Abstract
Background and Objectives
Problem gambling impacts substance use disorder (SUD) treatment outcomes. To support early intervention, this study investigated gambling behaviors associated with problem gambling among SUD patients.
Methods
Patients (N = 593) from 19 residential SUD treatment facilities in California were surveyed about gambling behaviors and services. Patients identified as “problem gamblers” screened positive for gambling disorder or reported gambling behaviors at high risk for progression to gambling disorder. Multivariate regression analyses compared problem gamblers to nonproblem gamblers on specific gambling behaviors and receipt of gambling services.
Results
Of the total sample, 223 (37.6%) gambled in the past year. Among past‐year gamblers, those meeting problem gambling criteria (n = 106) were more likely than nonproblem gamblers to endorse betting on “games of skill” including cards (adjusted odds ratio [AOR] = 2.35, 95% confidence interval [CI] = 1.46, 3.78) and participating in online/electronic gambling (AOR = 5.61, 95% CI = 2.27, 13.88) and casino games (AOR = 2.30, 95% CI = 1.42, 3.72). Patients with problem gambling reported more gambling activity types than nonproblem gamblers (mean difference = 1.82, 95% CI = 1.59, 2.05). Only 35% of patients with problem gambling received gambling services from the SUD treatment program.
Discussion and Conclusions
Specific gambling behaviors increase risk of problem gambling for SUD patients. SUD treatment programs should screen for and address problem gambling.
Scientific Significance
SUD patients may have lower risk thresholds for gambling disorder. This study provides new data on specific behaviors associated with problem gambling among SUD patients.
INTRODUCTION
People with substance use disorders (SUD) often have other addictive behaviors including problem gambling. 1 Patients in SUD treatment who have problem gambling behavior also may have worse SUD outcomes because they tend to have higher SUD severity 2 and report increased impulsivity 3 and substance cravings 4 compared to those without comorbid problem gambling. Among patients receiving methadone, for instance, problem gambling was associated with increased substance use during treatment and early treatment discontinuation. 5 Substance use and gambling are also mutually reinforcing behaviors, such that patients in SUD treatment report increased gambling while using substances and increased substance use while gambling. 6 Addressing problem gambling among patients in SUD treatment can improve treatment outcomes, including less substance use and lower psychological distress 7 following gambling interventions for those with gambling comorbidity. 8
The Diagnostic and Statistical Manual of Mental Disorders (DSM‐5) identifies gambling disorder as meeting at least four of nine DSM‐5 criteria which include symptoms such as persistent need to gamble, difficulty interrupting gambling, and chasing gambling losses. 9 While useful for diagnostic purposes, the narrow focus of DSM‐5 gambling disorder criteria may fail to capture the broad spectrum of gambling behaviors that can impact SUD recovery. People with SUD who display problem gambling behaviors are at high risk of progressing to gambling disorder without clinical intervention. 10 To capture the full scope of gambling risks, “problem gambling” is often examined in lieu of gambling disorder. 11 , 12
Problem gambling has been defined as “all gambling behavior patterns that compromise, disrupt, or damage personal, family, or vocational pursuits.” 13 This broad category includes both gambling disorder and subthreshold risky gambling behavior. This study focused broadly on problem gambling using behavioral markers of gambling intensity associated with gambling‐related harms. Among people with SUD, meta‐analyses report pooled prevalence rates for problem gambling of 19% 1 compared to less than 2% for general population samples. 14
Because patients with comorbid SUD and problem gambling are more likely to seek treatment for SUD than for gambling, 15 SUD treatment presents a strategic opportunity to identify and treat problem gambling. However, few SUD treatment programs address problem gambling due to factors such as lack of workforce training and funding to provide gambling services. 16
One barrier to effective screening and treatment for problem gambling in SUD care may be identification of who is most at risk. A recent meta‐analysis found that individuals with SUD who were male, had depressive and mood disorders, and met criteria for alcohol, tobacco, or cannabis use disorders had increased odds of problem gambling. 1 Other studies have identified more stimulant use 5 , 17 and more psychological distress 17 among patients with problem gambling, compared to those without.
