Conversion to psychosis in adolescents and adults: similar proportions, different predictors. Issue 12 (6th September 2021)

• Record Type: Journal Article
• Authors: Zhang, TianHong; Xu, LiHua; Chen, Ying; Wei, YanYan; Tang, XiaoChen; Hu, YeGang; Li, ZhiXing; Gan, RanPiao; Wu, GuiSen; Cui, HuiRu; Tang, YingYing; Hui, Li; Li, ChunBo; Wang, JiJun
• Is Part Of: Psychological medicine. Volume 51:Issue 12(2021)
• Page Start: 2003
• Page End: 2011
• Publication Date: 2021-09-06
• Subjects: Cohort study -- early onset -- schizophrenia -- transition -- ultra high risk; Psychiatry -- Periodicals; Medicine and psychology -- Periodicals; Clinical psychology -- Periodicals; 616.89
• DOI: 10.1017/S0033291720000756
• Languages: English
• ISSNs: 0033-2917

Abstract

Abstract: Background: Age effects may be important for improving models for the prediction of conversion to psychosis for individuals in the clinical high risk (CHR) state. This study aimed to explore whether adolescent CHR individuals (ages 9–17 years) differ significantly from adult CHR individuals (ages 18–45 years) in terms of conversion rates and predictors. Method: Consecutive CHR individuals ( N = 517) were assessed for demographic and clinical characteristics and followed up for 3 years. Individuals with CHR were classified as adolescent ( n = 244) or adult ( n = 273) groups. Age-specific prediction models of psychosis were generated separately using Cox regression. Results: Similar conversion rates were found between age groups; 52 out of 216 (24.1%) adolescent CHR individuals and 55 out of 219 (25.1%) CHR adults converted to psychosis. The conversion outcome was best predicted by negative symptoms compared to other clinical variables in CHR adolescents (χ 2 = 7.410, p = 0.006). In contrast, positive symptoms better predicted conversion in CHR adults (χ 2 = 6.585, p = 0.01). Conclusions: Adolescent and adult CHR individuals may require a different approach to early identification and prediction. These results can inform the development of more precise prediction models based on age-specific approaches.