Journal of Clinical and Experimental Neuropsychology, 2015 http://dx.doi.org/10.1080/13803395.2015.1018145
Increased impulsivity in pathological gambling: Considering nicotine dependence Anja Kräplin, Silke Behrendt, Stefan Scherbaum, Maja Dshemuchadse, Gerhard Bühringer, and Thomas Goschke Department of Psychology, Technische Universitaet Dresden, Dresden, Germany (Received 27 June 2014; accepted 8 February 2015) Introduction: It has been highlighted that increased impulsivity is an important etiological factor in pathological gambling (PG). However, the role of the highly prevalent comorbid mental disorders in PG remains unclear. This is of special concern as the highest comorbidity was found between PG and nicotine dependence (ND), which, in turn, has also been associated with heightened impulsivity. This study aimed to find out whether increased impulsivity in PG is a specific characteristic related to comorbid mental disorders in general or especially to ND. Method: A cross-sectional study was designed that included a healthy control group and three disorder groups with comorbid PG excluding ND, comorbid PG including ND, and ND alone according to the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition. Impulsivity was assessed according to the dimensions response and choice impulsivity applying behavioral and self-report measures. Results: We found (a) increased response impulsivity in the comorbid PG group when compared to the control group. Furthermore, increased choice impulsivity was found (b) in comorbid PG excluding ND and (c) in comorbid PG including ND when compared to ND alone. Other group differences did not reach significance. Conclusions: Our findings highlight that comorbid PG shares deficits in inhibitory control with ND. In contrast, maladaptive choices are a specific disorder characteristic of comorbid PG when compared to ND. If replicated in further studies, intervention strategies strengthening cognitive control skills might be effective for comorbid PG as well as ND whereas strategies enhancing maladaptive valuation of rewards might be specifically effective in comorbid PG. Keywords: Pathological gambling; Impulsivity; Comorbidity; Nicotine dependence; Decision making.
High impulsivity and high prevalence rates of comorbid mental disorders like nicotine dependence (ND) are well investigated phenomena in pathological gambling (PG; e.g., Kessler et al., 2008; van Holst, van den Brink, Veltman, & Goudriaan, 2010). This study aimed to extend knowledge on impulsivity in PG by considering comorbid mental disorders with a special focus on ND and applying a multidimensional assessment of impulsivity.
According to recent reviews and models on addictive behaviors, an imbalance of valuation and cognitive control brain networks may be associated with PG (Bickel, Jarmolowicz, Mueller, Gatchalian, & McClure, 2012; Bühringer, Wittchen, Gottlebe, Kufeld, & Goschke, 2008; Goldstein & Volkow, 2011; Goschke, 2014; Redish, Jensen, & Johnson, 2008; van Holst et al., 2010). A central indicator of this brain
The authors wish to thank René Mayer for programming the administered tasks and for his helpful comments on the methods. Anja Kräplin, Silke Behrendt, Stefan Scherbaum, Maja Dshemuchadse, and Thomas Goschke declare to have no conflict of interest. Gerhard Bühringer has received unrestricted research grants from the Bavarian State Ministry of Finance (regulatory authority for and operator of the state gambling monopoly) via the Bavarian State Ministry of the Environment and Public Health, from the German Federal Ministry of Economics and Technology (regulatory authority for the commercial gaming industry), and from the commercial gaming industry. No potential conflict of interest was reported by the author(s). This study was funded by the German Research Foundation [grant number GO-720/8-1], [grant number SFB 940/1 2013]. Address correspondence to: Anja Kräplin, Addiction Research Unit, Institute of Clinical Psychology and Psychotherapy, Technische Universitaet Dresden, Chemnitzer Strasse 46, D-01187 Dresden, Germany (E‑mail: [email protected]).
