In ‘Wake-up call for British psychiatry’ Craddock et al explained how recent attempts to improve psychosocial care for people with mental illness focus on non-specific psychosocial support. This has been at the expense of proper diagnostic assessment and prescription of treatment by psychiatrists aimed at treatment of specific disorders and recovery. They describe a creeping devaluation of psychiatry which is caricatured as narrow, biological, reductionist, oppressive, discriminatory and stigmatising. Some trusts have implemented ‘New Ways of Working for Psychiatrists’ in a way that undermines the central importance of psychiatrists in mental healthcare. Consequently, patients may be treated in secondary care without ever being seen by a psychiatrist. We consider a number of different changes that have interacted in unforeseen ways, with unintended adverse consequences for psychiatric services in England. We aim to continue the debate here.

We consider the properties of the highest posterior probability model in a linear regression setting. Under a spike and slab hierarchy we find that although highest posterior model selection is total risk consistent, it possesses hidden undesirable properties. One such property is a marked underfitting in finite samples, a phenomenon well noted for Bayesian information criterion (BIC) related procedures but not often associated with highest posterior model selection. Another concern is the substantial effect the prior has on model selection. We employ a rescaling of the hierarchy and show that the resulting rescaled spike and slab models mitigate the effects of underfitting because of a perfect cancellation of a BIC-like penalty term. Furthermore, by drawing upon an equivalence between the highest posterior model and the median model, we find that the effect of the prior is less influential on model selection, as long as the underlying true model is sparse. Nonsparse settings are, however, problematic. Using the posterior mean for variable selection instead of posterior inclusion probabilities avoids these issues.