The Brief Addiction Science Information Source (BASIS)

The WAGER, Vol. 15(1) – Disordered Gambling and Meaningful Correlates among Cambodian Refugees

Recent research shows that Cambodian refugees who settled in the United States suffer from disproportionately high rates of trauma (Marshall, Schell, Elliott, Berthold, & Chun, 2005). Other research shows that they also might be more prone to disordered gambling (Petry, Armentano, Kuoch, Norinth, & Smith, 2003). This week the WAGER reviews a study that examines whether socio-demographic variables and trauma are significant correlates of lifetime disordered gambling among a sample of Cambodian refugees (Marshall, Elliott, & Schell, 2009).

Methods

• Study participants (n=127) were a subsample of Cambodian refugees residing in Long Beach, California. The researchers recruited study participants by using a three stage random procedure:
  1) Screening: Census tracks identified areas with the highest concentration of Cambodian residents. The researchers screened 2,001 (97%) of the identified households for eligibility.
  2) Eligibility: Participants aged 35 to 75 were eligible if they previously lived in Cambodia during part of the Khmer Rouge regime. This process identified 719 Cambodians as eligible from 586 households (29%).
  3) Random Recruitment: One eligible person was selected randomly from each household; 490 participants agreed to participate.
• Subsample: The researchers added gambling-related questions to the core study after enrolling approximately 75% of participants. The 127 subsequent participants were used in this study.
• Face-to-face interviews assessed the following variables:
• Socio-demographic variables: age, gender, marital status, English proficiency (dichotomized as good or fair vs. poor or no ability) and household income (dichotomized as above or below the federal poverty line). 
• Pre-immigration trauma exposure: assessed 35 events using items from the 17-item Cambodian Harvard Trauma Questionnaire (Mollica et al., 1992) and the 46-item Bosnian version of the Harvard Trauma Questionnaire (Allden et al., 1998). 
• Disordered gambling classification (i.e., problem or potential pathological gambling) assessed using the South Oaks Gambling Screen (SOGS; Lesieur & Blume, 1987). A score of 3 or more indicated disordered gambling. 
• The researchers conducted bivariate and multivariate logistic regression analyses to examine the relationships between the socio-demographic and trauma variables and disordered gambling. They reported the results as odds ratios; Table 1 summarizes these findings.  

Table 1: Odds Ratios and 95% Confidence Intervals (CI) for Predicting Disordered Gambling (Adapted from Marshall et al, 2009)

 
Click image to enlarge, or adjust your browser's zoom setting.
¹Analyses are weighted and adjusted for clustering resulting from sample design
²Odds ratios are adjusted for variables with P-values less than 0.05 in bivariate models
³Per Standard Deviation
*P < 0.05
**P < 0.01

Results

• Eighteen participants (13.9%) met criteria for disordered gambling.
• Table 1 indicates that disordered gambling was significantly correlated with:
• Being male (bivariate analysis only)  
• Being married/cohabitating  
• Being exposed to high levels of trauma 

Limitations

• The rate of disordered gambling had large confidence intervals; anywhere from 4% to 24%.
• The results might not generalize to other populations, including other populations of Cambodian refugees.

Discussion

The results indicate lower rates (13.9%) of disordered gambling compared to the rates observed among Asian refugees in a similar study (i.e., 70.0%; Petry et al., 2003) that also used the SOGS with the same scoring classification. One difference between the two studies is that the Petry et al. study examined a convenience sample of refugees from three Asian communities (Laos, Cambodia, and Vietnam) at Asian community service organizations in Connecticut. Marshall et al. randomly selected a sample from a large Cambodian refugee community. Regardless, rates of disordered gambling appear to be meaningfully higher among Asian refugees than those observed in the general population (i.e., 1.4-2.9%; Kessler et al., 2008; Petry, Stinson, & Grant, 2005).

The correlation between traumatic events and disordered gambling in this study is consistent with other research (Kausch, Rugle, & Rowland, 2006); however, Marshall et al. found that marriage/cohabitation was significantly associated with disordered gambling. This finding contradicts other research (e.g., Petry et al., 2003; Petry et al., 2005), which reported positive associations between being divorced or single and disordered gambling. As a result, further examinations of disordered gambling and meaningful disordered gambling correlates among Cambodian refugees and other refugee communities are warranted.

-Tasha Chandler

What do you think? Please use the comment link below to provide feedback on this article.

References

Allden, K., Ceric, I., Kapetanovic, A., Lavelle, J., Loga, S., & Mathias, M. e. a. (1998). Harvard trauma manual: Bosnian-Herzegovina version. Cambridge, MA: Harvard Program for Refugee Trauma, Harvard Medical School.

Kausch, O., Rugle, L., & Rowland, D. Y. (2006). Lifetime histories of trauma among pathological gamblers. The American Journal on Addictions, 15(1), 35-43.

Kessler, R. C., Hwang, I., LaBrie, R. A., Petukhova, M., Sampson, N. A., Winters, K. C., et al. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine 38(9), 1351-1360.

Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of Psychiatry, 144(9), 1184-1188.

Marshall, G. N., Elliott, M. N., & Schell, T. L. (2009). Prevalence and correlates of lifetime disordered gambling in Cambodian refugees residing in Long Beach, CA. Journal of Immigrant Minority Health, 11(1), 35-40.

Marshall, G. N., Schell, T. L., Elliott, M. N., Berthold, S. M., & Chun, C. A. (2005). Mental health of Cambodian refugees 2 decades after resettlement in the United States. Jama, 294(5), 571-579.

Mollica, R. F., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). The Harvard Trauma Questionnaire. Validating a cross-cultural instrument for measuring torture, trauma, and posttraumatic stress disorder in Indochinese refugees. The Journal of Nervous and Mental Disease, 180(2), 111-116.

Petry, N. M., Armentano, C., Kuoch, T., Norinth, T., & Smith, L. (2003). Gambling participation and problems among South East Asian refugees to the United States. Psychiatric Services, 54(8), 1142-1148.

Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry, 66(5), 564-574.

WAGER Vol. 14 (10): Generation Gap: Difference in PG symptoms based on age of onset.

Research suggests that, like OCD, the clinical characteristics and symptoms of pathological gambling (PG) (e.g., symptom severity, changes in relationships, financial trouble, comorbid disorders) can vary based on age of onset. Research has indicated differences in PG-related clinical characteristics and symptoms based on gender (Ibanez, Blanco, Moreryra, & Saiz-Ruiz, 2003; Potenza, Steinberg, Wu, Rounsaville, & O'Malley S, 2006) and age when seeking treatment (Grant et al., 2007; Lucke & Wallace, 2006; Pietrzak & Petry, 2006). To further examine the clinical characteristics of PG, Grant and colleagues (Grant, Kim, Odlaug, Buchanan, & Potenza, 2009) studied a cohort of 322 adult pathological gamblers (PGs) and focused particularly on the clinical characteristics and symptoms evidenced among participants who developed PG as older adults.

