The current work provides a quantitative-descriptive characterization of patterns of lifetime comorbidity among common mental disorders exhibited by individuals in two large-scale epidemiological samples. Within each sample, LCA revealed five classes with contrasting disorder profiles that were highly similar across the two samples. The classes included a “few-disorders” class with low probabilities of diagnoses of all disorders, and four classes showing elevated probabilities of particular lifetime disorders. While sample size and replication across two different nationally-representative datasets collected a decade apart from each other represent significant strengths of the current study, weaknesses include the fact that diagnoses were determined by interviews administered by laypersons, lifetime diagnoses were based on retrospective recall, and that each sample consisted only of noninstitutionalized subjects. Additionally, certain disorders (PTSD in both the NCS and NCS-R; conduct disorder and substance use disorders in the NCS-R) were assessed on a follow-up basis in selected subsets of participants who met criteria for other diagnoses in an initial assessment. Thus, it is possible that such disorders showed evidence of greater comorbidity because they were selectively associated with classes exhibiting some other diagnosable form of psychopathology.
