The current teaching paradigm

Initially, I struggled to figure out what underlay my statement to Lilly's partner and why it bothered the both of us. I suspect that my knee-jerk reaction to assign Lilly's actions to her pathology comes from an approach to mental illness baked into medical education. During preclinical studies, with the DSM as our guide, we learned about psychiatric illnesses as clusters of symptoms with associated risk factors, purported biological mechanisms and standard treatments. More advanced lectures and readings described the biopsychosocial model and the various social, psychological and developmental factors contributing to mental illness. However, in most lectures and study materials an implicit comparison is made between psychiatric and medical illness: both have biological underpinnings; both have some risk factors, whether developmental, social or biological; and both have standard management. We were encouraged to break down clinical psychiatric vignettes in much the same way as medical ones, featuring social factors such as poverty or developmental factors such as childhood abuse in lists of pertinent positives and negatives to inform our differential diagnoses. In many ways, it was no surprise that on the wards we would make comparisons with diabetes or physical illness in psychoeducation of patients and families.

… and its fit with psychiatry

So why had I felt uncomfortable about my argument? Was it not true that psychiatric illness can be thought of as strongly biological in origin and patients with psychiatric illness were not to blame for their symptoms? Were they not the victims of biological and neurological dysfunction driving them to pathological cognitions and behaviour? In fact, some have correctly pointed out that the ‘medical model’ of psychiatric illness not only has a strong utility in describing mental illness, but also helps destigmatise otherwise long-stigmatised disorders such as borderline personality disorder (Novick Reference Novick and Ross2020). Among the public, patients’ families and service providers, describing psychiatric illness as similar to medical conditions can help onlookers build empathy with patients who can otherwise be ‘difficult’ to interact with (Novick Reference Novick and Ross2020). Then in speaking with Lilly's partner, why had my comparison failed?

I think now that the core of my discomfort lies in a single unanswerable question: ‘If this was Lilly's illness, then which parts were Lilly?’. Lilly reminded me of another patient I met, who had obsessive–compulsive disorder. He obsessively asked which of his thoughts were obsessions, the results of his illness, and which were his ‘true self’. As individuals, we form our sense of identity from the sum of our thoughts, emotions and behaviour. Perhaps my discomfort in blaming Lilly's diagnosis comes from my own personal, selfish need to believe that as individuals, myself included, we have agency. Alternatively, it may come from a true inability to clearly draw a line for Lilly and other patients between their illness and their selves. And most of all, perhaps in trying to generate empathy for Lilly, I felt I had deprived her of elements of her own identity.

The question of agency

Philosophical arguments aside, on a practical level there can be a danger to removing a patient's agency. For patients with personality disorders and other psychiatric conditions, an internal locus of control, or a self-perception of agency over their own actions, is protective against worse outcomes (Hashworth Reference Hashworth, Reis and Townsend2022). A sense that they control their own actions can play a strong role in psychotherapy, motivating patients to assess, become conscious of and alter their own thoughts and actions (Pickard Reference Pickard2011). In reducing thoughts and actions to biological pathology, we may be questioning the limits of a patient's identity but also jeopardising a protective internal locus of control.

Ultimately, the question of which emotions, beliefs and actions are the result of illness rather than personal identity is a complex one, with answers varying from person to person. A rich philosophical literature exists on the nature of responsibility and culpability in mental illness. Distinctions are made between types of disorder (such as psychotic illnesses, personality disorders, impulse control disorders and mood disorders) as well as psychiatric symptoms (such as faulty cognitions, impaired insight and impaired impulse control) and their relationships to personal agency (Pickard Reference Pickard2011; Ayob Reference Ayob2016; Szalai Reference Szalai2016). In some instances, these philosophical distinctions may serve as helpful tools for the experienced practitioner in parsing out what behaviours and thoughts fall outside of a person's agency and identity. A familiarity with this literature may help practitioners and students in approaching patients and helping them and their families understand the patients’ behaviour. But in Lilly's case and many others, the line between expression of illness and personal identity can be a challenging one to draw. The question of agency may be stronger than any particular answer. Although the desire to destigmatise mental illness by comparing it with a physical illness independent of the patient comes from a noble desire to help patients, perhaps it also does harm. Perhaps we can resist the urge to blame patients and instead generate empathy by recognising the powerful and diverse biological, developmental and engrained psychological factors driving behaviour and still cultivate in patients the conviction that they have the agency to change.