Methods

• Grant et al. (2009) recruited 322 adult participants that met current (past year) DSM-IV criteria for PG.

• Participants were recruited during a 5 year period using advertisements, and referrals from a cognitive-behavioral study, a pharmacological study, and outpatient treatment at a private or public hospital. 
• Researchers examined PG and PG-related clinical characteristics with the following instruments: 

• Structured Clinical Interview for DSM-IV Axis I Disorders and PG; 
• A semi-structured questionnaire to collect information about demographic and clinical features of PG and participant treatment history;
• The Yale Brown Obsessive Compulsive Scale Modified for PG (PG-YBOCS);
• The South Oaks Gambling Screen (SOGS);
• The Clinical Global Impression-Severity scale (CGI).
• Researchers grouped the participants by age of onset, defined as “the age when participants’ PG symptoms first met DSM-IV criteria” (Grant et al., 2009). The investigators placed them into the following three groups:

• Younger than 26 years old (N=63); 
• Between 26 and 54 years old (N=217);
• 55 years old and older (i.e., late onset; N=42).
• To test for differences among the three age groups, Grant et al. used Pearson’s chi-square, Kruskal-Wallis and post hoc Bonferroni analyses.

Results

• Compared to their younger counterparts, late onset PGs had the following significant characteristics (presented in Table 1 below with p values):

• Endorsed fewer PG-related criteria (i.e., lower SOGS scores);
• Less participation in gambling-related strategic games;
• Less participation in Gamblers Anonymous;
• More likely to seek treatment soon after experiencing PG symptoms;
• More likely to suffer from an anxiety disorder;
• Less likely to have a father with a history of a gambling problem.

Table 1. Significant differences in clinical characteristics of pathological gamblers of different ages.

Adapted from Table 2: Clinical characteristics of pathological gamblers (n=322) based on age of pathological gambling onset (Grant et al., 2009).

Characteristics	25 and younger (n=63)	26-54 (n=217)	55 and older (n=42)	p-Value
SOGS Scores, mean (± SD)	13.6 ±5.04	13.8 ±3.34	10.0 ±2.76	.004a, b
Strategic game playing, n (%)	42 (66.7)	102 (47.0)	9 (21.4)	<.001a
Gamblers Anonymous, n (%)	30 (47.6)	88 (40.6)	7 (16.7)	.004a
Seeking professional treatment, mean (± SD)	17.6 ± 11.6	8.4 ± 5.7	3.9 ± 3.7	.001a, b
Anxiety Disorder, n (%)	7 (11.1)	32 (14.7)	16 (38.1)	<.001a
Father with a gambling problem, n (%)	18 (29.0)	51 (23.7)	3 (7.1)	.025a

aBonferroni post hoc pairwise comparison (26-54 yrs) vs. (55 yrs and older) p < .05. 

bBonferroni post hoc pairwise comparison (26 and younger) vs. (55 yrs and older) p < .05. 

Limitations

• The sample lacked cultural and ethnic diversity (i.e., 94.4% were Caucasian).
• Participant history, including age of PG onset, was based on self-recall.
• Vague age of onset definition; it is unclear if participants were grouped based on the life time appearance of the first DSM criterion or the first criterion experienced in a 12-month period in which 5 or more DSM criteria were experienced.
• Using 55 years as a cut-off point is not evidence based; the authors explain that 55 years represents a reasonable extreme (13%) of the total sample and that other studies used this age as a cut off about older gamblers.

Conclusion

The study sample is limited in its diversity, which suggests a need for further research into PG-related clinical characteristics among more diverse samples of PGs, including those who experience late-onset PG. It also could be worthwhile for researchers to examine the triggers of late onset of any gambling. The study results indicate that those with late onset PG are more likely to have certain clinical characteristics, but the authors fail to explain why the two other study groups (26 and younger and 26-54) have higher rates of treatment seeking. Answering this question could explain a number of the current study results: likelihood of treatment in each group and the kinds of symptoms that each group experienced. Given the difference in the number of people per age group it’s possible that the majority of the participants (aged 54 and younger) were simply treatment seekers whose responses skewed results. 

-Ingrid Maurice

What do you think? Please use the comment link below to provide feedback on this article.

References

Grant, J. E., Kim, S. W., Odlaug, B. L., Buchanan, S. N., & Potenza, M. N. (2009). Late-onset pathological gambling: clinical correlates and gender differences. Journal of Psychiatric Research, 43(4), 380-387.

Grant, J. E., Mancebo, M. C., Pinto, A., Williams, K. A., Eisen, J. L., & Rasmussen, S. A. (2007). Late-onset obsessive compulsive disorder: clinical characteristics and psychiatric comorbidity. Psychiatry Research, 152(1), 21-27.

Ibanez, A., Blanco, C., Moreryra, P., & Saiz-Ruiz, J. (2003). Gender differences in pathological gambling. Journal of Clinical Psychiatry, 64(3), 295-301.

Lucke, S., & Wallace, M. (2006). Assessment and management of pathological and problem gambling among older adults. Geriatr Nurs, 27(1), 51-57.

Pietrzak, R. H., & Petry, N. M. (2006). Severity of gambling problems and psychosocial functioning in older adults. J Geriatr Psychiatry Neurol, 19(2), 106-113.

Potenza, M. N., Steinberg, M. A., Wu, R., Rounsaville, B. J., & O'Malley S, S. (2006). Characteristics of Older Adult Problem Gamblers Calling a Gambling Helpline. Journal of Gambling Studies.

The WAGER, Vol. 14(9) - Turning rats into gamblers: Modulating gambling behavior using laboratory animals.

Animal models of human behavior help scientists to understand pathology and develop new drugs. This week’s The WAGER reviews a study that examines an animal model of gambling behavior (Zeeb, Robbins, & Winstanley, 2009). Zeeb et al. sought to determine if rats are capable of “playing the odds” and if altering their brain chemistry (e.g., to mimic the neurochemistry of human gamblers) changes the rats’ decision-making, which according to the authors was a proxy for gambling strategy.

