Two hundred nine cases of intentional drug overdose were reviewed to determine the importance of discordance between drugs identified by toxicology analysis and those suspected clinically. The laboratory agreed exactly with the clinical impression in 47% of cases. Clinically unsuspected drugs were identified by the laboratory in 27% of cases; the characteristics of these patients did not differ significantly from those of other patients. Unexpected toxicology findings led to changes in therapy in only three cases, and none of these changes appeared to have a major impact on outcome. Although routine comprehensive toxicology screening frequently may identify unsuspected drugs, it is likely that a policy of more selective use of the laboratory would not compromise the care of patients with drug overdose.