Objective: Although childhood psychiatric diagnoses often rely on maternal reports, little is known about their long-term reliability and diagnostic accuracy. Thus, the authors sought to examine these psychometric features in a cohort of ADHD and control children.
Method: The sample consisted of 140 referred children with ADHD and 120 normal controls. The authors compared childhood diagnoses based on maternal reports of their children's psychopathology at this baseline assessment with those collected 1 year later.
Results: Both reliability and accuracy were excellent for ADHD. Reliability and specificity were also excellent for conduct disorder, oppositional defiant disorder, major depression, bipolar disorder, separation anxiety, and multiple anxiety disorders. Reliability and sensitivity were relatively low for simple phobia, social phobia, agoraphobia, and overanxious disorder.
Conclusions: With some exceptions, maternal reports of their children's psychopathology provided a reliable and accurate means of assessment. Generally, maternally derived diagnoses were less accurate for internalizing compared with externalizing disorders. However, specificity was high for all diagnoses, suggesting that mothers were not biased to report symptoms that had not occurred.