Are socioenvironmental factors associated with psychotic symptoms in people with first-episode psychosis? A cross-sectional study of a West London clinical sample

Objectives To determine whether neighbourhood-level socioenvironmental factors including deprivation and inequality predict variance in psychotic symptoms after controlling for individual-level demographics.

Design A cross-sectional design was employed.

Setting Data were originally collected from secondary care services within the UK boroughs of Ealing, Hammersmith and Fulham, Wandsworth, Kingston, Richmond, Merton, Sutton and Hounslow as part of the West London First-Episode Psychosis study.

Participants Complete case analyses were undertaken on 319 participants who met the following inclusion criteria: aged 16 years or over, resident in the study’s catchment area, experiencing a first psychotic episode, with fewer than 12 weeks’ exposure to antipsychotic medication and sufficient command of English to facilitate assessment.

Outcome measures Symptom dimension scores, derived from principal component analyses of the Scale for the Assessment of Positive Symptoms and Scale for the Assessment of Negative Symptoms, were regressed on neighbourhood-level predictors, including population density, income deprivation, income inequality, social fragmentation, social cohesion, ethnic density and ethnic fragmentation, using multilevel regression. While age, gender and socioeconomic status were included as individual-level covariates, data on participant ethnicity were not available.

Results Higher income inequality was associated with lower negative symptom scores (coefficient=−1.66, 95% CI −2.86 to –0.46, p<0.01) and higher levels of ethnic segregation were associated with lower positive symptom scores (coefficient=−2.32, 95% CI −4.17 to –0.48, p=0.01) after adjustment for covariates.

Conclusions These findings provide further evidence that particular characteristics of the environment may be linked to specific symptom clusters in psychosis. Longitudinal studies are required to begin to tease apart the underlying mechanisms involved as well as the causal direction of such associations.