These have included social psychiatry, epidemiological surveys, instrument development, the social causes and concomitants of depression and schizophrenia, psychological mechanisms in schizophrenia, cognitive behaviour therapy, and health service research.

In the last twenty-five years, my main research activity, with colleagues, has been in epidemiological and mental health service research, and in using psychological treatments to test psychological theories of psychosis. Much of the recent epidemiological research has involved analysis of the four National Surveys of Psychiatric Morbidity (1993-2014), together with the National Prisons Survey. In addition, we have just published the results of a psychiatric survey in Holloway and Pentonville prisons (2017). 

There have been major papers in the epidemiology of substance abuse and of suicidal ideation, and studies of social exclusion and of access to treatment for mental disorders. Most recently, we have used National Survey data to assess links between victimisation experiences and psychosis, and to examine the structure of quasi-psychotic experience in the normal population. We are also currently analysing the European Schizophrenia Cohort (EuroSC), a sample of 1208 people with schizophrenia from Germany, France and England, followed up in detail six-monthly for two years.

My mental health service research included evaluations of policy initiatives, with important RCTs of Crisis Resolution Teams and Assertive Outreach Teams. 

The Prevention of Relapse in Psychosis programme (funded by Wellcome 2001-2007) was a collaboration with KCL, UEA and the University of Manchester. It used a large multicentre RCT of CBT and family intervention in psychosis to test a detailed model of psychotic symptoms (ISRCTN 83557988. Over 300 patients were randomised and a large number of papers based on planned baseline analyses have been published. The results of the trial were published in June 2008, and we used the follow up data to test hypotheses relating to psychological processes associated with improvement. We have now obtained further funding in the light of these findings. 

This work led to a collaboration with Prof Mel Slater at UCL on a programme using virtual reality in experiments designed to identify psychological processes involved in paranoid ideation in both patients and normal members of the population.

The most recent initiative in this programme involved a collaboration with KCL, and the Universities of Oxford, Manchester and Sussex for a trial of cognitive behaviour therapy for psychosis designed to ameliorate the fast thinking that characterises delusions. The therapy (SlowMo) is supported by personalised phone apps.  The project received £1.4 million of EME funding ("A randomised controlled trial to evaluate the outcomes and mechanisms of a novel digital reasoning intervention for persecutory delusions"). Recruitment and retention of patients in the trial was very successful: 362 participants were randomised, The results have now been analysed, and the final report has been submitted (August 2020). The experimental group showed a range of improvements compared with controls. The base paper is being drafted.