Highly disparate research techniques have been employed to develop estimates of the extent of opiate use. Exclusive reliance upon public health and law enforcement statistics has been supplemented in recent years with data generated by survey research‘incidence and prevalence’epidemiological methods. State agencies attempting to measure the extent of opiate use as a requirement for the receipt of federal categorical grant formula funds utilize multiple indicators: opiate treatment, arrest, incarceration, serum hepatitis, narcotics overdose death, and survey data. To address the question of undetected opiate use in a major southwestern city, the authors have undertaken a study measuring the extent of law enforcement recognition of the addictive status of a random sample of patients selected from the city's two municipally operated methadone treatment programs. The results indicate that over one-half of the sample was unknown to the police as addicts. Descriptive attributes of the unknown group were identified through appropriate statistical analyses. These findings are comparable to those of a similar study conducted in 1973 in a major eastern city.‘Official drug statistics are like an iceberg—or, more precisely, an aerial view of that iceberg when the shadow of the information-gathering mechanism distorts what little reality a photograph might reveal’[1].