© 2015 Taylor & Francis
2
KRÄPLIN ET AL.
network dysfunction is heightened impulsivity (Bechara, 2005; Franken, van Strien, Nijs, & Muris, 2008; Verdejo-Garcia, Lawrence, & Clark, 2008), which has been clearly demonstrated in PG (for an overview see van Holst et al., 2010; Verdejo-Garcia et al., 2008). However, the exact underlying neurocognitive mechanisms and neuroanatomical bases as well as the question of separable dimensions of impulsivity are still under debate (Dawe & Loxton, 2004; Evenden, 1999; Fineberg et al., 2014; Meda et al., 2009; Moeller, Barratt, Dougherty, Schmitz, & Swann, 2001; Potenza & De Wit, 2010). The majority of studies have revealed at least two dimensions of impulsivity: response impulsivity and choice impulsivity (e.g., Bickel et al., 2012; Dalley, Everitt, & Robbins, 2011; Dick et al., 2010; Kim & Lee, 2011; Whiteside & Lynam, 2001). Response impulsivity is indicated by poor performance on tasks involving motoric inhibition like stop-signal tasks. Choice impulsivity refers to the discounting of delayed rewards and is operationalized with intertemporal choice tasks. Up to now, PG has been related to increased response impulsivity (e.g., Billieux et al., 2012; Fuentes, Tavares, Artes, & Gorenstein, 2006; Kertzman et al., 2008; Lawrence, Luty, Bogdan, Sahakian, & Clark, 2009; Whiteside, Lynam, Miller, & Reynolds, 2005) as well as increased choice impulsivity (e.g., Alessi & Petry, 2003; Benowitz, 2009; Dixon, Jacobs, & Sanders, 2006; Dixon, Marley, & Jacobs, 2003; Holt, Green, & Myerson, 2003; Kräplin, Dshemuchadse, et al., 2014). Crucially, another important disorder characteristic of PG has mainly been neglected in previous research: heightened comorbidity. Comorbid mental disorders are very common in PG, especially mood, anxiety, and substance use disorders (SUD), whereas PG is frequently secondary to the other mental disorders (Kessler et al., 2008; Petry, Stinson, & Grant, 2005). The missing consideration of comorbidity in PG may result from a focus on internal validity of results or from nonreporting. There are three important issues that arise from comorbidity as an important PG characteristic: The first issue concerns external validity since it is unclear whether the heightened impulsivity found in PG so far is generalizable to the typically comorbid PG population. This is important to know especially for the development of intervention strategies, as comorbidity is highly common in the treatment setting (Petry et al., 2005). The second issue concerns the specificity of heightened impulsivity dimensions as a disorder characteristic of PG without SUD compared to SUD. This study focused on ND as a comparison group as the first
step to answer the question of specificity. ND displays the highest comorbidity rates with PG with lifetime prevalence rates of around 60% (Kessler et al., 2008; Petry et al., 2005). Previous studies have demonstrated associations between heightened choice and response impulsivity with smoking (Mitchell, 2004; Reynolds, 2006; Ryan, MacKillop, & Carpenter, 2013) whereas no study diagnosing ND according to the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM–IV) has been conducted. As a third issue, findings on high impulsivity in PG may map an additive effect of PG plus a comorbid SUD like ND on impulsivity (Verdejo-Garcia et al., 2008). Up to now, two studies have shown an additive effect of problem gambling (as defined by the South Oaks Gambling Screen, SOGS; Lesieur & Blume, 1987) on impulsivity values in SUD (Petry, 2001b; Petry & Casarella, 1999). To address the issues listed above, our study aims were threefold: First we wanted to compare a comorbid PG group (without ND or other SUD) to healthy controls. We hypothesized that the comorbid PG group would show higher response and choice impulsivity than healthy controls, replicating findings of previous studies. Second, the study aimed to compare comorbid PG (without ND or other SUD) with ND alone. Up to now, one study demonstrated comparable performance and neuronal patterns concerning response impulsivity in PG and heavy smokers (de Ruiter, Oosterlaan, Veltman, van den Brink, & Goudriaan, 2012). Cognitive control dysfunctions may be related to both PG and ND, leading to impaired top-down regulation of addictive behaviors. In contrast, impaired choice behavior and a hyporesponsiveness of the reward system to monetary gains were found in PG compared to smokers (de Ruiter et al., 2009). These neuropsychological findings underline what seems obvious when comparing the clinical pictures of PG and ND: Dysfunctional changes of the valuation system leading to an increased incentive value of money may be specifically salient in PG. Therefore, our hypothesis was that PG is related to an increased choice impulsivity compared to ND, whereas response impulsivity may be comparable to ND. Our third aim was to compare impulsivity in comorbid PG (including ND) with ND alone. We hypothesized an additive effect of comorbid PG and ND on impulsivity as changes in neurotransmission related to PG and ND may interact and further shift the brain network imbalance in favor of immediate short-term rewards (e.g., Benowitz, 2009; Schweighofer et al., 2008). For an
IMPULSIVITY IN PATHOLOGICAL GAMBLING
Figure 1. Overview of the hypothesized impulsivity characteristics for the control group (CG), the pathological gambling (PG) group, the PG group with comorbid nicotine dependence (ND), and the ND only group.
overview, our hypotheses are summarized in Figure 1.