Specific gambling behaviors may also increase the risk for problem gambling. In general population surveys, higher gambling involvement (participation in more than one gambling activity type) and intensity (more frequent gambling, higher expenditure) were associated with problem gambling. 18 Additionally, gambling activity types and formats influence problem gambling risk. Casino gambling, especially skill‐based games (where the player applies strategy to influence outcomes), was found to predict problem gambling. 19 Online gambling also predicts problem gambling due to widespread availability and the potential for continuous play, especially for electronic card games (e.g., online poker) and sports betting. 20
Less is known about which gambling behaviors are riskiest for patients in SUD treatment. In one study, patients in residential SUD treatment who participated in a higher number of gambling activities were more likely to meet criteria for problem gambling. 21 In another study, patients receiving methadone who engaged in problem gambling behaviors reported higher expenditures than nonproblem gamblers when using electronic gambling machines (EGMs—e.g., slot machines) and casino table games. 22 However, these studies were conducted before the recent expansion of online gambling and focused on casino gambling rather than a full range of gambling activities and contexts.
This study investigated gambling activities and behaviors associated with problem gambling among patients in residential SUD treatment. Rather than focusing solely on gambling disorder, the study examined a spectrum of behaviors as risk factors for gambling‐related harms that could potentially impact SUD recovery. The study also explored differences in receipt of gambling services among patients who engaged in problem gambling versus those who did not. It was hypothesized that patients with problem gambling would be more likely than nonproblem gamblers to endorse stimulants as their primary substance; participate in multiple gambling activity types; and endorse casino and online/electronic gambling. It was also expected that patients with problem gambling would receive gambling‐related treatment services more often than those without problem gambling.
METHODS
Sample
The study sample was drawn from 19 residential SUD treatment facilities across California participating in a parent study about tobacco‐free policies in residential SUD programs. All programs were state‐licensed, publicly funded and served mostly low‐income patients covered by Medi‐Cal, California's Medicaid program. Program recruitment procedures for the parent study are described in Mccuistian et al., 2026. 23
Data collection
Beginning in 2023, patients completed structured cross‐sectional surveys. All patients enrolled in the residential program at the time of data collection were eligible to participate. For most programs, patient surveys were administered in person during scheduled site visits. Each participant received an iPad displaying the IRB‐approved study information sheet, which was reviewed by a team member with opportunities for questions. After providing consent via the iPad by clicking a checkbox, participants completed a 30‐min survey. Unique study IDs were applied to each survey to preserve anonymity. Each participant received a $20 gift card.
For one program, data collection began with in‐person administration, but due to scheduling conflicts the research team left printed materials with program staff to increase participation. The research team also mailed study materials to two programs that requested remote participation. Program staff distributed these materials to patients, who consented via checking a box and completed the paper surveys. Program staff then distributed gift cards to participants and returned completed materials to the research team within 1 week.
All study procedures were approved by the Institutional Review Board of the University of California, San Francisco.
Measures
Full sample
Demographics and other characteristics
All study participants were asked about past 12‐month gambling status (yes/no), receipt of gambling information or services in current treatment program (yes/no/I don't know) and demographics including age, gender (male, female, other), race/ethnicity (Hispanic or Latino, non‐Hispanic Black, non‐Hispanic White, multiracial/other), education (less than high school, high school/GED, more than high school), and sexual orientation (heterosexual, gay or lesbian, bisexual, something else). Surveys also asked about health insurance (yes/no), reason for treatment (SUD, substance use and mental health, mental health disorder/other), and primary drug (alcohol, stimulants, marijuana/cannabis, opioids, other).
Past‐year gamblers
Participants who endorsed gambling during the previous 12 months answered additional questions about gambling behaviors in the past 12 months, including frequency of gambling (ranging from less than once to daily), income lost to gambling (ranging from 5% to 30% or more of income), and emotional, social, and financial problems caused by gambling.
Problem gambling
The researchers used a subset of these questions to create a “problem gambling” measure that expands upon the DSM‐5 gambling disorder criteria to also capture individuals at high risk of progression to gambling disorder. 17 The composite measure used three criteria: screening positive on the Brief Biosocial Gambling Screen (BBGS) 24 OR having lost at least 5% of total income to gambling in the past 12 months OR gambling 3 or more days per week during the past 12 months. Gambling involvement (number of gambling activity types) was not included because involvement thresholds for problem gambling remain unclear. 25
The BBGS includes three questions derived from DSM‐5 criteria for gambling disorder, and a “yes” to any constitutes a positive screen. The BBGS identifies gambling disorder with high sensitivity (0.96) and specificity (0.99) and demonstrated greater accuracy for SUD patients as compared to similar gambling screens. 26
The “five percent lost income” threshold, which was the lowest nonzero response for the “income lost” item in the survey, was more conservative than low‐risk gambling thresholds established for general population samples, which range from 1% to 2% of personal income. 27 , 28
The gambling frequency cutoff of 3 or more days per week was also more conservative than thresholds established for the general population, which range from less than once to twice weekly). 12 , 28 This conservative threshold excluded participants with low‐risk behaviors, such as purchasing a weekly lottery ticket.
