- Email: [email protected]
The impact of impulsivity on cocaine use and retention in treatment
Journal of Substance Abuse Treatment 21 (2001) 193 – 198
Regular article
The impact of impulsivity on cocaine use and retention in treatment F. Gerard Moeller, M.D.a,*, Donald M. Dougherty, Ph.D.a, Ernest S. Barratt, Ph.D.b, Joy M. Schmitz, Ph.D.a, Alan C. Swann, M.D.a, John Grabowski, Ph.D.a a
Department of Psychiatry and Behavioral Sciences, University of Texas-Houston Health Science Center, Houston, TX, USA b Department of Psychiatry and Behavioral Sciences, University of Texas Medical Branch, Galveston, TX, USA Received 21 April 2000; received in revised form 14 August 2001; accepted 24 August 2001
Abstract To determine whether impulsivity was related to severity of drug use and treatment outcome, 50 cocaine dependent subjects underwent baseline measures of severity of current cocaine use and the Barratt Impulsiveness Scale (BIS-11). The hypothesis of the study was that there would be a significant correlation between impulsivity and cocaine use severity. As predicted, there was a significant correlation between BIS-11 total scores and self-reported average daily cocaine use as well as cocaine withdrawal symptoms. A subset of 35 patients underwent a 12-week double-blind placebo controlled trial of buspirone and group therapy. Subjects with high baseline impulsivity remained in the study a significantly shorter period than did subjects with lower baseline impulsivity. This study shows that impulsivity is a significant predictor of cocaine use and treatment retention, and suggests the need for targeting impulsivity in cocaine dependence treatment. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Cocaine; Impulsivity; Treatment; Substance abuse
1. Introduction The treatment of substance abuse has traditionally focused on the compulsive aspects of the disorder in which craving plays a central role (Mathew, Claghorn, & Largen, 1979). Recently, there has been an increased interest in the role of impulsivity in substance abuse. In this model of substance abuse, at least some drug use does not occur in response to chronic craving, but in a rapid, unplanned fashion. As reviewed in Brady, Myrick, and McElroy (1998), there is evidence for an association between impulsivity and substance abuse. Impulsive populations including individuals with intermittent explosive disorder (McElroy, Soutullo, Beckman, Taylor, & Keck, 1998), impulsive arsonists (Virkkunen, De Jong, Bartko, & Linnoila, 1989), and impulsively violent offenders (Linnoila, 1983) all have higher rates of substance abuse or dependence than the general population. Studies that have examined children at risk for substance abuse have also found higher rates of impulsivity. Using a * Corresponding author. 1300 Moursund, Houston, TX 77030, USA. Tel.: +1-713-500-2858; fax: +1-713-500-2634. E-mail address: [email protected] (F.G. Moeller).
model of behavioral self-regulation involving inattention, impulsivity/hyperactivity, and aggression, Dawes, Tarter, and Kirisci (1997) compared children with a family history of substance abuse to those without this history, and found that children with a family history of substance abuse scored higher on the behavioral self-regulation measures (which equated to higher impulsivity). Self-regulation of goal directed behavior was more impaired in another study comparing high-risk children to low risk children, and this measure predicted impulsive aggression in the high-risk group (Giancola, Moss, Martin, Kirisci, & Tarter, 1996). Most studies objectively measuring impulsivity find higher impulsivity scale scores in substance dependent individuals. Using the Barratt Impulsiveness Scale (BIS-11), two studies found that substance dependent individuals have higher total scores than controls (Allen, Moeller, Rhoades, & Cherek, 1998, Patton, Stanford, & Barratt, 1995). However, one study of alcoholic subjects with impulse control disorders did not find increased total BIS scores compared to controls (Lejoyeux, Feuche, Loi, Solomon, & Ades, 1998). In a study using the Eysenck Personality Inventory, drug-abusing patients were less sociable and more impulsive than controls (King, Jones, Scheuer, Curtis,
0740-5472/01/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 7 4 0 - 5 4 7 2 ( 0 1 ) 0 0 2 0 2 - 1
194
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
& Zarcone, 1990). Another study using the Psychopathic States Inventory showed higher impulsivity in antisocial substance dependent subjects than controls (Moss, Yao, & Panzak, 1990). A few studies have attempted to determine the impact of impulsivity on severity or mechanisms of substance abuse. Impulsivity as measured by the BIS-11 was positively correlated with the history of frequency of alcohol consumption in incarcerated individuals in one study (Fishbein & Reuland, 1994). In a second study, impulsivity as measured by the Impulsivity-Venturesomeness-Empathy scale was significantly correlated with self reported euphoria after intravenous cocaine administration (Cascella et al., 1994), implying that more impulsive cocainedependent individuals may be biologically at a greater risk for relapse. Thus, there is evidence of an association between impulsivity and substance abuse from a number of studies examining the issue from several perspectives. In order to determine the relationship between impulsivity and severity of cocaine dependence, a group of 50 cocaine dependent subjects completed the Barratt Impulsiveness Scale (BIS11), along with measures of quantity of cocaine use, craving and withdrawal. The hypothesis was that there would be a positive correlation between severity of cocaine use and impulsivity. In order to determine the impact of impulsivity on treatment for cocaine dependence, a subset of 41 subjects entered a 12-week treatment study in which baseline impulsivity was used as a predictor of subject dropout. It was hypothesized that subjects with higher baseline impulsivity would drop out of treatment significantly sooner than would subjects with lower impulsivity.
2. Materials and methods 2.1. Subjects Ninety-eight treatment-seeking cocaine-dependent subjects were screened for inclusion into the study after obtaining informed consent. Subjects were recruited from advertisements for research treatment of cocaine dependence, and from community referrals. Subjects were screened using the Structured Clinical Interview for DSM-IV (SCID) and (SCID-II) (First, Spitzer, Gibbon, & Williams, 1996), a physical examination, and routine blood work. Inclusion criteria included presence of current cocaine dependence by DSM-IV criteria, lack of current or past DSM-IV Axis I diagnosis other than substance dependence or substance induced mood disorder, and willingness to complete questionnaires. This study was approved by the Committee for the Protection of Human Subjects of the University of Texas Houston Health Science Center. Fifty cocaine dependent subjects, 39 male and 11 female, met the inclusion criteria for the baseline comparison study.