METHOD Design, recruitment, and screening A convenience sample was recruited for a crosssectional study divided in the following groups: individuals diagnosed with PG (PG group), with PG and ND (PG/ND group), or with ND alone (ND group), and these were compared to a control group. Participants were recruited through advertisements and postings on community boards. PG and ND had been diagnosed according to the DSM–IV, Text Revision (DSM–IV–TR; American Psychiatric Association, APA, 2000) within the previous 12 months. The PG and ND group were mutually exclusive: Participants in the PG group had no lifetime diagnosis of ND, had never smoked regularly for more than 4 weeks, and had not smoked in the previous 12 months. In the ND group, participants were non- or infrequent gamblers who displayed no lifetime DSM–IV criteria for PG. Participants in the control group had never been smokers (i.e., had smoked fewer than 20 cigarettes in their life; Pomerleau, Pomerleau, Snedecor, & Mehringer, 2004), and had not smoked at all in the previous 12 months. Participants in the control group had also never exceeded irregular, unproblematic gambling. All volunteers were screened for additional comorbid disorders with the Munich Composite International Diagnostic Interview (DIA-X/M-CIDI; Wittchen & Pfister, 1997). For all groups we excluded participants who fulfilled any of the following criteria: (a) aged under 18 or over 55 years, (b) use of psychotropic medication in the previous 2 weeks, (c) disorders that might
3
influence cognition or motor performance (e.g., attention-deficit hyperactivity disorder, ADHD), (d) having a language other than German as first language, (e) positive urine tests for amphetamines, cocaine, ecstasy, methamphetamines, opioids, or cannabis, and (f) currently undergoing treatment for mental or personality disorders. For the ND and the control groups, we additionally excluded individuals with (g) other mental disorders in the previous 12 months (i.e., somatoform, anxiety, affective, eating, substance use, obsessive compulsive, and psychotic disorders). Since we were interested in comorbid PG cases, the exclusion criterion (g) was not applied for the PG groups.
Final sample From 308 interested and contacted individuals, the final sample consisted of 13 participants in the PG group, 13 in the PG/ND group, 41 in the ND group, and 52 in the control group. General exclusion criteria included no further interest to participate or no availability (n = 153) and were not exclusively assigned to one of the groups (n = 31). Reasons for exclusion in the PG/ND and PG groups included no recent diagnosis of PG (n = 19), a neurological disorder (n = 3), and intake of substances or psychoactive medication (n = 3). In the ND group we excluded individuals due to no recent diagnosis of ND (n = 44), a comorbid neurological or mental disorder (n = 29), and intake of substances or psychoactive medication (n = 9). In the control group we excluded individuals because they did not fulfill our criteria of being never smokers (n = 3), had a comorbid neurological or mental disorder (n = 13), and took psychoactive substances or medication (n = 1). Demographical characteristics for the four groups are presented in Table 1. We tested group differences using t test or chi square test. First, the PG group differed from the control group in showing a higher proportion of male participants (χ2 = 8.14, p < .01) and more comorbid mental disorders (χ2 = 8.20, p < .01). Second, compared to the ND group, the PG group had also a significantly higher proportion of male participants (χ2 = 8.21, p < .01). Third, the PG/ND group had a significantly lower intelligence quotient (IQ; t = 3.09, p < .01), fulfilled more DSM–IV criteria for ND (t = –3.52, p < .01), and displayed more comorbid mental disorders (χ2 = 7.32, p < .01) than the ND group. As expected, we found an earlier onset for regular smoking than for first gambling problems: Participants in the PG/ND started daily smoking for more than 4 weeks at the age of 17.08 years
4
KRÄPLIN ET AL.
