In a pioneering new procedure, deep brain stimulation is being trialled as a treatment for the eating disorder anorexia nervosa. Neurosurgeons at the John Radcliffe Hospital in Oxford implanted electrodes into the nucleus accumbens of a woman suffering with anorexia to stimulate the part of the brain involved in finding food rewarding. Whilst reports emphasize that this treatment is ‘highly experimental’ and would ‘only be for those who have failed all other treatments for anorexia’, there appeared to be tentative optimism surrounding the potential efficacy of the procedure: the woman who had undergone the surgery was reportedly ‘doing well’ and had shown ‘a response to the treatment’.
It goes without saying that successful treatments for otherwise intractable conditions are a good thing and are to be welcomed. Indeed, a woman who had undergone similar treatment at a hospital in Canada is quoted as saying ‘it has turned my life around. I am now at a healthy weight.’ However, the invasive nature of the procedure and the complexity of the psychological, biological and social dimensions of anorexia should prompt us to carefully consider the ethical issues involved in offering, encouraging and performing such interventions. We here outline relevant considerations pertaining to obtaining valid consent from patients, and underscore the cautious approach that should be taken when directly modifying food-related desires in a complex disorder involving interrelated social, psychological and biological factors.
Acquiring valid informed consent from competent adult patients is a prerequisite for performing any type of medical intervention on them. A need to acquire such consent is entailed by the commitment to the principle of respecting and promoting patient autonomy, especially in relation to patients’ own health. The more invasive and/or riskier the procedure, the more robust the informed consent process must be, in ordinary cases. Whilst it is usually fairly straightforward to obtain valid consent from competent adult patients, things become more complicated when patients are not competent, or might not be competent in relation to certain aspects of the treatment decision. Cases involving eating disorders have sometimes been brought before judges who have the task of deciding whether the patient is competently making a decision to refuse treatment and/or whether the patient can be treated (often force-fed) against his or her will (or in the absence of ‘will’, where a patient’s competence is in doubt).
The procedures for obtaining consent from patients with anorexia – and assessment of whether that consent is valid – are clearly challenging and likely to be complicated by unique features of different cases. However, even where the patient is deemed to be competent, giving an anorexic patient the option of treatment in the form of deep brain stimulation could still present ethical problems: if the patient justifiably or even falsely believes that failure to give consent for the surgical intervention will result in doctors resorting to force-feeding further down the line, they might opt for the neurosurgical intervention believing that they will be able to continue resisting any new or enhanced urges to eat.
Unlike force-feeding, the neurosurgical intervention itself does not frustrate the anorexic patient’s primary goal to forgo food. If patients believe that they will have to endure one or the other intervention, and that surgery is in this respect ‘less bad’, then their decision to undergo the procedure in the perceived context of a ‘forced choice’, or coercion, will not be autonomous. Whilst we do not claim that it is impossible to autonomously consent to deep brain stimulation for anorexia, we simply raise this concern as one that should be considered in cases where patients might be likely to perceive the situation as less than fully free. What patients most want might be to forgo both food and DBS, and not be force-fed. But that option might not be available.
Desire manipulation and biopsychosocial complexity
The second issue we wish to discuss is the biopsychosocial complexity of an eating disorder such as anorexia, and how directly modifying first-order motivations must be thought about from the perspective of this broader context. To be clear, the point we raise is not a concern about this type of treatment somehow making people ‘less themselves’, as if the dysfunctions typical of anorexia are somehow essential to the person suffering from it. Rather, we intend to emphasise that attention should be paid to the agent’s complicated matrix of beliefs, goals and desires, and the multitude of factors – from biology to environment – that influence it.
Deep brain stimulation is primarily a biological intervention. It directly alters neurological activity in the brain, with psychological and behavioral consequences. One news report states, ‘Deep in the middle of the brain, the nucleus accumbens is thought to malfunction in some patients with the eating disorder – meaning they do not enjoy eating.’ Prof Aziz is quoted as saying, ‘What we are trying to do is to rewire the brain so that eating becomes a pleasurable experience…food becomes a very painful object for anorexics and the part of the brain that makes people enjoy eating does not work in the same way’. This approach essentially identifies a neurological pathology and attempts to remedy it.
