ORIGINAL RESEARCH
Age of Pathological Gambling Onset: Clinical and Treatment-Related Features Young-Chul Shin, MD, PhD, Sam-Wook Choi, MD, PhD, Juwon Ha, MD, Jung Yeon Mok, MA, Se-Won Lim, MD, PhD, Jung-Seok Choi, MD, PhD, and Dai-Jin Kim, MD, PhD
Objectives: This study examined differences in the clinical and treatment-related features of pathological gambling (PG) on the basis of the age of PG onset among pathological gamblers who sought treatment. Methods: A total of 702 male outpatients with a primary diagnosis of PG and who were treated in a clinical practice were assessed by retrospective chart review. We selected the age of 25 years and younger as the threshold for “group 1.” We then stratified the participants into 4 groups on the basis of the age of PG onset in 10-year intervals. Analysis of covariance with a covariant of age and the Pearson χ 2 test were used for analyses. Results: We found that the earlier-onset gamblers were less likely to be escape type (P < 0.05), used significantly more Internet-based gambling (P < 0.001), and were less likely to engage in nonstrategic gambling (P < 0.05) than the later-onset gamblers. In addition, the earlier-onset gamblers took anticraving medication, such as naltrexone, significantly more often (P < 0.05), and sought treatment significantly more slowly after the onset of PG than the later-onset group (P < 0.01). Regarding adherence to treatment, however, there was no significant difference among the 4 groups on the basis of the age of PG onset. Conclusions: The age of PG onset is associated with several important clinical and treatment features. More studies are needed to advance prevention and treatment strategies for each age group. From the Department of Psychiatry (YCS, SWL), Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea; Department of Psychiatry, Gangnam Eulji Hospital, Eulji Addiction Institute, and Department of Addiction Rehabilitation and Social welfare (SWC), Eulji University, Korea; Department of Psychiatry (JH), Sejong General Hospital, Bucheon, Korea; Eulji Addiction Institute (JYM), Eulji University, Korea; Department of Psychiatry (JSC), SMG-SNU Boramae Medical Center, Seoul, Korea; and Department of Psychiatry (DJK), Seoul St Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea. Received for publication October 10, 2013; accepted February 10, 2014. Supported by the Korean Health Technology R&D project, Ministry of Health and Welfare (A129157). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the article. The authors declare no conflicts of interest. Send correspondence and reprint requests to Sam-Wook Choi, MD, PHD, Department of Psychiatry, Gangnam Eulji Hospital, Eulji University, 202 Dosan-daero, Gangnam-gu 135–815, South Korea. E-mail: peaceinu@ hanmail.net. C 2014 American Society of Addiction Medicine Copyright ISSN: 1932-0620/14/0803-0205 DOI: 10.1097/ADM.0000000000000031
Key Words: age of onset, adherence to treatment, early-onset, lateonset, pathological gambling (J Addict Med 2014;8: 205–210)
P
athological gambling (PG) is characterized by persistent and recurrent maladaptive gambling behavior that leads to significant deleterious legal, financial, physical, and psychosocial consequences (Sadock and Sadock, 2007). Pathological gambling has many similarities to substance use disorders, not only in adverse consequences but also in their natural history and phenomenology (Grant et al., 2010). Therefore, according to the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV), PG was classified as an impulse-control disorder, but it was reclassified as an addictive disorder in May 2013 in accordance with the DSM-5 (American Psychiatric Association, 2013). Generally speaking, significant biopsychosocial and environmental variables may account for developing PG. Although this concept provides a comprehensive view of an addictive disorder, pathological gamblers are not actually a homogenous group. In fact, studies on alcohol use disorders, which share similarities with PG, revealed the underlying importance of the age of onset in defining emerging clinical features (Cloninger et al., 1981; Hingson et al., 2006). Although no consensus has been reached, several studies have identified different subtypes of pathological gamblers (Blaszczynski and Nower, 2002; Grant et al., 2009; Shin et al., 2009; Alvarez-Moya et al., 2010; Milosevic and Ledgerwood, 2010). In our previous study, early and late discrimination on the basis of the age of onset of PG defined PG subtypes in terms of temperament and character (Shin et al., 2009). According to several studies, late-onset PG has distinct clinical features and the age of onset is associated with various clinical characteristics (Potenza et al., 2001; Tavares et al., 2001; Lynch et al., 2004; Shin et al., 2009; Jimenez-Murcia et al., 2010). Late-onset gamblers were significantly less likely to play strategic games and sought treatment significantly more rapidly after the onset of PG compared with the early-onset group (Grant et al., 2009). However, few studies have examined the differential manifestation between early- and late-onset gamblers with respect to the clinical and treatment variables of PG. In particular, little research to date has examined the relation among treatment-related variables, such as the duration of treatment
J Addict Med r Volume 8, Number 3, May/June 2014
Copyright © 2014 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
205
J Addict Med r Volume 8, Number 3, May/June 2014
Shin et al.
