Viewpoint: How to avoid a dichotomy between autonomy and beneficence: from liberalism to communitarianism and beyond

Abstract

The history of medical ethics demonstrates how different values have been upheld at different times and how ethical frameworks have developed: how certain arguments have risen to prominence, while others have fallen into disrepute. The central values of the classical Hippocratic ethics were nonmaleficence (Do no harm!), beneficence and confidentiality. After World War II, medical ethics started to advocate patient autonomy in the guise of informed consent. However, over the last 20 years, there has been growing dissatisfaction with the individual rights-centred ethical framework. Within clinical ethics, it has been noted that an obligation to respect the patient’s self-determination has worked against beneficence and trust and has thus done a great deal of damage to the doctor–patient relationship. The opinion that the rights-based ethics is not in the best interest of the patient and that patient care is best served by a limited form of paternalism has been gathering support. The central moral concern of public health ethics is to specify the conditions that warrant paternatilistic interventions that override individual autonomy to prevent people from adopting unhealthy behaviours [1]. Likewise, in the context of research ethics, it has been argued that principles of individual autonomy and informed consent and privacy can seriously hamper medical research that aims to further the common good. It has become clear that prior, free, specifically informed and explicit consent may not be an achievable ideal in emergent research fields, such as biobank research. The need to develop new ethical frameworks focusing on more collective values, such as reciprocity, mutuality, solidarity, citizenry and universality, has been advocated [2]. In the last two decades, there has been talk of the ‘communitarian turn’ in ethics. The basic difference between liberal and communitarian ethical frameworks lies in the fact that liberalism claims that the individual is more important than the society in which he/she lives, whereas communitarianism regards society as more important than the individual. Liberalism stresses rights, communitarianism the common good. In the traditional communitarian model, individual rights and interests are subordinate to the common good; however, the more recent neo-communitarian approach tries to balance individual rights with community interests. Here, I will discuss the view that although autonomy and beneficence seem at times to be in conflict, there is no reason to see one or the other as predominant. A wider understanding of autonomy is crucial for understanding how to avoid a division between autonomy and beneficence. Traditional medical ethics, dating back to ancient times, has a virtue-based ethical framework, centred on crucial values of the physician’s practice, such as confidentiality and ‘doing good’. The relationship between doctor and patient was based on the patient’s trust and acknowledgement of the physician’s authority. The contemporary debate about patients’ rights and the protection of human subjects of research is based on reaction to the Nazi eugenic experiments and the outcome of the Nuremberg trials. Relying on the post-war principles of medical ethics outlined in various international declarations, it was argued that respect towards the interests and choices of an individual would be the key to never allow social interests or the paternalistic impulses of the medical community to override individual interests in medicine. Thus, the principle of informed consent was introduced in medical ethics mainly to protect individuals against possible harm. As it established an obligation of respect for the patient’s self-determination, it counteracted paternalism. Tauber [3] pointed out that informed consent also opposed beneficence and thus damaged the doctor–patient relationship, which had always been based on mutual trust. With regard to scientific research, whether the protection of individual rights has gone too far is frequently questioned; defence of autonomy and privacy has become an obstacle not only to the use of data in scientific research but also to the use of such information in the implementation of social goals. It has been claimed that epidemiological research is being obstructed, as statistical data cannot be collected without the subject’s explicit agreement [4]. Genetic databases, especially large population-based ones, also pose new challenges to the traditional medical ethics framework centred on the individual and his/her informed consent. Currently, there is growing support for broad consent (also described as open, generic or blanket consent) to be used for population genetic databases [5–9]. Two main arguments have been used to justify the use of broad consent rather than informed consent. The instrumental argument is that it is impossible to obtain fully informed consent, because at the time of collection, one does not know what kind of research the samples will be used for. The substantial argument stresses that there is public interest in keeping the samples for unlimited time and using them for different research projects [10]. In 2002, the Human Genome Organisation Ethics Committee [11–14] stated that human genomic databases are global public goods and humanity as a whole should be the beneficiary of these public resources. Thus, it has been argued that instead of emphasizing the individual’s rights (to refuse to participate in research, or to withdraw the samples or data from the database), one should see it as one’s ‘duty to facilitate research progress and to provide knowledge that could be crucial to the health of others’ [15]. Similarly, the issue of whether informed consent is deemed mandatory for the handing over of data to population-based databases has recently arisen in the context of the creation of electronic health records. It is hoped that collecting and holding health records in a national computer system will