Mistakes, Misguided Moments, and Manslaughter

On 23 June 2008 in the New South Wales District Court Murrell SC DCJ brought to its conclusion a legal saga that had begun at Kogarah over 5 years previously. This case should be of considerable interest to perfusionists because manslaughter charges against health professionals have hitherto been very uncommon in Australia, and it may be a sign of a wind of change on the medico-legal landscape internationally. Furthermore, it involved the administration of drugs that would typically be within the domain of an anesthetist by a health professional with very different background and training (the parallel with perfusion is obvious).

On 6 November 2002, Dr. P (a dentist) performed a dental implant procedure on SC, a 67-year-old overweight man with several relevant risk factors. Dr. P had completed a diploma course in sedation in 1999, and he had used intravenous sedation for >680 procedures. Dr. P administered 17 mg midazolam, 150 mg propofol, and nitrous oxide in oxygen (through a nasal mask) to supplement local anesthetic and allay SC’s anxiety. He was assisted by an experienced registered nurse (tasked with monitoring the patient) and two dental assistants. Pulse oximeter readings <90% were recorded on a number of occasions during the procedure. Dr. P had interrupted surgery on each of these occasions to extend the patient’s neck and open the airway until an acceptable saturation was restored; his evidence was that communication was maintained throughout the procedure.

Over the final 10 minutes of surgery the oxygen saturation fell steadily, until the nurse informed Dr. P that SC was seriously hypoxic and hypertensive. Surgery was terminated, and resuscitative procedures were instituted, but to no avail: SC suffered a “critical loss of blood oxygen saturation” and died of hypoxic brain damage 2 days later (1).

It seems reasonable to suggest that some errors were made in this case—but the problem facing the court was whether these errors amounted to a crime.

There is a substantial body of theoretic and empiric information concerning the nature of error (2–5), some of which is counterintuitive. Importantly, errors are not random events, but occur under circumstances that are well understood. They can be classified, and certain types of error can be predicted in particular situations. Errors are unintentional and are made by experts and novices; they do not necessarily imply carelessness in the normal sense of the word, and the corollary of this is that deterrence is ineffective in preventing them. The consequences of an error may be entirely out of proportion with the magnitude of the error itself, and it is a reflection of human nature that there is a strong tendency for punishment to be based on these consequences rather than on the underlying behavior that lead to them. For example, if one gives the wrong drug and is fortunate enough to cause no harm, the likelihood of punishment is low, but if the patient dies, punishment is almost inevitable and likely to be serious—yet the behavior is identical in these two cases.

Violations are quite distinct from errors. They involve choice, and the corollary of this is that it may be possible to deter violations. Violations may predispose to error and may make serious consequences more likely if error does occur. Some violations are routine, are tolerated as normal (this is common in healthcare), and may even be encouraged by those in authority, implicitly or explicitly. Not all violations are equally culpable, and it may be quite difficult to differentiate between certain violations and certain errors.

In legal terms, violations may be thought of as manifestations of recklessness and errors as examples of negligence. It could be said that violations manifest at least some degree of mens rea (literally, a guilty state of mind) and that errors (by definition) do not. Errors and violations may be without consequence but often contribute to harm (known as iatrogenic harm) caused by healthcare (4). In some cases, the harmed patients are very sick, with complex medical conditions (5), but this does not in itself exonerate those responsible. In others, relatively well people are involved, and either way, it is not surprising that health professionals who have made errors have faced criminal prosecution for alleged negligence or recklessness from time to time. Nevertheless, in most countries, there has, in general, been a reluctance to resort to criminal law for events arising out of the normal medical care of patients.

To understand why, it is necessary to compare the various processes by which society responds to negligence. Most practitioners fear civil actions (except in New Zealand, where the no fault compensation system makes this risk negligible). Without doubt, litigation can be very stressful to all concerned, but in the end, these law suits fall within the range of the normal interactions that take place between individuals within society and do not necessarily imply serious moral blameworthiness. Similarly, it is generally understood that disciplinary proceedings lie within the realm of professional regulation. Being prosecuted by the state for a serious crime, on the other hand, carries strong connotations of moral opprobrium. Typically, it involves being arrested, taken to the police station, and formally charged by police officers. Photographs and fingerprint impressions are required. Jail suddenly becomes a possibility, even pending possible conviction. An application for bail is required, and one’s passport must be surrendered. A conviction would result in restricted entry to many countries (to say the least). At depositions and hearings, one’s name may well appear on lists that include people accused of crimes such as assault, theft, and rape. None of these features are associated with civil or disciplinary proceedings.

