May 11th, 2017

Leaving Against Medical Advice (AMA): A Clinician’s Dilemma

Alexandra Godfrey, BSc PT, MS PA-C

Alexandra Godfrey, PA-C, practices emergency medicine in North Carolina.

“You’ll have to sign out against medical advice (AMA) . Your blood pressure is high.” The ER physician stood in the doorway of my room.

“What difference would it make now?” I asked.

The doctor fiddled with the cuffs of his white coat, then glanced at his cell phone.

I picked up my car keys.

“High blood pressure is dangerous,” he added.

I looked down at the floor, watching puddles form around my feet as the snow melted on my UGGs. A cold wind blew snow against the window, causing the blinds to rattle.

“I don’t know the cause of death,” I said.

The doctor tapped his foot on the floor; “You could have a stroke. And your insurance might not cover this visit, if you leave AMA.”

I leaned forward in my chair, holding my head in my hands. It was then that I saw the photograph tucked in the top of my purse. Perhaps this would help?

I reached into my purse and offered the picture to the doc – that grainy ultrasound image of my son.

“Do you want to see his picture?” The doctor shook his head. I crumpled in my seat. Tears gathered.

“High blood pressure might damage your kidneys or your heart,” he said, crossing his arms over his chest.

Damage my heart?

I bit down on my bottom lip, then formed my hands into a temple. My obstetrician had insisted I come to the ER.  I hadn’t wanted to come. The experience had been awful from the beginning.

I looked at the doctor through the arches of my fingers.

“Do you own a dictionary.” I asked.

“Yes – why?”

“You should look up the word empathy,” I replied in a quiet voice.

I got up then and walked out. I did not look back. Not once.

I had never left a hospital AMA before. And I have not left AMA since. Even now, years on, I know I made the right choice. It changed everything.

The AMA Dilemma

Certainly, patients discharged against medical advice are both a concern and a challenge for healthcare providers due to the increased risk of both litigation and adverse medical events. Furthermore, these patients are  more likely to be readmitted for the same or a related condition. Additionally, they frequently suffer from serious underlying pathology.

What does leaving against medical advice (AMA) mean?

When a patient leaves AMA, the patient is leaving before their treating physician recommends discharge or despite medical advice to the contrary.

This definition implies the patient received and understood the medical advice given. In practice, the term AMA is often used regardless of whether medical advice was given or not. Additionally, patients are frequently required to sign an AMA form. Unfortunately, these attestation forms are often confusing, coercive, and written in language beyond the reading level of the patient.

This creates two important questions:

  • If a patient does not have sufficient information or understanding to make informed decisions, are they in fact leaving against medical advice?
  • Shouldn’t health literacy be a consideration in AMA discharges?

Who leaves AMA?

Identifying those patients most at risk for leaving AMA is important if we are to design interventions to prevent it. Variables include:

  • Alcohol or substance abuse
  • Male sex
  • Low socioeconomic status
  • History of leaving AMA
  • Black race
  • Psychiatric disorders
  • Absence of health insurance
  • Homeless

AMA Healthcare Disparities

Certainly, the choice to leave a medical facility lies with the competent patient, but the choice to designate the discharge as AMA lies with the medical provider. Despite this, there is a lack of data about providers who discharge patients AMA. To fully understand AMA predictors, we need to know more about clinicians.

It is well known that clinician attitudes vary based on patient characteristics. Such variability may shape the quality of care given. Stigma may result in lower quality care resulting in poorer outcomes. Furthermore, if you consider the groups most at risk for AMA discharge, there’s clear intersectionality between them. For instance, a black sexual minority male is more likely to be homeless, less likely to have insurance, and at higher risk of depression or other mental illness. Consequently, a disproportionate number of patients leaving AMA come from stigmatized or marginalized minority groups.

Why do patients leave AMA?

Recognition of the reasons patients leave can help us negotiate these encounters. They may include:

  • Personal and financial obligations
  • Breakdown of communication between the patient and medical staff
  • Dissatisfaction with care
  • Lack of trust in the healthcare provider
  • Not understanding the need for further testing or treatment

Medico-legal Considerations

Above all, providers need to be aware that the simple signing of an AMA form does not confer absolute medico-legal protection. Furthermore, not giving discharge instructions, follow-up or medications could be construed as coercion, negligence or unwillingness to consider alternative options for the patient. Moreover, advising a patient that their insurance will not cover their visit if they leave AMA is not only incorrect but could also be considered coercion.

