Eagles et al (Reference Eagles, Klein and Gray2001) emphasise the problems inherent in the psychiatrist's role with regard to suicide prevention. We sympathise with their view that unrealistic expectations may lead to psychiatrists being unfairly criticised. Nevertheless, it would be a pity if we were to lose enthusiasm, without good reason, for engaging in the challenge of suicide prevention.

In quoting our paper (Reference Vassilas and MorganVassilas & Morgan, 1993), Eagles et al refer to our finding that younger suicides (aged <35 years) had a relatively low general practitioner (GP) contact rate (20%) in the last month before they died. However, it is also important to realise that the contact rate varied greatly between subgroups of our sample. For instance, 52% of older suicides made such contact, and more specifically 68% of those over 65 years of age did so (Reference Vassilas and MorganVassilas & Morgan, 1994). In a later paper (Reference Vassilas and MorganVassilas & Morgan, 1997) we showed that 39% of all males and 76% of all females made contact with any service within the last month of their lives. This index concerned contact with either primary or secondary services rather than just with mental health services, which is the criterion used by Appleby et al (Reference Appleby, Shaw and Amos1999). Our wider index of contact presents a more encouraging, yet we believe realistic, view of our potential role in suicide prevention.

It is clear that in practice it can be difficult to coordinate primary and secondary care, but close collaboration between psychiatrist and GP, with regard to the assessment and management of suicide risk, surely warrants special attention in our attempts to become more effective in suicide prevention. The acquisition of relevant clinical skills is, as Eagles et al points out, as crucial as the epidemiological approach in this field.