Gambling activities
Past‐year gamblers were asked about specific gambling activities and formats during the previous 12 months including instant win, scratch‐off, or daily lotteries; lottery tickets; bingo; cards or board games for money; VLTs (video lottery terminals) outside a casino; casino games other than VLTs or slot machines; internet/online/electronic gambling; horse racing; sports betting (including fantasy sports); stock market; betting on games of skill; and other. Response options were informed by evidence‐based categorization of gambling activities by chance‐ versus skill‐based formats 29 and higher‐ versus lower‐risk contexts. 18
Due to small cell sizes for certain gambling activity variables, closely related activities were merged to create broader categories. For example, both card and board games are often categorized as “games of skill” because they require the player to apply strategy or “pick a winner” as opposed to “games of chance” such as slot machines or the lottery. 30 The researchers therefore created composite categories of “chance” and “skill.” Sports betting remained separate due to literature reports of increased risk for problem gambling. 20 Gambling activity variables in the final analysis were games of chance (scratch‐off, lottery, bingo, VLTs); cards/board games/other games of skill; casino; online/electronic; sports betting; and other.
Data analysis
The study first examined descriptive data on gambling status and behaviors for all cases (N = 593). Among past‐year gamblers (N = 223), patients meeting problem gambling criteria were compared to nonproblem gamblers on demographics, health insurance, reason in treatment, and primary drug.
Next, patients meeting problem gambling criteria were compared to nonproblem gamblers on gambling activities, gambling involvement, and receipt of gambling treatment services. For all univariate comparisons, chi‐square or Fisher's Exact tests were used for categorical variables, and t‐tests for continuous variables.
Multivariate regression analyses then assessed whether unadjusted differences remained significant after adjusting for other factors. Logistic regression models were used for dichotomous outcomes and linear regression models for continuous outcomes. All regression models adjusted for demographics (age, race/ethnicity, education) which differed between patients meeting problem gambling criteria and non‐problem gamblers at p ≤ .05 in univariate analyses. All models also adjusted for nesting of participants within clinics.
RESULTS
Full sample
In the full sample (N = 593), mean age was 38.5 years and about 26% identified as female (Table 1). Nearly 39% identified as Hispanic or Latino, followed by about 30% non‐Hispanic White, 17% non‐Hispanic Black, and 15% multiracial or “other.” While 36% percent had completed some postsecondary education, 26% had completed less than high school. About 12% identified as non‐heterosexual. Sixty‐five percent endorsed being in treatment for SUD only, and 21% reported being in treatment for both SUD and mental health disorders. Stimulants were endorsed as the primary drug of use by nearly 38% of the sample, followed by alcohol (22%) and opioids (20.6%). Of the full sample (N = 593), 37% endorsed past‐year gambling (Table 2).
Table 1.
Demographics, reason in treatment, and primary drug among SUD treatment patients by problem gambling status (past‐year gamblers only) (N = 223).