Of those, 1 had 6 years or less of formal education, 14 were high school dropouts, 24 had graduated from high school or received their GED, 10 had some college, and 1 was a four year college graduate. All subjects completed a drug use questionnaire, a cocaine craving scale based on Halikas, Kuhn, Crosby, Carlson, and Crea (1991) and a 13-item cocaine withdrawal scale based on the symptoms of cocaine withdrawal listed in DSM-III-R (Moeller et al., 1997). This scale is a self-report questionnaire with levels ranging from 0 (not at all) to 5 (severely) on the following items: Depressed; Irritable; Anxious; Tired; Sleeping too much; Not sleeping enough; Restless or fidgety; Trembling or twitching; Sweating; Nausea or vomiting; Diarrhea or stomachache; Appetite increased; and Unpleasant dreams. All subjects submitted a urine drug screen for determination of benzoylecgonine levels. Semiquantitative measurement of benzoylecgonine was accomplished using a modified EMIT which uses a 5 point standard curve (Behring Diagnostics Inc., 1996). Urine creatinine was also measured and used to compensate for differences in urine dilution (Wilkins, 1997). An Alcosensor III was used to screen for alcohol consumption. The measure of impulsivity used in this study was the BIS-11 (Patton et al., 1995). This is a 30-item questionnaire that has been validated in several different populations, including substancedependent individuals (Patton et al., 1995, Allen et al., 1998), incarcerated individuals with antisocial personality disorder (Barratt, Stanford, Felthous, & Kent, 1997), and patients with Bipolar Disorder (Swann, Anderson, Dougherty, & Moeller, 2001). The BIS-11 includes three subscales labeled attentional, non-planning, and motor, based on a principal component analysis (Patton et al., 1995). These three subscales have been replicated in an Italian (Fossati, Di Ceglie, Acquarini, & Barratt, 2001) and a Japanese (Someya et al., 2001) version of the BIS. Examples of items used in the BIS-11 include: ‘‘I act on impulse’’ and ‘‘I plan tasks carefully.’’ A subset of 41 subjects agreed to participate in a 12-week double-blind placebo-controlled trial of buspirone for the treatment of cocaine dependence. Subjects who were randomized to buspirone were initially started on a dose of 5 mg twice daily, with a titration to 15 mg three times daily over the course of 2 weeks. All subjects were seen twice weekly for behavioral testing and underwent urine drug screens at each visit. Female subjects underwent urine pregnancy testing at each visit. Subjects were given take home doses of buspirone twice weekly with enough medication to last until their next visit. Subjects also underwent weekly group therapy sessions focusing on relapse prevention. All subjects received monetary compensation for attendance and completion of behavioral testing (approximately $20 per visit). Compensation was not contingent on providing a drug-free urine. As part of a separate study examining the effects of buspirone on aggression, subjects
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
also underwent behavioral laboratory testing twice weekly. This testing consisted of two sessions of the Point Subtraction Aggression Paradigm (PSAP) (Cherek, 1981). Of the 41 subjects who entered the study, 6 subjects were terminated from the study prematurely. One subject was unable to return due to being admitted to an inpatient treatment program as a condition of his parole. One subject was hospitalized for a nonpsychiatric medical illness for several weeks. Four subjects were dropped from the study because they were unable to complete the behavioral laboratory testing. The 35 remaining subjects were divided into two groups based on baseline BIS-11 scores. Subjects with a baseline BIS-11 score above the score 74 were designated as high impulsivity (HIBIS) and subjects with a baseline BIS-11 score equal to or below 74 were designated as low impulsivity (LOWBIS). This score was the median for a larger sample of cocaine-dependent subjects who we have screened in our laboratory using the BIS-11 to date (n = 79), including subjects who did not participate in this study due to not meeting inclusion criteria or lack of willingness to participate. A larger sample of scores was used to determine the cut off since it was more representative of cocaine-dependent individuals in general. Twelve of the 35 subjects were included in the HIBIS group, and 23 of the subjects were included in the LOWBIS group. Cocaine craving was measured weekly using a visual analog scale based on Halikas et al. (1991). Anxiety was also measured weekly using the Hamilton Anxiety Rating Scale (Hamilton, 1959). Drug and alcohol usage was monitored by obtaining urine and expired air samples on each morning of participation in the research. Testing was performed using the Syva EMIT systems. Once a week, the quantity of cocaine used was estimated by semiquantitative determination of benzoylecgonine in urine samples.
195
and presence of ASPD were included as covariates due to differences in these factors between HIBIS and LOWBIS groups. The effect of treatment assignment on cocaine craving, and anxiety was determined using an end point analysis with a repeated measures ANCOVA. Baseline measures and length of time in the study were used as covariates. Differences between the HIBIS and LOWBIS groups on time until dropout were analyzed using the Kaplan-Meier survival statistic. The number of negative urine drug screens was compared between treatment groups (Buspirone vs. Placebo) and impulsivity groups (HIBIS vs. LOWBIS) using an ANCOVA with length of time in treatment as a covariate.
3. Results 3.1. Baseline measures The average age of the subjects in the baseline study was 39.3 ± 6.6 years of age. Subjects reported spending an average of $61 per day on cocaine, with a range of $5-$400 per day. The range and distribution of the BIS-11 scores is shown in Table 1. As can be seen from Table 2, impulsivity was positively correlated with several cocaine use measures. There was a significant positive correlation between BIS-11 total scores and log transformed reported average quantity of cocaine use. There was also a significant positive correlation between BIS-11 total scores and severity of withdrawal symptoms. Although there was a positive correlation between BIS-11 scores and log transformed urinary benzoylecgonine levels as well as between BIS-11 scores and cocaine craving, these correlations did not achieve statistical significance.
2.2. Statistical analysis Spearman correlations were used to test the relationship between BIS-11 scores, quantity of cocaine used (in dollars per day), creatinine corrected urinary benzoylecgonine levels, strength of cocaine craving, and strength of cocaine withdrawal symptoms. Due to non-normal distribution of the data, self reported quantity of cocaine used and benzoylecgonine levels were log transformed prior to statistical analyses. Due to multiple correlations, probability for statistical significance was set to p < 0.01. Baseline sociodemographic differences between HIBIS and LOWBIS groups were determined using a two-tailed t-test. The correlation between impulsivity scores and length of time spent in the study was determined using a Spearman correlation coefficient. The effect of impulsivity (HIBIS vs. LOWBIS) on the number of weeks in the study was determined using an analysis of covariance (ANCOVA) with treatment group (Buspirone vs. Placebo), age, and presence of Antisocial Personality Disorder (ASPD) included as covariates. Age
Table 1 Distribution of BIS-11 scores n
Range
Mean
Median
Total BIS Score Baseline Group HIBIS LOWBIS
50 12 23
37 – 112 76 – 112 37 – 74
69.7 ± 12.4 84.7 ± 11.1 63.6 ± 8.5
72.5 80.5 64.0
Motor Baseline Group HIBIS LOWBIS
50 12 23
12 – 36 22 – 37 12 – 25
21.5 ± 5.2 28.6 ± 5.0 19.0 ± 3.3
22.0 27.5 19.0
Nonplanning Baseline Group HIBIS LOWBIS
50 12 23
17 – 47 33 – 47 17 – 38
30.6 ± 5.1 36.0 ± 3.9 29.0 ± 4.7
31.0 35.0 30.0
Attentional Baseline Group HIBIS LOWBIS
50 12 23
8 – 30 14 – 30 8 – 21
17.6 ± 4.2 20.1 ± 4.4 15.7 ± 3.5
18.0 18.5 16.0
196
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
Table 2 Correlations between baseline BIS-11 scores and cocaine use variables
Total Nonplanning Attentional Motor
Cocaine usea
Benzoylecgonineb
Craving
Withdrawal
0.373* 0.235 0.086 0.441*
0.211 0.084 0.115 0.237
0.288 0.086 0.103 0.377*
0.575* 0.394* 0.490* 0.478*
a
Log of the average dollar amount spent per day. Log of the Creatinine corrected urinary level. * p < .01.
