Duration of untreated psychosis and long-term outcomes: A systematic review examining the critical window hypothesis.

BACKGROUND

The "critical window hypothesis" posits that prolonged duration of untreated psychosis (DUP) causes irreversible neurobiological damage during a time-sensitive period, leading to worse long-term outcomes. While widely accepted as justification for early intervention psychiatry, empirical evidence supporting this hypothesis remains contentious. This systematic review evaluates DUP-outcome relationships across symptomatic, functional, cognitive, and neurobiological domains.

METHODS

Following PRISMA guidelines, we searched five databases through December 2025 for longitudinal studies examining DUP and outcomes at ≥12 months in first-episode psychosis. Study quality was evaluated using Newcastle-Ottawa Scale; findings were synthesized narratively due to substantial heterogeneity.

RESULTS

From 8011 citations, 30 studies encompassing approximately 9800 first-episode psychosis patients were included. Longer DUP consistently associated with worse symptomatic outcomes at 1-2 years (median r = 0.20-0.35), but associations weakened substantially at 5-10 years. Functional outcomes showed modest associations (r = 0.15-0.25), primarily in social domains. Cognitive effects were minimal and inconsistent. Neuroimaging revealed gray matter reductions, though directionality unclear. Critical threshold analyses suggested inflection points at 6-12 months but varied considerably. Major limitations included retrospective DUP assessment, variable definitions, survivor bias, and inadequate confounder control (only 37% controlled premorbid functioning), substantially limiting causal inference.

CONCLUSIONS

While DUP shows consistent short-term associations, evidence for irreversible damage through critical window mechanisms remains inconclusive. Temporal attenuation suggests DUP may reflect illness severity and premorbid vulnerability rather than neurotoxicity. The field requires prospective studies with standardized measurement, adequate confounder control, and extended follow-up. Early intervention remains clinically justified through humanitarian principles, even if the critical window hypothesis lacks robust empirical support.