These differences are reflected in the required standard of proof: civil cases are determined on the balance of probabilities, whereas criminal charges must be proven beyond reasonable doubt (the requirements for disciplinary proceedings vary between countries). In general, they are also reflected by a high barrier to prosecution, and until recently, New Zealand was the outstanding exception to this.

In New Zealand, the section of the Criminal Code usually cited in manslaughter charges against health professionals is s155.

155. Duty of persons doing dangerous acts: Every one who undertakes (except in the case of necessity) to administer surgical or medical treatment, or to do any other lawful act the doing of which is or may be dangerous to life, is under a legal duty to have and to use reasonable knowledge, skill, and care in doing any such act, and is criminally responsible for the consequences of omitting without lawful excuse to discharge that duty.

This requirement for reasonable knowledge, skill, and care is usually associated with the civil standard of negligence. However, in New Zealand, it was explicitly established in the early (motoring) cases of R v Dawe (6) and R v Storey (7) that the standard of negligence required for a criminal conviction under s155 was no greater than that needed for civil actions.

In 1987, an anesthetist came to New Zealand to work at a small hospital. During his first year there, under difficult circumstances, he mistakenly administered an ampoule of dopamine to a patient instead of doxapram (of similar appearance). He was open and honest about his mistake, but the patient died. The doctor was subsequently convicted of manslaughter (8). In summing up, the judge explained that the case against the doctor was based entirely on the contention that he had made a mistake, through carelessness, on one occasion. This hardly sounds like a good basis for a manslaughter prosecution, and the many mitigating circumstances in this tragic case have been outlined elsewhere (9), notably the fact that the error had been made in circumstances that could be considered an emergency. The New Zealand Courts appeared to take the position (in effect) that emergencies do not occur in hospitals, but only in situations such as (for example) motor vehicle accidents at the side of the road (10) and that the various mitigating factors were primarily relevant to sentencing rather than to the question of whether or not the doctor should be convicted.

By 2004, 10 health professionals had faced charges for manslaughter for alleged negligence in the normal conduct of their work; New Zealand has <4,000,000 inhabitants, perhaps a 5th of Australia and a 15th of the United Kingdom. All of these prosecutions related to mistakes— errors or what one might call misguided moments; none involved crimes in the common sense (e.g., fraud, rape, or assault), or even willful violations of expected standards of professional behavior (such as working under the influence of alcohol).

Not surprisingly some doctors (and other health professionals) became anxious about the implications of these prosecutions. One of the requirements for a manslaughter charge in the context of healthcare is the death of a patient, and those working in high-risk specialties felt particularly vulnerable. For example, two of the people prosecuted were cardiothoracic surgeons and several were anesthetists.

The New Zealand Medical Law Reform Group was established in response to this increased anxiety (11) to coordinate initiatives from organizations of the medical profession seeking “a proper balance between the criminal code and other means of accountability in medicine” (12). A key principle identified early in the campaign of this group was that any change in the law should apply to everyone; it is not the profession of the person involved in these events that should determine whether or not criminal prosecution is appropriate but rather the level of blameworthiness (13).

In 1997, the New Zealand crimes act was amended to require “a major departure” from the standard of care expected of a reasonable person in the circumstances. This aligned the New Zealand position with those countries (notably Australia, the United Kingdom, the United States, and Canada) in which the requirement is generally expressed by reference to “gross negligence.”

Not everyone will immediately agree that this change in the law was a good thing, and so it is worth considering in more detail the appropriate response when a patient is inadvertently harmed by healthcare. It is widely perceived that many patients who have been harmed want some or all of the following things (14): an acknowledgment that something has gone wrong; an apology; an assurance that steps will be taken to avoid the same harm to someone else; compensation; and punishment of the healthcare professional who caused the harm.