improve the quality of care and efficiency and reduce paperwork. Of course, easier access to such a complex resource also creates vulnerabilities for patients; health records contain sensitive private information the disclosure of which can potentially harm the patient. The essence of the ethical dilemma is claimed to lie in the fact that ‘explicit informed consent preserves freedom of choice at the cost of less health and welfare while strong paternalism, without the possibility to opt out, promotes health and welfare at the cost of freedom’ [16]. The ‘opt-in’ model would respect the lessons learnt from the history of medical ethics –that efforts to promote the common good must be constrained by liberty and informed consent, that those affected have the right to accept or reject the programme and that patients have a legitimate interest in privacy and preventing third parties from gaining access to sensitive data. The ‘opt-out’ policy of data sharing, whereby an individual will still have a right to prevent disclosure to treating physicians, will limit individual autonomy but is likely to guarantee that the data of the majority of the population will be available in the database for administrative and statistical purposes and guarantee the availability of a sound research base. As the benefits of electronic health record databases have been seen for various social uses (the exploitation of health data for scientific purposes and statistics, and for the auditing, planning and management of health care services), the public interest argument has in most countries come to prevail over individual interests of liberty, privacy and self-determination, and opt-out has been favoured by most countries. This also shows clearly a turn from rights-based discourse to that based on common good. For those who reject the return to the paternalistic model of ethics, the main questions are where and how individual autonomy fits into the framework of bioethics. Attempts have been made to strike a balance between autonomy and beneficence, or rights and common good, which to most ethicists seem to be in conflict. In clinical ethics, the best known example of balancing autonomy and beneficence is Pellegrino and Thomasma’s account of the doctor–patient relationship as ‘beneficence-in-trust’ [17]. In their view, a rights ethic is a minimalist ethic, whereas one based on beneficence embraces more fully the nuances of the patient’s best interests. Beneficence is not in conflict with autonomy (whereas paternalism is) but rather is a powerful means of supporting autonomy and preserving the dignity of patients. They argue that ‘the patient seeks not only to be protected from harm, but also to be healed and to have health restored or improved, pain and anxiety relieved, disability lessened. The patient desires these good ends within some definition of the good life that is uniquely and personally his or her own’ [18]. Attention to the patient’s individual account of the ‘good life’ presupposes respect for his or her individual autonomy. So if autonomy is seen as one of the individual goods that the patient seeks to secure besides the desire to be helped, the dichotomy between beneficence and autonomy can be avoided. Tauber [3] proposes a reconfiguration of personal autonomy and a renewed commitment to an ethics of care. He argues that beneficence should not be confused with medical paternalism. In his understanding, beneficence and responsibility are moral principles that are not only compatible with patient autonomy but also strengthen it. As regards research ethics, it is important to note that communitarians are not arguing for the abandonment of individual rights of research participants, only that these should not be viewed as absolutes and a balance should be found between the principles of autonomy and privacy, on the one hand, and the risks and benefits of human research, on the other. If there is a clear need to undertake research for the public good, rather than for individual benefit, it may be acceptable that the individual liberty of the research subjects is restricted. At the same time, it has been stressed that even in carrying out research for public good, there needs to be adequate data protection to safeguard against misuse, discrimination and stigmatization. Indeed, Chadwick and Berg [15] claim that ‘it is arguably neither morally nor practically feasible to expect individuals to show solidarity in the absence of such protections’. However, we should be aware of the problem that common good/public interest is difficult to define and it may be easily manipulated and used to further particular interests in the guise of common interests. In actuality, the common good is not merely the equivalent of an aggregation of all private goods in a society. It implies those goods that serve all members of the given community in the long term [19]. At the same time, it may sometimes be helpful to reconceptualize the debate between public interest and individual interest in less dichotomized terms. Is it not possible to express the argument concerning public interest in efficient health care as one also concerning personal interest in better health? As regards electronic health databases, we might care more about health than privacy and think that asking for explicit consent will require considerable resources (both time and money) that could be used for medical treatment. We also understand that if too few patients are included, the social uses of the databases will be lost, and we thus voluntarily give up our privacy to promote health – our own as well as that of others. We can even try to reconceptualize electronic health projects as attempts to advance or enhance individual autonomous functioning instead of seeing them as a means of making the individual rights to autonomy and privacy less important than the common good (better health care). During the times when health records were kept in paper form or on doctors’ individual computers, the individual patient had almost no control over the processing of his/her medical data. With