TABLE 1 Demographical data for the control group, the pathological gambling group, the PG group with comorbid nicotine dependence, and the ND only group Group Control (n = 52) Variables Age (years) Estimated IQ DSM–IV criteria for PG DSM–IV criteria for ND Male Income
Increased impulsivity in pathological gambling: Considering nicotine dependence Anja Kräplin, Silke Behrendt, Stefan Scherbaum, Maja Dshemuchadse, Gerhard Bühringer, and Thomas Goschke Department of Psychology, Technische Universitaet Dresden, Dresden, Germany (Received 27 June 2014; accepted 8 February 2015) Introduction: It has been highlighted that increased impulsivity is an important etiological factor in pathological gambling (PG). However, the role of the highly prevalent comorbid mental disorders in PG remains unclear. This is of special concern as the highest comorbidity was found between PG and nicotine dependence (ND), which, in turn, has also been associated with heightened impulsivity. This study aimed to find out whether increased impulsivity in PG is a specific characteristic related to comorbid mental disorders in general or especially to ND. Method: A cross-sectional study was designed that included a healthy control group and three disorder groups with comorbid PG excluding ND, comorbid PG including ND, and ND alone according to the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition. Impulsivity was assessed according to the dimensions response and choice impulsivity applying behavioral and self-report measures. Results: We found (a) increased response impulsivity in the comorbid PG group when compared to the control group. Furthermore, increased choice impulsivity was found (b) in comorbid PG excluding ND and (c) in comorbid PG including ND when compared to ND alone. Other group differences did not reach significance. Conclusions: Our findings highlight that comorbid PG shares deficits in inhibitory control with ND. In contrast, maladaptive choices are a specific disorder characteristic of comorbid PG when compared to ND. If replicated in further studies, intervention strategies strengthening cognitive control skills might be effective for comorbid PG as well as ND whereas strategies enhancing maladaptive valuation of rewards might be specifically effective in comorbid PG. Keywords: Pathological gambling; Impulsivity; Comorbidity; Nicotine dependence; Decision making.
High impulsivity and high prevalence rates of comorbid mental disorders like nicotine dependence (ND) are well investigated phenomena in pathological gambling (PG; e.g., Kessler et al., 2008; van Holst, van den Brink, Veltman, & Goudriaan, 2010). This study aimed to extend knowledge on impulsivity in PG by considering comorbid mental disorders with a special focus on ND and applying a multidimensional assessment of impulsivity.
According to recent reviews and models on addictive behaviors, an imbalance of valuation and cognitive control brain networks may be associated with PG (Bickel, Jarmolowicz, Mueller, Gatchalian, & McClure, 2012; Bühringer, Wittchen, Gottlebe, Kufeld, & Goschke, 2008; Goldstein & Volkow, 2011; Goschke, 2014; Redish, Jensen, & Johnson, 2008; van Holst et al., 2010). A central indicator of this brain
The authors wish to thank René Mayer for programming the administered tasks and for his helpful comments on the methods. Anja Kräplin, Silke Behrendt, Stefan Scherbaum, Maja Dshemuchadse, and Thomas Goschke declare to have no conflict of interest. Gerhard Bühringer has received unrestricted research grants from the Bavarian State Ministry of Finance (regulatory authority for and operator of the state gambling monopoly) via the Bavarian State Ministry of the Environment and Public Health, from the German Federal Ministry of Economics and Technology (regulatory authority for the commercial gaming industry), and from the commercial gaming industry. No potential conflict of interest was reported by the author(s). This study was funded by the German Research Foundation [grant number GO-720/8-1], [grant number SFB 940/1 2013]. Address correspondence to: Anja Kräplin, Addiction Research Unit, Institute of Clinical Psychology and Psychotherapy, Technische Universitaet Dresden, Chemnitzer Strasse 46, D-01187 Dresden, Germany (E‑mail: [email protected]).