Method

•  Investigators trained rats (N = 32) to choose between four trays (P1 – 4) to nose-poke for food (see Figure 1).
• • Each time a rat “won”, it received one or more sugar pellets as a reward. 
  • Each time it “lost”, there was a timeout during which it could not play. 
  • In terms of long-term gain, the optimal choice was to choose P2, which was getting 2 sugar pellets with moderate waiting time. 
  • P1 represented the choice with the lowest reward (1 pellet) and the lowest probability of timeout; P4 was the choice with the highest reward (4 pellets) and the highest probability of timeout.

• After at least one week of stable baseline behavior, the researchers injected various drugs that affected (i.e., increased or decreased) the level of dopamine in brain to alter the animals' brain chemistry to mimic that of problem gamblers. Examples of drugs are:
• • Amphetamine: potentially increases the level of dopamine in the brain.
  • Eticlopride: blocks dopamine receptors.
  • SKF 81297: triggers dopamine response by the brain cell.
• Investigators conducted repeated measures ANOVA to measure how the preferences were changed following the drug injection.

Figure 1. Trial structure for the Rat Gambling Task.*

Notes:* The location of the trays was counterbalanced across rats. A picture represents one possible setting.

Results

• At baseline, animals consistently showed a preference for the optimal choice (P2), associated with a modest gain and moderate punishment.
• Amphetamine significantly increased selection of the non-optimal choice P1 (p < .01), which is associated with the least punishment but also with the least reward.
• Eticlopride significantly increased selection of the optimal choice (p < .01).
• SKF 81297 significantly increased the selection of P4 (risky with largest probability of punishment) and significantly decreased the selection of the optimal choice (p < .01). 

Limitations

• The "Rat gambling task" is not a complete analogy for human gambling behavior (e.g., rats cannot really “lose” food).
• The "decision making" task is not necessarily a proxy for gambling.
• It is not clear if dopamine leads to the avoidance of punishment or awareness of rewards.
• It is not clear if rats are relying on memory to make their choice, rather than basing their preference on the perception of risk or punishment tolerance.

Conclusion

    This study indicates that rats are capable of ‘playing the odds’. Moreover, they seem to be “risk-averse” animals, similar to human beings. The obtained data suggest that changing brain chemistry produces subsequent changes in animals “similar-to-gambling” behavior. Specifically, dopaminergic agents (drugs that influence dopamine activity in brain) can impair or improve gambling performance. The results imply that high levels of dopamine might pre-dispose non-optimal decision making, creating a risk factor for gambling-related problems. The mechanism of dopamine as a risk factor is not yet clear. However, the results are consistent with previous studies that found dopamine treatment to increase gambling drive among human patients (Abler, Hahlbrock, Unrath, Gron, & Kassubek, 2009). Taken together, these findings indicate that the rats gambling task might be a useful tool to study biological aspects of gambling.

-Julia Braverman

What do you think?  Please use the comment link below to provide feedback on this article.

References

Abler, B., Hahlbrock, R., Unrath, A., Gron, G., & Kassubek, J. (2009). At-risk for pathological gambling: imaging neural reward processing under chronic dopamine agonists. Brain: A Journal of Neurology, 132(9), 2396-2402.

Zeeb, F. D., Robbins, T. W., & Winstanley, C. A. (2009). Serotonergic and Dopaminergic Modulation of Gambling Behavior as Assessed Using a Novel Rat Gambling Task. Neuropsychopharmacology, 34(10), 2329-2343.

The WAGER, Vol. 14(8) - Video Game Playing and Gambling-Related Problems: Is there a Connection?

Research has begun to focus on how gambling impacts adolescents (e.g., Delfabbro, Lahn, & Grabosky, 2005; Welte, Barnes, Tidwell, & Hoffman, 2009). However, researchers rarely have studied the correlation between adolescent video game playing and disordered gambling behavior. This week’s WAGER reviews a study (Delfabbro, King, Lambos, & Puglies, 2009) that examined associations between gambling-related problems and types and frequency of video game play among Australian adolescents.

Methods

• Participants (N= 2,669) were students between the ages 12 and 17 attending six secondary schools (i.e., grades 8-13) in South Australia.
  • Delfabbro et al. purposely sampled schools; they selected four schools to obtain representation from each of the four major statistical districts in the metropolitan area and selected two from major regional centers.
  • All students who were present on the survey’s administration day completed the survey with the exception of students who were withdrawn by parents (i.e., withdrawers). The absentee rate reported by the schools was approximately 10%; the withdrawer rate and completion rate were not reported.
• The survey measured video-game involvement during the past year and DSM-IV Pathological Gambling (PG) items.
  • Video-game involvement measures included frequency of play, number of hours played, and type of game played, including: TV games (e.g., Xbox, Nintendo, Playstation); phone-based games; hand-held games (e.g., Nintendo, Gameboy); PC games; and arcade games.
  • Researchers assessed PG using the DSM-IV Juvenile Criteria (DSM-IV-J; Fisher, 1992); a screening instrument for gambling-related problems among adolescents modified from the DSM-IV criteria for PG.
    • A 12-item yes/no scale.
      • According to Fisher (1992), a total score of four or more indicates pathological gambling.
      • The screen asked about current problems. Five items specify “in the past year,” six items are asked in the present tense, and one item was more ambiguous (Have you fallen out with members of your family, or close friends, because of your gambling behavior?).
• The participants’ DSM-IV-J scores permitted the investigators to establish three groups: Not at risk (i.e., endorsed no items on the scale); At-risk (i.e., endorsed 1-3 items on the scale); and PGs (i.e., endorsed 4 or more items on the scale).
• The authors conducted a series of hierarchical multiple regression analyses to examine associations between gambling-related problems and the frequency and type of video game playing.

Results

• Table 1 indicates the gambling frequency (i.e., the number of video-game sessions per year) of various types of video game play among the three gambling groups.
  • The results indicate that the PG and At Risk groups played arcade, phone, hand-held, and TV games more frequently than the Not at Risk group and the At Risk group played PC games more frequently than the Not at Risk group.
• Participants who played video-games daily (n=574) had significantly higher DSM-IV-J scores (p<.001) than their counterparts who did not play or played video games less frequently (M=0.41, SD=1.25 and M=0.16, SD=0.78 respectively).

Table 1 Gambling Groups in Relation to Frequency of Video-Game Play in the Past Year (adapted from Delfabbro et al., 2009).

