Robert F. Leeman PhD
Assistant Professor of Psychiatry
Alcohol; Alcohol-related problems; Smoking; Disinhibition; Impulsivity; Impaired control over alcohol use; Young adults; Human laboratory research; Treatment; Clinical trials;
- Developing and validating a human laboratory paradigm to model impaired control over alcohol use in heavy drinking young adults and utilizing this paradigm to test the preliminary efficacy of novel interventions.
- Testing the efficacy of a very brief (10 minutes or less), web-based alcohol reduction intervention for undergraduates
- Analyses using self-report, survey data to predict high-risk drinking in undergraduates from measures of trait disinhibition, expectancies of alcohol's disinhibiting effects and impaired control over alcohol use.
- Project Choice, a double-blind, placebo-controlled clinical trial of naltrexone plus BASICS counseling for heavy drinking reduction in young adults.
My main research interests concern young adult alcohol use and smoking. Regarding young adult alcohol use, I am particularly interested in associations between alcohol use and problems with self-control. An example of this is impaired control over alcohol use, which is the difficulty that some drinkers have with adhering to limits they have placed on their alcohol consumption. My colleagues and I have developed a novel human laboratory paradigm to model impaired control (Leeman et al., in press). We have also adapted a very brief (10 minutes or less), web-based alcohol reduction intervention for undergraduates that has been utilized successfully in Australia for use in the United States. I also collaborated on a clinical trial to test the efficacy of naltrexone and individual counseling for young adults (Project Choice). My research interests in smoking concern interactions between alcohol and smoking and relationships between weight concerns/food deprivation and smoking.
Extensive Research Description
Overview of Research Interests
Addictive behaviors are an important public health problem, associated with morbidity, mortality and high financial cost to society. My research concerns psychosocial predictors of risk and mechanisms underlying addictive behaviors, principally high-risk alcohol use in young adults and smoking. The psychosocial risk factors and mechanisms of greatest interest to me concern disinhibition, both at trait and state levels. The decision to abuse substances is inherently impulsive in that it represents a choice of immediate reinforcers (e.g., euphoria from substance use) over larger, delayed reinforcers (e.g., avoiding adverse consequences). Therefore, the strength of association between addictive behaviors and aspects of disinhibition is not surprising.
State and Trait Disinhibition
“Disinhibition” has been defined as the loss of restraint over some form of behaviour” (Bond, 1998, p. 42). Trait disinhibition can be manifested as any number of traits including types of impulsivity, sensation seeking and a propensity toward risk taking. These traits have been associated with alcohol use and increased risk of problem drinking in young adults (e.g., Anderson, Smith,& Fischer, 2003; Grekin & Sher, 2006). We also know that alcohol and other addictive substances can induce states of disinhibition, the behavioral output of which resembles that of trait disinhibition (Bond 1998; Lyvers, 2000). We know less, however, about alcohol-induced disinhibition as a predictor of problem drinking. We also know little about whether there are discernable types of alcohol-induced disinhibition. These issues are addressed in a recent review article (Leeman, Grant, & Potenza, 2009).
Survey Research Concerning Disinhibition Expectancies and Impaired Control
To address these gaps in the literature, as part of my dissertation research, I developed the Drinking-Induced Disinhibition Scale (DIDS; Leeman, Toll, & Volpicelli, 2007). The DIDS is a reliable, valid measure of expectancies of three types of alcohol’s disinhibiting effects (i.e., euphoric/social, dysphoric and sexual). In other words, this scale measures drinkers’ beliefs that they will experience certain effects to a greater extent when they drink than when not drinking. Since the initial work on the measure, we have completed a prospective study in which the DIDS, along with other variables assessed during freshman year, were included in models to predict problem drinking reported during senior year by a cohort of undergraduates (Leeman, Toll, Taylor, & Volpicelli, 2009). The euphoric/social subscale—which measures expectancies that one will be more fun and social when drinking than when not—assessed during freshman year predicted unique variance in heavy episodic drinking reported during senior year. The DIDS has been translated into Danish (Hesse & Tutenges, 2008). The convergent validity and the factor structure of the measure were confirmed in the Dutch study. At present, we are working on a new subscale to assess aggressive disinhibition.
