Blindness and reliability in lifetime psychiatric diagnosis

Area of Science:

• Psychiatry
• Genetics
• Epidemiology

Background:

• Accurate lifetime psychiatric diagnoses are essential for genetic studies of psychiatric disorders.
• Structured interviews are a common method for obtaining psychiatric histories.
• Assessing the reliability of diagnostic assessments is critical for research validity.

Purpose of the Study:

• To evaluate the reproducibility of lifetime psychiatric diagnoses obtained through structured interviews.
• To assess the credibility of diagnostic assignments in the context of genetic research.
• To determine the impact of interviewer blinding on diagnostic accuracy.

Main Methods:

• Two paradigms were employed: simultaneous coding by multiple interviewers and test-retest interviews with a mean interim of 6.7 months.
• Inter-rater reliability was assessed using simultaneous coding of interview forms.
• Diagnostic agreement was measured using the kappa (k) statistic.

Main Results:

• Both simultaneous coding and test-retest paradigms demonstrated high reliability for psychiatric diagnoses.
• The overall kappa value for diagnostic agreement was .79, indicating substantial consistency.
• Minor depression was the only diagnostic category that showed low reliability.
• Interviewer blinding did not significantly affect the unbiased nature of the interview.

Conclusions:

• Structured interviews provide a reliable method for assessing lifetime psychiatric diagnoses, supporting their use in genetic studies.
• High diagnostic reproducibility over time suggests stability in psychiatric conditions.
• While interviewers need not be blind, independent diagnosticians should determine final diagnoses for optimal accuracy.