Abstract The National Survey on Drug Use and Health (NSDUH) and the Arrestee Drug Abuse Monitoring (ADAM) Program provide information on alcohol and drug use by individuals who have recently been arrested. The studies differ in their target populations (civilian, noninstitutionalized individuals vs. arrestees in 39 sites recently booked into jails) and data collection methods. This study uses 2003 ADAM and 2002–2008 NSDUH data for adult males living in the 39 ADAM sites who reported a past year arrest and 2002–2008 Uniform Crime Reporting (UCR) data to examine how well NSDUH covers the arrestee population and to compare estimates of drug and alcohol use and substance abuse or dependence. In general, ADAM estimates of rates of self-reported drug use were higher. The magnitude of these differences cannot be accounted for by undercoverage in NSDUH. Other possible reasons for these differences and their implications for interpreting NSDUH and ADAM data are discussed.