Item-level psychometric data define continuous diagnostic and symptom fields across two transdiagnostic cohorts

Background. Psychiatric diagnoses remain clinically useful, but symptom burden, impairment and comorbidity vary continuously within and across disorders. We tested whether item-level psychometric data could be used to define a low-dimensional participant space on which diagnoses and other clinical variables vary as smooth, clinically interpretable gradients. Methods. We analysed two independent public transdiagnostic cohorts with structured diagnostic assessment and item-level questionnaire/interview data. Within each cohort, we derived a two-dimensional psychometric embedding without using diagnostic labels. We then modelled diagnoses, symptom severity, functioning and treatment burden as smooth functions of embedding position using out-of-fold generalised additive models, and assessed the extent to which residual item-level features improved prediction beyond the common low- dimensional structure. Results. In both cohorts, participants occupied a continuous two-dimensional psychometric space. Diagnoses were distributed as broad, overlapping probability gradients rather than sharply separated regions. Symptom severity, functioning and treatment burden varied systematically across the same space. For common diagnoses, out-of-fold models based on embedding position were generally well calibrated and captured most of the predictable signal, with residual features adding modest gains for a limited subset of outcomes. Although axis orientation differed across cohorts, the overall pattern of continuous geometry and overlapping clinical fields recurred. Conclusions. Routine item-level psychometric data can define a low-dimensional participant space on which multiple psychiatric outcomes vary as smooth, overlapping gradients.