Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
You have not opened the wrong journal!
The police have a duty to protect the public and to investigate any, and all, serious crimes. The article by Lynøe and Leijonhufvud raises important issues about the interaction between hospital staff and police in cases in which suggested medical negligence crosses into the arena of serious legal offences, which range from murder and homicide to serious assault.1
Although arising in Sweden, the issues raised in this case are generalisable. While our understanding is limited to the journal article there seem several uncontested facts: a children's intensive care doctor was arrested, charged and acquitted of the mercy killing of an infant in a Swedish hospital. The baby had sustained severe brain injury prior to this clinician's involvement, either due to birth asphyxia or to an accidental overdose of salt. The baby died after withdrawal of life-sustaining treatment (WLST) with parental agreement. At a later stage, the family asked the police to investigate the initial salt incident. Despite the team attending the baby after WLST stating that only a morphine infusion was administered, a 25-day postmortem femoral artery sample showed ‘an extremely high level of thiopentone’ (a powerful barbiturate), although no control sample (eg, intracardiac) was analysed.
Healthcare staff are subject to the laws of their country. However, Lynøe and Leijonhufvud raise the issue of a conflict between what they term ‘soft law’—the rules, regulations and ethics of healthcare practice and the law of the land. However, if the two conflict generally the former must bow to the latter. There are of course cases in which the latter is either morally suspect or interpreted in such a way by corrupt regimes, but that is not the case here. However, Lynøe and Leijonhufvud argue that Swedish national law, as with many jurisdictions, lacks the subtlety to deal with complex healthcare end-of-life cases.
The authors identify two main errors:
Thiopentone was given to an infant by at ‘least one other physician’ and ‘probably several physicians’ to treat convulsions and induce anaesthesia before WLST. However, ‘contrary to regulations, these thiopentone administrations were never recorded’.
There was a failure of the prosecuting authorities to review the possible limitations of postmortem sampling so long after death, to consider all possible mechanisms of the thiopentone level and to arrange an initial peer review of the medicolegal report before police action.
I would add a third:
3. Investigation by hospital authorities and the National Board of Health and Welfare ought to have occurred after death, including a full postmortem.
In the UK, family doctor Dr Harold Shipman is thought to have murdered hundreds of older patients, ostensibly providing ‘compassionate end-of-life care’. Initial defenders of the ‘highly regarded popular family man’ suggested he was merely using powerful drugs to ‘palliate’ patients who were dying. Only when questions lead to full postmortem examinations was he stopped. While excellent palliative care is a right,2 so to is protection of the dying from rogue professionals. Accurate prescribing, observation of administration and recording of such medications used is mandatory.4
However, prescribing for patients at the end-of-life has been described as ‘full of ethical anxiety for the prescriber’ with concerns about boundaries between palliative care, palliative sedation, doctor-assisted suicide and euthanasia.5
The doctrine of double effect (DDE) describes the permissibility of an action known to cause serious harm as the side effect of an intended primary good end. Thomas Aquinas has been attributed with first portraying the principle of DDE when he justified the killing an assailant in self-defence, as long as killing was not the main intention.6 In healthcare, this approach permits the administration of high-dose opioids and sedatives in palliative care to relieve patient pain and discomfort without the intention of causing death, even though death might occur sooner. Of course the use of such medication to deliberately end life is euthanasia and illegal in most countries.
While in combat Aquinas’ approach seems valid, there are reasons it appears to me unnecessary in healthcare, as outlined below.
Firstly, prescribers routinely select drugs and doses while considering multiple risk/benefit issues—effects, side effects, drug interactions and complex host factors such as organ failure. How is it logical for doctors to ignore one such side effect—respiratory depression—in palliative opiate prescription, whereas this is usually unacceptable, for instance when prescribing chemotherapy?
Secondly, ‘palliative sedation’ does not shorten overall survival, obviating the need for the ‘doctrine’ to routinely excuse compassionate end-of-life care.7
The medicolegal expert's suggestion that the 25-day postmortem thiopentone level indicated a mercy killing meant prosecutors and police had a clear duty to investigate. While this ought to have commenced with scientific questions about the accuracy of the sample8 and a peer review of the opinion, interviews with all those involved with this baby's care—including the palliative component—are entirely justified. It is hard to understand why any thiopentone was detected, given the drug administration records.
However, it would appear that the investigative process commenced with an arrest of a clinician on shift, rather than as might be expected a series of sensitive staff interviews.
Of course if the safety of current patients is compromised or there is the need to prevent evidence being compromised, this is justifiable; however, the undermining of the ‘innocent’ person in front of colleagues and patients affects their future ability to work in that environment again.
Clearly this must remain at police discretion, but the arrest of doctors on duty always causes added distress to ‘the medical tribe’.9 However, doctors are not a special case. Airline pilots, policeman themselves or the judiciary have equal claims to having important roles in society, but must remain subject to the law. The only residual argument is that of patient safety and care, but surely if any doctor on duty becomes unwell, systems exist to cover them?
While medics are no different in being subject to the law, the involvement of police in a complex end-of-life case has troubled the Swedish medical establishment. The biggest concern voiced seems to be that in future inadequate palliative therapies may be used.
This argument seems to be that those caring for the dying will use less analgesia/distress-relieving medication if they feel at risk from subsequent prosecution. But there is little evidence to support this, and the demand for permissive palliative use of such drugs with antecedent immunity from prosecution with a vague threat that people will be left to die in pain seems empty and unnecessary.
Healthcare staff will not abandon patients, but the protection demanded will be best achieved by codification of good end-of-life practices through excellent professional standards with open and transparent care, rather than any immunity from state sanction.
Doctors must remain accountable under law for their actions, and advanced immunity from prosecution when delivering end-of-life care cannot be justified.
‘Soft law’ or those professional ethical codes we operate under must answer to the law of the country. Therefore, for healthcare professionals the former, recently clarified in the US and Sweden,10 i must be protection enough, but as the Shipman case demonstrated the supremacy of the law must prevail.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.
↵i In 2008, the American Medical Association Council on Ethical and Judicial Affairs approved an ethical policy regarding the practice of palliative sedation.