This paper reviews issues related to the measurement of drug use and other behaviors in studies of human immunodeficiency virus (HIV) infection in intravenous drug users. These issues include: (1) choice of variables, (2) study design, (3) time frame of measurement, (4) categories of measurement, (5) interviewer effects, and (6) validity of measurement. Difficulties and approaches for measuring drug use and other intimate behaviors in intravenous drug users are discussed. Attempts to come to terms with these measurement issues in the context of HIV infection in intravenous drug users should lead to the further development of methods for use in the general context of drug abuse research.