Relationship between extent of cocaine use and dependence among adolescents and adults in the United States.

The relationships between intensity of cocaine use, route of administration and past year dependence were investigated in a nationally representative sample of past year cocaine users aged 12 and over (N = 2,349) from three aggregated surveys (1991-1993) of the National Household Survey on Drug Abuse (NHSDA). An approximate measure of DSM-IV dependence criteria was developed from self-reported symptoms of dependence and drug-related problems. The model and the data provided no basis for rejecting the hypothesis that the logit of the probability of dependence increased linearly with the logarithm of the frequency of cocaine use in the past year and with quantity of cocaine use in the past 30 days. The associations between frequency and quantity of cocaine use and cocaine dependence varied significantly by an age-by-gender interaction and race/ethnicity. African-American users, who had higher rates of cocaine dependence than whites, used cocaine more frequently, used it by more addictive routes (smoking or injection), and were more likely to use crack. Adolescent females, who reported higher rate of cocaine dependence than males, used cocaine more frequently and reported more symptoms at low doses of cocaine use. Multivariate logistic regressions indicated that frequency and quantity of use, as well as route of administration, retained unique associations with cocaine dependence. Implications of the findings for the epidemiological study of cocaine use and dependence are discussed.