Trend analyses of the U.S. monitoring data systems (the National Household Survey on Drug Abuse and the Monitoring the Future Study) and of the country's surveillance program, the Community Epidemiology Work Group (CEWG), indicate that several new "drug abuse" patterns have emerged over the past several years. For adolescents, drug use rates are converging for females and males, the mean age at which youngsters initiate drug use has declined, and more young adolescents are reporting using drugs. Furthermore, emergent new drug use patterns are being observed by the CEWG. The use of drugs such as Rohypnol, the injecting of crack-cocaine, and the spread of methamphetamines by new traffickers challenge our existing knowledge and understanding of drug use and its prevention. The National Institute on Drug Abuse has funded several large longitudinal studies that follow selected children and adolescents into their twenties, and some into their thirties. This research has been a rich source of information on the determinants of initiating and continuing drug abuse. Yet the findings from the surveys have not been well explored by the longitudinal studies, nor have the findings from the longitudinal studies been used in the surveys to better understand the observed changing trends in drug use patterns. This paper addresses six issues that have been observed from the findings from analyses of data from the surveys or macroepidemiologic studies. Information from the sub-population or micro-epidemiologic studies are reviewed for possible hypotheses to explain each issue. Suggestions for further research and implications for prevention also are presented.