Click image to enlarge, or adjust your browser's zoom setting.
*p<0.01 and **p<0.001
Effect sizes: η²: 0.01-0.06 = small effect size; 0.07-0.13 = moderate effect size; 0.14+ = large effect size

Limitations

• This survey relies on self-report.
• The researchers used a cross-sectional design that does not allow for exploration of causal relationships between video-game playing and gambling. This does not take common factors (e.g., interaction between players and technology, and structural characteristics of the game such as lighting and sounds) into consideration that might account for the association between video gaming and gambling.
• Because of the selection procedures, the participants might not be representative of their schools, their classes, youthful gamblers, or youthful video game players.
• The authors used the number of students as the total degrees of freedom in their statistical analyses.  However, the cluster sampling within schools yields fewer degrees of freedom and without adjustment, the power of the statistical tests is exaggerated.
• The DSM-IV-J items shift time frames (i.e., past year, presently, and lifetime).  As a result, it is not possible to determine if gambling-related problems are current or in the past.

Conclusion

The results indicated significant associations between various types of video-gaming and gambling-related problems. However, the effect sizes were very small; this suggests that frequent video-game playing accounts for only a small part of the relationship between video game playing and gambling-related problems among adolescents. The inconsistency within the results (e.g., PC games were protective of PG, but hand-held games did not show a difference between no risk and at risk) suggests that other factors might better explain the association between video games and gambling-related problems. Future research should consider exploring additional factors (e.g., social/family influences, personality, beliefs etc.) that extend beyond frequency of playing video games that may explain why some adolescents experience gambling-related problems.

-Tasha Chandler

What do you think?  Please use the comment link below to provide feedback on this article.

References

Delfabbro, P. H., King, D., Lambos, C., & Puglies, S. (2009). Is video-game playing a risk factor for pathological gambling in Australian adolescents? Journal of Gambling Studies, 25(3), 391-405.

Delfabbro, P. H., Lahn, J., & Grabosky, P. (2005). Adolescent gambling: A report on recent ACT research. Canberra: ACT Gambling and Research Commission.

Welte, J. W., Barnes, G. M., Tidwell, M.-C. O., & Hoffman, J. H. (2009). The association of form of gambling with problem gambling among American youth. Psychology of Addictive Behaviors, 23(1), 105-112.

The WAGER, Vol. 14(7) – College Gambling Activities: Gambling Groups by Games Played

Longitudinal studies have focused on adolescents’ impulsive activities and have found this to be a period during which youth explore and engage in gambling (Vitaro, Wanner, Ladouceur, Brendgen, & Tremblay, 2004; Wanner, Vitaro, Ladouceur, Brendgen, & Tremblay, 2006). However, very few longitudinal studies have focused on the gambling activities of young adults. This week, the WAGER reviews a study (Goudriaan, Slutske, Krull, & Sher, 2009) that examined clusters of gambling activities among college students over a four year period to examine whether there were changes in gambling behavior/participation over time.

Methods

• Incoming students from the University of Missouri-Columbia between the ages of 17.5 and 19.5 years old at baseline were recruited to participate in a longitudinal study during the summer prior to their freshman year.
• A subsequent online survey was administered to participants each spring semester.
  • 3,720 students completed one or more surveys: year 1 (n=2450); year 2 (n=2482); year 3 (n=2357); and year 4 (n=2250). 
• The survey measured participation in ten gambling activities each year: playing cards for money, betting on horses/dog/animal races, betting on sports, playing dice games, casino gambling, lottery, bingo, slot/poker/gambling machines, playing games of skill for money, and Internet gambling.
• The researchers used latent class analyses (LCAs) to derive gambling group membership defined by games played and stability of group membership over time.

Results

• LCAs identified four gambling groups: the low-gambling group gambled sporadically or not at all; the card gambling group included card, lottery, and non-regulated forms of gambling (e.g., sports betting and games of skill for money); the casino/slot gambling group engaged in more formal gambling (e.g., slot machine and casino); and the extensive gambling group engaged in all or most of the gambling activities.
• Table 1 shows the percentage of participants in each group at each year.

Figure 1: LCAs of Gambling Group Percentages each Year Including Number of Participants (adapted from Goudriaan et al., 2009).

Click image to enlarge, or adjust your browser's zoom setting.
*Please note that the figure encompasses students based on estimated frequencies including estimated data of students missing at one or more data collection points.

• Participants’ gambling activities changed over time. Transitions from group to group showed:
  • The gambling groups did not significantly change during the first two years of the study; this is a time when most students are legally not allowed in casinos.
  • A comparison of gambling behavior/participation between the second year and the fourth year yields many notable findings.
    • Membership decreased by 26.6% (59.7% to 43.8%) among the low gambling group and increased by 20.9% (5.3% to 6.7%) within the extensive gambling group.
    • Membership in the group that gambles in more formal venues (i.e., casino/slot gambling group) increased 94.9% (2.2% to 43.4%) – that is more than 13 times.
    • Membership decreased by 81.2% (32.9% to 6.2%) within the card gambling group as membership within the casino/slot gambling group increased, indicating a transition in preference from non-regulated to formal gambling.   

Limitations

• The results of this study might not generalize to all populations because the participants were recruited from a single-site and were predominantly white (90%).
• The gambling activities were ambiguous in terms of venue and activity. As a result, participants might have endorsed more than one activity for the same behavior (e.g., card gambling for money can happen with friends or in a casino).

Conclusion

This study’s findings indicate that gambling activities cluster into two main types among college students: non-regulated and formal.  In addition, gambling preferences change between the first two years of college and the second two years of college; during this time, most college students become of legal age to enter and gamble in casino environments. According to these findings, this newfound gambling opportunity significantly impacts student gambling preference (i.e., non-regulated to formal gambling). Being cognizant of these preference shifts toward gambling activities among college students should help researchers design better intervention/education programs. This study suggests that a flexible intervention program that changes as a student progresses through college would be most appropriate. Perhaps colleges should emphasize gambling education focusing on non-regulated form of gambling for underclassmen and gambling education about formal forms of gambling for upperclassmen.

- Tasha Chandler

What do you think?  Please use the comment link below to provide feedback on this article.

References

Goudriaan, A. E., Slutske, W. S., Krull, J. L., & Sher, K. J. (2009). Longitudinal patterns of gambling activities and associated risk factors in college students. Addiction, 104(7), 1219-1232.

Vitaro, F., Wanner, B., Ladouceur, R., Brendgen, M., & Tremblay, R. E. (2004). Trajectories of gambling during adolescence. Journal of Gambling Studies, 20(1), 47-69.

Wanner, B., Vitaro, F., Ladouceur, R., Brendgen, M., & Tremblay, R. E. (2006). Joint trajectories of gambling, alcohol and marijuana use during adolescence: A person- and variable-centered developmental approach. Addictive Behaviors, 31(4), 566-580.

