Medical error

Our understanding of medical errors need not be complex nor controversial; however, the analysis to follow will be made clearer by establishing several basic points. First, following the Institute of Medicine, error can be known as the ‘failure of a planned action to be completed as intended…or the use of a wrong plan to achieve an aim’.1 Second, errors in healthcare settings often lead to adverse events, namely, harms ‘caused by medical management rather than the underlying condition of the patient’.1 Third and finally, unlike the widely cited report, I will not assume that all adverse events occurring as a result of medical error are preventable. Indeed, some cases show that a patient’s injury or death may be an extremely adverse event brought about by medical error, and one that could not have been prevented given the circumstances.9

Consider Nurse Martha Warren, an oncology nurse at an understaffed hospital.10 On a particularly unfortunate shift, Nurse Warren finds herself responsible for tending to five patients, two of which suddenly need immediate life-saving attention. Naturally, she cannot be in two places at the same time, and she chooses to tend to patient 1 while patient 2 has a cardiac arrest and dies. While patient 2’s death is an adverse event to the extent that it was caused by medical management, it is not obvious that Nurse Warren should apologise for her actions. After all, it was simply not possible for her to fulfil every competing demand that had been placed on her. Accordingly, she may be tempted to blame the system, say, for continually allowing critical understaffing. But how exactly would blaming the system help the grieving family? Given that the death of patient 2 occurred on her watch, it seems she owes a personal explanation. I will return to this example, below, in arguing that the explanation would be best received where Nurse Warren takes the blame.

For the moment, consider a second example. Doctor Jean Robert administers insulin to her patient with diabetes after failing to see that an attending nurse has already administered the appropriate dosage.11 While overdosing can have lasting and even fatal effects, Doctor Robert’s patient experiences moderate but temporary discomfort. Unlike the previous case, here it is obvious that the adverse event could have been prevented, say, by Doctor Robert double-checking with the attending nurse. Again, there may be a temptation to pass-off blame onto larger, systemic issues at hand. No doubt, the prospect of letting oneself off the hook may seem even more tempting—and quite feasible—considering the doctor’s position of power. Perhaps the nurse could have noted more clearly that the dosage had been administered. Yet, deflecting blame in this way appears somehow wrong, particularly when one’s ability to do so is directly correlated with an imbalance in power relations. Intuitively, some circumstances more than others allow one to appropriately deflect blame away from oneself. Yet, recent literature suggests that we should shift altogether away from a culture of blame. These suggestions have created a peculiar dichotomy concerning our responses to medical errors.

Blame or safety?

As the examples above suggest, we might think blame is entirely fitting, at least in cases where the practitioner could have reasonably prevented the adverse event. On the other hand, blame itself is often seen as contributing to the ongoing harms. A common theme to the arguments supporting the shift away from blame is the idea that it may actually hinder our attempts at systemic change. It is claimed that medical professionals fear being blamed due to the risk of malpractice lawsuits and the stigma attached to having one’s expertise called into question. Clinical nurse specialist Lisa Day, for example, says ‘if providers fear a punitive response from management they will be more likely to hide their errors…blame-free error and the elimination of shame by shifting inquiry to the system is an important step toward a safer hospital environment’.12

Because blame may allow errors to remain hidden, many authors support more comprehensive reforms, such as non-punitive reporting and collective accountability.6 8 12–17 However, it is not clear that removing or avoiding blame in medical settings is possible, nor is it necessary for improvements in safety or patient-practitioner relations. Furthermore, removing notions of blame threatens to drastically reduce the efficacy of apologies and forgiveness in cases where such exchanges can help to move forward. In other words, blame-free errors in healthcare are not possible or necessary, nor are they desirable.

Taking the blame

While I have argued for the importance of self-blame, in some cases responses like guilt and regret appear to be somewhat irrational, namely where a practitioner could not have prevented the adverse event. Nonetheless, those who take the blame are in the best position to disclose errors, apologise and to demonstrate a commitment to improvement for themselves or on behalf of their institution. In this final section, I explain what it means to take the blame, why it should often be encouraged, and how to account for the negative psychological effects for individual practitioners.

Consider that there are cases where one is not strictly liable in the sense of owing reparations, but at the same time, a harm has been inflicted and it is somehow connected to the agent.25 Bernard Williams’s widely discussed response known as agent-regret, surveyed above, illustrates the reaction we might expect. In Williams’s famous example of moral luck, a lorry driver ‘through no fault of his, runs over a child’.22 Of course, the driver is not blameworthy for the unfortunate state of affairs; yet he feels appropriately bad in a way that differs from the bad feelings of any bystanders. Importantly, with his unique connection to the events—being causally but not morally responsible—the driver is in a privileged position to help the existing victims or to inflict further harm. That is, he can express his genuine sense of regret and allow, say, the child’s parents to find solace and begin moving forward. Alternatively, he might convey—truly yet callously—that he was not at fault.