b
For BIS-11 subscales, there was a significant positive correlation between withdrawal symptoms and the nonplanning, as well as attentional subscale. However, the subscale that was most associated with severity of cocaine use was the motor subscale. The motor impulsivity subscale was significantly correlated with log transformed reported average quantity of cocaine use, strength of craving, and withdrawal scale scores (See Table 2.) 3.2. Treatment measures Subjects in the HIBIS group were slightly but significantly older (t = -2.1, df = 33, P = 0.043), and included significantly more subjects with ASPD (Pearson Chi Square = 4.7, df = 1.0, P = 0.03). There was no significant difference between the groups in education (Mann-Whitney U Statistic = 104.5, P = 0.217), gender (Pearson Chi Square = 0.005, df = 1.0, P = 0.944), or in baseline cocaine craving (t = 1.015, df = 33, P = 0.317). As can be seen in Table 3, impulsivity was predictive of subject dropout.
Table 3 Demographic and behavioral characteristics of subjects in treatment phase Variable n
HIBIS 12
LOWBIS 23
Placebo 18
Buspirone 17
Demographics # Women Agea Educationc
3 36.2 ± 7.3 2.6 ± 0.9
6 41.2 ± 6.3 3.0 ± 1.0
4 39.0 ± 7.3 2.8 ± 1.2
5 40.1 ± 6.8 3.0 ± 0.7
Diagnostic # ASPD BIS Scorea,b
7 84.7 ± 11.1
5 63.7 ± 8.5
7 76.4 ± 14.1
5 65.0 ± 10.9
Outcome % Completeda Weeks in Studya Baseline Craving Endpoint Craving Percent Coc. Neg.
25.0 5.2 ± 4.6 6.0 ± 3.3 3.7 ± 3.0 23.4 ± 34.8
69.6 9.6 ± 3.9 5.0 ± 2.2 3.1 ± 2.0 11.4 ± 16.3
38.9 6.6 ± 4.7 5.1 ± 2.0 4.1 ± 2.4 14.0 ± 28.5
70.6 9.8 ± 4.0 5.7 ± 3.2 2.4 ± 2.1 17.2 ± 20.0
a Denotes significant difference between HIBIS and LOWBIS groups ( p < 0.05). b Denotes significant difference between placebo and buspirone groups ( p < 0.05). c Based on SCID categories (1 = grad 6 or less, 2 = grade 7-12, 3 = H.S. grad, 4 = part college, 5 = 2 yr. college grad, 6 = 4 yr. college grad, 7 = part graduate school, 8 = advanced degree).
Since HIBIS and LOWBIS groups differed in age and number of subjects with ASPD, these variables along with treatment medication (buspirone or placebo) were included as covariates in an ANCOVA comparing HIBIS and LOWBIS subjects. With age, presence of ASPD, and treatment medication as covariates, there was a significant effect of group on length of time in treatment ( F = 5.07, df = 1,30, P = 0.032). There was also a significant inverse correlation between BIS scores and length of time in treatment (Spearman r = -0.342, P = 0.044) Results of the Kaplan-Meier survival analysis showed that high impulsivity was a significant predictor of subject dropout (Log Rank Statistic = 8.44, df = 1, P = 0.004) (See Fig. 1). The effect of buspirone on subject retention, urine drug screens, and cocaine craving is also shown in Table 3. With baseline BIS scores included as a covariate, the effect of buspirone on length of time in treatment was not significant ( F = 1.96, df = 1,32, P = 0.171). With regards to cocaine craving, there was no main effect of treatment on craving between subjects ( F = 1.00, df = 1,31, P = 0.323). Within subjects, there was no effect of length of time in treatment ( F = 1.66, df = 12,372, P = 0.104), and there was no time by treatment interaction ( F = 1.42, df = 12,372, P = 0.185). Although the total number of negative urine drug screens was greater for subjects treated with buspirone than those treated with placebo (see Table 3), with number of weeks of treatment as a covariate in the analysis, there was no significant difference between the two groups ( F = 1.48, df = 1,32, P = 0.233). In order to determine if there was any difference between treatment groups in the quantity of cocaine used, a separate analysis was done using the creatinine corrected semiquantitative cocaine metabolites. In this analysis, there was no main effect of treatment ( F = 0.249, df = 1,31, P = 0.621), nor was there an effect of the number of weeks of treatment ( F = 0.550, df = 12,372, P = 0.685), or a time by treatment interaction ( F = 1.47, df = 12,372, P = 0.219).
Fig. 1. Retention by High vs. Low impulsivity group.
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
4. Discussion As predicted, impulsivity as measured by the BIS-11 was significantly correlated with severity of cocaine use as measured by self-reported quantity of cocaine use, and with severity of cocaine withdrawal symptoms. The association between the BIS-11 and cocaine severity was strongest for the motor impulsivity subscale. This subscale was significantly correlated with self-reported cocaine use, craving, and withdrawal symptoms. In the treatment phase of the study, subjects with higher baseline impulsivity as measured by the BIS-11 were more likely to drop out of treatment and remained in treatment significantly fewer weeks. There was also an effect of buspirone on length of time in treatment, however, this effect appeared to have been mediated by impulsivity differences between groups. Although there was no difference in age, gender, or presence of ASPD between subjects treated with buspirone and placebo, there was a significant difference in BIS-11 scores. When BIS-11 scores were included as covariates in the statistical analysis, the effect of buspirone on length of time in treatment was no longer significant. This study lends support to the importance of impulsivity in cocaine use. While individuals with cocaine dependence may use cocaine as a result of craving or withdrawal symptoms, it appears that reacting to internal or external stimuli without regards to the consequences is also important in cocaine use. This supports the hypothesis of Jentsch and Taylor (1999) that impaired inhibition is important in drug seeking behavior, along with the augmenting incentive motivational aspects of drug use. One question that could not be addressed by this study is whether cocaine use directly increases impulsivity possibly through impairment in frontal lobe function as Jentsch and Taylor also hypothesized. It is also possible the association is due to impulsivity leading to cocaine use without cocaine use directly increasing impulsivity. Although other studies have examined variables that may be related to impulsivity with regards to their effect on treatment outcome, to our knowledge this is the first published study to use an impulsivity scale to determine the effect of impulsivity on treatment outcome. The finding that impulsivity predicts a poor outcome in treatment of cocaine dependence supports the importance of targeting impulsivity in cocaine dependence treatment. One of the most consistently effective treatments currently available for cocaine dependence involves increasing, through the use of monetary or other types of incentives, short-term rewards for abstinence from cocaine use (Grabowski, Higgins, & Kirby, 1993; Higgins et al., 1991; Higgins & Budney, 1993). By increasing short-term rewards for abstinence, contingency management therapy targets the lack of planning for the future (i.e., impulsivity) that is common in cocaine dependent subjects. The main shortcoming of this study is the relatively small number of subjects in the two treatment groups. The small
197
number of subjects limits the conclusions that can be drawn from negative results, due to the possibility of a type II error. It is possible buspirone could have had an effect on its own, or could have interacted with impulsivity to improve treatment retention given a larger sample size. Likewise, a larger sample size could have resulted in an effect of impulsivity or buspirone on urine toxicology, given the fact there were nonsignificant differences in urine toxicology between groups. It should also be pointed out that the subjects included in the statistical analysis were free of DSM-IV Axis I disorders other than substance abuse/dependence, which limits the generalizability of these findings. With these limitations in mind, this study does show that impulsivity has an impact on cocaine use and retention in treatment. Treatments for cocaine dependence should target reduction of impulsive as well as compulsive aspects of cocaine use.