the creation of electronic health record databases, there will be a trace left from every enquiry made about patients’ health data. Thus, the patient will have the opportunity through the patient portal to track all access to his/her data. In this way, the autonomy and privacy of people can be significantly enhanced by the electronic health projects. Both individual and public interests are also at stake in the case of population-based genetic databases. Whereas talk of the genomic databases as global public goods refers to scientific knowledge as a common good, health can be seen as having both individual and a communal value that should not be set in opposition. Thus, the choice is often not between two opposing individual and communal values, but between values that can be conceptualized in both ways. Although there is a growing dissatisfaction with the individual rights-centred ethical framework with patients’ autonomy as its governing philosophical principle, in reality the main target of this criticism has been the concept of informed consent, which has been faulted for being individual centred to the detriment of collective or community interests, as well as too formal and unnecessarily cumbersome to obtain. For good reasons, the concept of informed consent has been modified (e.g. open or broad consent in genetic database projects) or, in some areas of application (e.g. in electronic health record databases and epidemiological studies), abandoned altogether. However, informed consent requirements for epidemiological, observational or interventional studies are still interpreted too narrowly, and this is limiting the ability of studies to provide new medical knowledge that would be beneficial for patients. Hansson has convincingly argued against too strict an interpretation of consent in studies of low risk of harm to the individual. He suggests that research participants should have access to indirect means of exercising autonomy through institutions (e.g. ethical review boards) that have to find the right balance between the different interests of the individual and the public. Respect for autonomy, in his view, does not imply deciding for oneself in isolation from others [20]. Several others have seen it as a problem that autonomy is understood too narrowly. Manson and O’Neill have pointed out that in current bioethical discussions, autonomy has been understood mainly in the sense of autonomous decisions. On the basis of this understanding, informed consent has been conceptualized as disclosure of information by those who seek consent, and as decision-making by those whose consent is sought [21]. They argue that this narrow focus ignores or underplays what is actually needed for effective communication and commitments between the parties. Rehbock has stressed that autonomy should not be reduced to autonomous decisions. Her thesis is that interpretation of autonomy in the sense of capacity for autonomous decisions does not support respect of autonomy but can lead instead to paternalistic tendencies in medicine. Rehbock distinguishes between moral and psychological understandings of autonomy. The moral concept of autonomy implies that our beneficent actions should not be paternalistic but instead always take into account the other person’s will [22]. Respect for autonomy should not be dependent on whether or not one can get over the threshold of competence. Although in individual cases it may be very difficult to recognize another person’s will, much can be determined on the basis of our common human nature. However, we should be careful not to suppose that the other’s will is similar to our own. Therefore, we should pay attention to the other person’s desires, preferences and understanding of the good life when we make estimations about his/her will. If we see general human will as the basis of reciprocal moral duties of help and avoiding harm between people, the close connection between autonomy and beneficence becomes clear. Throughout the history of medical ethics, we have seen that philosophers have favoured ethical frameworks that are centred on a single core value. The pendulum has swung from beneficence to autonomy and now back again. Although autonomy and beneficence seem at times to be in conflict, there is no reason to see one or the other as dominant. The key for understanding how to avoid a dichotomy between autonomy and beneficence is a wider understanding of autonomy. Beneficence without respect of an individual’s autonomy may lead to actions that are not really beneficial as they try to promote the other person’s well-being against his/her will. Respect for autonomy (for example in cases of mental incapacity) without beneficence may remain ignorant of the person’s real desire to be helped by others. Thus, both autonomy and beneficence are needed, but their specific interdependence depends on the particular situation and on social and political context. A similar contextualized approach can be applied in the case of a seeming conflict between two different value systems (e.g. liberal individualism and communitarianism). Instead of debating whether one should prefer individual rights or common-good-based arguments, one should rather consider the value conflicts in specific socio-political situations and keep in mind the larger picture and the long-term perspective. Time will show whether medical ethics will succeed in overcoming the dichotomy of autonomy and beneficence or autonomy and solidarity and accommodate both the values of liberal individualism and communitarianism. No conflict of interest was declared. This work was supported in part by an Estonian Science Foundation grant (in the European Economic Area/Norwegian financial mechanisms scheme) ‘New ethical frameworks for genetic and electronic health record databases’ funded from 2008 to 2010. I thank Tiina Kirss for help with my English language and Laura Lilles for technical assistance. I owe many thanks to Kadri Simm and Theda Rehbock for useful suggestions. In addition, I thank Mats Hansson for his encouragement and patience.