Abstract
BACKGROUND AND HYPOTHESIS: Duration of untreated psychosis (DUP) has been linked to worse mental health outcomes in psychotic disorders. We meta-analytically studied the relationship between "long" vs. "short" DUP and mental health outcomes. STUDY DESIGN: This PRISMA/MOOSE-compliant meta-analysis searched for nonoverlapping individual studies from database inception until November 01, 2023, reporting data from author-defined "short"/"long" DUP (according to author's definition) in patients with first-episode psychosis (FEP). We compared differences between "short"/"long" DUP groups at baseline and/or follow-up in continuous and binary outcomes. We conducted random-effects meta-analyses, stratified analyses, heterogeneity analyses, meta-regression analyses, and quality assessment (PROSPERO: CRD42023479321). STUDY RESULTS: From 16,055 citations, 34 studies were included (n = 6,425, age = 27.5 ± 7.1 years, males = 60.4%, white = 70.2%, DUP: mean = 60.8 ± 43.8 weeks, median = 52.5, interquartile range = 31.3, 68.0 weeks, follow-up = 19.2 ± 35.0 months). The definition of "short"/"long" varies significantly between the studies. Compared to "short" DUP (mean = 10.2 ± 11.2 weeks), "long" DUP (mean = 58.8 ± 76.4 weeks) was associated with higher baseline Positive and Negative Syndrome Scale (PANSS) negative (k = 14, ES = 0.45, 95%CI = 0.16, 0.74) and Scale for the Assessment of Negative Symptoms (k = 7, ES = 0.29, 95%CI = 0.11, 0.47) scores, lower remission (k = 7, OR = 0.40, 95%CI = 0.24, 0.67) and more suicide attempts (k = 4, OR = 2.01, 95%CI = 1.36, 2.96). At follow-up, compared to "short" DUP, "long" DUP was associated with lower Global Assessment of Functioning (k = 4, ES = -0.63, 95%CI = -0.83, -0.43) and higher PANSS negative subscale scores (k = 5, ES = 0.66, 95%CI = 0.05, 1.27). CONCLUSIONS: In FEP, longer DUP is related to greater baseline negative symptoms, less remission, and more suicide attempts, as well as greater postbaseline negative symptom severity and functional disability. To what degree longer DUP contributes to poorer outcomes or whether DUP only correlates with these outcomes requires further study. A greater consensus on the definition of long DUP is needed to make comparisons between studies more feasible.