and pharmacotherapeutic characteristics, and the age of onset in PG (Jimenez-Murcia et al., 2010). The objective of this study was to examine differences in the clinical and treatment-related variables of PG on the basis of the age of onset among pathological gamblers who sought treatment. Knowing differential manifestation of PG on the basis of the age of onset might be helpful to advance our understanding of PG etiology and facilitate the study of neurobiological mechanism underlying PG and have significant implications for treatment outcome.
METHODS Participants Participants were drawn from a retrospective chart review of consecutive enrollments in a PG clinic at Kangbuk Samsung Hospital between 2002 and 2011. All of the participants sought treatment because of their problems related to PG. Participants were included if they met the following inclusion criteria: a primary diagnosis of current DSM-IV PG, a score of at least 5 on the South Oaks Gambling Screen, and an age of 18 years or older. The South Oaks Gambling Screen is a 20item, self-report screening tool for gambling-related problems. A score of greater than 5 indicates probable PG. The exclusion criteria included the presence of mental retardation, substance use disorder except alcohol and nicotine dependence, dementia or the inability to understand, and consent to participate in the study. Participants were screened by a trained psychiatrist with the aid of the Structured Clinical Interview for DSM-IV Axis I Disorders. The study was approved by the Ethics Committee of Kangbuk Samsung Hospital and the Eulji University. Written informed consent could not be obtained because this study was conducted using a retrospective chart review design.
Clinical Evaluation and Treatments Sociodemographic variables assessed in the PG clinic included sex, education, marital, and employment status. Gambling behavior was assessed as follows: types of gambling activity, use of Internet gambling, gambling debts at first visit, and age of onset of PG. “Age of onset” was defined as the age at which participants first met DSM-IV criteria for PG. The data used in our study were gathered through semistructured clinical interviews, which specifically dealt with, but were not confined to, gambling in early life, the intensity of urge to gamble, boredom proneness, mood and anxiety state, coping strategy for stress, motivation to gamble, and history of impulsiveness. In this study, impulsivity may be defined as a tendency to engage in unplanned action before consciously weighing its consequences. Although psychometric scales were not applied, these interviews were consistently applied to all patients. The authors then classified those who reported no or low impulsivity, no or mild urge to gamble, depressive and anxiety symptoms and gambling as a coping strategy as “escape gamblers” (McCormick, 1987; Ledgerwood and Petry, 2006; Milosevic and Ledgerwood, 2010). Those with a history of impulsive behaviors, gambling for enhancement, and moderate to severe urge to gamble were categorized as “action gamblers” (McCormick, 1987; Lesieur and Blume, 1991; Milosevic and Ledgerwood, 2010). The gambling types were
206
coded into “strategic” (eg, horse racing, bicycle racing, motor boat racing, stock-trading, sports betting, and card games), “nonstrategic” (eg, lotteries, scratch card, and slot machine), and mixed (Potenza et al., 2001; Grant et al., 2009). For substance use, patients were categorized as never, former, and current smokers on the basis of their answers to questions on lifetime and current smoking status. Patients were also asked about their current alcohol drinking status and average drinking days in a week. The PG clinic provides individual outpatient treatment on the basis of motivational enhancement therapy, cognitivebehavioral therapy, and pharmacotherapy. The clinic also provides group therapy on the basis of motivational and cognitivebehavioral approaches. The patients received medication if needed, and the choice of medication was based on the patient’s presentation; patients who reported significant cravings for gambling were started on an anticraving drug. Participants who gambled to cope with anxiety or depressed mood were started on an antidepressant medication. The patients simultaneously received distinct therapeutic alternatives, such as Gamblers Anonymous (GA), if they chose to participate. Adherence to treatment was measured by the treatment duration and follow-up rate at 6 and 12 months after the first visit to the clinic.
