As a physician training in obstetrics in an Australian hospital, I (J.C.) found myself faced with situations in which patients, despite increased risk of harm, chose to discharge themselves from the hospital against medical recommendations that they remain. My experiences were not isolated ones; these incidences of discharge against medical advice (DAMA) have remained relatively stable over several decades at 1 to 2 percent of all hospital discharges. Footnote 1,Footnote 2
Existing ethical work on DAMA tends to focus on the question of whether a patient’s decision to discharge himself or herself ought to be respected. This question is commonly addressed using a principlist approach. Footnote 3,Footnote 4,Footnote 5,Footnote 6 DAMA is cast as a conflict between the patient’s autonomy and the physician’s goal of beneficence. Invariably, such accounts find that once a patient’s decisionmaking capacity is established and he or she is sufficiently informed, the physician ought to respect the patient’s decision. Little, if anything, is said about what ought to happen next. This approach to the ethics of DAMA is operationalized, from an institutional perspective, in the form of a legal waiver that the patient is required to sign to indicate that he or she understands the situation and accepts responsibility for any adverse outcome.
The physician’s role is reduced to one of assessing competence and providing information, rather than engaging with the full scope of the practical challenges that a physician encounters in a DAMA situation (“What should I say to this patient?” “Why has this happened?” “How do I manage my frustration?” “How should I document this?”). Standard ethical analyses focus only on whether a DAMA is justifiable, providing insufficient ethical insight into how a DAMA may proceed in an ethical way. Empirical evidence suggests that physicians involved in DAMA feel self-doubt and uncertainty about potential missed opportunities, and struggle with real-world constraints and their sense of responsibility for the patient’s ongoing well-being. Footnote 7 Hence there is a pressing need to explore a different and additional ethical question about DAMA: not just “should a physician respect a patient’s decision to discharge himself or herself?” but also “how can a physician facilitate an ethical DAMA?”
The case of “Lorraine” described in the next section is based on my own experiences and is an amalgam of specific encounters I had with patients during my training. Although details have been altered to maintain patients’ anonymity, the circumstances and remarks contained within the case are real.
Lorraine
I was asked to see Lorraine, a woman who had been admitted to hospital into our unit and now wanted to go home early. She had presented the day before with abdominal pain and was 33 weeks pregnant. It was unclear at the time if she was in early labor or not. She had had scant care during her pregnancy, and from what we could gather, she had previously used intravenous drugs, had contracted a blood borne virus, and was addicted to heroin, for which she was now on methadone-substitute therapy. She was quite agitated, and this limited our ability to assess her fully.
We gave Lorraine a course of steroids to minimize the risk of neonatal complications should the baby be born early, and arranged for her to be monitored closely while the pain continued. Blood tests were sent to screen for other non-pregnancy causes of her abdominal pain. We arranged for her daily methadone, which had been missed for some time, to be given. Her pain subsided and she was moved to the ward from our birth suite when we felt that it was unlikely that she was in labor or had another more sinister cause for her episode of abdominal pain.
Given Lorraine’s scant antenatal care, we took the opportunity of her admission to complete some tests that had not been performed, including an ultrasound scan to assess the fetus’s growth and well-being. Our team of midwives and addiction medicine specialists ensured that Lorraine would be able to receive methadone appropriately when she went home. Our social workers were able to ensure that she had safe and stable accommodation to go home to. All this was accomplished in less than a day. From a medical perspective, she was stable and, having had a course of steroids, her fetus was as prepared as it could be for early delivery, should that occur.
I was asked to see Lorraine late in the afternoon. She wanted to go home; she said that she had more important things to do than hang around in the hospital “waiting for nothing to happen.” I reviewed the interim ultrasound report. Lorraine’s baby was small, and potentially growth restricted. One measurement was abnormal, a potential sign of early placental dysfunction. Certainly, the fetus was not showing any signs of distress; in fact it was otherwise perfectly well. There was nothing on the ultrasound to suggest that the fetus required imminent delivery; close surveillance was the prudent choice.
I communicated all this to Lorraine, who appeared to take it all in, because after I’d finished she said, “So what you’re telling me is that something might be wrong or nothing might be wrong and that you want to scan the baby again in a few days… so I’m still sitting here waiting for nothing to happen.” She wanted to go home. I said that the prudent thing would be to stay another night and revisit her situation in the morning. I reassured her that there did not appear to be any sinister cause of her initial pain, and I struggled in my mind to see why in her case staying another 12 hours would be likely to change things significantly.
Lorraine apologized for her behaviour on admission and said, “I’ve got nothing against you, doc, but I’ve got other things to do.” The more she said that, the more I thought, “I’ve got other things to do, too.” As Lorraine began unplugging her mobile phone charger and packing her bags, it became apparent that this was no longer a negotiation. I explained to Lorraine that my advice was to stay another night and that she’d have to sign the Discharge Against Medical Advice waiver that the hospital required. If she had stayed until the next morning, we would have let her go home and booked in her extra ultrasound scans and an appointment in the clinic. So I did both; Lorraine signed the form and I arranged for her follow-up as if she were simply going home.
Discussion
I’ve only recently thought about the patients I’ve known to have “discharged against medical advice.” Sometimes I’ve been asked to see them after hours, after their treating team has gone home. Often I’ve grappled with brief medical notes, trying to piece together the treatment plan and make a rapid assessment of the harm that might befall these patients if they left at that moment. And always I’ve been troubled by the simplicity of the Discharge Against Medical Advice waiver. It seems to absolve our responsibility to these patients. Surely we owe them something more? Addressing that “something more” is our concern here.
