Inpatient use and area-level socio-environmental factors in people with psychosis

Purpose There is consistent evidence that socio-environmental factors measured at an area-level, such as ethnic density, urban environment and deprivation are associated with psychosis risk. However, whether area-level socio-environmental factors are associated with outcomes following psychosis onset is less clear. This study aimed to examine whether the number of inpatient days used by people presenting to mental health services for psychosis was associated with five key area-level socio-environmental factors: deprivation, ethnic density, social capital, population density and social fragmentation. Methods Using a historical cohort design based on electronic health records from the South London and Maudsley NHS Trust Foundation electronic Patient Journey System, people who presented for the first time to SLAM between 2007 and 2010 with psychosis were included. Structured data were extracted on age at presentation, gender, ethnicity, residential area at first presentation and number of inpatient days over 5 years of follow-up. Data on area-level socio-environmental factors taken from published sources were linked to participants’ residential addresses. The relationship between the number of inpatient days and each socio-environmental factor was investigated in univariate negative binomial regression models with time in contact with services treated as an offset variable. Results A total of 2147 people had full data on area level outcomes and baseline demographics, thus, could be included in the full analysis. No area-level socio-environmental factors were associated with inpatient days. Conclusion Although a robust association exists between socio-environmental factors and psychosis risk, in this study we found no evidence that neighbourhood deprivation was linked to future inpatient admissions following the onset of psychosis. Future work on the influence of area-level socio-environmental factors on outcome should examine more nuanced outcomes, e.g. recovery, symptom trajectory, and should account for key methodological challenges, e.g. accounting for changes in address.