Statistical Analysis We gathered clinical data from 760 participants but included only 740 participants because of missing data for the age of PG onset. There was no consensus in previous studies as to what ages define either “early-onset” or “late-onset.” In this study, we selected the age of 25 years or younger as the threshold for “group 1” on the basis of previous studies of PG (Grant et al., 2009; Shin et al., 2009) and other studies related to alcoholism typology (Penick et al., 1990; Lim et al., 2008). From the age of 25, we divided participants into the following 4 groups on the basis of the age of PG onset: (1) group 1: up to 25 years old; (2) group 2: older than 25 years and up to 35 years; (3) group 3: older than 35 years and up to 45 years; and (4) group 4: older than 45 years. We thought that stratified categorical groups have a benefit to find the effect of age of PG onset on various variables than dichotomous division. Continuous variables were assessed with an analysis of covariance after controlling for age differences. Categorical variables were assessed with the Pearson χ 2 test. All analyses were performed using SPSS 18 for Windows. The alpha level for significance was 0.05, and all tests were 2-tailed.
RESULTS Comparisons of Sociodemographic Data Among 740 participants, female patients accounted for 5.1% (n = 38). According to previous studies, significant sex differences exist in the clinical presentation of PG, with men frequently beginning gambling early in life and reporting a slow emergence of problems compared with women. Therefore, we included only men (n = 702) in the current study to avoid obscure differences on the basis of the age of PG onset because of different male-to-female ratios. C
2014 American Society of Addiction Medicine
Copyright © 2014 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
J Addict Med r Volume 8, Number 3, May/June 2014
Age of Pathological Gambling Onset
Table 1 shows the sociodemographic data for the 4 groups of participants. As shown, 192 of the 702 participants with PG (27.4%) reported an onset of PG symptoms on or before 25 years of age (group 1), with a mean age of PG onset of 21.1 years compared with 28.9 years (group 2, 52.9%), 38.4 years (group 3, 13.7%), and 50.6 years (group 4, 6.1%) for each group (P < 0.001). The earlier-onset group was more likely than groups 2, 3, and 4 to have never been married (χ 2 = 97.91; P < 0.001) and groups 1 and 4 were more likely than groups 2 and 3 to be unemployed (χ 2 = 30.10; P < 0.001).
Comparisons of Gambling-Related and Clinical Data Table 2 shows the gambling-related and clinical variables. There was no difference between groups for types of gambler (χ 2 = 14.06; P = 0.120). However, when we conducted the Pearson χ 2 test for only the action and escape types, we found a significant difference between groups (χ 2 = 8.18; P < 0.05). The ratio of action gamblers in each group was the highest within each group (63.4∼75.5%), and the 2 earlieronset groups were less likely to be escape gamblers than the 2 later-onset groups (14.4%, 13.5% vs 24.7%, 20.9%). The preferred type of gambling was strategic gambling across all groups, and the earlier-onset group was less likely to engage in nonstrategic gambling than the later-onset group (χ 2 = 14.36; P = 0.026). The earlier-onset gamblers used significantly more Internet-based gambling than the later-onset gamblers (χ 2 = 22.01; P < 0.001). In regard to substance use, the earlier-onset gamblers were significantly more likely to be current smokers (χ 2 = 13.21; P = 0.040), but there was no difference between groups for drinking days in a week (F = 0.17; P = 0.916). All groups showed similar amounts of gambling debts at first visit to the PG clinic (F = 1.01; P = 0.388).
Comparisons of Treatment-Related Data In terms of self-help groups, approximately 30% of gamblers in each group attended GA (χ 2 = 0.61; P = 0.895) and approximately 5% to 19% of gamblers in each group participated group therapy at the PG clinic (χ 2 = 7.90; P = 0.246). The earlier-onset gamblers took anticraving medication, such
as naltrexone, significantly more often than the later-onset gamblers (χ 2 = 8.91; P = 0.031), but antidepressants medications were similarly prescribed among all groups (χ 2 = 6.810; P = 0.730). The pathological gamblers in group 1 sought treatment significantly more slowly after the onset of PG than the other groups (F = 4.53; P = 0.004). The earlier-onset groups were more likely to adhere to visiting the outpatient clinic than the later-onset groups, but this difference was not statistically significant (F = 0.48; P = 0.696). The duration of treatment for each group was 9.5, 8.0, 7.3, and 6.7 months, respectively. When only participants who stayed in treatment for a minimum of 60 days were included in the analyses, the mean duration of treatment for each group was 18.5, 15.2, 12.9, and 9.8 months, respectively, and such durations were not significantly different (F = 1.03; P = 0.38). The mean follow-up rate for each group at 6 and 12 months after first clinic visit was not significantly different (F = 0.98, P = 0.40, and F = 0.59, P = 0.62, respectively). These findings are consistent regardless of GA attendance. In summary, there was no significant difference among the 4 groups on the basis of age of PG onset for adherence to treatment. Detailed information about these results is presented in Table 3.