This article provides an excellent review of medico-legal standards.

In Conclusion

For me, leaving the ER AMA gave me precious time alone with my son. Moreover, it removed me from a hostile and unpleasant situation. I returned to the hospital the next day to deliver my son. He had died suddenly in the second trimester. No amount of ER medicine would have changed that outcome. My experience that night precipitated my return to education and pursuit of a career as an emergency medicine PA.

Certainly, the onus is on all providers to carefully examine the role that their bias, language, and behavior might play in triggering an AMA discharge. Furthermore, patients must understand the medical advice given and the potential sequelae of their decisions if we are to document AMA. Providers should take time to explore their patient’s thinking and engage them in shared decision making. Repairing the patient-clinician alliance may help facilitate communication and rebuild trust. And of course, you don’t get told to read the dictionary…

Register Now for more NEJM Journal Watch Content

24 Responses to “Leaving Against Medical Advice (AMA): A Clinician’s Dilemma”

  1. Larsen, MD says:

    This article is a perfect example of why PAs should not practice medicine: you think a tragedy is a reason to forgo safe medical care. I’m absolutely speechless. Would you really recommend a patient “take all the time she needed” and hope that her medical condition magically sorts itself out?

    Me being empathetic to a patient is helping her receive appropriate and complete medical care– you know, treating the “whole patient.”

    • Alexandra says:

      Larsen: I would not recommend a patient takes all the time she likes to grieve for a lost child in the context of a potentially life threatening or dangerous complication. That wasn’t my point. The anecdote is more an illustration of what we don’t want to happen.

      The point here is there was a failure in communication between the patient and the medical team. NO – this isn’t ideal and not the recommendation I make. It’s an anecdote and an illustration. The onus of the article is for clinicians to consider how and why AMA discharges happen AND to point out why they are dangerous. They are dangerous because patients have a higher mortality rate and dangerous because readmission is common and often patients have serious pathology. I wrote this to help us prevent them. The clinician here was an MD not a PA but I am certain all medical providers have experienced AMA discharges. The anecdote is designed to illustrate what can happen when communication breaks down. We all need to work on that … and it has little to do with PA versus MD practice. I am a PA and clearly I have the capacity to evaluate why AMA discharges happen and how these might be prevented and what we can do to prevent them. I think that says quite a lot about PAs. And also – I question clinician bias and intersectionality – which means we all need to evaluate our own practice and thinking for implicit and explicit bias. It seems to me bias abounds and it prevents us from learning from our experiences. All this means I put great thought into my clinical practice.

  2. Joe Wilms says:

    This is very confusing. Why did the author leave AMA? Because the doctor didn’t want to see pictures of her son’s ultrasound and was awkwardly looking at his phone? How did leaving only to return the next day change everything? Why were blinds rattling as if you were inside, but snow melting on your boots? Why was there no concern about a dangerously high blood pressure?

    • Alexandra says:

      The reasons I left AMA were complex and not appropriate for here. They were powerful ones. Certainly you can have a draft strong enough to rattle blinds but not enough to cool the air temp of the room. Was I concerned about my high Bp at that point ? Not really. I didn’t know what that meant. This was a long time ago and before I knew about such things. Improved communication, education and some empathy might have changed that outcome. Why did it change everything ? Because I got time with my unborn child, which to a mother who has just lost a much wanted son might be everything. And I also realized how much caring for the “whole patient” mattered. If that had happened, I probably wouldn’t have left.

  3. Eman Zaher says:

    Very nice article, I wish every health care worker take a course or workshop about communication skills and how to empathize with patients ….
    Sometimes we ( physicians) are the main reasons of patients non compliance to thief treatment plan …
    I personally speak from a physician who was disappointed many times by colleagues in medicine when I was a patient ….

