Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Hughes, John


Many, if not most, drug abuse counselors and treatment programs recommend abstinence from all psychoactive substances, in part, because of a fear that clients who decrease or stop their use of one drug will substitute another. Research to confirm this notion of substitution, however, mostly fails to show that abstinence from one drug increases use of another. A within-subjects study investigated whether consumption of alcohol and other substances changed during marijuana abstinence. Using an ABA design, 28 individuals who met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSMIV; American Psychiatric Association [APA], 2000) criteria for either cannabis dependence or abuse and were not trying to stop their marijuana use completed an 8-day baseline period in which they used marijuana and other drugs as usual, then a 13-day marijuana abstinence period, and finally a 7-day return-to-baseline period. Marijuana abstinence was induced by a previously-validated contingent compensation schedule. Participants called a voicemail system daily to provide self-report of marijuana and alcohol use and visited the laboratory twice per week to provide self-report of caffeine, cigarette, and other illicit drug use, to complete self-report measures on psychological symptoms such as withdrawal and craving, and to submit urine samples to biochemically verify marijuana abstinence. Alcohol use significantly increased from a mean of 2.6 drinks/day (SD=1.0) during the baseline period to 3.0 drinks/day (SD=1.0) during the marijuana abstinence period (p=0.03), a 15% increase. Alcohol use then significantly decreased to 2.5 drinks/day (SD=1.3) during the return-to-baseline period (p=0.03), a 17% decrease. Although alcohol substitution occurred during marijuana abstinence, substitution of cigarettes, caffeine, and non-marijuana illicit drugs did not occur. Individuals with a diagnosis of past alcohol abuse or dependence substituted alcohol to a greater degree (52% increase) than those without this past history (3% increase). Increases in alcohol drinks/day correlated with increases in marijuana withdrawal discomfort scores and with increases in alcohol craving scores from the baseline to the marijuana abstinence period. Problems related to alcohol did not significantly increase from baseline to marijuana abstinence. This study provides empirical validation of the clinical notion of drug substitution and suggests that clinicians’ concerns about drug substitution may be valid, but this study’s results need to be replicated in individuals who seek treatment for marijuana problems. Whether substitution reduces the ability to abstain from marijuana also needs to be tested. If alcohol substitution does occur and interferes with the ability to quit marijuana, this would be important empirical support for the clinical practice of recommending abstinence from all substances.