Impact of substance-induced and general medical condition exclusion criteria on the prevalence of common mental disorders as defined by the CIDI

Objectives: It has been previously argued that the methodology used by the Composite International Diagnostic Interview version 2.1 to assess the substance-induced and general medical condition exclusion criteria are inadequate. As a result prevalence estimates generated from epidemiological studies using this interview may be underestimated. The purpose of the current study was to examine the substance-induced and general medical condition exclusion criteria in the Australian National Survey for Mental Health and Well-being and determine the impact that they have on prevalence estimates of the common mental disorders. Method: Data from the 1997 Australian National Survey of Mental Health and Well-being were analysed. Frequencies were generated as an indication of how many respondents believed that their psychiatric symptoms were always due to a substance or general medical condition. New DSM-IV prevalence estimates were calculated ignoring the application of the substance-induced and general medical condition exclusion criteria and compared to standard DSM-IV prevalence estimates. Results: The effect of the substance-induced and general medical condition exclusion criteria on final prevalence rates were minimal, with approximately a 0.1% increase when the exclusions were ignored. This equates to a relative difference ranging from no difference for generalized anxiety disorder to an increase of 12% of the base prevalence estimate for agoraphobia. Conclusions: In surveys that use the Composite International Diagnostic Interview version 2.1 the substance-induced and general medical condition exclusion criteria have a minor impact on determining final case definition in the majority of mental disorders.