It is clear from cases such as this one that the primary ethical question for a physician in the situation described in the previous section is not whether to respect Lorraine’s decision to discharge herself. At a certain point in the narrative, it becomes clear that the patient is definitely going to leave regardless. The ethical issue then is how ought the physician to deal with the fact that this patient is going to leave?
If patient autonomy is justifiably being respected at some possible cost to patient well-being, minimizing that cost is an appropriate guiding idea in this context. This is particularly appropriate in the obstetric context, where fetal well-being may also be at risk. We can thus conceptualize an ethical DAMA as one in which such risks of harm are minimized. The current organizational response of simply requiring the patient to sign a waiver focuses attention only on potential harms to the institution. We suggest that a more justifiable organizational response would be one that also encourages physicians to minimize potential harms to the patient (and fetus) that could result from the DAMA.
A crucial way in which these harms can be minimized is to facilitate the patient accessing follow-up care. Follow-up care minimizes the health risks associated with the early discharge. Secondarily, it also contributes to the physician’s sense of having fulfilled his or her obligations to the patient, even in the context of the patient’s decision to discharge. From the physician’s perspective, facilitating follow-up care may include clarifying any outstanding investigations, requesting the investigations be conducted once the patient has left the hospital, and enabling the patient to be reviewed by an appropriate physician in a timely way; for example, by making a clinic appointment on his or her behalf.
Lorraine’s situation highlights several potential barriers to follow-up care. One barrier is the patient’s complex needs, such as substance misuse. Complex needs can impair patient access to follow-up care. A patient’s complex needs may include housing, financial issues, language barriers, and specific cultural needs. Any of these, or a combination, may impair a person’s ability to access care, as exemplified by Lorraine’s “scant antenatal care.” Patients with complex care needs often benefit from a multidisciplinary approach; that is, they are cared for by a number of individuals with expertise in a particular area as was the case in Lorraine’s situation. However, complex needs are often unable to be resolved before the patient leaves, and can make it more difficult for the patient to access follow-up care once discharged.
A second important barrier to patients’ accessing follow-up care is the team context in which hospital care is delivered. In Lorraine’s case, the discussions around discharge were the first time that the physician had met this patient. Lorraine’s stay in an Australian public hospital is typical in that she was admitted to the hospital under the care of a treating unit, which commonly comprises several junior physicians who operate with varying degrees of autonomy under the guidance of a more senior physician. Often there will be multiple specialists who are involved in a patient’s care, depending on the degree of medical complexity. A team approach is useful in improving efficiency, and provides training opportunities, with more experienced physicians performing more complex tasks. However, the team approach also raises specific challenges when a patient chooses to discharge against medical advice. First, these events are unpredictable and unplanned, meaning that the physician called to see a patient may not be the physician most familiar with the patient, the most experienced physician on the team or, indeed, the physician with the best rapport with the patient. He or she may be the most junior member of the team who, likely, also has “other things to do, too.” Events can progress rapidly, as they did in this case, limiting the ability of junior physicians to seek help from more senior colleagues. Dealing with a DAMA after hours seems to exacerbate the junior physician’s challenges of lack of time, experience, and familiarity, and the unavailability of more senior colleagues.
These barriers to patients’ accesing follow-up care after a DAMA are not well addressed by an organizational approach that involves only a waiver. We suggest a broader checklist as a more appropriate organizational mechanism for supporting an ethical DAMA. Such a checklist works as a tool that prompts physicians faced with a patient requesting a DAMA to consider the potential harms to the patient and to facilitate optimal follow-up. We acknowledge that there may be challenges in completing this type of checklist, especially for physicians working after hours who are not familiar with the patient or that patient’s circumstances. In these situations, it is necessary to alert the treating team of the patient’s discharge and the need for follow-up to be arranged as soon as practicable. We suggest that, in such situations, the discharging physician carry a responsibility for facilitating follow-up, either directly or via alerting and reminding the treating team. Given the importance of communication to the doctor–patient relationship, we suggest that a copy of the complete checklist also be provided to the patient.
-
1) Patient name, date of birth, hospital identifier, contact details
-
2) Treating team
-
3) Date of admission
-
4) Reason for admission
-
5) Other issues identified during admission
-
6) Other teams involved in this patient’s admission (e.g., social work)
-
7) Relevant results
-
8) Investigations outstanding or incomplete
-
9) Follow-up arranged/needs to be arranged
-
10) Completing physician
-
11) Copy provided to patient
-
12) Copy provided to treating team, if completed after hours
This kind of organizational mechanism encourages a more holistic approach to DAMA than a waiver. It encourages and enables staff to focus on minimizing the risks of the DAMA to the patient involved.
Conclusion
An ethical DAMA involves much more than respecting a patient’s decision to leave. It requires an awareness of the patient’s needs, an appreciation of other health professionals who may be involved in that person’s care, and an undertaking to arrange appropriate follow-up. Physicians struggle with DAMAs; they are challenged by a DAMA’s complexity and the real-world constraints of working in a complex health system. In this context, a systematic approach to the patient who decides to leave is required. To this end, we have proposed a checklist as an organizational mechanism for supporting an ethical DAMA; one that seeks to meet the needs of physicians, the patients they care for, and the organizations in which they work.