In fact, the priority is to ensure that the patient is adequately cared for. The staff, often referred to as the second victims, must also be looked after, if only to ensure that they do not compound matters or harm other patients while distressed by the primary event (15). One can then move on to an acknowledgment that something has gone wrong. This is obviously the right thing to do, morally, but the requirement for “open disclosure” has now also been given legal force in many jurisdictions. Apologies are appropriate but need to be genuine. An engaged attempt to understand what went wrong and to prevent the same thing from happening again is clearly essential. One of the greatest weaknesses of criminal law is that prosecutions tends to focus on establishing negligence and causation, and wider issues relevant to the overall standard of care and how this may be improved in the future are often neglected.

Compensation is also often appropriate and is the major objective of civil litigation. Even when no blame is attributable to the practitioner, it is usually also the case that none is attributable to the patient either. Litigation seeks a fairer outcome than simply leaving the consequences to lie with the injured person. In no fault systems of compensation, such as that in New Zealand, an attempt is made to distribute loss arising from an accident through the community. Of course, some forms of harm are inevitable and will not be covered—the normal scars of surgery for example. Compensation should be reserved for situations in which there is an element of the unexpected. In tort (in contrast to no fault systems of compensation), it is also necessary to establish an element of fault, so some degree of blameworthiness is necessarily implied by the award of damages. Nevertheless, punishment is not usually the primary objective of the civil law.

What should one think about the final element in this list—the desire for the healthcare professional to be punished? The punishment of wrongdoing is accepted as appropriate in most cultures. However, there is a fine line between punishment that is appropriate and vengeance, which most civilized societies would hold is not. It is widely accepted that punishment should be reserved for actions that are blameworthy and that the innocent should not be punished. Punishment should reflect behavior and not the profession of the person concerned or the outcome of the actions in question. Harm per se is not necessarily evidence of blameworthiness. Because action undertaken with intent to cause harm is culpable whether it succeeds in causing harm or not, it should be punished. Conversely, serious harm may follow innocent actions. For example the administration of a drug to a patient with no known allergy may result in an anaphylactic reaction and death; this outcome can clearly be attributed to the action of administering the drug, but few people would suggest that punishment was appropriate in these circumstances.

The law must be just and grounded in reality. These concepts are generally reflected in the well-accepted legal principle of reasonableness. The expected standard of care is that of the reasonable practitioner in the circumstances of the case. Unfortunately, to ask whether an error is reasonable is to misunderstand the construct because there is nothing reasonable about error. It is not the action, but the person carrying out the action, that needs to be considered. For example, it is well established that anesthetists regularly administer the wrong drug (16). This is not a reasonable thing to do, but it clearly is something done by reasonable doctors. That is the nature of error, captured in the following simple English definition:

An error is when someone is trying to do the right thing but actually does the wrong thing (15).

In the law, this idea was explored in Bateman, in which the judge said:

“You should only convict a doctor of causing a death by negligence if you think he did something which no reasonably skilled doctor should have done.” (13)

In Marshall vs Lindsey County Council, the judge clarified the point, saying:

“What is reasonable in a world not wholly composed of wise men and women must depend on what people presumed to be reasonable constantly do.” (17)

A further explanation was given by Lord Denning. He said that an error of judgment (in effect, a misguided moment) in a professional context did not amount to negligence. To test this, he said,

“one might ask the average competent and careful practitioner: ‘Is this the sort of mistake that you yourself might have made?’ If he says: ‘Yes, even doing the best I could, it might have happened to me’, then it is not negligent.”

This passage was “corrected” in the House of Lords by Lord Fraser who courteously suggested that what Lord Denning had meant to say was that an error of judgment was not necessarily negligent.

