Environment in Association With Symptoms of Psychosis

young man furrowed brow peeking from a window blind
Researchers gleaned information from local data in West London, United Kingdom, which showed some correlation between environment and symptoms in a cohort of 319 patients with first-time psychosis that included 61.3% of patients with schizophrenia.

Associations may exist between environmental characteristics and psychosis symptoms and symptom clusters, according to a study recently published in BMJ Open. Although causal mechanisms have not been identified, these findings lend support to calls for dimension-based models of psychosis.

This cross-sectional study included 195 patients with schizophrenia (61.3%) of 319 participants, (median age, 24.16 years; 65.83% men) all with first-time psychosis. Participants were aged 16 years or older, lived in West London, and had taken antipsychotic medication for <12 weeks. The demographic data collected on the participants did not include ethnicity. Each participant was characterized using 3 symptom-dimension scores for disorganization, positive, and negative symptoms, using the Scale for the Assessment of Positive Symptoms and the Scale for the Assessment of Negative Symptoms. Each individual’s place of residence was also assessed for measures of urbanicity, inequality, deprivation, and social capital. To determine whether neighborhood characteristics predicted psychotic symptoms, multilevel linear regression was used to regress symptom dimension scores on neighborhood-level predictors. 

Related Articles

Higher income inequality was associated with less severe negative symptoms, with an adjusted correlation coefficient of -1.66 (95% CI, -2.86 to -0.46; P <.01). Furthermore, an association was identified between higher ethnic minority segregation and lower severity of positive symptoms, with an adjusted correlation coefficient of -2.36 (95% CI, -4.2 to -0.52; P =.01). Neighborhood characteristics did not predict disorganization symptoms or variance in global hallucinations or paranoid delusions (P >.05).

Limitations of this study include the cross-sectional design and consequent inability to determine causality, potential measurement error due to the broad study window, a lack of data on individual ethnicity, inherent limitations to the assumption that voter turnout approximates the degree of social cohesion, and a relatively small number of participants per neighborhood. 

The study researchers conclude that “at least some of the association between psychosis and the environment may be operating at the level of symptoms,” with correlations identified between certain environmental predictors and “specific symptoms or symptom clusters.” These findings support the need “in both clinical and research practice for dimension-based models of psychosis.”


Tibber MS, Kirkbride JB, Mutsatsa S, et al. Are socioenvironmental factors associated with psychotic symptoms in people with first-episode psychosis? A cross-sectional study of a West London clinical sample [published online September 18, 2019]. BMJ Open. doi: 10.1136/bmjopen-2019-030448