The Wager Vol. (14)6 - BBGS vs. NODS-CLiP: Which Brief Screen for Pathological Gambling Wins the Battle of Psychometrics?

The process of screening for addictive behaviors is moving from research labs into the mainstream of public health practice (Anderson, Aromaa, Rosenbloom, & Enos, 2008). Most people and programs that consider administering public health screens face significant temporal and financial restrictions. Consequently, the development of brief screens that can identify the most people in need of treatment, without also generating many false positives, is critical. Among the general population, pathological gambling (PG) is a relatively rare disorder (Kessler et al., 2008; Petry, Stinson, & Grant, 2005). As a result, public health workers tend to screen for and detect PG less often than other more prevalent expressions of addiction, such as alcohol and other drug dependence. Until now, there were only two brief screens for detecting PG among the general population (i.e., the Lie/Bet Questionnaire, and the MAGS 7, respectively; Johnson et al., 1997; Shaffer, LaBrie, Scanlan, & Cummings, 1994). This week the WAGER compares two new, independent and nearly simultaneously published reports focusing on the development of brief screens for PG (Gebauer, LaBrie, & Shaffer, in press; Toce-Gerstein, Gerstein, & Volberg, in press).

Methods

• The Brief Bio-Social Gambling Screen (BBGS; Gebauer et al., in press)
  •  Gebauer et al. developed the BBGS using past year DSM-IV PG items from the Alcohol Use Disorder and Associated Disabilities Interview Schedule IV (AUDADIS-IV; Grant, Dawson et al., 2003) that was included within the National Epidemiological Survey on Alcohol and Related Conditions (NESARC; Grant, Moore, & Kaplan, 2003). 
    • The NESARC survey collected information from a United States nationally representative random sample of individuals (N=43,093) from the general household population.
    • Gebauer et al. targeted participants who endorsed five or more DSM-IV symptoms or signs as the group of pathological gamblers (PGs) to be distinguished from participants who failed to meet these criteria. 
  • The researchers used data analytic procedures, including step-wise entry, step-wise elimination, and combinations of minimal sets of DSM-IV criteria, to identify the subset of DSM-IV criteria that was sufficiently sensitive to identify PGs correctly, and specific enough to exclude non-PGs from being incorrectly classified as PGs (i.e., false positives).
• The NODS-CliP (Toce-Gerstein et al., in press)
  • Toce-Gerstein et al. (in press) developed the NODS-CLiP using lifetime DSM-IV PG items from the NORC Diagnostic Screen for Gambling Disorders (NODS; Gerstein et al., 1999). 
    • Researchers administered the NODS to participants (N=17,180) in eight different general adult population field studies.
    • Toce-Gerstein et al. targeted participants who endorsed five or more DSM-IV symptoms or signs as the group of PGs to be distinguished from participants who failed to meet these criteria. 
  • The researchers developed the NODS-CliP by analyzing how well each possible subset of 2-4 NODS items identified PG.

 Results

• Both the NODS-CLiP and the BBGS offer three items as a brief screen that can identify PGs from the general population.
  • NODS-CLiP 
    •  Loss of Control: Have you ever tried to stop, cut down, or control your gambling?
    • Lying: Have you ever lied to family members, friends or others about how much you gamble or how much money you lost on gambling?
    • Preoccupation: Have there been periods lasting 2 weeks or longer when you spent a lot of time thinking about your gambling experiences, or planning out future gambling ventures or bets?
  • BBGS
    • Withdrawal: During the past 12 months, have you become restless, irritable or anxious when trying to stop/cut down on gambling?
    • Deceive: During the past 12 months, have you tried to keep your family or friends from knowing how much you gambled?
    • Bailout/Need Money: During the past 12 months, did you have such financial trouble that you had to get help with living expenses from family, friends, or welfare?
• Psychometric Values
  • Table 1 summarizes the psychometric values for both the NODS-CLiP and the BBGS (i.e., Sensitivity, Specificity and Positive Predictive Values). 

Table 1. Comparing the Psychometric Values of the NODS-CLiP and the BBGS

*Click image to enlarge, or adjust your browser's zoom setting.

Limitations

• The NODS-CliP identified lifetime PG; the BBGS identified past-year PG.
  • Using a lifetime context for symptom clustering is problematic. This strategy yields more endorsed problems than past-year time frames for symptom identification; further, using a lifetime frame for screening is inconsistent with the process of making a clinical diagnosis. For example, research shows that gambling related problems evidence considerable waxing and waning from year to year (LaPlante, Nelson, LaBrie, & Shaffer, 2008). As a result, the NODS-CliP might not reflect current problems that often associate with treatment seeking. The BBGS features current problem identification. The different PPVs, in part, reflect this circumstance. 
  • Because the NODS-CLiP uses a lifetime context for symptom identification, the PGs who screened positive might not present as PGs currently. As a result, the NODS-CLiP’s sensitivity, its apparent strength (i.e., 0.99 for the NODS-CLiP vs. 0.96 for the BBGS), is uncertain for applications seeking to identify current PG.
• Both studies were limited because both teams of researchers selected PG items from those that currently were available in the DSM-IV.
  •  The DSM-IV criteria do not represent an exhaustive list of sequelae associated with PG and there is no empirical evidence that, despite its reliability, the DSM-IV includes conceptually valid diagnostic criteria. In fact, the efficacy of both screens serves to demonstrate that some of the DSM diagnostic criteria for PG fail to discriminate it adequately from those who do not suffer with this problem.
• Both studies are subject to the problems associated with data obtained exclusively from self-report.

Conclusions

Which screen wins? One clear advantage of the BBGS is the assessment of past year PG, which is consistent with clinical practice, compared to lifetime PG. Beyond this, determining the answer really depends on the goal of screening. For most clinical programs, time and money are precious resources that cannot be wasted. If an agency was to conduct follow-up evaluations with those identified as PGs by the NODS-CLiP, we would expect only one in eight to be identified correctly as a PG; however, of those identified as a PG by the BBGS, approximately one in three would be identified correctly as a PG. A screen that yields many false positives (e.g., 7 for every 8 screened), and does not maximize specificity, is best to garner liberal estimates of problems, but has limited clinical utility. Therefore, for clinical programs, a screen with superior specificity and high sensitivity makes the most sense (i.e., BBGS). Nevertheless, for researchers and others who simply want to identify the most PGs possible, without care for false positives, the slight sensitivity advantage evidenced by the NODS-CLiP might make it more attractive. Ultimately, you will have to decide the goal of screening, and then pick the measure that best achieves that goal.

