How do we measure psychiatric diagnoses? Implications of the choice of instruments in epilepsy

- The DIS-IV assessed lifetime and current DSM-IV-TR diagnoses in epilepsy.
- The DIS-IV was compared to screening instruments and single questions across diagnoses.
- Specificity was high and sensitivity low for the DIS-IV in participants with epilepsy.
- The CBCL and ASR DSM-oriented diagnoses do not adequately identify DSM diagnoses.
- Instrument selection must consider the study questions and instrument costs.

We evaluated several commonly used screening instruments for the detection of mood disorders, anxiety disorders, and attention-deficit hyperactivity disorder (ADHD). These were compared to a criterion-based standardized questionnaire, the Diagnostic Interview Survey (DIS)-IV, designed to make DSM-IV-TR diagnoses in the community-based study of childhood-onset epilepsy. The DIS-IV was administered to young adult cases with epilepsy at a 15-year follow-up assessment and compared to symptom screens administered at the same visit, and at a previous 9-year assessment. Among cases, the specificity of the DIS-IV ranged from 0.77 to 0.99 and the predictive value of a negative psychiatric diagnosis was similarly high. Sensitivity was lower, ranging from 0 to 0.77, with correspondingly low predictive value of a positive diagnosis. Symptom-based instruments assess current symptom burden and are useful for determining associations with ongoing seizures or quality of life. Criterion-based standardized interviews, such as the DIS-IV, provide psychiatric diagnoses over the lifetime, which is most useful in studies of epilepsy genetics and studies of comorbidities and prognosis of epilepsy.