Acknowledgments This study was supported by National Institutes on Drug Abuse grants #DA-R0108425, and DA-K02000403.
References Allen, T. J., Moeller, F. G., Rhoades, H. M., & Cherek, D. R. (1998). Impulsivity and history of drug dependence. Drug and Alcohol Dependence, 50, 137 – 145. Barratt, E. S., Stanford, M. S., Felthous, A. R., & Kent, T. A. (1997). The effects of phenytoin on impulsive and premeditated aggression: a controlled study. Journal of Clinical Psychopharmacology, 17, 341 – 349. Behring Diagnostics Inc. (1996). EMIT II Cocaine Metabolite Assay Package Insert. Cupertino, CA: Behring Diagnostics Inc. Brady, K. T., Myrick, H., & McElroy, S. (1998). The relationship between substance use disorders, impulse control disorders, and pathological aggression. American Journal of Addictions, 7, 221 – 230. Cascella, N. G., Nagoshi, C. T., Muntaner, C., Walter, D., Haertzen, C. A., & Kumor, K. M. (1994). Impulsiveness and subjective effects of intravenous cocaine administration in the laboratory. Journal of Substance Abuse, 6, 355 – 366. Cherek, D. R. (1981). Effects of smoking different doses of nicotine on human aggressive behavior. Psychopharmacology (Berl.), 75, 339 – 345. Dawes, M. A., Tarter, R. E., & Kirisci, L. (1997). Behavioral self-regulation: correlates and 2 year follow-ups for boys at risk for substance abuse. Drug and Alcohol Dependence, 45, 165 – 176. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. (1996). Structured clinical interview for DSM-IV Axis I disorders patient edition. New York: Biometrics Research Department, New York State Psychiatric Institute. Fishbein, D. H., & Reuland, M. (1994). Psychological correlates of frequency and type of drug use among jail inmates. Addictive Behaviors, 19, 583 – 598. Fossati, A., Di Ceglie, A., Acquarini, E., & Barratt, E. S. (2001). Psychometric properties of an Italian version of the Barratt Impulsiveness Scale-11 (BIS-11) in nonclinical subjects. Journal of Clinical Psychology, 57, 815 – 828. Giancola, P. R., Moss, H. B., Martin, C. S., Kirisci, L., & Tarter, R. E. (1996). Executive cognitive functioning predicts reactive aggression in
198
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
boys at high risk for substance abuse: a prospective study. Alcoholism. Clininical and Experimental Research, 20, 740 – 744. Grabowski, J., Higgins, S. T., & Kirby, K. C. (1993). Behavioral treatments of cocaine dependence. NIDA Research Monograph, 133 – 149. Halikas, J. A., Kuhn, K. L., Crosby, R., Carlson, G., & Crea, F. (1991). The measurement of craving in cocaine patients using the Minnesota Cocaine Craving Scale. Comprehensive Psychiatry, 32, 22 – 27. Hamilton, M. (1959). The assessment of anxiety states by rating. British Journal of Medical Psychology, 32, 50 – 55. Higgins, S. T., & Budney, A. J. (1993). Treatment of cocaine dependence through the principles of behavior analysis and behavioral pharmacology. NIDA Research Monograph, 97 – 121. Higgins, S. T., Delaney, D. D., Budney, A. J., Bickel, W. K., Hughes, J. R., Foerg, F., & Fenwick, J. W. (1991). A behavioral approach to achieving initial cocaine abstinence. American Journal of Psychiatry, 9, 1218 – 1224. Jentsch, J. D., & Taylor, J. R. (1999). Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli. Psychopharmacology, 146, 373 – 390. King, R. J., Jones, J., Scheuer, J. W., Curtis, D., & Zarcone, V. P. (1990). Plasma cortisol correlates of impulsivity and substance abuse. Personality and Individual Differences, 11, 287 – 291. Lejoyeux, M., Feuche, N., Loi, S., Solomon, J., & Ades, J. (1998). Impulsecontrol disorders in alcoholics are related to sensation seeking and not to impulsivity. Psychiatry Research, 81, 149 – 155. Linnoila, M. (1983). Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Science, 33, 2609 – 2614.
Mathew, R. J., Claghorn, J. L., & Largen, J. (1979). Craving for alcohol in sober alcoholics. American Journal of Psychiatry, 136, 603 – 606. McElroy, S. L., Soutullo, C. A., Beckman, D. A., Taylor, P. J., & Keck, P. E. J. (1998). DSM-IV intermittent explosive disorder: a report of 27 cases. Journal of Clinical Psychiatry, 59, 203 – 210. Moeller, F. G., Dougherty, D. M., Rustin, T., Swann, A. C., Allen, T. J., Shah, N., & Cherek, D. R. (1997). Antisocial personality disorder and aggression in recently abstinent cocaine dependent subjects. Drug and Alcohol Dependence, 44, 175 – 182. Moss, H. B., Yao, J. K., & Panzak, G. L. (1990). Serotonergic responsivity and behavioral dimensions in antisocial personality disorder with substance abuse. Biological Psychiatry, 28, 325 – 338. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768 – 774. Someya, T., Sakado, K., Seki, T., Kojima, M., Reist, C., Tang, S. W., & Takahashi, S. (2001). The Japanese version of the Barratt Impulsiveness Scale, 11th version (BIS-11): its reliability and validity. Psychiatry and Clinical Neurosciences, 55, 111 – 114. Swann, A. C., Anderson, J., Dougherty, D. M., & Moeller, F. G. (2001). Measurement of interepisode impulsivity in bipolar disorder. Psychiatry Research, 101, 195 – 197. Virkkunen, M., De Jong, J., Bartko, J., & Linnoila, M. (1989). Psychobiological concomitants of history of suicide attempts among violent offenders and impulsive fire setters. Archives of General Psychiatry, 46, 604 – 606. Wilkins, J. N. (1997). Quantitative urine levels of cocaine and other substances of abuse. NIDA Research Monograph, 175, 235 – 252.