What do you think?  Please use the comment link below to provide feedback on this article.

References

Anderson, P., Aromaa, S., Rosenbloom, D., & Enos, G. (2008). Screening and Brief Intervention: Making a Public Health Difference. Boston: Join Together.

Gebauer, L., LaBrie, R. A., & Shaffer, H. J. (in press). Optimizing DSM-IV classification accuracy: A brief bio-social screen for gambling disorders among the general household population. Canadian Journal of Psychiatry.

Gerstein, D., Murphy, S., Toce, M., Hoffmann, J., Palmer, A., Johnson, R., et al. (1999). Gambling Impact and Behavior Study: Report to the National Gambling Impact Study Commission. Chicago: National Opinion Research Center.

Grant, B., Dawson, D., Stinson, F., Chou, P., Kay, W., & Pickering, R. (2003). The Alcohol Use Disorder and Associated Disabilities Interview Schedule-IV (AUDADIS-IV): Reliability of alcohol consumption, tobacco use, family history of depression and psychiatric diagnostic modules in a general population sample. Drug & Alcohol Dependence, 71, 7-16.

Grant, B., Moore, T., & Kaplan, K. (2003). Source and Accuracy Statement: Wave 1 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism.

Johnson, E. E., Hamer, R., Nora, R. M., Tan, B., Eisenstein, N., & Engelhart, C. (1997). The Lie/Bet Questionnaire for screening pathological gamblers. Psychological Reports, 80, 83-88.

Kessler, R. C., Hwang, I., LaBrie, R. A., Petukhova, M., Sampson, N. A., Winters, K. C., et al. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine 38(9), 1351-1360.

LaPlante, D. A., Nelson, S. E., LaBrie, R. A., & Shaffer, H. J. (2008). Stability and progression of disordered gambling: Lessons from longitudinal studies. Canadian Journal of Psychiatry, 53(1), 52-60.

Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry, 66(5), 564-574.

Shaffer, H. J., LaBrie, R., Scanlan, K. M., & Cummings, T. N. (1994). Pathological gambling among adolescents: Massachusetts gambling screen (MAGS). Journal of Gambling Studies, 10(4), 339-362.

Toce-Gerstein, M., Gerstein, D., & Volberg, R. (in press). The NODS-CLiP: A rapid screen for adult pathological and problem gambling. Journal of Gambling Studies, epub ahead of print.

 

The WAGER, Vol. 14(5) - What’s in a game? Are certain types of gambling more likely to lead to disordered gambling?

At present, research findings are mixed about the impact of specific forms of gambling on gambling-related problems. Using limited samples, some studies have found an association between specific gambling activity and disordered gambling (e.g., Wood & Griffiths, 1998), often implicating gambling machines as especially “addictive” because of their high rate of gambling opportunities. Treatment seekers often report disproportionate rates of play on specific games. However, treatment seekers are not representative of others with similar problems. Therefore, it is not surprising that some researchers (e.g., National Research Council, 1999; Welte, Barnes, Wieczorek, Tidwell, & Parker, 2004) have found that the number of gambling activities is more predictive of problem gambling.  This week’s WAGER reviews a study that examines the relationship between specific forms of gambling and gambling-related problems among a nationally representative sample of U.S. youths (Welte, Barnes, Tidwell, & Hoffman, 2009).

Methods

• Participants were U.S. youths (ages 14-21) selected via random digit dialing of household telephone numbers who completed the National Survey of Youth and Gambling (N = 2,274).
• Interviews assessed past year gambling behavior and gambling-related problems.
  •  Investigators used a researcher-developed survey to assess participant involvement in 15 gambling activities or settings.
  • Investigators evaluated gambling-related problems using the South Oaks Gambling Screen Revised for Adolescents (SOGS-RA).

Results

• Sixty-eight percent of study participants (N=1,535) reported gambling during the previous twelve months; the highest percentage of these gamblers reported engaging in card games (33%), office pools and charitable gambling  (30%), and the lottery (29%).
• Table 1 summarizes the risks for gambling-related problems and gambling versatility (i.e., mean number of types of gambling in which respondents engaged) for each gambling activity/setting.
  • Overall, higher levels of gambling versatility were associated with more gambling-related problems.
• Researchers performed incidence risk ratio analyses to investigate the impacts of engaging in particular forms of gambling while controlling for all other types of gambling behavior.
  • After controlling for involvement in other games/settings, card games, games of skill, gambling at casinos, and ‘other’ gambling were associated with an increased risk of gambling-related problems.  
  • Other gambling types, including Internet gambling, were not associated with increased risk when controlling for involvement in other gambling activities.

Table 1.  Prevalence of Any Gambling Symptoms and Gambling Versatility on All Forms of Gambling for Gamblers Who Played Each Form of Gambling, U.S. Youth and Gambling Survey (Adapted from Welte et al., 2009)

  

Limitations

• Self-report of gambling activity involvement and potential problem gambling behavior can be associated with recall and reporting bias.
• Data was collected using random digit dialing of household phones, which only occasionally included cell phones.  This strategy generally limits the study sample to youths with a landline home telephone. This group might not be representative of the youth population segment.
• Study questions did not distinguish gambling settings from specific forms of gambling (e.g., questions were asked about casino gambling and about gambling machines not in a casino but not about gambling machines in casinos).

Conclusion
The findings from this study do not support the idea that the most rapid forms of gambling, such as gambling machines, are the most problematic type of game. The finding that gamblers who had higher gambling versatility (i.e., experience with more types of gambling) were more likely to experience gambling problems is, however, consistent with previous research (e.g., National Research Council, 1999; Welte et al., 2004) showing that overall gambling involvement is a more useful predictor of gambling-related problems than participation in specific gambling types.  A high correlation between a specific type of gambling (e.g., Internet gambling) and problem gambling does not reflect a causal influence; this association might mean that devotees of that type of gambling also engage in a lot of other gambling. Youth gambling prevention and harm reduction programs might benefit by incorporating the findings of this study into their practices.  Specifically, by focusing less on particular types of gambling and focusing more on the potential risks of high levels of gambling engagement, such programs might better prevent or reduce the harms associated with youth problem gambling behavior.

-Erica Marshall

What do you think?  Please use the comment link below to provide feedback on this article.

References
National Research Council. (1999). Pathological gambling: A critical review. Washington D.C.: National Academy Press.

Welte, J., Barnes, G., Tidwell, M., & Hoffman, J. (2009). The association of form of gambling with problem gambling among American youth. Psychology of Addictive Behaviors, 23(1), 105-112.