Impairment in self-control can also manifest itself as difficulties in limiting one’s substance use. Impaired control over substance use (i.e., difficulties in limiting substance use, especially once initial use has begun) has long been viewed as a hallmark of addiction (Levine, 1978; O’Brien et al., 2006). My colleagues and I established that impaired control predicted unique variance in heavy episodic drinking and alcohol-related problems (i.e., negative consequences of alcohol use and aspects of alcohol dependence) in cross-sectional models using the freshman year wave of the aforementioned prospective study (Leeman, Fenton, & Volpicelli, 2007). Subsequently, we found that impaired control assessed during freshman year predicted unique variance in alcohol-related problems during senior year (Leeman, Toll, Taylor, & Volpicelli, 2009).
Human Laboratory Models of Addictive Behaviors
Another area of my research is in human laboratory models of aspects of addictive behaviors. Specifically, I am interested in laboratory models that further my research interests in links between disinhibition and addictive behaviors. My colleagues and I completed a study in which we used a human laboratory model of smoking lapse behavior (McKee, 2009) to examine the combined effect of nicotine and food deprivation, compared to nicotine deprivation alone, on ability to resist smoking and on subsequent ad-libitum (i.e., open access) smoking behavior (Leeman, O’Malley, White, & McKee, 2010). We found that when daily smokers were deprived of food in addition to nicotine, they had more difficulty resisting smoking their first cigarette than smokers who were nicotine deprived only. Recent findings have suggested, like substance use, deprivation can also induce disinhibition. Mitchell (2004) found that nicotine-dependent individuals were more likely to favor immediate cigarettes over delayed money when nicotine deprived than when not deprived. The laboratory model used in our study could be viewed as a model of deprivation-induced disinhibition.
I am in the process of developing a human laboratory model of impaired control over alcohol use in young adults (Leeman et al., in press). Once validated, this model could be used for a number of purposes, including as a means of testing the preliminary efficacy of treatments to reduce heavy drinking in young adults (e.g., counseling approaches). Given the prevalence (Wechsler, Lee, Nelson, & Kuo, 2002) and deleterious results (Perkins, 2002) of heavy drinking among young adults, new, efficacious harm reduction approaches are needed for this population. A time and cost-effective method to evaluate these treatment approaches would be valuable to the field.
Interventions to Reduce Heavy Drinking in Young Adults
I collaborated on a randomized, double-blind, placebo-controlled trial to test the combined efficacy of the FDA-approved, opiate antagonist naltrexone and BASICS (Brief Alcohol Screening and Intervention for College Students; Dimeff et al., 1999) to reduce heavy drinking in 18-25 year olds. We conducted an open-label pilot study of this approach, the findings of which suggested that this line of research is feasible and potentially efficacious (Leeman, Palmer, Corbin, Romano, Meandzija, & O'Malley, 2008). This five-year NIAAA-funded study began in September 2007. The study contains a comprehensive battery of psychosocial predictors collected both intermittently throughout the study and on a daily basis during the screening and treatment periods. Inclusion of these assessments will permit analyses to identify mechanisms underlying the effects of this treatment approach.
Currently, we are in the first year of a randomized, clinical trial to test a very brief (10 minutes or less), web-based alcohol reduction intervention for college students. This intervention, called THRIVE (Kypri et al., 2009) has been utilized successfully in Australia. We have adapted the intervention for use in the United States (US-THRIVE). In addition to attempting to replicate the Kypri et al. findings, we are conducting novel tests of focused sets of protective behavioral strategies included as part of this intevention. This intervention has great dissemination potential, given its accessibility, very brief duration and evidence for its efficacy.