Regular article
The impact of impulsivity on cocaine use and retention in treatment F. Gerard Moeller, M.D.a,*, Donald M. Dougherty, Ph.D.a, Ernest S. Barratt, Ph.D.b, Joy M. Schmitz, Ph.D.a, Alan C. Swann, M.D.a, John Grabowski, Ph.D.a a
Department of Psychiatry and Behavioral Sciences, University of Texas-Houston Health Science Center, Houston, TX, USA b Department of Psychiatry and Behavioral Sciences, University of Texas Medical Branch, Galveston, TX, USA Received 21 April 2000; received in revised form 14 August 2001; accepted 24 August 2001
Abstract To determine whether impulsivity was related to severity of drug use and treatment outcome, 50 cocaine dependent subjects underwent baseline measures of severity of current cocaine use and the Barratt Impulsiveness Scale (BIS-11). The hypothesis of the study was that there would be a significant correlation between impulsivity and cocaine use severity. As predicted, there was a significant correlation between BIS-11 total scores and self-reported average daily cocaine use as well as cocaine withdrawal symptoms. A subset of 35 patients underwent a 12-week double-blind placebo controlled trial of buspirone and group therapy. Subjects with high baseline impulsivity remained in the study a significantly shorter period than did subjects with lower baseline impulsivity. This study shows that impulsivity is a significant predictor of cocaine use and treatment retention, and suggests the need for targeting impulsivity in cocaine dependence treatment. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Cocaine; Impulsivity; Treatment; Substance abuse
1. Introduction The treatment of substance abuse has traditionally focused on the compulsive aspects of the disorder in which craving plays a central role (Mathew, Claghorn, & Largen, 1979). Recently, there has been an increased interest in the role of impulsivity in substance abuse. In this model of substance abuse, at least some drug use does not occur in response to chronic craving, but in a rapid, unplanned fashion. As reviewed in Brady, Myrick, and McElroy (1998), there is evidence for an association between impulsivity and substance abuse. Impulsive populations including individuals with intermittent explosive disorder (McElroy, Soutullo, Beckman, Taylor, & Keck, 1998), impulsive arsonists (Virkkunen, De Jong, Bartko, & Linnoila, 1989), and impulsively violent offenders (Linnoila, 1983) all have higher rates of substance abuse or dependence than the general population. Studies that have examined children at risk for substance abuse have also found higher rates of impulsivity. Using a * Corresponding author. 1300 Moursund, Houston, TX 77030, USA. Tel.: +1-713-500-2858; fax: +1-713-500-2634. E-mail address: [email protected] (F.G. Moeller).
model of behavioral self-regulation involving inattention, impulsivity/hyperactivity, and aggression, Dawes, Tarter, and Kirisci (1997) compared children with a family history of substance abuse to those without this history, and found that children with a family history of substance abuse scored higher on the behavioral self-regulation measures (which equated to higher impulsivity). Self-regulation of goal directed behavior was more impaired in another study comparing high-risk children to low risk children, and this measure predicted impulsive aggression in the high-risk group (Giancola, Moss, Martin, Kirisci, & Tarter, 1996). Most studies objectively measuring impulsivity find higher impulsivity scale scores in substance dependent individuals. Using the Barratt Impulsiveness Scale (BIS-11), two studies found that substance dependent individuals have higher total scores than controls (Allen, Moeller, Rhoades, & Cherek, 1998, Patton, Stanford, & Barratt, 1995). However, one study of alcoholic subjects with impulse control disorders did not find increased total BIS scores compared to controls (Lejoyeux, Feuche, Loi, Solomon, & Ades, 1998). In a study using the Eysenck Personality Inventory, drug-abusing patients were less sociable and more impulsive than controls (King, Jones, Scheuer, Curtis,
0740-5472/01/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 7 4 0 - 5 4 7 2 ( 0 1 ) 0 0 2 0 2 - 1
194
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
& Zarcone, 1990). Another study using the Psychopathic States Inventory showed higher impulsivity in antisocial substance dependent subjects than controls (Moss, Yao, & Panzak, 1990). A few studies have attempted to determine the impact of impulsivity on severity or mechanisms of substance abuse. Impulsivity as measured by the BIS-11 was positively correlated with the history of frequency of alcohol consumption in incarcerated individuals in one study (Fishbein & Reuland, 1994). In a second study, impulsivity as measured by the Impulsivity-Venturesomeness-Empathy scale was significantly correlated with self reported euphoria after intravenous cocaine administration (Cascella et al., 1994), implying that more impulsive cocainedependent individuals may be biologically at a greater risk for relapse. Thus, there is evidence of an association between impulsivity and substance abuse from a number of studies examining the issue from several perspectives. In order to determine the relationship between impulsivity and severity of cocaine dependence, a group of 50 cocaine dependent subjects completed the Barratt Impulsiveness Scale (BIS11), along with measures of quantity of cocaine use, craving and withdrawal. The hypothesis was that there would be a positive correlation between severity of cocaine use and impulsivity. In order to determine the impact of impulsivity on treatment for cocaine dependence, a subset of 41 subjects entered a 12-week treatment study in which baseline impulsivity was used as a predictor of subject dropout. It was hypothesized that subjects with higher baseline impulsivity would drop out of treatment significantly sooner than would subjects with lower impulsivity.
2. Materials and methods 2.1. Subjects Ninety-eight treatment-seeking cocaine-dependent subjects were screened for inclusion into the study after obtaining informed consent. Subjects were recruited from advertisements for research treatment of cocaine dependence, and from community referrals. Subjects were screened using the Structured Clinical Interview for DSM-IV (SCID) and (SCID-II) (First, Spitzer, Gibbon, & Williams, 1996), a physical examination, and routine blood work. Inclusion criteria included presence of current cocaine dependence by DSM-IV criteria, lack of current or past DSM-IV Axis I diagnosis other than substance dependence or substance induced mood disorder, and willingness to complete questionnaires. This study was approved by the Committee for the Protection of Human Subjects of the University of Texas Houston Health Science Center. Fifty cocaine dependent subjects, 39 male and 11 female, met the inclusion criteria for the baseline comparison study.