Although we do not deny that there may be biological treatments for psychological disorders in some cases, we want to emphasize the complexity that emerges from a model according to which social, psychological and biological factors are casually interrelated in the production of mental disorder. Even if it is the case that the most seriously affected anorexic patients manifest neurological malfunction, a lack of desire to eat is not the only contributory factor. It is widely agreed that sometimes anorexia emerges where the individual feels a need to exert control over some aspect of their life. Indeed, his recent Leverhulme lecture series on the science of self-control, Prof Neil Levy suggested that some eating disorders might involve an excess of self-control. Anorexia is also sometimes comorbid with body dysmorphic disorder, an anxiety disorder resulting from a mixture of genetic, psychological and environmental factors (such a bullying), which causes a person to have a distorted perception of how they look. Given this complex picture of the etiology and manifestation of anorexia, it seems that simply making food more appealing may leave much of the disorder unaddressed, or could in some cases even exacerbate it: if the anorexic person has a strong desire to control what they eat, or has the illusion that they are much larger than they in fact are, then finding food more rewarding at the appetitive level may do little to address these and other underlying factors. In either case, the new appetite may not be welcomed and may cause additional psychological distress. Higher-order goals, values and perceptions of one’s body are unlikely to reduce to simple appetitive drives, although they will interact.
One feature of deep brain stimulation is that it gives patients control over stimulation. They can decide whether to terminate stimulation or when and how to use it. For example, patients with Parkinson’s Disease or Depression are given control over their stimulation device. In this way, if there are unpleasant side effects, they can terminate stimulation
On the face of it, this is empowering. Patients control their treatment. But we should not blithely accept that this enhances autonomy or respects it. Stimulation may impose a change on the agent’s values and deep-seated desires in a way that makes the choice to continue inauthentic.
The most extreme case is addiction. Pleasure is addictive – the fact that a person continues to stimulate a part of the pleasure pathway does not necessarily mean that the choice is an expression of reflection, deep-seated values, or anything to do with that agent’s own authentic self. It is possible that such stimulation alienates the agent, and this will need careful consideration.
The Wrong Kind of Treatment?
Another consideration is that many people will see that such an approach is using the wrong kind of treatment modality. As we have said, even if there is some kind of brain pathology or malfunction in anorexia (of course there would be as the person eats abnormally and that behaviour must stem from brain activity), the causes are primarily psychosocial. Thus the best treatment is psychosocial.
This is fallacious reasoning. Even if the cause is psychosocial, it is an open question whether the best treatment is biological, psychological or social. For this we need scientific evaluation, for example, through randomised controlled trials.
Whether deep brain stimulation is the best treatment for refractory anorexia is a scientific question, to be answered once we have a clear idea of the values that we are seeking to achieve. This requires good science. But it also requires good ethics. We need to examine what kind of desires an intervention should foster, and how these can be reconciled with the wider goals and desires of the patient. We must also ask to what extent the goals of an anorexic person can autonomously be held and acted in accordance with These are ethical questions which we must answer before conduct our scientific evaluation.
So, although we welcome research into these promising interventions, we suggest that close attention must be paid to obtaining valid consent if neurosurgery were to be offered as a treatment option in ‘last resort’ cases. We also encourage neuroscientists to work alongside psychiatrists and psychologists to ensure that the neurobiological effects are understood and monitored in the broader framework that incorporates psychological and environmental factors, which are likely to be interrelated in the etiology and prognosis of a complex disorder such as anorexia. Finally, we need to address ethical questions pertaining to the limits of anorexic patients’ autonomous control over their various desires and pursuit of their fulfillment.
- This article was originally published on the Practical Ethics blog
This opinion piece reflects the views of the author, and does not necessarily reflect the position of the Oxford Martin School or the University of Oxford. Any errors or omissions are those of the author.