The objective of this study was to determine (1) the type and extent of self-help efforts among patients presenting for treatment of substance use disorders, and (2) the association of self-help with demographicand clinical characteristics. A retrospective report of life self-help methods, current demographic characteristics, and current and lifetime clinical characteristics was used. Six hundred and forty-two patients in treatment for substance use disorder were interviewed at one of two university medical centers witb Alcohol-Drug Programs located within departments of psychiatry. A research associate (RA) interviewed patients regarding seven types of self-help involving specific, mutually exclusive behaviors and rated the patient's lifetime self-help methods. The patient, RA, and addiction psychiatrists provided demographic, familial, and clinical information. Most patients (78%) had tried one or more types of self-help, with a mean of 2.7 methods per patient. They more frequently chose methods related to the substance (decreasing amounts or frequency, or changing substance type) or joining a self-help group than methods that involved changing friends, residence, or occupation/job/school. Certain patterns of self-help tended to occur together, (eg, changing substance frequency and dose), whereas others appeared more independent (eg, joining a self-help group). Some self-help approaches occurred mostly in association with other methods rather than alone (eg, changing occupation/job/school). More self-help was associated with higher socioeconomic class, more relatives with substance use disorder, greater severity of substance use disorder, and more treatment for substance use disorder. Self-help tends to occur more often after exposure to addicted relatives or addiction treatment. Clinicians and public adult education should promulgate self-help methods in the general population.