Of those, 1 had 6 years or less of formal education, 14 were high school dropouts, 24 had graduated from high school or received their GED, 10 had some college, and 1 was a four year college graduate. All subjects completed a drug use questionnaire, a cocaine craving scale based on Halikas, Kuhn, Crosby, Carlson, and Crea (1991) and a 13-item cocaine withdrawal scale based on the symptoms of cocaine withdrawal listed in DSM-III-R (Moeller et al., 1997). This scale is a self-report questionnaire with levels ranging from 0 (not at all) to 5 (severely) on the following items: Depressed; Irritable; Anxious; Tired; Sleeping too much; Not sleeping enough; Restless or fidgety; Trembling or twitching; Sweating; Nausea or vomiting; Diarrhea or stomachache; Appetite increased; and Unpleasant dreams. All subjects submitted a urine drug screen for determination of benzoylecgonine levels. Semiquantitative measurement of benzoylecgonine was accomplished using a modified EMIT which uses a 5 point standard curve (Behring Diagnostics Inc., 1996). Urine creatinine was also measured and used to compensate for differences in urine dilution (Wilkins, 1997). An Alcosensor III was used to screen for alcohol consumption. The measure of impulsivity used in this study was the BIS-11 (Patton et al., 1995). This is a 30-item questionnaire that has been validated in several different populations, including substancedependent individuals (Patton et al., 1995, Allen et al., 1998), incarcerated individuals with antisocial personality disorder (Barratt, Stanford, Felthous, & Kent, 1997), and patients with Bipolar Disorder (Swann, Anderson, Dougherty, & Moeller, 2001). The BIS-11 includes three subscales labeled attentional, non-planning, and motor, based on a principal component analysis (Patton et al., 1995). These three subscales have been replicated in an Italian (Fossati, Di Ceglie, Acquarini, & Barratt, 2001) and a Japanese (Someya et al., 2001) version of the BIS. Examples of items used in the BIS-11 include: ‘‘I act on impulse’’ and ‘‘I plan tasks carefully.’’ A subset of 41 subjects agreed to participate in a 12-week double-blind placebo-controlled trial of buspirone for the treatment of cocaine dependence. Subjects who were randomized to buspirone were initially started on a dose of 5 mg twice daily, with a titration to 15 mg three times daily over the course of 2 weeks. All subjects were seen twice weekly for behavioral testing and underwent urine drug screens at each visit. Female subjects underwent urine pregnancy testing at each visit. Subjects were given take home doses of buspirone twice weekly with enough medication to last until their next visit. Subjects also underwent weekly group therapy sessions focusing on relapse prevention. All subjects received monetary compensation for attendance and completion of behavioral testing (approximately $20 per visit). Compensation was not contingent on providing a drug-free urine. As part of a separate study examining the effects of buspirone on aggression, subjects
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
also underwent behavioral laboratory testing twice weekly. This testing consisted of two sessions of the Point Subtraction Aggression Paradigm (PSAP) (Cherek, 1981). Of the 41 subjects who entered the study, 6 subjects were terminated from the study prematurely. One subject was unable to return due to being admitted to an inpatient treatment program as a condition of his parole. One subject was hospitalized for a nonpsychiatric medical illness for several weeks. Four subjects were dropped from the study because they were unable to complete the behavioral laboratory testing. The 35 remaining subjects were divided into two groups based on baseline BIS-11 scores. Subjects with a baseline BIS-11 score above the score 74 were designated as high impulsivity (HIBIS) and subjects with a baseline BIS-11 score equal to or below 74 were designated as low impulsivity (LOWBIS). This score was the median for a larger sample of cocaine-dependent subjects who we have screened in our laboratory using the BIS-11 to date (n = 79), including subjects who did not participate in this study due to not meeting inclusion criteria or lack of willingness to participate. A larger sample of scores was used to determine the cut off since it was more representative of cocaine-dependent individuals in general. Twelve of the 35 subjects were included in the HIBIS group, and 23 of the subjects were included in the LOWBIS group. Cocaine craving was measured weekly using a visual analog scale based on Halikas et al. (1991). Anxiety was also measured weekly using the Hamilton Anxiety Rating Scale (Hamilton, 1959). Drug and alcohol usage was monitored by obtaining urine and expired air samples on each morning of participation in the research. Testing was performed using the Syva EMIT systems. Once a week, the quantity of cocaine used was estimated by semiquantitative determination of benzoylecgonine in urine samples.
195
and presence of ASPD were included as covariates due to differences in these factors between HIBIS and LOWBIS groups. The effect of treatment assignment on cocaine craving, and anxiety was determined using an end point analysis with a repeated measures ANCOVA. Baseline measures and length of time in the study were used as covariates. Differences between the HIBIS and LOWBIS groups on time until dropout were analyzed using the Kaplan-Meier survival statistic. The number of negative urine drug screens was compared between treatment groups (Buspirone vs. Placebo) and impulsivity groups (HIBIS vs. LOWBIS) using an ANCOVA with length of time in treatment as a covariate.
3. Results 3.1. Baseline measures The average age of the subjects in the baseline study was 39.3 ± 6.6 years of age. Subjects reported spending an average of $61 per day on cocaine, with a range of $5-$400 per day. The range and distribution of the BIS-11 scores is shown in Table 1. As can be seen from Table 2, impulsivity was positively correlated with several cocaine use measures. There was a significant positive correlation between BIS-11 total scores and log transformed reported average quantity of cocaine use. There was also a significant positive correlation between BIS-11 total scores and severity of withdrawal symptoms. Although there was a positive correlation between BIS-11 scores and log transformed urinary benzoylecgonine levels as well as between BIS-11 scores and cocaine craving, these correlations did not achieve statistical significance.
2.2. Statistical analysis Spearman correlations were used to test the relationship between BIS-11 scores, quantity of cocaine used (in dollars per day), creatinine corrected urinary benzoylecgonine levels, strength of cocaine craving, and strength of cocaine withdrawal symptoms. Due to non-normal distribution of the data, self reported quantity of cocaine used and benzoylecgonine levels were log transformed prior to statistical analyses. Due to multiple correlations, probability for statistical significance was set to p < 0.01. Baseline sociodemographic differences between HIBIS and LOWBIS groups were determined using a two-tailed t-test. The correlation between impulsivity scores and length of time spent in the study was determined using a Spearman correlation coefficient. The effect of impulsivity (HIBIS vs. LOWBIS) on the number of weeks in the study was determined using an analysis of covariance (ANCOVA) with treatment group (Buspirone vs. Placebo), age, and presence of Antisocial Personality Disorder (ASPD) included as covariates. Age
Table 1 Distribution of BIS-11 scores n
Range
Mean
Median
Total BIS Score Baseline Group HIBIS LOWBIS
50 12 23
37 – 112 76 – 112 37 – 74
69.7 ± 12.4 84.7 ± 11.1 63.6 ± 8.5
72.5 80.5 64.0
Motor Baseline Group HIBIS LOWBIS
50 12 23
12 – 36 22 – 37 12 – 25
21.5 ± 5.2 28.6 ± 5.0 19.0 ± 3.3
22.0 27.5 19.0
Nonplanning Baseline Group HIBIS LOWBIS
50 12 23
17 – 47 33 – 47 17 – 38
30.6 ± 5.1 36.0 ± 3.9 29.0 ± 4.7
31.0 35.0 30.0
Attentional Baseline Group HIBIS LOWBIS
50 12 23
8 – 30 14 – 30 8 – 21
17.6 ± 4.2 20.1 ± 4.4 15.7 ± 3.5
18.0 18.5 16.0
196
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
Table 2 Correlations between baseline BIS-11 scores and cocaine use variables
Total Nonplanning Attentional Motor
Cocaine usea
Benzoylecgonineb
Craving
Withdrawal
0.373* 0.235 0.086 0.441*
0.211 0.084 0.115 0.237
0.288 0.086 0.103 0.377*
0.575* 0.394* 0.490* 0.478*
a
Log of the average dollar amount spent per day. Log of the Creatinine corrected urinary level. * p < .01.