Referring to these sorts of cases as ‘ambiguous agency’, in a recent work, Mason makes clear that it is often up to us to indicate how a morally significant loss should be taken.25 Many of our relationships, Mason claims, entail duties that require an ‘attitudinal back-up’. We must be invested in our duties to others, in the sense that fulfilling or failing to fulfil them will be accompanied by distinct reactions and feelings. In particular, apologies involving expressions of remorse allow us to show others that we are willing to ‘own’ certain actions. However, in doing so, we not only take responsibility, as Mason holds; for indeed responsibility may belong to us in part as things stand. By taking ownership of harms to others and for damages entirely beyond our control, we show a willingness to take the blame.

But why exactly would we do this? As we can imagine, where we are not fully responsible for harms that befall those within our care, an expression of remorse (or guilt and related responses) might simply open us up to external blame and further punitive measures. On Mason’s account, seeing an action as our own allows us to ‘secure the respect and trust of others’.25 We should be willing to take the blame by expressing negative moral emotions and owning certain actions because of the goods to be achieved for oneself and the institution one represents, and for the substantial benefits imparted on those to whom we have a duty.

In medical settings, patients and families can be reassured that harms are being taken seriously, that practitioners are more concerned with the damages, and how to prevent the recurrence of similar tragedies, than with their own innocence. As Mason notes, we can limit what will ‘count as an excuse’, even where excuses are clearly available.25 Doing so stands to provide immediate consolation as well as hope for an improved future. Additionally, if taking the blame succeeds in providing some degree of consolation and hope for improvements in the delivery of healthcare, practitioners stand to effectively re-establish the respect and trust coveted within the patient-practitioner relationship. Granted, Mason has in mind the ambiguous harms occurring in personal relationships, cases which are unlike the lorry driver and unlike situations involving medical errors. Nonetheless, the mechanism by which we might apologise for errors—including those beyond our control—likewise encompasses our investments in professional duties. Several considerations support this expansion.

First, it is clear that some professional domains entail duties that bear great moral significance. As Williams points out, ‘lawyers and doctors have elaborate codes of professional ethics…because their clients need to be protected, and be seen to be protected, in what are particularly sensitive areas of their interests’.31 Indeed, the duties held while occupying certain offices are such that their failure to be fulfilled may result in catastrophic harm to those who depend on their fulfilment. Given this sort of dependence, there will be situations in which one’s failure to deliver cannot be let off the hook. In short, the harms demand an explanation and sincere offering of reassurance. Notably, this demand is plausibly made in cases of deliberate or inadvertent harms, and it is often most effectively met by an expression of some negative emotional response.ⁱ

Furthermore, professional actions involve ambiguous agency, perhaps even more frequently and more severely than in personal domains. With this recognition, the account established here can be seen as yielding practical recommendations concerning healthcare professionals’ conduct. For example, practitioners may often have a sense that they alone are not entirely responsible for certain harms that befall patients. In the case of Nurse Warren, as I claimed above, it seems reasonable to suppose she owes to the family of patient 2 at least an explanation of what happened. She can acknowledge her role in the adverse event while making clear that it was the result of tragic circumstances. That being said, her explanation of the tragedy to those most profoundly affected by it should be accompanied by a negative affective experience, such as guilt, regret or remorse. With such a heartfelt explanation, the door is then open for the family to find solace and offer their understanding, as all who are affected by the tragedy can work to support one another and see that systematic improvements are carried out.

Of course, it may be that those who could have done no better in playing their part within a tragic causal chain are entirely without some negative response like guilt or regret. Very often we see practitioners who grow ‘cold’ to the harms and losses that occur on their watch.40 To be sure, the notion of ‘compassion fatigue’ has received considerable attention in recent literature, and understandably so, as the psychological effects on healthcare professionals are significant causes for alarm. Compassion fatigue is typically seen as a form of traumatic stress, where caregivers have given too much.41 The excessive degree or duration of their caring has, in a very real way, exhausted their ability to provide additional care. Certainly, it is of the utmost importance that we address these unfortunate realities with policies aimed at promoting practitioners’ psychological health, at preventing burnout and staffing shortages, which of course have negative effects on the overall quality of patient care. Still, the claim to be made here is that where practitioners are entirely unaffected by morally significant losses, they are no longer in a position to effectively apologise to those who care for the victims. Without some degree of self-blame, however irrational it may be, practitioners are less able to show that harms are being taken seriously, and thereby less able to allow patients and families to forgive and move forward than those who take the blame.