The paper by Craddock et al ^{Reference Craddock, Antebi, Attenburrow, Bailey, Carson, Cowen, Craddock, Eagles, Ebmeier, Farmer, Fazel, Ferrier, Geddes, Goodwin, Harrison, Hawton, Hunter, Jacoby, Jones, Keedwell, Kerr, Mackin, McGuffin, MacIntyre, McConville, Mountain, O'Donovan, Owen, Oyebode, Phillips, Price, Shah, Smith, Walters, Woodruff, Young and Zammit1} and the subsequent eLetters illustrate the variety of opinions that attracted me to psychiatry. I work in a multi-agency service and our assessments and interventions can be carried out by professionals in Mind, in social services and in the National Health Service (NHS). In our service we share responsibilities. This allows me (a consultant psychiatrist) to pursue a resurgent interest in psychopharmacology, treatment adherence and the harm caused by side-effects of medication. Although I appreciate the academic endeavours in biomedical science, I believe it is very important to contextualise them for non-academics. Randomised controlled trials don't speak to clinicians as well as naturalistic studies. I have noticed that some of my psychiatric colleagues (and myself at times) shy away from precise diagnosis, acutely aware of how diagnoses are deliberately used to stigmatise people by individuals outside mental health services (as well as within). This is happening at a time when case definitions are becoming important to health service managers. Perhaps some psychiatrists are uncomfortable in their traditional territory. However, if psychiatrists step back too far, then others will move in. I expect that senior managers, rather than other clinicians or service users, are likely to move into the spaces that we vacate. Psychiatrists should not support the replacement of ‘doctor knows best’ with ‘manager knows best’. New Ways of Working may end up doing exactly that. Instead of being a shot in the arm, it may be a shot in the foot. Four trusts in the north of England are already constructing their own diagnostic systems to use alongside or instead of existing diagnostic schemes as a currency for payment by results. Assigning patients to pseudo-diagnostic ‘care clusters’ could be something all staff do, not just the doctors. If psychiatrists step back from diagnosis, then diagnosis may change from a clinical concept with an associated evidence base, to a financial planning tool. There are other drivers of change too. In the prevalent atmosphere of anxiety and blame, risk assessment, not diagnosis, is now arguably the main gateway into acute mental health services. This means that some very ill people may have to wait for treatment, while people who seem to be at acute risk are attended to first.

Times change and if psychiatrists of any persuasion want to retain some influence they have to put up, not shut up; so well done for making the biomedical case. Biomedical psychiatry complements psychosocial psychiatry and is uniquely part of medical doctors' expertise. The Royal College of Psychiatrists should take this issue up with its members.