The shifting sands of patient autonomy and public interest considerations in health care

Much has been written in academic literature about the interplay between public interest and autonomy in a democratic society. Obvious limits to an individual's freedom to choose their actions as they wish are where they interfere with the choices of others, but the definitions of these limits vary between societies and over time. In this paper, we highlight some recent examples within medicine that illustrate the shifting sands of this balance and hope to stimulate a wider discussion in clinical circles.

The freedom of self-determination, to ‘write one's own story’, is termed as autonomy. In health care, this finds expression in a patient's central role in providing consent; to legitimize either a clinical intervention or the disclosure of personal information. Occasionally, such interventions can also be legitimized by demonstrating that the ‘public interest’ overrides these expressions of autonomy: immunization programmes, for example, have relied on a public interest argument.

Over the past 30 years, it seems that autonomy has ascended to a prime position over other ethical cornerstones of clinical practice. To ‘do good’; and to ‘do no harm’; and the need to incorporate justice in health-care provision were considered emblematic ethical principles. But the ascendancy of autonomy within these principles is reflected by Raanan Gillon's statement that ‘autonomy is the first amongst equals’. ¹ Perhaps in an attempt to remove the image of paternalistic doctors, the central role of consent as a direct expression of personal autonomy could be considered to have eclipsed other ethical aspects of patient care.

Demonstrating that a breach of confidentiality is in the public interest has always been an exception to the apparent primacy of autonomy in health care, but a clear definition of what constitutes a significant enough public interest has been more elusive. The term has had no consistent definition in professional guidelines or academic literature but generally refers to a ‘common wellbeing’ or ‘general welfare’. We have encountered several examples, from several different fields of clinical medicine, where a justification of ‘in the public interest’ has been used to dispense with either the need for consent for a clinical intervention; or argued for disclosure of confidential information. These examples illustrate the shifting sands of the limits of autonomy in health care.

National campaigns

The proposed introduction of a national summary care record and a system of implied consent for organ donation has caused difficulties for those keen to implement such innovations. The travails of these projects are well documented and continue. ² Both systems wished to shift the onus away from opting-in with individual consent, to an arrangement more likely to result in higher recruitment – an opt-out system where communities or individuals were asked to object to their inclusion in the system.

A common ground of objection to both was that consent was finessed away. Neither system could ensure that each and every person was provided with the necessary disclosure to inform their decision-making; thus, there was no certainty of the existence of informed consent. ³ Those who supported the drive for an opt-out system for electronic summary records or for organ donation were nevertheless balancing public interest against individual autonomy. The over-arching good that they hoped to achieve is undoubted. Increasing the supply of donor organs would be of immeasurable benefit for those who need them, and to have patient information readily available to all clinicians would clearly be advantageous. These public benefits had to be weighed against the harm that could be caused by altering a fundamental presumption: that until the citizen shows evidence of their consent to a proposed touch or disclosure, the law will protect him or her from such a transgression. If the certainty is removed, that an uninformed minority have had an opportunity to exercise their choice, the original presumption that they have given consent evaporates.

Creutzfeldt–Jakob disease

The Creutzfeldt–Jakob disease (CJD) Incidents Panel was set up by the Department of Health (DH) in 2000 as an expert panel to advise on the management of instances of the potential transmission of CJD. Among many other tasks, the panel considered what information should be collected on patients who may have been exposed to this disease. In its 2005 report, ⁴ the group of patients who had a low or unknown level of exposure was addressed. The Panel considered that this group required ‘special attention and efforts, as so little is known about the risk of infection they may have experienced’. However, when deliberating whether an anonymized database of such patients should be created, the Panel acknowledged the ‘strong argument that people… should be able to choose whether or not they are told about their possible exposure’. To ‘enable’ this choice, the Panel proposed that patients should be entered onto the database without their consent. In doing so, it was argued, the patients' choice of not being told about their exposure was preserved.

Public advertisements would make the general public aware of the option to find out if they were on the database, and the significance of the entry. The Panel's report did not record whether there was a belief that the benefit to the public interest from accumulating the data justified the unconsented data collection, but this could be inferred by the creation of a database. The report illustrates a balancing exercise between the harm that a patient's notification (of being a person of low, but finite, risk of exposure to CJD) might cause, against the uncertain and possibly modest public health benefits that might flow from accrual of the data. An explicit balancing exercise between public interests and an individual's control of their own data is neatly avoided. Instead, the high chance of individual suffering is compared with the rather small chance that any public benefit will emerge from a notification exercise.

The Panel's reluctance to cause individuals anxiety is commendable and appears to be autonomy-defending, in that it defends a potential right not to know. But this decision gives effect to the proposition that the public interest in collecting the data trounces the patient's right to control the disclosure of his or her personal information.

Violence and health care

There is also a public interest in ensuring that citizens are protected from harm. The arrival of a victim of knife or gunshot injury in an emergency department raises the possibility that the perpetrator is still at liberty, and may strike again. The wounded patient may be at risk of a further attack, during which other people at the scene of the crime or in the hospital may be harmed. For these reasons, ⁵ the General Medical Council (GMC) recommends that doctors in the department should inform the police of the arrival of any patient with a gunshot wound and any non-accidental knife wound inflicted as the result of an attack. It is also recommended that no personal information should be disclosed; such a disclosure should be left to the discretion of the patient. Exceptions to this recommendation occur when the doctor believes that the disclosure would be in the public interest: for example, where failing to disclose the information may put others at risk of death or serious harm; or where ‘disclosure is likely to help in the prevention, detection or prosecution of crime’. Non-disclosure remains the default position, but this can be altered, depending on the circumstances.

