Feasibility and Acceptability of a Smartphone App to Assess Early Warning Signs of Psychosis Relapse

BACKGROUND Around 80% of those treated for a first episode of psychosis relapse within five years, with cumulative relapse rates of 78% and 86% for second and third relapses during this period. Relapses can be devastating for the individual and their family, may lead to a deteriorating course of illness and frequently require hospital admission, the principal source of schizophrenia's annual direct cost to the NHS of over £3.9 billion. Given the prevalence and considerable negative consequences of relapse, it is clear that relapse prevention strategies for those with psychosis are a priority. There is growing evidence that interventions monitoring 'early signs' can be effective in preventing relapses of psychosis. Such interventions work on the premise that timely prediction of relapses will allow preventative action to be taken, minimizing the chance of full relapse occurring. The patient is assisted in identifying and monitoring early signs of relapse, and in developing concrete action plans for dealing with them (e.g. short term medication increases, stress reduction techniques, intensive psychological support). Early signs reported to emerge in the weeks before a relapse include: anxiety, dysphoria, insomnia, poor concentration, attenuated psychotic symptoms (Early Signs Scale, ESS) and fear of relapse (Fear of Recurrence Scale, FoRSe). However, such checklists are only modestly predictive of relapse so they could be improved by including more specific psychopathology. Evidence suggests that 'basic symptoms' may be useful relapse indicators that could be added to checklists of conventional early warning signs to improve predictive power. Studies in individuals at high risk of psychosis have characterised basic symptoms as subtle, sub-clinical, qualitative disturbances in one's experience of oneself and the world which are predictive of transition to first episode psychosis. Typical basic symptoms include: changes in perceptions, such as increased vividness of colour vision; mild subjective cognitive problems; impaired tolerance to certain stressors; subjective difficulty finding or understanding common words. Two retrospective studies examining service users' experiences in the run up to a recent relapse of psychosis provide preliminary evidence that basic symptoms occur prior to relapse. AIMS The long term aim is to conduct a definitive study to prospectively investigate the predictive value of basic symptoms as early signs of psychosis relapse using a mobile phone application to monitor these within individuals' everyday lives. In line with the Medical Research Council guide for developing complex interventions a feasibility study will be conducted first. This study has four phases. In Phase 1 the investigators will design a measure of basic symptoms, assessed via smart-phone, and adapt it as applicable following feedback from participants. Phase 2 begins with a screening interview to identify participants with at least one basic symptom; these individuals will be eligible for Phase 3 (since past basic symptoms are likely to predict future basic symptoms). Cross-sectional assessments will also be conducted in Phase 2; by comparing those with and without basic symptoms the investigators will begin to characterise the sub-group of individuals with whom the basic symptom assessment can be used. In Phase 3 a prospective, longitudinal design will be used to investigate the feasibility of using a mobile phone application to regularly measure basic symptoms, conventional early signs and relapse over an extended period. Finally, in Phase 4, participants' experiences of using the phone application will be explored using qualitative interviews (acceptability).