DSM-IV's strong empirical base has yielded an instrument with good to excel
lent reliability and improved validity. Diagnostic reliability depends on b
oth the clarity and validity of diagnostic criteria and the changeability o
f disorders over time: The reliability of schizophrenic spectrum disorders,
personality disorders, and some childhood and adolescent disorders remains
problematic. Findings on diagnostic validity appear paradoxical: Attempts
to validate schizophrenic spectrum disorders with neurobiological and genet
ic-familial validators have been only modestly successful, whereas the trip
artite personality trait model has differentiated a range of depressive and
anxiety disorders. Research on comorbidity has identified several highly c
omorbid disorders (substance-related disorders, personality disorders, depr
ession, and anxiety) as well as some adverse consequences of comorbidity. T
he advantages of dimensional approaches to diagnosis have largely been demo
nstrated conceptually; ultimate conclusions about the strengths and weaknes
ses of dimensional and syndromal methods await substantial additional empir
ical research.