Fusar-Poli, P., Manchia, M., et al. (2022, October).
European Neuropsychopharmacology, 63, 17–34.
https://doi.org/10.1016/j.euroneuro.2022.08.001
Abstract
Precision psychiatry is an emerging field with transformative opportunities for mental health. However, the use of clinical prediction models carries unprecedented ethical challenges, which must be addressed before accessing the potential benefits of precision psychiatry. This critical review covers multidisciplinary areas, including psychiatry, ethics, statistics and machine-learning, healthcare and academia, as well as input from people with lived experience of mental disorders, their family, and carers. We aimed to identify core ethical considerations for precision psychiatry and mitigate concerns by designing a roadmap for research and clinical practice. We identified priorities: learning from somatic medicine; identifying precision psychiatry use cases; enhancing transparency and generalizability; fostering implementation; promoting mental health literacy; communicating risk estimates; data protection and privacy; and fostering the equitable distribution of mental health care. We hope this blueprint will advance research and practice and enable people with mental health problems to benefit from precision psychiatry.
From the Results section
3.1. Ethics of precision psychiatry: Key concepts
Broadly speaking, ethical issues concern the development of ‘practical ought claims’ (Sheehan and Dunn, 2013) (i.e. normative claims that are practical in nature), which arise when we face ethical uncertainty in precision psychiatry. These practical claims come schematically like this: how should somebody or a group of people act in relation to a particular issue when they face certain circumstances? For example, how should researchers inform patients about their individualised risk estimates after running a novel clinical prediction model? To address these questions, four overarching ethical principles have been suggested (by Beauchamp and Childress) (Beauchamp and Childress, 2019), which include autonomy, beneficence, non-maleficence and justice. These can be applied to precision psychiatry, complemented by an extra principle of “explainability/interpretability” (Panel 1) which has been specifically introduced for artificial intelligence (Floridi et al., 2018) (for a more detailed discussion of ethical platforms for big data analytics see eSupplementary 1).
Although these four principles have become the cornerstones of biomedical ethics in healthcare practice, they have been criticised as they are often conflicting with no clear hierarchy and are not very specific (i.e. these principles are somewhat implicit, representing general moral values), leading to “imprecise ethics” that may not fit the needs of precision psychiatry (Table 1). Rather we should ask ourselves “why” a certain act may be harmful or beneficial. For example, let's imagine having a risk assessment; what would that mean for the individual, their family planning, workplace, choosing their studies, or their period of life? Alternatively, let's imagine that the risk assessment is not performed; what would be the results in a few years’ time? To address these sorts of questions, this study will consider ethical values in a broader sense, for example, by taking into account some of the different principles present in the charter of fundamental rights of the European Union – starting from dignity, freedom, equality, solidarity, citizens’ rights and justice (Table 1) (European Union, 2012; Hallinan, 2021). In particular, human dignity and human flourishing are the most crucial elements from an ethical point of view that are tightly linked to autonomy and self-determination (which is modulated by several factors such as physical health, psychological state, sociocultural environment, as well as values and beliefs). The loss of insight associated with some psychiatric disorders may incapacitate the individual to make autonomous decisions. For example, autonomy emerged as the driving decision component for undergoing risk prediction testing among young populations (Mantell et al., 2021a), regardless of whether a person would decide for or against risk profiling. Finally, it is important to highlight that unique ethical considerations may be associated with the historically complex socio-political perceptions and attitudes towards severe mental disorders and psychiatry (Ball et al., 2020a; Manchia et al., 2020a).