Welte, J., Barnes, G., Wieczorek, W., Tidwell, M., & Parker, J. (2004). Risk factors for pathological gambling. Addictive Behaviors, 29(2), 323-335.

Wood, R. T., & Griffiths, M. D. (1998). The acquisition, development and maintenance of lottery and scratchcard gambling in adolescence. Journal of Adolescence, 21(3), 265-273.

 

The WAGER, Vol. 14(4) - Pay Attention: Does Persistent ADHD Lead to Gambling Problems in Young Adults?

A number of studies (Carlton & Manowitz, 1992; Carlton et al., 1987; Rugle & Melamed, 1993) have shown an association between symptoms of Attention Deficit Hyperactivity Disorder (ADHD) in childhood and pathological gambling behavior during adulthood.  The majority of these studies have retrospective designs, and examine past ADHD symptoms after the development of pathological gambling behavior. These study designs are susceptible to considerable recall bias and leave the causal relationship between ADHD and problem gambling unclear. This week’s WAGER reviews a prospective study (Breyer et al., 2009), which attempts to illuminate the relationship between persistent ADHD and gambling behaviors from childhood through young adulthood .

Methods
The study sample included participants involved with a childhood longitudinal ADHD study (i.e., the Minnesota Competence Enhancement Program).

• At baseline (ages 7-11):
  • N (ADHD) = 318; N (Control) = 144
•  At follow-up (ages 18- 24):
  • N (ADHD) = 142, a 45% follow-up (FU) rate; N (Control) = 93, a 65% FU rate. 
• Significantly fewer ADHD cases than controls participated in the follow-up assessment (X² = 11.7, p = 0.008)
• Investigators assessed childhood ADHD status using the Revised Parent Version of the Diagnostic Interview for Children and Adolescents; at follow-up they assessed ADHD status using the Adolescent Diagnostic Interview.
• At follow-up, researchers identified two subgroups of ADHD cases, ADHD “persisters” (i.e., those with a diagnoses both at baseline and follow-up; N = 47) and ADHD “desisters” (i.e., those with a diagnoses at baseline but not at follow-up; N = 95).
• Researchers used the South Oaks Gambling Screen – Revised for Adolescents (SOGS-RA) to assess gambling behavior at follow-up.
  • Investigators categorized participants with scores of 0-1 as “social gamblers” and those with scores of 2+ as “possible problem gamblers” (PPGs).
• Researchers performed ANOVAs and Chi-square analyses to examine relationships between ADHD status and gambling outcome variables.

Results

• There were no significant subgroup differences in gambling participation or frequency between persisters, desisters, and controls.  
• Persisters had significantly higher rates of PPG. 
  • A significantly higher percentage of persisters (19.1%) met the criteria for PPG than desisters (5.3%) and controls (5.4%; X² (4) = 10.05; p < 0.05). See Graph 1.

Figure 1.  Gambling participation and problem severity by ADHD status
               (Adapted from Breyer et al, 2009)

Limitations

• Gambling behavior and ADHD status were assessed via self report.
• Sample was predominantly Caucasian (95%).
• Significant differential loss to follow-up between ADHD cases and controls between baseline and reassessment.
• Study did not examine ADHD co-occurring disorders, particularly externalizing disorders (e.g., those that effect impulsivity and self-control), which might confound the relationship between gambling behavior and ADHD.

Conclusion
Despite the limitations of this study, this research provides important longitudinal evidence about the relationship between persistent ADHD and disordered gambling. Among study participants, those with persistent ADHD have a higher prevalence of PPG during young adulthood; however, a causal relationship between ADHD and gambling problems remains uncertain.  We cannot conclude a causal relationship because the factors contributing to ADHD or the characteristics of persistent ADHD (i.e., potential mediators and moderators) might also increase vulnerability to PPG.  Further, the low occurrence of PPG makes it difficult to evaluate these potential mediators and moderators.  Future research will need to address the limitations of this study (e.g., larger, more ethnically/racially diverse sample, assessment of ADHD co-occurring disorders) to fully understand the causal relationship between persistent ADHD and disordered gambling behaviors.

- Erica Marshall
What do you think?  Please use the comment link below to provide feedback on this article.

References

Breyer, J. L., Botzet, A. M., Winters, K. C., Stinchfield, R. D., August, G., & Realmuto, G. (2009). Young Adult Gambling Behaviors and their Relationship with the Persistence of ADHD. Journal of Gambling Studies.

Carlton, P. L., & Manowitz, P. (1992). Behavioral restraint and symptoms of attention deficit disorder in alcoholics and pathological gamblers. Neuropsychology, 25, 44-48.

Carlton, P. L., Manowitz, P., McBride, H., Nora, R., Swartzburg, M., & Goldstein, L. (1987). Attention deficit disorder and pathological gambling. The Journal of Clinical Psychiatry, 48(12), 487-488.

Rugle, L., & Melamed, L. (1993). Neuropsychological assessment of attention problems in pathological gamblers. The Journal of Nervous and Mental Disease, 181(2), 107-112.

The WAGER, Vol. 14(3) - The early bird is even earlier: predicting gambling behavior in kindergarten

Youth gambling is a potential precursor and a possible indication of pathological gambling later in life (Pagani, Derevensky, & Japel, 2009; Proimos, DuRant, Pierce, & Goodman, 1998). In a prospective longitudinal study, Pagani et al. (2009) examine the correlation between early childhood (i.e., kindergarten) impulsive behavior and youth (i.e., sixth grade) gambling activities.

Methods

• Researchers recruited the children and parents from the kindergarten cohort from the Montreal Longitudinal Preschool Study
  • The authors followed 377 children exclusively selected from intact families
  • The final sample (n= 163) included children with complete youth and parental gambling involvement information from both kindergarten and sixth grade
• Youth Gambling was the dependent variable
  • Assessed via self-report by telephone interview
  • Measured via a 5-item scale to determine past year youth gambling involvement
  • How many times have you done the following: cards; bingo; bought lottery (instant or sports); played video games for money; placed bets at sports venues
  • Scoring: 0 (never) to 3 (once per week or more)
• Early Impulsive Behavior was the independent variable
  • Assessed via completion of the Social Behavior Questionnaire (assessment of children’s behavioral adjustment and current psychosocial adjustment) by children’s kindergarten teacher
  • Impulsivity was measured in a 9-item subscale (α=.91) (combination of inattentiveness, distractibility, hyperactive factors)
  • Scoring: reverse scoring of 1 (often or very true) to 3 (never or not true) followed by summation
• Researchers used two models to measure the correlation
  • The simple model used impulsivity alone
  • The controlled model combined impulsivity with the following covariates:
  • Gender: there are gender differences for ADHD and gambling
  • Maternal education: can represent other unmeasured parental characteristics and explains the home environment and its effect on youth behavior
  • Family dysfunction: correlated with ADHD and gambling
  • Parental gambling: if present it can easily be an example to children
  • Emotional distress (e.g., depression symptoms): ADHD and depression, and gambling and depression can coexist together

Results

• Early childhood impulsivity is significantly correlated to youth gambling
  • According to both models, a 1-unit increase in teacher-reported early childhood impulsivity associated with a corresponding 25% increase in later self-reported youth gambling activities (see Table 1)
• When controlling for impulsivity, others variables (i.e., maternal education, family dysfunction, parental gambling, emotionally distressed behavior (e.g., depression symptoms)) are not significant.