| Mean (SD) or n (%) | ||||
|---|---|---|---|---|
| All participants (N = 223) | No problem gambling (n = 117) | Problem gambling (n = 106) | p ‐valuea | |
| Age | 38.5 (11.0) | 40.2 (11.3) | 36.7 (10.4) | .0181 |
| Gender | .4530 | |||
| Male | 157 (70.4) | 79 (67.5) | 78 (73.6) | |
| Female | 59 (26.5) | 33 (28.2) | 26 (24.5) | |
| Other | 7 (3.1) | 5 (4.3) | 2 (1.9) | |
| Race/ethnicity | .0023 | |||
| Hispanic or Latino | 86 (38.7) | 38 (32.5) | 48 (45.7) | |
| Non‐Hispanic Black | 38 (17.1) | 16 (13.7) | 22 (21.0) | |
| Non‐Hispanic White | 65 (29.3) | 47 (40.2) | 18 (17.1) | |
| Multiracial/Other | 33 (14.9) | 16 (13.7) | 17 (16.2) | |
| Education | .0076 | |||
| Less than high school | 58 (26.0) | 21 (18.0) | 37 (35.0) | |
| High school/GEDb | 84 (37.7) | 45 (38.5) | 39 (36.8) | |
| More than high school | 81 (36.3) | 51 (43.6) | 30 (28.3) | |
| Sexual orientation | .8215 | |||
| Straight/Heterosexual | 191 (88.0) | 100 (88.5) | 91 (87.5) | |
| Gay/lesbian/bisexual/something else | 26 (12.0) | 13 (11.5) | 13 (12.5) | |
| Has health insurance | 217 (97.3) | 116 (99.1) | 101 (95.3) | .0751 |
| Reason in treatment | .1171 | |||
| Substance use disorder | 146 (65.5) | 73 (62.4) | 73 (68.9) | |
| Both substance use and mental health disorders | 47 (21.1) | 23 (19.7) | 24 (22.6) | |
| Mental health disorder/other | 30 (13.5) | 21 (18.0) | 9 (8.5) | |
| Primary drug of use | .0965 | |||
| Alcohol | 49 (22.0) | 32 (27.4) | 17 (16.0) | |
| Stimulants | 84 (37.7) | 39 (33.3) | 45 (42.5) | |
| Marijuana/Cannabis | 16 (7.2) | 6 (5.1) | 10 (9.4) | |
| Opioids | 46 (20.6) | 22 (18.8) | 24 (22.6) | |
| Other | 28 (12.6) | 18 (15.4) | 10 (9.4) | |
p‐values are reported for comparisons of demographic variables by past‐year problem gambling status. From Chi‐Square tests/Fisher's Exact test for categorical variables and t‐test for the continuous variable. Bolded values represent p ≤ .05.
General Educational Development (GED) is a high school equivalency diploma program in the United States.
Table 2.
Gambling behaviors among SUD treatment patients (N = 593).
| All participants (N = 593) | Mean (SD) or n (%) |
|---|---|
| Gambled in past 12 months | 223 (37.6) |
| Past‐year gamblers (N = 223) | |
| Brief Biosocial Gambling Screen (BBGS)a | |
| Became restless, irritable, or anxious when trying to stop/cut down on gambling in past 12 months | 40 (18.0) |
| Tried to keep family/friends from knowing how much gambled | 44 (19.9) |
| Had to get help with living expenses due to financial trouble from gambling | 37 (16.7) |
| Gambling Intensity and Frequency | |
| How much money lost gambling in past 12 months | |
| 5% of income | 82 (47.4) |
| 6%–9% of income | 20 (11.6) |
| 10%–20% of income | 15 (8.7) |
| 21%–30% of income | 11 (6.4) |
| 30% or more of income | 14 (8.1) |
| Won more money than lost | 31 (17.9) |
| How often gambled for money in the past 12 months | |
| Not at all | 44 (19.7) |
| Less than once a month | 66 (29.6) |
| 1‐3 days per month | 50 (22.4) |
| 1‐2 days per week | 25 (11.2) |
| 3‐6 days per week | 17 (7.6) |
| Daily | 21 (9.4) |
| Number of gambling activity types in the past 12 months | |
| None | 29 (13.0) |
| One | 109 (48.9) |
| Two | 40 (17.9) |
| Three | 24 (10.8) |
| Four | 7 (3.1) |
| Five or more | 14 (6.3) |
| Gambling activities and formats | |
| Instant win, scratch‐off, or daily lotteries | 102 (45.7) |
| Lottery tickets | 87 (39.0) |
| Bingo | 33 (14.8) |
| Cards/board games for money | 25 (11.2) |
| Video lottery terminals (VLTs) outside a casino | 11 (4.9) |
| Casino games other than VLTs or slot machines | 56 (25.1) |
| Internet/online/electronic gambling | 34 (15.3) |
| Horse racing | 9 (4.0) |
| Sports betting (including fantasy sports) | 23 (10.3) |
| Stocks | 18 (8.1) |
| Betting on games of skill | 22 (9.9) |
| Other | 17 (7.6) |
| Received information or counseling about how gambling can affect SUD recovery | 74 (33.3) |
| Met composite criteria for problem gambling | 106 (47.5) |
A response of “yes” to any of the BBGS items constitutes a positive screen for gambling disorder (Gebauer et al., 2010).