b
For BIS-11 subscales, there was a significant positive correlation between withdrawal symptoms and the nonplanning, as well as attentional subscale. However, the subscale that was most associated with severity of cocaine use was the motor subscale. The motor impulsivity subscale was significantly correlated with log transformed reported average quantity of cocaine use, strength of craving, and withdrawal scale scores (See Table 2.) 3.2. Treatment measures Subjects in the HIBIS group were slightly but significantly older (t = -2.1, df = 33, P = 0.043), and included significantly more subjects with ASPD (Pearson Chi Square = 4.7, df = 1.0, P = 0.03). There was no significant difference between the groups in education (Mann-Whitney U Statistic = 104.5, P = 0.217), gender (Pearson Chi Square = 0.005, df = 1.0, P = 0.944), or in baseline cocaine craving (t = 1.015, df = 33, P = 0.317). As can be seen in Table 3, impulsivity was predictive of subject dropout.
Table 3 Demographic and behavioral characteristics of subjects in treatment phase Variable n
HIBIS 12
LOWBIS 23
Placebo 18
Buspirone 17
Demographics # Women Agea Educationc
3 36.2 ± 7.3 2.6 ± 0.9
6 41.2 ± 6.3 3.0 ± 1.0
4 39.0 ± 7.3 2.8 ± 1.2
5 40.1 ± 6.8 3.0 ± 0.7
Diagnostic # ASPD BIS Scorea,b
7 84.7 ± 11.1
5 63.7 ± 8.5
7 76.4 ± 14.1
5 65.0 ± 10.9
Outcome % Completeda Weeks in Studya Baseline Craving Endpoint Craving Percent Coc. Neg.
25.0 5.2 ± 4.6 6.0 ± 3.3 3.7 ± 3.0 23.4 ± 34.8
69.6 9.6 ± 3.9 5.0 ± 2.2 3.1 ± 2.0 11.4 ± 16.3
38.9 6.6 ± 4.7 5.1 ± 2.0 4.1 ± 2.4 14.0 ± 28.5
70.6 9.8 ± 4.0 5.7 ± 3.2 2.4 ± 2.1 17.2 ± 20.0
a Denotes significant difference between HIBIS and LOWBIS groups ( p < 0.05). b Denotes significant difference between placebo and buspirone groups ( p < 0.05). c Based on SCID categories (1 = grad 6 or less, 2 = grade 7-12, 3 = H.S. grad, 4 = part college, 5 = 2 yr. college grad, 6 = 4 yr. college grad, 7 = part graduate school, 8 = advanced degree).
Since HIBIS and LOWBIS groups differed in age and number of subjects with ASPD, these variables along with treatment medication (buspirone or placebo) were included as covariates in an ANCOVA comparing HIBIS and LOWBIS subjects. With age, presence of ASPD, and treatment medication as covariates, there was a significant effect of group on length of time in treatment ( F = 5.07, df = 1,30, P = 0.032). There was also a significant inverse correlation between BIS scores and length of time in treatment (Spearman r = -0.342, P = 0.044) Results of the Kaplan-Meier survival analysis showed that high impulsivity was a significant predictor of subject dropout (Log Rank Statistic = 8.44, df = 1, P = 0.004) (See Fig. 1). The effect of buspirone on subject retention, urine drug screens, and cocaine craving is also shown in Table 3. With baseline BIS scores included as a covariate, the effect of buspirone on length of time in treatment was not significant ( F = 1.96, df = 1,32, P = 0.171). With regards to cocaine craving, there was no main effect of treatment on craving between subjects ( F = 1.00, df = 1,31, P = 0.323). Within subjects, there was no effect of length of time in treatment ( F = 1.66, df = 12,372, P = 0.104), and there was no time by treatment interaction ( F = 1.42, df = 12,372, P = 0.185). Although the total number of negative urine drug screens was greater for subjects treated with buspirone than those treated with placebo (see Table 3), with number of weeks of treatment as a covariate in the analysis, there was no significant difference between the two groups ( F = 1.48, df = 1,32, P = 0.233). In order to determine if there was any difference between treatment groups in the quantity of cocaine used, a separate analysis was done using the creatinine corrected semiquantitative cocaine metabolites. In this analysis, there was no main effect of treatment ( F = 0.249, df = 1,31, P = 0.621), nor was there an effect of the number of weeks of treatment ( F = 0.550, df = 12,372, P = 0.685), or a time by treatment interaction ( F = 1.47, df = 12,372, P = 0.219).
Fig. 1. Retention by High vs. Low impulsivity group.
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
4. Discussion As predicted, impulsivity as measured by the BIS-11 was significantly correlated with severity of cocaine use as measured by self-reported quantity of cocaine use, and with severity of cocaine withdrawal symptoms. The association between the BIS-11 and cocaine severity was strongest for the motor impulsivity subscale. This subscale was significantly correlated with self-reported cocaine use, craving, and withdrawal symptoms. In the treatment phase of the study, subjects with higher baseline impulsivity as measured by the BIS-11 were more likely to drop out of treatment and remained in treatment significantly fewer weeks. There was also an effect of buspirone on length of time in treatment, however, this effect appeared to have been mediated by impulsivity differences between groups. Although there was no difference in age, gender, or presence of ASPD between subjects treated with buspirone and placebo, there was a significant difference in BIS-11 scores. When BIS-11 scores were included as covariates in the statistical analysis, the effect of buspirone on length of time in treatment was no longer significant. This study lends support to the importance of impulsivity in cocaine use. While individuals with cocaine dependence may use cocaine as a result of craving or withdrawal symptoms, it appears that reacting to internal or external stimuli without regards to the consequences is also important in cocaine use. This supports the hypothesis of Jentsch and Taylor (1999) that impaired inhibition is important in drug seeking behavior, along with the augmenting incentive motivational aspects of drug use. One question that could not be addressed by this study is whether cocaine use directly increases impulsivity possibly through impairment in frontal lobe function as Jentsch and Taylor also hypothesized. It is also possible the association is due to impulsivity leading to cocaine use without cocaine use directly increasing impulsivity. Although other studies have examined variables that may be related to impulsivity with regards to their effect on treatment outcome, to our knowledge this is the first published study to use an impulsivity scale to determine the effect of impulsivity on treatment outcome. The finding that impulsivity predicts a poor outcome in treatment of cocaine dependence supports the importance of targeting impulsivity in cocaine dependence treatment. One of the most consistently effective treatments currently available for cocaine dependence involves increasing, through the use of monetary or other types of incentives, short-term rewards for abstinence from cocaine use (Grabowski, Higgins, & Kirby, 1993; Higgins et al., 1991; Higgins & Budney, 1993). By increasing short-term rewards for abstinence, contingency management therapy targets the lack of planning for the future (i.e., impulsivity) that is common in cocaine dependent subjects. The main shortcoming of this study is the relatively small number of subjects in the two treatment groups. The small
197
number of subjects limits the conclusions that can be drawn from negative results, due to the possibility of a type II error. It is possible buspirone could have had an effect on its own, or could have interacted with impulsivity to improve treatment retention given a larger sample size. Likewise, a larger sample size could have resulted in an effect of impulsivity or buspirone on urine toxicology, given the fact there were nonsignificant differences in urine toxicology between groups. It should also be pointed out that the subjects included in the statistical analysis were free of DSM-IV Axis I disorders other than substance abuse/dependence, which limits the generalizability of these findings. With these limitations in mind, this study does show that impulsivity has an impact on cocaine use and retention in treatment. Treatments for cocaine dependence should target reduction of impulsive as well as compulsive aspects of cocaine use.
