In a recent article, Henri Wijsbek discusses the 1991 Chabot "psychiatric euthanasia" case in the Netherlands, and argues that Chabot was justified in helping his patient to die. Dutch legislation at the time permitted physician assisted suicide when the patient's condition is severe, hopeless, and unbearable. The Dutch Supreme Court agreed with Chabot that the patient met these criteria because of her justified depression, even though she was somatically healthy. Wijsbek argues that in this case, the patient's integrity had been undermined by recent events, and that this is the basis for taking her request seriously; it was unreasonable to expect that she could start again. In this paper, I do not challenge the Dutch euthanasia criteria in the case of somatic illness, but I argue that both Chabot and Wijsbek are wrong because we can never be sufficiently confident in cases of severe exogenous depression to assist the patient in her irreversible act. This is partly because of the essential difference between somatic and mental illness, and because of the possibility of therapy and other help. In addition, I argue that Wijsbek's concept of integrity cannot do the work that he expects of it. Finally, I consider a 2011 position paper from the Royal Dutch Medical Association on euthanasia, and the implications it might have for Chabot-style cases in the future.