professional, those who would be unwilling to perform abortions under any circumstances for religious reasons are not well suited for reproductive health care.

When objection is not conscientious

While there circumstances such as the ones I described earlier in which physicians can and should decline to provide treatment, the so-called conscience rule goes too far in its allowances. For example, if a pregnant woman comes to the emergency room at night in distress due to what doctors subsequently deem a life-threatening complication of pregnancy and they recommend termination because her fetus is not yet viable, members of the on-call team cannot morally refuse to assist in her abortion. In this urgent situation, unnecessary delays in care from trying to call in additional staff or refer her to another facility may cause her irreparable harm.

It is not a physician’s job to tell patients how to live according to the physician’s personal code of ethics, whether religious or secular. Nor should a physician withhold treatment from patients simply because they fail to adhere to his or her personal standards of morality. Rather, a physician’s duty is to promote patients’ wellness and flourishing through the application of evidence-based medicine to the best of his or her professional ability. Personal beliefs, religious or otherwise, must not interfere with that.

There is nothing conscientious about doctors objecting to caring for patients when we simply disagree with how our patients live their lives. It is unethical for doctors to bully patients in the name of our personal convictions — a blatant violation of our professional duty. We owe it to ourselves and to our patients to hold our profession to a higher standard.

Sarah C. Hull, M.D. is a cardiologist at Yale School of Medicine and associate director of its Program for Biomedical Ethics.

  • This is still a tricky issue as treating the fetus as the biologic being whose needs supersede those of the mother puts us back on some shaky ground. Having said that, I am strongly in favor of only transferring one embryo.
    I am distressed, Ms. (Dr?) Showalter, at your comment: ” I am distressed at how infertility treatment has morphed from treating disease regarding infertility to supporting social activism in the redefinition of a family unit. Gestation has become a corporate venture involving up to 5 “parents” bound by contractual obligations, payments and designer babies.”
    Modern fertility treatment is neither “social activism” nor a “corporate venture.” It is recognizing that all members of our society, regardless of their sexual orientation, have the right to reproduce. We have had contractual relationships regarding parentage ever since marriage became a legal institution, a few thousand years ago. We simply have the means now of helping everyone with a will to do so reproduce responsibly and legally providing for the needs of their offspring by clearly defining these relationships.
    I sincerely hope you don’t believe we should not help LGBTQ persons become parents. That would be a giant step backwards.

  • As a fertility specialist, would it be a violation of patient autonomy for me to refuse to transfer more than one embryo at a time to patient at her request due to higher – primarily prematurity – risks to the offspring? I admit that I tell patients up front that I will never transfer 3 embryos at a time, but the double embryo transfer can be a bit of a sticky wicket, for example, when a gay male couple wants a surrogate of theirs to receive an embryo created with an egg donor and sperm from each partner. Surrogacy is already so very expensive so two separate pregnancies is very costly. Does my conscience prevail in this situation?

    • It is my understanding that there is data showing improved outcomes with single embryo transfers, and this is standard of care is some countries. It has not become standard of care here because of cost. That said, twin pregnancies can significantly increase cost of care because of complications and risks of preterm delivery, so the total cost, not just cost of ART should be considered. When a patient is seeking pregnancy, it is my belief that the interests of the offspring should come first. It is my understanding that if you make a policy based on sound medical judgement that is uniformly administered, you can make such a policy for your practice. While we respect patient autonomy, physicians are not automatons obligated to do whatever the patient wants regardless of ethical considerations. I am distressed at how infertility treatment has morphed from treating disease regarding infertility to supporting social activism in the redefinition of a family unit. Gestation has become a corporate venture involving up to 5 “parents” bound by contractual obligations, payments and designer babies.

  • To quote Dr Hull:

    “those who would be unwilling to perform abortions under any circumstances for religious reasons are not well suited for reproductive health care.”

    Brava. I couldn’t agree more.

Comments are closed.