DISCUSSION Several studies have reported that PG seems to be heterogeneous, but few studies have examined the differential manifestation of pathological gamblers on the basis of the age of PG onset with respect to the clinical and treatment variables of PG. In particular, few studies to date have examined the relation among treatment-related variables, such as the duration of treatment and pharmacotherapeutic characteristics, and the age of onset in PG. This study included a clinical sample who visited the first gambling clinic in Korea for treatment of gambling problems between 2002 and 2011. Therefore, we could examine clinical measures using not only baseline data but also the retrospectively analyzed results of treatment. In this study, earlier-onset gamblers were younger and more likely to never have been married. For example, the mean
TABLE 1. Comparison of Sociodemographic Features in 4 Groups of Pathological Gamblers Based on Age of Pathological Gambling Onset (N = 702) Variables Age of onset, mean (±SD) Age, in years, mean (±SD) Education, in years, mean (±SD) Marital status, n (%) Single Married Divorced Widower Unknown Employment status, n (%) Full-time Part-time Unemployment Unknown C
Group 1 (n = 192)
Group 2 (N = 371)
Group 3 (N = 96)
Group 4 (N = 43)
P
F
Post Hoc
21.1 (2.3) 33.1 (8.5) 14.2 (2.3)
28.9 (2.7) 38.4 (7.5) 14.3 (2.4)
38.4 (2.7) 46.2 (6.7) 14.0 (3.5)
50.6 (5.4) 57.0 (6.7) 13.4 (3.5)
Age of Pathological Gambling Onset: Clinical and Treatment-Related Features Young-Chul Shin, MD, PhD, Sam-Wook Choi, MD, PhD, Juwon Ha, MD, Jung Yeon Mok, MA, Se-Won Lim, MD, PhD, Jung-Seok Choi, MD, PhD, and Dai-Jin Kim, MD, PhD
Objectives: This study examined differences in the clinical and treatment-related features of pathological gambling (PG) on the basis of the age of PG onset among pathological gamblers who sought treatment. Methods: A total of 702 male outpatients with a primary diagnosis of PG and who were treated in a clinical practice were assessed by retrospective chart review. We selected the age of 25 years and younger as the threshold for “group 1.” We then stratified the participants into 4 groups on the basis of the age of PG onset in 10-year intervals. Analysis of covariance with a covariant of age and the Pearson χ 2 test were used for analyses. Results: We found that the earlier-onset gamblers were less likely to be escape type (P < 0.05), used significantly more Internet-based gambling (P < 0.001), and were less likely to engage in nonstrategic gambling (P < 0.05) than the later-onset gamblers. In addition, the earlier-onset gamblers took anticraving medication, such as naltrexone, significantly more often (P < 0.05), and sought treatment significantly more slowly after the onset of PG than the later-onset group (P < 0.01). Regarding adherence to treatment, however, there was no significant difference among the 4 groups on the basis of the age of PG onset. Conclusions: The age of PG onset is associated with several important clinical and treatment features. More studies are needed to advance prevention and treatment strategies for each age group. From the Department of Psychiatry (YCS, SWL), Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea; Department of Psychiatry, Gangnam Eulji Hospital, Eulji Addiction Institute, and Department of Addiction Rehabilitation and Social welfare (SWC), Eulji University, Korea; Department of Psychiatry (JH), Sejong General Hospital, Bucheon, Korea; Eulji Addiction Institute (JYM), Eulji University, Korea; Department of Psychiatry (JSC), SMG-SNU Boramae Medical Center, Seoul, Korea; and Department of Psychiatry (DJK), Seoul St Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea. Received for publication October 10, 2013; accepted February 10, 2014. Supported by the Korean Health Technology R&D project, Ministry of Health and Welfare (A129157). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the article. The authors declare no conflicts of interest. Send correspondence and reprint requests to Sam-Wook Choi, MD, PHD, Department of Psychiatry, Gangnam Eulji Hospital, Eulji University, 202 Dosan-daero, Gangnam-gu 135–815, South Korea. E-mail: peaceinu@ hanmail.net. C 2014 American Society of Addiction Medicine Copyright ISSN: 1932-0620/14/0803-0205 DOI: 10.1097/ADM.0000000000000031