© 2015 Taylor & Francis
2
KRÄPLIN ET AL.
network dysfunction is heightened impulsivity (Bechara, 2005; Franken, van Strien, Nijs, & Muris, 2008; Verdejo-Garcia, Lawrence, & Clark, 2008), which has been clearly demonstrated in PG (for an overview see van Holst et al., 2010; Verdejo-Garcia et al., 2008). However, the exact underlying neurocognitive mechanisms and neuroanatomical bases as well as the question of separable dimensions of impulsivity are still under debate (Dawe & Loxton, 2004; Evenden, 1999; Fineberg et al., 2014; Meda et al., 2009; Moeller, Barratt, Dougherty, Schmitz, & Swann, 2001; Potenza & De Wit, 2010). The majority of studies have revealed at least two dimensions of impulsivity: response impulsivity and choice impulsivity (e.g., Bickel et al., 2012; Dalley, Everitt, & Robbins, 2011; Dick et al., 2010; Kim & Lee, 2011; Whiteside & Lynam, 2001). Response impulsivity is indicated by poor performance on tasks involving motoric inhibition like stop-signal tasks. Choice impulsivity refers to the discounting of delayed rewards and is operationalized with intertemporal choice tasks. Up to now, PG has been related to increased response impulsivity (e.g., Billieux et al., 2012; Fuentes, Tavares, Artes, & Gorenstein, 2006; Kertzman et al., 2008; Lawrence, Luty, Bogdan, Sahakian, & Clark, 2009; Whiteside, Lynam, Miller, & Reynolds, 2005) as well as increased choice impulsivity (e.g., Alessi & Petry, 2003; Benowitz, 2009; Dixon, Jacobs, & Sanders, 2006; Dixon, Marley, & Jacobs, 2003; Holt, Green, & Myerson, 2003; Kräplin, Dshemuchadse, et al., 2014). Crucially, another important disorder characteristic of PG has mainly been neglected in previous research: heightened comorbidity. Comorbid mental disorders are very common in PG, especially mood, anxiety, and substance use disorders (SUD), whereas PG is frequently secondary to the other mental disorders (Kessler et al., 2008; Petry, Stinson, & Grant, 2005). The missing consideration of comorbidity in PG may result from a focus on internal validity of results or from nonreporting. There are three important issues that arise from comorbidity as an important PG characteristic: The first issue concerns external validity since it is unclear whether the heightened impulsivity found in PG so far is generalizable to the typically comorbid PG population. This is important to know especially for the development of intervention strategies, as comorbidity is highly common in the treatment setting (Petry et al., 2005). The second issue concerns the specificity of heightened impulsivity dimensions as a disorder characteristic of PG without SUD compared to SUD. This study focused on ND as a comparison group as the first
step to answer the question of specificity. ND displays the highest comorbidity rates with PG with lifetime prevalence rates of around 60% (Kessler et al., 2008; Petry et al., 2005). Previous studies have demonstrated associations between heightened choice and response impulsivity with smoking (Mitchell, 2004; Reynolds, 2006; Ryan, MacKillop, & Carpenter, 2013) whereas no study diagnosing ND according to the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM–IV) has been conducted. As a third issue, findings on high impulsivity in PG may map an additive effect of PG plus a comorbid SUD like ND on impulsivity (Verdejo-Garcia et al., 2008). Up to now, two studies have shown an additive effect of problem gambling (as defined by the South Oaks Gambling Screen, SOGS; Lesieur & Blume, 1987) on impulsivity values in SUD (Petry, 2001b; Petry & Casarella, 1999). To address the issues listed above, our study aims were threefold: First we wanted to compare a comorbid PG group (without ND or other SUD) to healthy controls. We hypothesized that the comorbid PG group would show higher response and choice impulsivity than healthy controls, replicating findings of previous studies. Second, the study aimed to compare comorbid PG (without ND or other SUD) with ND alone. Up to now, one study demonstrated comparable performance and neuronal patterns concerning response impulsivity in PG and heavy smokers (de Ruiter, Oosterlaan, Veltman, van den Brink, & Goudriaan, 2012). Cognitive control dysfunctions may be related to both PG and ND, leading to impaired top-down regulation of addictive behaviors. In contrast, impaired choice behavior and a hyporesponsiveness of the reward system to monetary gains were found in PG compared to smokers (de Ruiter et al., 2009). These neuropsychological findings underline what seems obvious when comparing the clinical pictures of PG and ND: Dysfunctional changes of the valuation system leading to an increased incentive value of money may be specifically salient in PG. Therefore, our hypothesis was that PG is related to an increased choice impulsivity compared to ND, whereas response impulsivity may be comparable to ND. Our third aim was to compare impulsivity in comorbid PG (including ND) with ND alone. We hypothesized an additive effect of comorbid PG and ND on impulsivity as changes in neurotransmission related to PG and ND may interact and further shift the brain network imbalance in favor of immediate short-term rewards (e.g., Benowitz, 2009; Schweighofer et al., 2008). For an
IMPULSIVITY IN PATHOLOGICAL GAMBLING
Figure 1. Overview of the hypothesized impulsivity characteristics for the control group (CG), the pathological gambling (PG) group, the PG group with comorbid nicotine dependence (ND), and the ND only group.
overview, our hypotheses are summarized in Figure 1.