Table 1. Prediction of Youth Gambling Behavior by Early Childhood Impulsivity (from Pagani et al., 2009)

*Click image to enlarge, or adjust your browser's zoom setting.
a For the simple model, R2=0.06; for the fully controlled model, R2=0.16
b The CIs, based on unstandardized βs, are derived for the fully controlled model

Limitations

• Self-report of gambling activity (by parents and children at six-year follow-up)
• Proxy report of early childhood impulsivity by kindergarten teachers
• Study did not include clinical diagnoses of ADHD (or other disorders associated with gambling problems)
• Low follow-up rate (i.e., 48%)

Conclusion
Youth gambling is a public health issue. Youth gambling can also be a precursor to later pathological gambling (Pagani et al., 2009; Proimos et al., 1998). Youth gambling is a dangerous activity because it can lead to (1) situations that young adolescents might not have the experience or maturity to handle and (2) gambling related problems later in life. The study of kindergarten behavior could potentially lead to an over analysis of simple childhood activities (i.e., instead of appreciating a child’s exuberance, parents will start to fear the advent of compulsive gambling). However this study’s findings, that excessive childhood impulsivity has a possible correlation to youth gambling, are promising because these results might lead to better early prevention methods. Despite its limitations, Pagani et al.’s work provided important longitudinal evidence; it will be interesting to see if the suggested correlation between impulsivity and gambling will continue to manifest itself as the participants grow older.

- Ingrid Maurice

What do you think? Please use the comment link below to provide feeback on this article.

References
Pagani, L. S., Derevensky, J. L., & Japel, C. (2009). Predicting gambling behavior in sixth grade from kindergarten impulsivity: a tale of developmental continuity. Arch Pediatr Adolesc Med, 163(3), 238-243.
Proimos, J., DuRant, R. H., Pierce, J. D., & Goodman, E. (1998). Gambling and other risk behaviors among 8th- to 12th-grade students. Pediatrics, 102(2), 1-6.

The WAGER, Vol. 14(3) - Lottery Revenue Increases during Government Check Week: Coincidence?

“Wake up, wake up, wake up it's the 1st of the month
To get up, get up, get up, so cash your checks and get up”
    --Bone Thugs-n-Harmony, “1st of tha [sic] Month,” from the album E 99 Eternal

The timing of government checks, such as social security, coincides with check recipients’ increase in consumption of essential goods and services.  But receiving government checks also coincides with an increase in recipient spending on non-essential goods and services, such as entertainment expenditures (Stephens, 2003).  This week’s WAGER reviews a study, which examined whether the timing of government subsidy distribution corresponds with an increase in lottery revenues, and which type of game revenue tends to increase the most.

Methods

• Weinbach & Paul (2008) used lottery sales data from four years in seven states for different types of games (data from LaFleur, 1993; 1994; 1995; & 1996).
  • Pick 3 games were less expensive to play, and had small yet frequent payouts. 
  • Pick 5 games were more expensive to play, and had large yet infrequent payouts.
• In each state, government transfers of funds for three programs (i.e., disability, social security, and Aid to Families with Dependent Children) occurred during the first week of the month.
• The authors used multiple regression analyses to examine if there was a relationship between lottery revenue and weeks in which the government distributed these checks.

Results

• Table 1 shows that in all 7 states, lottery revenues increased significantly during the first week of the month for Pick 3 games.
• Table 1 also shows that in 6 of 7 states, lottery revenues did not correspondingly increase for Pick 5 games.

Table 1. Pick 3 and Pick 5 revenue change during the 1st week of the month (Adapted from Weinbach et al., 2008).

*p value <.05
**p value <.01

Limitations

• Data was available for both Pick 3 and Pick 5 games in only seven states, so the number of states available for comparison is limited.
• This study examined large consumer trends; therefore it cannot distinguish recipients of government payments from other consumers.
  • For instance, some employers also could distribute paychecks during the first week of the month.  If so, the increase in Pick 3 revenues during the first week of the month might be attributed to both government and employer check distribution. 
  • However, there were no lottery revenue increases during the middle of the month, another common time for employers to distribute checks.  Therefore we can assume that first week of the month lottery revenue increases are partially due to government check distribution.

Conclusion
Weinbach and Paul (2008) found an increase in Pick 3 lottery expenditures during the first week of the month; this result is consistent with the finding that government check distribution occurs in tandem with increased spending on essential and non-essential goods and services (Stephens, 2003).  However, the authors did not observe such an increase in Pick 5 expenditures.  Taken together, these findings indicate that among individuals on a fixed interval reinforcement schedule (such as those receiving government subsidies), games with more frequent, less prolific payouts (e.g., Pick 3 games) more strongly reinforce the purchase of subsequent tickets than games with less frequent, more prolific payouts (e.g., Pick 5 games).

What do you think? Comments can be addressed to Leslie Bosworth.

References

LaFleur, T. (1993). LaFleur's 1993 World Lottery Almanac. Boyds, Maryland: TLF Publications, Inc.

LaFleur, T. (1994). LaFleur's 1994 World Lottery Almanac. Boyds, Maryland: TLF Publications, Inc.

LaFleur, T. (1995). LaFleur's 1995 World Lottery Almanac. Boyds, Maryland: TLF Publications, Inc.

LaFleur, T. (1996). LaFleur's 1996 World Lottery Almanac. Boyds, Maryland: TLF Publications, Inc.

Stephens, M. (2003). 3rd of the month: Do social security recipients smooth consumption between checks? American Economic Review, 93(1), 406-422.

Weinbach, A. P., & Paul, R. J. (2008). Running the Numbers on Lotteries and the Poor: An Empirical Analysis of Transfer Payment Distribution and Subsequent Lottery Sales. Atlantic Economic Journal, 36, 333-344.