Key Words: age of onset, adherence to treatment, early-onset, lateonset, pathological gambling (J Addict Med 2014;8: 205–210)
P
athological gambling (PG) is characterized by persistent and recurrent maladaptive gambling behavior that leads to significant deleterious legal, financial, physical, and psychosocial consequences (Sadock and Sadock, 2007). Pathological gambling has many similarities to substance use disorders, not only in adverse consequences but also in their natural history and phenomenology (Grant et al., 2010). Therefore, according to the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV), PG was classified as an impulse-control disorder, but it was reclassified as an addictive disorder in May 2013 in accordance with the DSM-5 (American Psychiatric Association, 2013). Generally speaking, significant biopsychosocial and environmental variables may account for developing PG. Although this concept provides a comprehensive view of an addictive disorder, pathological gamblers are not actually a homogenous group. In fact, studies on alcohol use disorders, which share similarities with PG, revealed the underlying importance of the age of onset in defining emerging clinical features (Cloninger et al., 1981; Hingson et al., 2006). Although no consensus has been reached, several studies have identified different subtypes of pathological gamblers (Blaszczynski and Nower, 2002; Grant et al., 2009; Shin et al., 2009; Alvarez-Moya et al., 2010; Milosevic and Ledgerwood, 2010). In our previous study, early and late discrimination on the basis of the age of onset of PG defined PG subtypes in terms of temperament and character (Shin et al., 2009). According to several studies, late-onset PG has distinct clinical features and the age of onset is associated with various clinical characteristics (Potenza et al., 2001; Tavares et al., 2001; Lynch et al., 2004; Shin et al., 2009; Jimenez-Murcia et al., 2010). Late-onset gamblers were significantly less likely to play strategic games and sought treatment significantly more rapidly after the onset of PG compared with the early-onset group (Grant et al., 2009). However, few studies have examined the differential manifestation between early- and late-onset gamblers with respect to the clinical and treatment variables of PG. In particular, little research to date has examined the relation among treatment-related variables, such as the duration of treatment
J Addict Med r Volume 8, Number 3, May/June 2014
Copyright © 2014 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
205
J Addict Med r Volume 8, Number 3, May/June 2014
Shin et al.
and pharmacotherapeutic characteristics, and the age of onset in PG (Jimenez-Murcia et al., 2010). The objective of this study was to examine differences in the clinical and treatment-related variables of PG on the basis of the age of onset among pathological gamblers who sought treatment. Knowing differential manifestation of PG on the basis of the age of onset might be helpful to advance our understanding of PG etiology and facilitate the study of neurobiological mechanism underlying PG and have significant implications for treatment outcome.
METHODS Participants Participants were drawn from a retrospective chart review of consecutive enrollments in a PG clinic at Kangbuk Samsung Hospital between 2002 and 2011. All of the participants sought treatment because of their problems related to PG. Participants were included if they met the following inclusion criteria: a primary diagnosis of current DSM-IV PG, a score of at least 5 on the South Oaks Gambling Screen, and an age of 18 years or older. The South Oaks Gambling Screen is a 20item, self-report screening tool for gambling-related problems. A score of greater than 5 indicates probable PG. The exclusion criteria included the presence of mental retardation, substance use disorder except alcohol and nicotine dependence, dementia or the inability to understand, and consent to participate in the study. Participants were screened by a trained psychiatrist with the aid of the Structured Clinical Interview for DSM-IV Axis I Disorders. The study was approved by the Ethics Committee of Kangbuk Samsung Hospital and the Eulji University. Written informed consent could not be obtained because this study was conducted using a retrospective chart review design.
