Abstract

When clinical decision support systems fail 70-90% of mental health referrals, the problem isn't implementation , it's foundational assumptions. Through reflexive thematic analysis with ten Norwegian GPs, we reveal how CDSS perpetuate rather than solve referral crises by assuming certainty where none exists. Our Design Science Research uncovers four systemic failures: (1) pervasive uncertainty across diagnostic, information, prognostic, and process dimensions that systems deny; (2) an experience paradox where expertise breeds withdrawal, not wisdom; ( 3) workflow impossibilities forcing dangerous shadow IT adoption; (4) communication breakdowns requiring strategic dishonesty for patient access. Most critically, we identify a "gray zone" , patients falling between service levels, invisible to binary syste m architectures. These findings drive transformative requirements that rethink CDSS fundamentals: from deterministic to probabilistic interfaces with confidence intervals, from judgment to pattern -based learning, from documentation burden to workflow integ ration, from gatekeeping to bidirectional collaboration. Mental health CDSS must embrace, not eliminate, irreducible clinical uncertainty

Download

Share

COinS