  4. Max Voysey says:

    It is a TWO person issue – not a one person issue. It stands as a sign of the breakdown of a relationship usually, although I have had pleasant AMA discharges.
    EVERY patient who fails to appear, is LTF (Lost to Follow up as the researchers like to report), storms out, or starts litigation against a physician is “AMA” in some manner, as are all antivaxers, and many alternative consumers.
    In your case perhaps it should have been ATD – agree to disagree, or NFMAA (No further medical advice applicable).

    • Alexandra says:

      I agree with you Max. It is a TWO person issue. Shared decision making, taking time to listen, reparing the physician – patient alliance and so on. I like your proposal of agree to disagree.

  5. Dr FJJ Douie says:

    Poignant article, and very apt. I am reminded of the film “the doctor” where the arrogant know-it-all surgeon becomes the patient of the man whose style he has frequently ridiculed to his students, and the subsequent awakening of a mild glimmer of humanity in him.
    If we did not do this job because we care, we will always be disappointed by what it gives back.
    As a side issue, I also feel that most people whose blood pressures are elevated in the ER have an adrenal system reacting appropriately to a recognition of their current environment. I worry about those who are normotensive in ER . . .

    • Alexandra Godfrey says:

      Dr Douie –

      I love the film “The Doctor”. He does indeed realize through his experiences the need for more humanity in medicine – and how arrogant his prior practice has been.

      My experience was almost the reverse. I realized by being a patient how arrogance and lack of caring can cause unwanted outcomes. It was important learning.

      High stress and ER presentations can definitely elevate BP.

      Thank you for your response.

  6. Sara Lolar, PA says:

    I get frustrated when clinicians “punish” a patient for leaving AMA. Like not giving needed pain meds or antibiotics. I find there is always a REASON patients leave. Oftentimes it is a social problem (need to lock up house, pick
    Up kids etc) or a misunderstanding/disagreement on the proposed treatment plan. I can almost always work out a “deal” with a patient so they come back. In my opinion, a patient leaving AMA is usually a failure on my part.

    • Alexandra says:

      I agree with you Sarah. AMAs often take on a punitive tone. Patients leave for a reason and it is up to us as clinicians to work out the safest plan for the patient.
      Thank you for your response.

  7. john schmitz says:

    I work as a PA in a community wellness center. We send many patients to the nearby ER since we can only do simple stuff. A lot of them, maybe half, leave AMA (or often without medical advice). Frequently, this is without ever getting out of the waiting room. I think most of the time, this is due to a few causes similar to those mentioned in the article
    – absent or insufficient communication with medical staff plus other commitments. Some of the shelters have deadlines for when you have to be there to get a bed. Long waits sitting without knowing what Is going on will drive people out.
    – lack of trust of the hospital. These are poor women who have been screwed by big institutions all their lives.
    – mental illness / substance abuse. People with mental illness and substance abuse get sick and injured also. But they often do not cooperate very well. or understand the need to wait around.

    If we have something serious and a particularly difficult patient who has developed rapport with one of us, we will sometimes accompany them and sit with them either in the waiting room or even in the ER to make sure they get the care they need.

    • Alexandra says:

      John Schmitz – thank you for your response. You really summarized the problems faced by the AMA population well. Poor communication, lack of trust, mental illness, poverty – these are all strong predictors for AMA. We try very hard in our facility to consider all these variables and decrease the likelihood of poor outcomes. Thank you for your insights and continue what sounds like excellent work.

  8. Jennifer says:

    What is the opposite of this- like when a severely sick individual is discharged from the hospital with a pcv of 11 despite multiple blood transfusions, due to severe prolonged and non-medically manageable uterine hemorrhage, even though the patient explains that they live alone and have no support system and do not feel safe being discharged? Because that is what happened to me and I am a doctor and capable of recognizing the serious risk I was being forced to take.

  9. Dr. J says:

    Sorry for your loss. I liked the review of AMA discharges. Sometimes what we do is not meeting the patient’s perception of their needs. Occasionally the patient is right and the medical team is minimizing their own risk and fulfilling an institutional policy with the patient’s needs not being well addressed. I don’t like the coercion that is used to try to get patients to comply. The linked article was eye opening.