“The true position,” he (Lord Fraser) said, “is that an error of judgment may, or may not, be negligent; it depends on the nature of the error.” (18)

The law should also reflect society’s needs. To this end, the overwhelming need for society in the context of iatrogenic harm is improved patient safety. A utilitarian argument could be made that justice should be secondary to preventing bad things from happening in the future (thereby increasing the overall happiness of society). In practice, there is unlikely to be a major disconnect between justice and utility: because error is unintentional, recurrence cannot be prevented through the deterrent effect of punishment. Indeed, punishing error is likely to be counterproductive. When those who steal, rape, murder, and commit fraud face criminal prosecution, there is a sense that justice is being done and also at least the possibility that those contemplating such actions might be deterred in the future. Neither applies when health professionals face criminal prosecution because of mistakes in the clinical management of their patients. First, in sharp contrast with the former group, they did not wake up intending to harm someone. Few practitioners look at those who rape and murder and say “there but for the grace of God go I,” but many do in relation to so-called “medical manslaughter” (10). Second, punishment may well inhibit open disclosure, discourage participation in high-risk specialties, and promote defensive medicine, but it is unlikely to stop error.

Violations are intentional, so deterrence might possibly be effective, but this does not distract from the argument that punishment must be just: to the extent that the practitioner has been aware of risk to patients and disregarded that risk, there will be little difficulty supporting the justice of punishment.

It seems reasonable to conclude, therefore, that aligning the New Zealand law with that in the United Kingdom and Australia was enlightened, in the interests of society, and the right thing to do. Disappointingly, it seems that the approach in England may now have changed. Between 1900 and 1989, only 15 doctors were prosecuted for manslaughter in the United Kingdom. In the next 15 years, 38 were prosecuted and 34 after 1994 (19). After a careful analysis of 22 cases in which a conviction was obtained, Ferner and McDowell (20) concluded that 14 involved an error rather than a violation. Once again, we are dealing with mistakes and misguided moments but not egregiousness.

Although several health professionals in Australia have faced criminal prosecution in recent years (21–24), it is worth returning to the case of Dr. P before responding to this development with too much alarm. The judge explained the elements of this case as follows:

“In order to establish the offence, the Crown must prove beyond reasonable doubt that: (1) the accused owed a duty of care to the deceased; (2) by his act or omission, the accused negligently breached that duty of care; (3) the accused’s negligent act/omission caused the deceased’s death; and (4) considering the extent by which the accused’s conduct fell short of a reasonable standard of care and the associated level of risk of death, the degree of the accused’s negligence was so “gross” that it amounted to a crime.” (1)

There was no dispute over the duty of care or that the accused’s acts/omissions caused the death of the deceased. The judge accepted that it was not beyond reasonable doubt that the doses of drugs administered were acceptable. However, the management of the hypoxic events were a different matter. He said:

“Given the recurring low readings, lack of sustained recovery, and the serious risks associated with oxygen deprivation, a reasonable dentist in the accused’s situation would have terminated the procedure well before 9:45 AM. I am satisfied beyond reasonable doubt that, in this respect, the accused was negligent.”

The key issue therefore was the degree of negligence. This was the judge’s decision:

“The accused…knew that repeated low oxygen saturation readings should cause a dentist to terminate a procedure, but, in the face of low readings, he negligently failed to terminate the procedure. However, that was probably because he did not fully appreciate the extent of the medical (as opposed to dental) crisis that was developing. One could not expect that a reasonable general dentist practicing sedation would have been better informed that (sic) the accused. The deficiency was largely a deficiency in training and accreditation. The accused’s negligent conduct fell well short of that which would “amount to a crime against the State and conduct deserving of punishment.”

The words “well short” should give comfort to Australian perfusionists. It seems that a simple error remains unlikely to result in a criminal conviction in Australia, even if the consequences are tragic.

This is a just verdict, but where does it leave the family and society? I think that it leaves everyone better off than an unjust conviction, but the process as a whole does not seem satisfactory. A criminal trial of this sort involves a substantial investment in money, time, and emotional trauma for all concerned, and it is a huge indictment that it seldom seems to produce constructive actions to make healthcare safer. Healthcare is economically constrained everywhere, and safety initiatives are often neglected on the grounds of cost. In the end, the costs are all borne by society, including those of the legal aftermath of medical accidents. Not only is the mere fact of a criminal prosecution punishing in itself, but the money could usually be better spent on improving patient safety.

So what should we, as practitioners, do to manage the medico-legal risk that arises from the ever-present possibility of an error? Quite simply, the answer lies in focusing on the care of our patients. When things go well, there is no legal risk (or very little). When they go wrong, a high standard of care is a sound defense. The great merit of this approach is that, in the end, it simply means doing the right thing.