Past‐year gamblers
Differences between problem gamblers and non‐problem gamblers
Univariate analyses showed some demographic differences between past‐year gamblers who met problem gambling criteria (n = 106) versus those identified as nonproblem gamblers (n = 117) (Table 1). Patients meeting problem gambling criteria were younger (p = .0181) than nonproblem gamblers and had less education (p = .0076). Problem gambling also was associated with race/ethnicity (p = .0023), such that patients meeting problem gambling criteria identified less often as non‐Hispanic White compared to non‐problem gamblers. The two groups did not differ on gender, sexual orientation, health insurance status, reason for treatment, or primary drug.
Gambling behaviors
Among past‐year gamblers (n = 223), about 20% screened positive on the BBGS (Table 2). The most frequently endorsed item was trying to keep family or friends from knowing how much the participant had gambled (19.9%); followed by experiencing restlessness, irritability, or anxiety when trying to control their gambling (18%); and having to get help with living expenses due to financial trouble from gambling (16.7%).
About 47% of past‐year gamblers reported losing at least 5% of their income in the past year. About 28% reported weekly or more frequent gambling, with 9.4% endorsing daily gambling. About 20% reported no gambling for money in the past year. This could signify either gambling for nonmonetary rewards (e.g., possessions or favors), or that they mistakenly endorsed being a past‐year gambler. Most past‐year gamblers (48.9%) participated in only one gambling activity type in the past year, while 38% participated in two or more types. Thirteen percent reported no activity types, which could either mean they misunderstood the question or mistakenly endorsed being a past‐year gambler. Conflicting responses (e.g., reporting past‐year gambling activities and also reporting no activity types) were not included in the problem gambling calculation.
The most popular gambling format was lottery tickets or instant win (46%), followed by lottery tickets (29%), casino games (25%), and internet/electronic gambling (15%). Sports betting and betting on other games of skill were each endorsed by about 10% of past‐year gamblers.
Thirty‐three percent of past‐year gamblers reported receiving information or counseling about how gambling could affect their SUD recovery.
Problem gambling group
When composite “problem gambling” criteria were applied, 47.5% of past‐year gamblers (and 17.9% of the full sample) met criteria (Table 2). Included were past‐year gamblers who screened positive on the BBGS; or had lost at least 5% of total income to gambling in the past 12 months; or had gambled 3 or more days per week during the past 12 months.
Gambling activities and service receipt
Compared to nonproblem gamblers, the problem gambling group more often endorsed past‐year participation in cards/board games/other games of skill (p = .0242), online/electronic gambling (p = .0002), and casino games (p = .0961) (Table 3). Patients meeting problem gambling criteria also participated in nearly twice as many gambling activities over the past year compared to nongamblers (p = .0003). There was no difference in receipt of gambling services between the two groups.
Table 3.
Gambling behaviors and services among problem versus nonproblem gamblers (N = 223).
| Mean (SD)/n (%) | p valuea | ||
|---|---|---|---|
| No problem gambling | Problem gambling | ||
| All past year gamblers (N = 223) | n = 117 | n = 106 | |
| Gambling activities | |||
| Games of chance (scratch‐off, lottery, bingo, VLT) | 80 (68.4) | 81 (76.4) | .1809 |
| Cards/board games/other games of skill | 15 (12.8) | 26 (24.5) | .0242 |
| Casino | 24 (20.5) | 32 (30.2) | .0961 |
| Online/electronic | 8 (6.8) | 26 (24.5) | .0002 |
| Sports betting | 9 (7.7) | 14 (13.2) | .1763 |
| Other | 17 (14.5) | 21 (19.8) | .2949 |
| Number of gambling activities endorsed | |||
| M (SD) | 1.3 (1.0) | 2.1 (2.0) | .0003 |
| Gambling treatment | |||
| Received info or counseling | 37 (31.9) | 37 (34.9) | .6347 |
From Chi‐Square tests for categorical variables and t‐tests for continuous variables. Bolded values represent p ≤ .10.
In multivariate regression analyses comparing the two groups on gambling activities (Table 4), the problem gambling group was more likely to endorse betting on “games of skill” including cards (adjusted odds ratio [AOR] = 2.35, 95% confidence interval [CI] = 1.46, 3.78) and participating in online/electronic gambling (AOR = 5.61, 95% CI = 2.27, 13.88) and casino games (AOR = 2.30, 95% CI = 1.42, 3.72). The problem gambling group also reported a higher mean number of past‐year gambling activities in comparison to nonproblem gamblers (mean difference = 1.82, 95% CI = 1.59, 2.05).