    • Alexandra says:

      Hi Dr J –

      I agree. Sometimes it is more about medicolegal risk than it is the patient. I found that article interesting too. It really helps you appreciate some of the nuances of what we do … and whether we should be doing them at all. Thank you for commenting.

  10. J. Mitchiner says:

    Nice article, Alex. I have found I can talk patients out of leaving AMA about 70% of the time, just by listening to them & finding out what their concerns & misunderstandings are.

  11. esdee says:

    “In practice, the term AMA is often used regardless of whether medical advice was given or not. Additionally, patients are frequently required to sign an AMA form. Unfortunately, these attestation forms are often confusing, coercive, and written in language beyond the reading level of the patient.”

    Is it true that ‘AMA’ is used whether medical advice is given or not? I’d be curious to see the data. I have seen many disturbing things in my 12 years of practicing medicine, but I’ve never seen a patient coerced into signing out AMA. On the contrary, I’ve seen members of the clinical team bend over backwards to ensure that a patient understands the consequences of her/his decision. It’s actually what lifts my spirit sometimes in the hospital, because I know that for that nurse/doctor/NP/PA/pharmacist, it would be MUCH easier to just let the patient sign and walk. But they take the time to exhibit true compassion and caring.

    Your experience was obviously different in the story you related. I’m sorry for your loss.

    • Alexandra Godfrey says:

      Hi Esdee –

      I too have seen clinicians “bend over backwards” to ensure patients understand that consequences of their actions. Definitely this happens but our experiences do not always reflect widespread practice. I am sure like me you have experienced practice that ranges from excellence to subpar.

      I think it is hard to study and quantify what is and isn’t appropriate medical advice for patients given the diversity in complaints, presentations, medical knowledge and rationale. However, there’s several excellent articles written about AMA discharges that highlight some of the misconceptions, methods of coercion or problems in managing these patients. My goal in writing this blog is stimulate discussion that help all clinicians consider how we can improve outcomes for all involved, I am also concerned about how bias and intersectionality creates an AMA population that disproportionately represents some minority groups.

      I have attached some studies that I found useful as I wrote this blog. You’ll find some applicable data in these and further citations that may answer your question. Certainly, we need more data about AMA discharges if we are to improve outcomes and our practice.

      Choi M, Kim H, Qian H, Palepu A. Readmission Rates of Patients Discharged against Medical Advice: A Matched Cohort Study. Laks J, ed. PLoS ONE. 2011;6(9):e24459. doi:10.1371/journal.pone.0024459.

      • esdee says:

        Thanks for the references. They are interesting, but only focus on documentation – we know that what occurs is often different than what is documented.

        Also, none of these provide evidence that AMA discharges are often ‘coercive’. That is a hefty charge to throw around, and one that I have never observed.

        I appreciate the dialogue here. Keep it up!

  12. Liz Brennon says:

    In reference to the post that they leave due to having to pick up kids, etc. – sometimes they don’t come in at all because they don’t have child care, can’t leave them alone, and also recognize they might be hospitalized and then what will they do? When kids land in social services it is extremely difficult to get them back and traumatic to all. So they take a risk and hope things will turn out OK. Not the best decision for their own health but sometimes they see it as their only viable choice.

    When, particularly, a single parent comes in and then wants to leave, it may be worth having the hospital’s social services help them figure out child care logistics, picking up kids from day care at closing… or even suggesting that they leave, pick up the kids, drop them off at a friend’s and then come back if they can’t find someone to go get them (and recognizing that often child care will refuse to let a child leave with someone “not on the list”). The later isn’t idea but it can be better than leaving and not coming back… or having young kids get off the school bus, the parent(s) aren’t home and then social services gets called…

  13. lwp says:

    So sorry for your loss.

  14. Andrew Will says:

    I am really sorry to your loss as well.

NP/PA Bloggers

NP/PA Bloggers

Elizabeth Donahue, RN, MSN, NP‑C
Alexandra Godfrey, BSc PT, MS PA‑C
Emily F. Moore, RN, MSN, CPNP‑PC, CCRN

Advanced practice clinicians treating patients in a variety of settings and specialties

Learn more about In Practice: Reflections from NPs and PAs.