Clinical Evaluation and Treatments Sociodemographic variables assessed in the PG clinic included sex, education, marital, and employment status. Gambling behavior was assessed as follows: types of gambling activity, use of Internet gambling, gambling debts at first visit, and age of onset of PG. “Age of onset” was defined as the age at which participants first met DSM-IV criteria for PG. The data used in our study were gathered through semistructured clinical interviews, which specifically dealt with, but were not confined to, gambling in early life, the intensity of urge to gamble, boredom proneness, mood and anxiety state, coping strategy for stress, motivation to gamble, and history of impulsiveness. In this study, impulsivity may be defined as a tendency to engage in unplanned action before consciously weighing its consequences. Although psychometric scales were not applied, these interviews were consistently applied to all patients. The authors then classified those who reported no or low impulsivity, no or mild urge to gamble, depressive and anxiety symptoms and gambling as a coping strategy as “escape gamblers” (McCormick, 1987; Ledgerwood and Petry, 2006; Milosevic and Ledgerwood, 2010). Those with a history of impulsive behaviors, gambling for enhancement, and moderate to severe urge to gamble were categorized as “action gamblers” (McCormick, 1987; Lesieur and Blume, 1991; Milosevic and Ledgerwood, 2010). The gambling types were
206
coded into “strategic” (eg, horse racing, bicycle racing, motor boat racing, stock-trading, sports betting, and card games), “nonstrategic” (eg, lotteries, scratch card, and slot machine), and mixed (Potenza et al., 2001; Grant et al., 2009). For substance use, patients were categorized as never, former, and current smokers on the basis of their answers to questions on lifetime and current smoking status. Patients were also asked about their current alcohol drinking status and average drinking days in a week. The PG clinic provides individual outpatient treatment on the basis of motivational enhancement therapy, cognitivebehavioral therapy, and pharmacotherapy. The clinic also provides group therapy on the basis of motivational and cognitivebehavioral approaches. The patients received medication if needed, and the choice of medication was based on the patient’s presentation; patients who reported significant cravings for gambling were started on an anticraving drug. Participants who gambled to cope with anxiety or depressed mood were started on an antidepressant medication. The patients simultaneously received distinct therapeutic alternatives, such as Gamblers Anonymous (GA), if they chose to participate. Adherence to treatment was measured by the treatment duration and follow-up rate at 6 and 12 months after the first visit to the clinic.
Statistical Analysis We gathered clinical data from 760 participants but included only 740 participants because of missing data for the age of PG onset. There was no consensus in previous studies as to what ages define either “early-onset” or “late-onset.” In this study, we selected the age of 25 years or younger as the threshold for “group 1” on the basis of previous studies of PG (Grant et al., 2009; Shin et al., 2009) and other studies related to alcoholism typology (Penick et al., 1990; Lim et al., 2008). From the age of 25, we divided participants into the following 4 groups on the basis of the age of PG onset: (1) group 1: up to 25 years old; (2) group 2: older than 25 years and up to 35 years; (3) group 3: older than 35 years and up to 45 years; and (4) group 4: older than 45 years. We thought that stratified categorical groups have a benefit to find the effect of age of PG onset on various variables than dichotomous division. Continuous variables were assessed with an analysis of covariance after controlling for age differences. Categorical variables were assessed with the Pearson χ 2 test. All analyses were performed using SPSS 18 for Windows. The alpha level for significance was 0.05, and all tests were 2-tailed.
RESULTS Comparisons of Sociodemographic Data Among 740 participants, female patients accounted for 5.1% (n = 38). According to previous studies, significant sex differences exist in the clinical presentation of PG, with men frequently beginning gambling early in life and reporting a slow emergence of problems compared with women. Therefore, we included only men (n = 702) in the current study to avoid obscure differences on the basis of the age of PG onset because of different male-to-female ratios. C
2014 American Society of Addiction Medicine
Copyright © 2014 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
J Addict Med r Volume 8, Number 3, May/June 2014
Age of Pathological Gambling Onset
Table 1 shows the sociodemographic data for the 4 groups of participants. As shown, 192 of the 702 participants with PG (27.4%) reported an onset of PG symptoms on or before 25 years of age (group 1), with a mean age of PG onset of 21.1 years compared with 28.9 years (group 2, 52.9%), 38.4 years (group 3, 13.7%), and 50.6 years (group 4, 6.1%) for each group (P < 0.001). The earlier-onset group was more likely than groups 2, 3, and 4 to have never been married (χ 2 = 97.91; P < 0.001) and groups 1 and 4 were more likely than groups 2 and 3 to be unemployed (χ 2 = 30.10; P < 0.001).