To assist clinicians in making these difficult decisions over the disclosure of confidential information, the DH produced some supplementary guidance ⁶ to its existing Code of Practice. Using the arrival of the victim of a knife attack in the emergency department as an ‘exemplar case’, the DH provides an alternative presumption that ‘a knife attack may be sufficient to justify … disclosure of confidential information even when consent is not given …’ Although not a dramatic change in wording, a shift has nevertheless occurred, from a starting point where a patient's confidential information will be withheld, to an initial position where such disclosure may be considered to be in the public interest. No ‘automatic’ disclosure is suggested, but it is reasonably foreseeable that patients will have to persuade their clinicians to withhold disclosure of their personal information.

Pregnancy

Another mandatory change in hospital practice is to seek a declaration of possible pregnancy prior to anaesthesia and surgery. A rapid response report from the National Patient Safety Agency ⁷ stated that preoperative assessment policies must include a check of possible pregnancy in women of childbearing age in the immediate preoperative period, followed by a pregnancy test if necessary. Checking the pregnancy status of teenage girls could be considered an altogether more sensitive matter.

There is no uniform guidance about checking the preoperative pregnancy status in young girls. When the National Institute for Clinical Excellence published guidance on preoperative testing in 2003, ⁸ the Guideline Development Group (GDG) recommended that pregnancy testing should be carried out in female patients of reproductive age who say that they might be pregnant. The GDG stated that informed consent should be obtained. An invited group of panellists had provided a consensus that informed the GDG. The panellists acknowledged the sensitivities of obtaining consent in female patients under the age of 16. This age group may be reluctant to admit that pregnancy is possible, particularly if the question is asked in consultation with their parents – either out of embarrassment, or fear of recrimination in the knowledge that sexual activity is illegal under the age of 16. Consent to test a patient under the age of 16 may be obtained without the knowledge of the parents if the patient is judged to be competent, but at the risk of undermining the relationship between health-care staff and the parents if the test is positive.

In a recent survey, ⁹ the pregnancy status of only 36 out of 53 (68%) teenage patients was successfully recorded during the consent process. It is accepted that this may be simply reflective of clinicians' disinclination to broach this intimate subject with either the girl or her parents. Nevertheless, it also illustrates that such conversations are difficult, for both the young person and her doctor, particularly in the presence of her parents.

We acknowledge the importance of preventing the harm that anaesthesia and surgery may inflict upon the fetus. But we should also at least consider whether requiring a young person to disclose her private information in these circumstances is a proportionate response to the risk to her unborn child.

It is settled law that a pregnant woman may make choices concerning her fetus irrespective of whether such choices are in its best interests. But although it may not be possible to demand (in the legal sense) that the pregnant girl discloses her pregnancy in the interests of the fetus, some may consider that the welfare of the fetus should be given a higher priority. If it is concluded that such disclosure is both necessary and proportionate, we also have to conclude that it is the public interest that is trumping her rights of confidentiality.

Genetic medicine

The tension between an individual's right to consent to the use of their information and a public interest in using such information in the absence of consent is also evident in genetic medicine, where the patient's rights must be weighed against those of existing or future relatives. The discovery of a patient's potentially fatal, preventable, inherited disorder may necessitate alerting relatives – who may be ignorant of this risk – even in the absence of the index person's consent. ¹⁰ The GMC advises that ‘a patient might refuse to consent to the disclosure of information that would benefit others … In these circumstances, disclosure might still be justified in the public interest’.

Again, a subtle shift is seen: no automatic disclosure is recommended, but doctors may infer that in any serious genetic condition, relatives should also be told about their risks. The availability of evidence-based interventions to those to whom disclosures are made is likely to play a part in such balancing exercises. The risk of a disease which is difficult to quantify, or is untreatable, may rank lower in the public interest stakes than a ‘serious’ or life-threatening risk (for example, an inherited cancer). Assessing the harms being prevented may be difficult: arguably ‘harm’ accrued by genetic predispositions has occurred at birth, rather than through the non-disclosure of relatives' results; although if treatment – either curative or ameliorative – could be provided, failure to disclose does look like a harm of sorts.

Legislation

Most recently, the advent of the Health and Social Care Bill highlights the tension between the state's desire for access to our personal information (possibly in the public interest) and the detrimental effect that this will have on an individual's right to control their own information. ¹¹ The Bill makes little reference to anonymity or consent in relation to data processing; the British Medical Association has noted that the Bill provides ‘broad powers to a range of bodies' to disclose patient information. ¹²

It could be argued that this will set the scene for a progressive replacement of personal autonomy in favour of the state's interpretation of what may be in the public interest. It is paradoxical that this is occurring at a time when the importance of an individual's right to control their own destiny apparently preoccupies our political leaders. We do not take a position on the desirability of this shift. When  countervailing public and private interests conflict, balancing these ‘rights’ requires an assessment of the harms and benefits that will be gained or lost, depending on whether autonomy or the public interest prevails in any particular case. The autonomy of individuals in communities is therefore to some extent inextricably linked to the common wellbeing of that community. Autonomos (Gk) referred to ‘self rule’; but interestingly the word was only applied to self-rule of communities rather than individuals. It may be that personal ‘autonomy’ and the public interest have more in common than it first appears.

It is simplistic and unhelpful to create a general proposition of whether personal autonomy or public interest should take priority. Despite the discussion on such issues in the philosophical literature, ¹³ there has been relatively little by the medical profession, and we feel it merits further debate.