Acknowledgments This study was supported by National Institutes on Drug Abuse grants #DA-R0108425, and DA-K02000403.
References Allen, T. J., Moeller, F. G., Rhoades, H. M., & Cherek, D. R. (1998). Impulsivity and history of drug dependence. Drug and Alcohol Dependence, 50, 137 – 145. Barratt, E. S., Stanford, M. S., Felthous, A. R., & Kent, T. A. (1997). The effects of phenytoin on impulsive and premeditated aggression: a controlled study. Journal of Clinical Psychopharmacology, 17, 341 – 349. Behring Diagnostics Inc. (1996). EMIT II Cocaine Metabolite Assay Package Insert. Cupertino, CA: Behring Diagnostics Inc. Brady, K. T., Myrick, H., & McElroy, S. (1998). The relationship between substance use disorders, impulse control disorders, and pathological aggression. American Journal of Addictions, 7, 221 – 230. Cascella, N. G., Nagoshi, C. T., Muntaner, C., Walter, D., Haertzen, C. A., & Kumor, K. M. (1994). Impulsiveness and subjective effects of intravenous cocaine administration in the laboratory. Journal of Substance Abuse, 6, 355 – 366. Cherek, D. R. (1981). Effects of smoking different doses of nicotine on human aggressive behavior. Psychopharmacology (Berl.), 75, 339 – 345. Dawes, M. A., Tarter, R. E., & Kirisci, L. (1997). Behavioral self-regulation: correlates and 2 year follow-ups for boys at risk for substance abuse. Drug and Alcohol Dependence, 45, 165 – 176. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. (1996). Structured clinical interview for DSM-IV Axis I disorders patient edition. New York: Biometrics Research Department, New York State Psychiatric Institute. Fishbein, D. H., & Reuland, M. (1994). Psychological correlates of frequency and type of drug use among jail inmates. Addictive Behaviors, 19, 583 – 598. Fossati, A., Di Ceglie, A., Acquarini, E., & Barratt, E. S. (2001). Psychometric properties of an Italian version of the Barratt Impulsiveness Scale-11 (BIS-11) in nonclinical subjects. Journal of Clinical Psychology, 57, 815 – 828. Giancola, P. R., Moss, H. B., Martin, C. S., Kirisci, L., & Tarter, R. E. (1996). Executive cognitive functioning predicts reactive aggression in
198
F.G. Moeller et al. / Journal of Substance Abuse Treatment 21 (2001) 193–198
boys at high risk for substance abuse: a prospective study. Alcoholism. Clininical and Experimental Research, 20, 740 – 744. Grabowski, J., Higgins, S. T., & Kirby, K. C. (1993). Behavioral treatments of cocaine dependence. NIDA Research Monograph, 133 – 149. Halikas, J. A., Kuhn, K. L., Crosby, R., Carlson, G., & Crea, F. (1991). The measurement of craving in cocaine patients using the Minnesota Cocaine Craving Scale. Comprehensive Psychiatry, 32, 22 – 27. Hamilton, M. (1959). The assessment of anxiety states by rating. British Journal of Medical Psychology, 32, 50 – 55. Higgins, S. T., & Budney, A. J. (1993). Treatment of cocaine dependence through the principles of behavior analysis and behavioral pharmacology. NIDA Research Monograph, 97 – 121. Higgins, S. T., Delaney, D. D., Budney, A. J., Bickel, W. K., Hughes, J. R., Foerg, F., & Fenwick, J. W. (1991). A behavioral approach to achieving initial cocaine abstinence. American Journal of Psychiatry, 9, 1218 – 1224. Jentsch, J. D., & Taylor, J. R. (1999). Impulsivity resulting from frontostriatal dysfunction in drug abuse: implications for the control of behavior by reward-related stimuli. Psychopharmacology, 146, 373 – 390. King, R. J., Jones, J., Scheuer, J. W., Curtis, D., & Zarcone, V. P. (1990). Plasma cortisol correlates of impulsivity and substance abuse. Personality and Individual Differences, 11, 287 – 291. Lejoyeux, M., Feuche, N., Loi, S., Solomon, J., & Ades, J. (1998). Impulsecontrol disorders in alcoholics are related to sensation seeking and not to impulsivity. Psychiatry Research, 81, 149 – 155. Linnoila, M. (1983). Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Science, 33, 2609 – 2614.
Mathew, R. J., Claghorn, J. L., & Largen, J. (1979). Craving for alcohol in sober alcoholics. American Journal of Psychiatry, 136, 603 – 606. McElroy, S. L., Soutullo, C. A., Beckman, D. A., Taylor, P. J., & Keck, P. E. J. (1998). DSM-IV intermittent explosive disorder: a report of 27 cases. Journal of Clinical Psychiatry, 59, 203 – 210. Moeller, F. G., Dougherty, D. M., Rustin, T., Swann, A. C., Allen, T. J., Shah, N., & Cherek, D. R. (1997). Antisocial personality disorder and aggression in recently abstinent cocaine dependent subjects. Drug and Alcohol Dependence, 44, 175 – 182. Moss, H. B., Yao, J. K., & Panzak, G. L. (1990). Serotonergic responsivity and behavioral dimensions in antisocial personality disorder with substance abuse. Biological Psychiatry, 28, 325 – 338. Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768 – 774. Someya, T., Sakado, K., Seki, T., Kojima, M., Reist, C., Tang, S. W., & Takahashi, S. (2001). The Japanese version of the Barratt Impulsiveness Scale, 11th version (BIS-11): its reliability and validity. Psychiatry and Clinical Neurosciences, 55, 111 – 114. Swann, A. C., Anderson, J., Dougherty, D. M., & Moeller, F. G. (2001). Measurement of interepisode impulsivity in bipolar disorder. Psychiatry Research, 101, 195 – 197. Virkkunen, M., De Jong, J., Bartko, J., & Linnoila, M. (1989). Psychobiological concomitants of history of suicide attempts among violent offenders and impulsive fire setters. Archives of General Psychiatry, 46, 604 – 606. Wilkins, J. N. (1997). Quantitative urine levels of cocaine and other substances of abuse. NIDA Research Monograph, 175, 235 – 252.