Comparisons of Gambling-Related and Clinical Data Table 2 shows the gambling-related and clinical variables. There was no difference between groups for types of gambler (χ 2 = 14.06; P = 0.120). However, when we conducted the Pearson χ 2 test for only the action and escape types, we found a significant difference between groups (χ 2 = 8.18; P < 0.05). The ratio of action gamblers in each group was the highest within each group (63.4∼75.5%), and the 2 earlieronset groups were less likely to be escape gamblers than the 2 later-onset groups (14.4%, 13.5% vs 24.7%, 20.9%). The preferred type of gambling was strategic gambling across all groups, and the earlier-onset group was less likely to engage in nonstrategic gambling than the later-onset group (χ 2 = 14.36; P = 0.026). The earlier-onset gamblers used significantly more Internet-based gambling than the later-onset gamblers (χ 2 = 22.01; P < 0.001). In regard to substance use, the earlier-onset gamblers were significantly more likely to be current smokers (χ 2 = 13.21; P = 0.040), but there was no difference between groups for drinking days in a week (F = 0.17; P = 0.916). All groups showed similar amounts of gambling debts at first visit to the PG clinic (F = 1.01; P = 0.388).
Comparisons of Treatment-Related Data In terms of self-help groups, approximately 30% of gamblers in each group attended GA (χ 2 = 0.61; P = 0.895) and approximately 5% to 19% of gamblers in each group participated group therapy at the PG clinic (χ 2 = 7.90; P = 0.246). The earlier-onset gamblers took anticraving medication, such
as naltrexone, significantly more often than the later-onset gamblers (χ 2 = 8.91; P = 0.031), but antidepressants medications were similarly prescribed among all groups (χ 2 = 6.810; P = 0.730). The pathological gamblers in group 1 sought treatment significantly more slowly after the onset of PG than the other groups (F = 4.53; P = 0.004). The earlier-onset groups were more likely to adhere to visiting the outpatient clinic than the later-onset groups, but this difference was not statistically significant (F = 0.48; P = 0.696). The duration of treatment for each group was 9.5, 8.0, 7.3, and 6.7 months, respectively. When only participants who stayed in treatment for a minimum of 60 days were included in the analyses, the mean duration of treatment for each group was 18.5, 15.2, 12.9, and 9.8 months, respectively, and such durations were not significantly different (F = 1.03; P = 0.38). The mean follow-up rate for each group at 6 and 12 months after first clinic visit was not significantly different (F = 0.98, P = 0.40, and F = 0.59, P = 0.62, respectively). These findings are consistent regardless of GA attendance. In summary, there was no significant difference among the 4 groups on the basis of age of PG onset for adherence to treatment. Detailed information about these results is presented in Table 3.
DISCUSSION Several studies have reported that PG seems to be heterogeneous, but few studies have examined the differential manifestation of pathological gamblers on the basis of the age of PG onset with respect to the clinical and treatment variables of PG. In particular, few studies to date have examined the relation among treatment-related variables, such as the duration of treatment and pharmacotherapeutic characteristics, and the age of onset in PG. This study included a clinical sample who visited the first gambling clinic in Korea for treatment of gambling problems between 2002 and 2011. Therefore, we could examine clinical measures using not only baseline data but also the retrospectively analyzed results of treatment. In this study, earlier-onset gamblers were younger and more likely to never have been married. For example, the mean
TABLE 1. Comparison of Sociodemographic Features in 4 Groups of Pathological Gamblers Based on Age of Pathological Gambling Onset (N = 702) Variables Age of onset, mean (±SD) Age, in years, mean (±SD) Education, in years, mean (±SD) Marital status, n (%) Single Married Divorced Widower Unknown Employment status, n (%) Full-time Part-time Unemployment Unknown C
Group 1 (n = 192)
Group 2 (N = 371)
Group 3 (N = 96)
Group 4 (N = 43)
P
F
Post Hoc
21.1 (2.3) 33.1 (8.5) 14.2 (2.3)
28.9 (2.7) 38.4 (7.5) 14.3 (2.4)
38.4 (2.7) 46.2 (6.7) 14.0 (3.5)
50.6 (5.4) 57.0 (6.7) 13.4 (3.5)
