In order to enhance rates of disclosure of domestic violence by service users, Rose et al ^{Reference Rose, Trevillion, Woodall, Morgan, Feder and Howard1} argue for additional specialist training for mental health professionals. I would argue that this is the wrong level at which to pitch training. I would also suggest that to precipitate any real shift in health workers’ attitudes, and therefore practice, we need to see changes in ICD-11.

With ICD-11 still in development, Rose et al's excellent paper should be mandatory reading for the Revision Steering Group. If, as the World Health Organization maintains, the ICD-11 aims to serve ‘not only… as a classification system but also as a building block for health’ (www.who.int/classifications/icd/ICDRevision.pdf), the Revision Steering Group would do well to reflect on the comments captured within this research. Medicine's ambivalence about accepting domestic violence as a key determinant of health is amply highlighted by the absence in our current ICD of any code for domestic violence. Whereas abuse of children can be recorded with a range of different Z codes, the abuse of adults remains non-existent in terms of axis V coding. This position surely validates both those in this study who do not see domestic violence as their business, but also goes some way towards promulgating the idea that this is a condition beyond the realms of ordinary practitioners’ experience and therefore competence.

Training regarding domestic violence needs to happen at university level. Domestic violence is not just something that affects mental health service users, and it is something that medical students can be trained to ask about, think about and feel comfortable enough to approach. I base my comments on training I co-deliver with a service user to 5th-year medical students. The training takes place in the context of practising interviewing skills.

During the course of providing the history, the service user mentions ‘being in a very violent relationship’. Medical students often freeze at this point, or say something such as ‘I am very sorry’, before moving swiftly on to another topic. At the end of the interview slot, the service user talks with the student group about how important it is to be able to ask about and listen to this kind of material, and how the student's desire to move away from the topic leaves her feeling this is something bad/dirty/unmentionable. She tells them how liberating it has been for her to be able to talk about this experience with others, and we both remind them of how common domestic violence is in our society, regardless of class or race or religion. Our work has not been evaluated in terms of whether the students who pass through our module go on to be better at facilitating discussion about domestic violence, but this would perhaps be a useful area of study for medical schools or other professional training centres.