If ODD is a precursor to CD its onset should occur earlier in development. Studies in childhood samples tend to support this pattern (Loeber, Green, Lahey, Christ, & Frick, 1992). The National Comorbidity Survey Replication (NCSR), assessing childhood disorders from retrospective adult reports, also found that ODD onset pre-dated onset of a range of other disorders including CD (Nock, Kazdin, Hiripi, & Kessler, 2007). Age-trends in prevalence suggest that rates of CD diagnoses are low in childhood, but rise steeply from late childhood/the early teens, whereas rates of ODD are relatively stable from early childhood to adolescence (Maughan, Rowe, Messer, Goodman, & Meltzer, 2004). In addition, comorbidity between CD and ODD in childhood is undoubtedly strong. In clinical samples, rates of ODD in cases of CD have been reported to be as high as 96% (Frick et al., 1992). Overlaps in general population samples are lower: approximately 60% of CD cases typically meet full criteria for ODD (Maughan et al., 2004; Rowe, Maughan, Pickles, Costello, & Angold, 2002) and others often show elevated levels of sub-threshold ODD symptoms (Rowe et al., 2002).