Table 4.
Comparison of gambling behaviors among SUD treatment patients by problem gambling status.
| Past‐year problem gambler versus nonpast year problem gambler (ref) a | ||
|---|---|---|
| OR (95% CI) b | p c | |
| Cards/Boards/Other games of skill | 2.35 (1.46, 3.78) | .0004 |
| Online/electronic | 5.61 (2.27, 13.88) | .0002 |
| Casino games | 2.30 (1.42, 3.72) | .0007 |
| Mean difference (95% CI)b | ||
| Number of gambling activities | 1.82 (1.59, 2.05) | <.0001 |
Abbreviations: CI, confidence interval; OR, odds ratio.
Adjusted for demographics (age, race/ethnicity, education); and also controlled for nesting of participants within clinics.
Presented are odds ratios for categorical data and mean differences for continuous data.
Bolded values represent p ≤ 0.05.
DISCUSSION
Comorbid problem gambling is highly prevalent among patients in SUD treatment and impacts SUD recovery outcomes. This study contributes new data on gambling activities and contexts associated with increased risk for problem gambling among patients in SUD treatment. These findings support early identification of risky gambling behaviors and clinical interventions for problem gambling among this population.
Past‐year gambling prevalence in the sample (37.6%) was lower than the global pooled prevalence rate for adults in a recent meta‐analysis (46.2%). 14 The prevalence of problem gambling behaviors in the sample (17.9%) was more than double that of the adult problem gambling prevalence found in that study (8.7%), 14 and similar to the 19% problem gambling prevalence rate among adults with SUD from a different meta‐analysis focused on comorbidity. 1 Among past‐year gamblers in the current sample, nearly 8% gambled more than twice a week, compared to 9.5% of US adults who gambled twice or more per week. 31 Notably, 9.4% of past‐year gamblers in the current sample reported daily gambling. Additionally, 8% of past‐year gamblers reported 30% income losses to gambling, which well exceeds the 1% limit for low‐risk gambling often used in gambling research. 27
Among past‐year gamblers in the study sample, patients who met problem gambling criteria were more likely to endorse engaging in games of skill. In general population surveys, participation in games of skill is also associated with higher rates of problem gambling. 30 One reason may be that games of skill foster a greater sense of control than games of chance. This “illusion of control” is associated with gambling pathology. 32 However, the level of risk for specific gambling activities may differ according to the format in which the activity is provided, such that even “games of chance” such as slot machines or lottery may become riskier when provided through electronic continuous play formats. Among individuals seeking treatment for gambling disorder, use of electronic gaming machines (i.e., slot machines, video lottery terminals) was associated with more severe gambling disorder. 33
Gambling activity has moved from venue‐based to online over the past several years as online formats have expanded. Online formats may encourage problem gambling due to immediate accessibility via cell phones and structural characteristics such as instant deposit/cash out and micro‐bets that foster continuous engagement and a perception of low investment/low risk activity. 34 In the study sample, online formats were strongly associated with meeting problem gambling criteria, even when the activity was not specified. This finding may be especially relevant for clinical interventions in SUD treatment since patients in this study had access to cell phones and could have continued gambling while in residential treatment. The availability of online gambling during residential SUD treatment may hinder treatment goals and warrants further study.
This study identified a direct positive association between gambling involvement (number of gambling activities) and problem gambling. This finding reflects general population surveys showing that participation in multiple gambling activities predicted problem gambling more strongly than gambling types, despite risk differentials associated with specific activities. 35
Casino gambling was also associated with problem gambling in the study sample. This was expected given previously identified associations between casino proximity and higher gambling intensity among SUD patients with comorbid problem gambling. 21 , 22 Casinos may encourage problem gambling through multiple continuous play and skilled gaming formats which encourage individuals to increase their gambling involvement quickly.