METHOD Design, recruitment, and screening A convenience sample was recruited for a crosssectional study divided in the following groups: individuals diagnosed with PG (PG group), with PG and ND (PG/ND group), or with ND alone (ND group), and these were compared to a control group. Participants were recruited through advertisements and postings on community boards. PG and ND had been diagnosed according to the DSM–IV, Text Revision (DSM–IV–TR; American Psychiatric Association, APA, 2000) within the previous 12 months. The PG and ND group were mutually exclusive: Participants in the PG group had no lifetime diagnosis of ND, had never smoked regularly for more than 4 weeks, and had not smoked in the previous 12 months. In the ND group, participants were non- or infrequent gamblers who displayed no lifetime DSM–IV criteria for PG. Participants in the control group had never been smokers (i.e., had smoked fewer than 20 cigarettes in their life; Pomerleau, Pomerleau, Snedecor, & Mehringer, 2004), and had not smoked at all in the previous 12 months. Participants in the control group had also never exceeded irregular, unproblematic gambling. All volunteers were screened for additional comorbid disorders with the Munich Composite International Diagnostic Interview (DIA-X/M-CIDI; Wittchen & Pfister, 1997). For all groups we excluded participants who fulfilled any of the following criteria: (a) aged under 18 or over 55 years, (b) use of psychotropic medication in the previous 2 weeks, (c) disorders that might
3
influence cognition or motor performance (e.g., attention-deficit hyperactivity disorder, ADHD), (d) having a language other than German as first language, (e) positive urine tests for amphetamines, cocaine, ecstasy, methamphetamines, opioids, or cannabis, and (f) currently undergoing treatment for mental or personality disorders. For the ND and the control groups, we additionally excluded individuals with (g) other mental disorders in the previous 12 months (i.e., somatoform, anxiety, affective, eating, substance use, obsessive compulsive, and psychotic disorders). Since we were interested in comorbid PG cases, the exclusion criterion (g) was not applied for the PG groups.
Final sample From 308 interested and contacted individuals, the final sample consisted of 13 participants in the PG group, 13 in the PG/ND group, 41 in the ND group, and 52 in the control group. General exclusion criteria included no further interest to participate or no availability (n = 153) and were not exclusively assigned to one of the groups (n = 31). Reasons for exclusion in the PG/ND and PG groups included no recent diagnosis of PG (n = 19), a neurological disorder (n = 3), and intake of substances or psychoactive medication (n = 3). In the ND group we excluded individuals due to no recent diagnosis of ND (n = 44), a comorbid neurological or mental disorder (n = 29), and intake of substances or psychoactive medication (n = 9). In the control group we excluded individuals because they did not fulfill our criteria of being never smokers (n = 3), had a comorbid neurological or mental disorder (n = 13), and took psychoactive substances or medication (n = 1). Demographical characteristics for the four groups are presented in Table 1. We tested group differences using t test or chi square test. First, the PG group differed from the control group in showing a higher proportion of male participants (χ2 = 8.14, p < .01) and more comorbid mental disorders (χ2 = 8.20, p < .01). Second, compared to the ND group, the PG group had also a significantly higher proportion of male participants (χ2 = 8.21, p < .01). Third, the PG/ND group had a significantly lower intelligence quotient (IQ; t = 3.09, p < .01), fulfilled more DSM–IV criteria for ND (t = –3.52, p < .01), and displayed more comorbid mental disorders (χ2 = 7.32, p < .01) than the ND group. As expected, we found an earlier onset for regular smoking than for first gambling problems: Participants in the PG/ND started daily smoking for more than 4 weeks at the age of 17.08 years
4
KRÄPLIN ET AL.
TABLE 1 Demographical data for the control group, the pathological gambling group, the PG group with comorbid nicotine dependence, and the ND only group Group Control (n = 52) Variables Age (years) Estimated IQ DSM–IV criteria for PG DSM–IV criteria for ND Male Income
