The formulation of the relationship between ODD and CD in DSM-IV was largely based on findings from clinical samples. Findings from community studies such as GSMS show a somewhat different pattern. Although there are strong overlaps between CD and ODD and developmental continuity at least in boys, these links are much weaker than those found in clinically referred samples. At least in late childhood and adolescence, many cases of CD will emerge without marked histories of ODD, and many cases of ODD will not progress to CD. As we have seen, however, this in no sense implies that cases meeting full criteria for ODD represent benign or transient disorders. Indeed, as our and other findings are increasingly underlining, ODD may be ‘in a class by itself’ (Copeland et al., 2009) in indexing risk for both emotional and behavioral pychiatric disorders later in development. Further studies are needed to clarify the processes underlying these links. At this stage, however, our findings point clearly to the utility of treating CD and ODD as fully separate disorders in DSM-V. As noted by Burke
