Best-estimate versus structured interview-based diagnosis in first-admission psychosis

Area of Science:

• Psychiatry
• Clinical Psychology
• Psychopathology

Background:

• Accurate diagnosis is crucial for effective treatment of psychotic disorders.
• The Structured Clinical Interview for DSM-III-R (SCID) is a widely used diagnostic tool.
• Evaluating diagnostic reliability in first-admission psychotic patients is essential.

Purpose of the Study:

• To compare psychiatrist best-estimate diagnoses with SCID algorithm diagnoses in first-admission psychotic patients.
• To assess the reliability and validity of the SCID for diagnosing various psychotic disorders.
• To determine the necessity of the best-estimate procedure in psychosis research.

Main Methods:

• Comparison of psychiatrist best-estimate diagnoses with SCID algorithm diagnoses at study entry and 6-month follow-up.
• Calculation of sensitivity, specificity, and kappa agreement for schizophrenia, bipolar disorder with psychosis, major depression with psychosis, and organic psychoses.
• Inclusion of 310 patients at entry and 228 at follow-up.

Main Results:

• Satisfactory diagnostic agreement for schizophrenia and bipolar disorder with psychosis.
• Moderate agreement for major depression with psychosis.
• Mixed results for organic psychoses, with lower sensitivity and agreement.
• Reasons for disagreement included organic factors and clinical judgment versus structured interview rules.

Conclusions:

• The SCID, administered by trained clinicians with supplementary information, reliably diagnoses schizophrenia and bipolar disorder.
• The best-estimate diagnostic procedure is necessary for studies involving a broad range of psychoses, especially when substance use is not an exclusion criterion.
• Diagnostic accuracy for organic psychoses remains a challenge for structured interviews alone.