Roychowdhury & Adshead should be thanked for raising the issue of the ethics of the use of actuarial risk assessment in psychiatry. ^{Reference Roychowdhury and Adshead1} These ethics might at first appear obvious: medical practitioners must have an overriding duty to protect the public from serious crime. It follows that they must do everything possible to accurately assess the risk of such crime, including the use of these assessment instruments. However, as Roychowdhury & Adshead point out, these instruments will produce misleading results if the prevalence of the serious crime being considered in the relevant population is low or unknown. Indeed, they point out: ‘A key challenge in psychiatry is that base rates [of the prevalence of serious crime] are often not known, are low and vary for different types of violence.’ So if doctors use these assessments they risk wrongly identifying their patient as at high risk of committing a serious crime, and then act in a way that is not in the best interests of that patient. Such an act would of course be inconsistent with the duties of a doctor as set out by the General Medical Council (GMC) in Good Medical Practice. ² It follows that while the prevalence of particular serious crimes in various patient populations is unknown or is known to be low, the use of these actuarial risk assessments will remain unethical. As Roychowdhury & Ashhead conclude: ‘[structured professional judgement] tools used as checklists of risk factors without construction of risk scenarios or a risk management plan remains harmful and unethical practice.’ In my opinion psychiatrists would value guidance on this issue from the GMC.