Tobacco and cannabis use are highly prevalent among individuals with psychotic disorders. Tobacco use has been consistently linked to poorer clinical and cognitive outcomes, whereas findings regarding cannabis use remain mixed. The present study was a secondary analysis of cross-sectional data from the Bipolar-Schizophrenia Network on Intermediate Phenotypes 2, a large multi-site psychosis spectrum sample with cognitive and clinical assessments. The primary aim was to more precisely characterize how tobacco or cannabis use, co-use, and non-use differentially relate to clinical and cognitive function. The final sample included 884 individuals with psychotic disorders without diagnosis of substance abuse in the past 30 days or substance dependence in the past 3 months for any substance except tobacco. Among individuals with psychosis, tobacco-only users demonstrated greater symptom burden and worse cognitive impairment than several comparison psychosis substance groups. In contrast, cannabis use showed more nuanced associations. Greater adult cannabis use was tied to worse general psychopathology for current tobacco users but fewer general symptoms among those who never used tobacco. More adult cannabis use was linked to better cognitive performance for current tobacco users but was not linked to working memory among past tobacco users. These findings highlight tobacco use as a potential marker of greater disease severity and cognitive impairment among individuals with psychosis, while also demonstrating the importance of considering patterns of co-occurring use of substances.

Loneliness is highly prevalent in psychosis and is associated with poorer clinical and quality-of-life outcomes. In nonclinical populations, loneliness is linked to a range of social and occupational difficulties. Difficulties with socio-occupational functioning (SOF) are common in psychosis and are consistently identified as a key recovery priority. However, the relationship between loneliness and SOF in psychosis remains unclear. A PROSPERO-registered systematic review was conducted. Database searches (PubMed, Scopus, PsycINFO and CINAHL) identified 1118 records, of which 15 met inclusion criteria and were synthesised narratively. Studies examined a range of SOF domains, including global, social, community and everyday functioning across various designs. Greater loneliness was generally associated with poorer SOF, although several studies reported non-significant relationships. Associations were clearest for social participation and community integration, while evidence for everyday functioning was more limited. Qualitative findings described reciprocal relationships between withdrawal, barriers to participation and feeling disconnected. Several psychosocial factors, including internalised stigma and social support, appeared to influence the loneliness-SOF relationship. Overall, the findings suggest that loneliness is associated with social and community aspects of functioning, although the relationship varies across measures and samples. Clarifying directionality and testing whether changes in loneliness and SOF influence one another should be priorities for future longitudinal and intervention research.

Most individuals affected by schizophrenia manifest negative symptoms. Because negative symptoms have a persistent deleterious effect on the social and vocational performance of the affected individuals, investigators in academia and pharmaceutical industry are making efforts to develop treatments for this condition. To be successful basic science must expand beyond attempts to interfere with the brain dopaminergic activity and clinical science must improve trials methodology.

Astrocytic dysfunction has been implicated in schizophrenia, but biomarkers of astrocytic activity have shown inconsistent patterns. Studies on disease heterogeneity, specifically treatment-resistant schizophrenia (TRS) and ultra-treatment-resistant schizophrenia (UTRS) are lacking. In this study, plasma glial fibrillary acidic protein (GFAP) was measured using an ultrasensitive Simoa assay in 174 participants (58 controls, 35 first-degree relatives, and 81 patients with TRS, according to Treatment Response and Resistance in Psychosis (TRRIP) criteria. A further n = 35 fulfilled TRRIP criteria for UTRS. Data included symptom severity, cognition, and functioning assessments. Cortical thickness was measured with magnetic resonance imaging (MRI). Associations between GFAP and demographic, clinical, cognitive, and imaging variables were examined using regression models adjusted for age, sex, and BMI. Plasma GFAP concentrations did not differ between controls and TRS or UTRS participants. GFAP showed no associations with symptom severity, cognition, functioning, or cortical thickness. Duration of illness showed a significant interaction with UTRS diagnosis associated to higher GFAP levels (b = 3.67, 95% CI 1.62-5.72, p < 0.001), while individuals with TRS showed the opposite pattern. The axonal marker neurofilament light chain (NfL), showed a positive relationship with disease duration in both disease categories. The results suggest that plasma GFAP does not act as diagnostic or severity marker in schizophrenia, but can highlight astroglial-related processes where UTRS and TRS show different trajectories.

Hypervigilance and suspiciousness are conceptually overlapping constructs included in measures of paranoia spectrum symptoms. However, no studies have examined the structural relationship between these constructs. Additionally, few studies have investigated whether ethnoracial differences in hypervigilance and suspiciousness are confounded by measurement properties. We hypothesized that the structure of hypervigilance and suspiciousness would be best conceptualized by a hierarchical model. We also tested measurement differences across ethnoracial groups. Community individuals (nFull = 8081, nAsian = 1794, nBlack = 863, nHispanic = 1182, nWhite = 3580) aged 16-30 ascertained from the Multi-site Assessment of Psychosis-risk (MAP) study completed self-report assessments of hypervigilance (PTSD Checklist-Civilian Version) and suspiciousness (Prodromal Questionnaire). Confirmatory factor analyses tested a hierarchical model and a correlated two-factor model. Measurement invariance analyses were conducted on the optimal latent factor structure. Although both the hierarchical and correlated two-factor models showed good fit in the full sample and across ethnoracial groups, only the correlated two-factor model found theoretically consistent loadings on their respective factors. Configural, metric, scalar, and strict measurement invariance was found across ethnoracial groups. Findings suggest that hypervigilance and suspiciousness are distinct, but moderately correlated constructs, which may help to improve the assessment of paranoia spectrum symptoms among community individuals. Measurement invariance tests showed the PTSD Checklist-Civilian Version and Prodromal Questionnaire respectively assess hypervigilance and suspiciousness similarly across ethnoracial groups, suggesting that these screening tools are valid self-reports among ethnoracially diverse community individuals.