In this study, contrary to hypothesis, there was no difference in primary drug between problem and nonproblem gamblers. This finding differed from previous research on patients in residential SUD treatment which identified higher rates of stimulant use in those who gambled compared to those who did not. 17
Study limitations include a sample drawn from residential SUD programs in a single US state, which may limit generalizability. The researchers also were unable to collect longitudinal data to explore the relationship between gambling problems and length of stay or SUD treatment outcomes and lacked access to medical charts to compare data such as SUD severity or other symptom scores. While the study's measure of problem gambling encompasses evidence‐based thresholds of gambling risks using the BBGS and may offer a more sensitive measurement of risky gambling behavior than the DSM‐5 criteria for gambling disorder, it has not been psychometrically validated to date and warrants future research. Gambling involvement analyses may have been impacted by collinearity or different interpretations of response options by study participants (e.g., one might endorse both “casino gambling” and “online/electronic gambling” if they participated in online “casino style” games). “Games of skill” and “games of chance” are also common but contested categories in gambling research, considering that gambling activities can be categorized in diverse ways depending on temporal and other structural characteristics, as well as the gambler's perception of how their own skill can influence the outcome of even “games of chance” such as lotteries. 29 Additionally, there were some inconsistencies in survey responses on gambling behaviors such that some respondents both specified past‐year gambling activities and also reported (in a different question) that they had participated in zero types of gambling in the past year. While the analysis plan prioritized endorsement of specific gambling activities in the past year for inclusion in the “past‐year gambling” sample, and cases with inconsistent responses were excluded from the “problem gambling” subsample, it is important to acknowledge that they introduce some ambiguity. Despite these limitations, this study presents new data on gambling activities associated with problem gambling among SUD treatment patients that can support identification of at‐risk gambling behaviors in SUD care.
CONCLUSIONS
People with SUD have disproportionate rates of problem gambling. Study findings indicate that patients in SUD treatment who engage in specific gambling activities are at higher risk for problem gambling, which may impact SUD recovery. Screening and clinical interventions for gambling, therefore, are important secondary or tertiary prevention strategies for SUD treatment. To support early intervention, payers should reimburse SUD treatment providers for gambling screening and services, even for risky gambling behaviors that are subthreshold for gambling disorder. At the workforce level, SUD clinicians should be trained in screening and treatment of problem gambling. Residential SUD treatment programs should establish clear policies limiting gambling during treatment, including participation in gambling activities during recreational time and via personal electronic devices. Improved detection and treatment of problem gambling among patients in SUD treatment will support better recovery outcomes and reduce harms associated with both addictive disorders.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
ACKNOWLEDGMENTS
This work was supported by the University of California Tobacco‐Related Disease Research Program (T32KT5241) and the National Institute on Drug Abuse (K23DA060073).
Pagano A, Lisha NE, Le JT, McCuistian C. Gambling behaviors associated with problem gambling among patients in treatment for substance use disorder. Am J Addict. 2026;35:748‐756. 10.1111/ajad.70203
[Correction added on 15 October 2026, after first online publication: The bolded p‐values in Tables 1 and 3 have been revised in this version.]
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1. Armoon B, Griffiths MD, Fleury M‐J, et al. The global prevalence of problem and pathological gambling and its associated factors among individuals with substance use disorders: a meta‐analysis. Int J Mental Health Addict. 2025;23(2):1241‐1274. 10.1007/s11469-023-01167-y] [DOI] [Google Scholar]
- 2. Rush BR, Bassani DG, Urbanoski KA, et al. Influence of co‐occurring mental and substance use disorders on the prevalence of problem gambling in Canada. Addiction. 2008;103(11):1847‐1856. 10.1111/j.1360-0443.2008.02338.x] [DOI] [PubMed] [Google Scholar]
- 3. Krmpotich T, Mikulich‐Gilbertson S, Sakai J, et al. Impaired decision‐making, higher impulsivity, and drug severity in substance dependence and pathological gambling. J Addict Med. 2015;9(4):273‐280. 10.1097/ADM.0000000000000129] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Di Nicola M, Tedeschi D, De Risio L, et al. Co‐occurrence of alcohol use disorder and behavioral addictions: relevance of impulsivity and craving. Drug Alcohol Depend. 2015;148:118‐125. 10.1016/j.drugalcdep.2014.12.028] [DOI] [PubMed] [Google Scholar]
- 5. Ledgerwood DM, Downey KK. Relationship between problem gambling and substance use in a methadone maintenance population. Addict Behav. 2002;27(4):483‐491. 10.1016/S0306-4603(01)00187-3] [DOI] [PubMed] [Google Scholar]
- 6. Thomas Sherba R, Martt NJ. Overall gambling behaviors and gambling treatment needs among a statewide sample of drug treatment clients in Ohio. J Gambl Stud. 2015;31(1):281‐293. 10.1007/s10899-013-9406-5] [DOI] [PubMed] [Google Scholar]
- 7. Baxley C, Weinstock J, Rash CJ, et al. Changes in secondary outcomes associated with brief interventions for problem gambling in methadone patients. Addict Behav. 2021;120:106953. 10.1016/j.addbeh.2021.106953] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Petry NM, Rash CJ, Alessi SM. A randomized controlled trial of brief interventions for problem gambling in substance abuse treatment patients. J Consult Clin Psychol. 2016;84(10):874‐886. 10.1037/ccp0000127] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. (DSM‐5). American Psychiatric Association Publishing; 2013. 10.1176/appi.books.9780890425596] [DOI] [Google Scholar]
- 10. Potenza MN, Balodis IM, Derevensky J, et al. Gambling disorder. Nat Rev Dis Primers. 2019;5(1):51. 10.1038/s41572-019-0099-7] [DOI] [PubMed] [Google Scholar]
- 11. Currie SR, Hodgins DC, Williams RJ, et al. Predicting future harm from gambling over a five‐year period in a general population sample: a survival analysis. BMC Psychiatry. 2021;21(1):15. 10.1186/s12888-020-03016-x] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Quilty LC, Avila Murati D, Bagby RM. Identifying indicators of harmful and problem gambling in a Canadian sample through receiver operating characteristic analysis. Psychol Addict Behav. 2014;28(1):229‐237. 10.1037/a0032801] [DOI] [PubMed] [Google Scholar]
- 13. National Council on Problem Gambling . Help and Treatment FAQ. ncpgambling.org. Accessed March 22; 2022. [Google Scholar]
- 14. Tran LT, Wardle H, Colledge‐Frisby S, et al. The prevalence of gambling and problematic gambling: a systematic review and meta‐analysis. Lancet Public Health. 2024;9(8):e594‐e613. 10.1016/S2468-2667(24)00126-9] [DOI] [PubMed] [Google Scholar]
- 15. Gooding NB, Williams JN, Williams RJ. The differences between gamblers and substance users who seek treatment. Can J Addict. 2023;14(1):26‐34. 10.1097/CXA.0000000000000171] [DOI] [Google Scholar]
- 16. Leavens E, Marotta J, Weinstock J. Disordered gambling in residential substance use treatment centers: an unmet need. J Addict Dis. 2014;33(2):163‐173. 10.1080/10550887.2014.909697] [DOI] [PubMed] [Google Scholar]
- 17. Pagano A, Le T, Fong T, et al. Gambling, tobacco use, and health among individuals in substance use disorder treatment. Am J Addict. 2023;32(6):563‐573. 10.1111/ajad.13458] [DOI] [PubMed] [Google Scholar]
- 18. Mazar A, Zorn M, Becker N, et al. Gambling formats, involvement, and problem gambling: which types of gambling are more risky? BMC Public Health. 2020;20(1):711. 10.1186/s12889-020-08822-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nelson SE, LaPlante DA, Gray HM, et al. Already at the table: patterns of play and gambling involvement prior to gambling expansion. J Gambl Stud. 2018;34(1):275‐295. 10.1007/s10899-017-9711-5] [DOI] [PubMed] [Google Scholar]
- 20. Phillips JG, Ogeil R, Chow Y‐W, et al. Gambling involvement and increased risk of gambling problems. J Gambl Stud. 2013;29(4):601‐611. 10.1007/s10899-012-9325-x] [DOI] [PubMed] [Google Scholar]
- 21. Toneatto T, Ferguson D, Brennan J. Effect of a new casino on problem gambling in treatment‐seeking substance abusers. Can J Psychiatry. 2003;48(1):40‐44. 10.1177/070674370304800108] [DOI] [PubMed] [Google Scholar]
- 22. Palmer‐Bacon J, Miles‐McLean H, Welsh C, et al. Impact of a casino opening on gambling behaviors of people engaged in methadone maintenance. J Gambl Stud. 2017;33(2):461‐472. 10.1007/s10899-016-9621-y] [DOI] [PubMed] [Google Scholar]
- 23. McCuistian C, Le JT, Samayoa E, et al. Examining the sustained impact of California's tobacco free for recovery initiative: qualitative perspectives from program directors. Subst Use Misuse. 2026:1‐9. 10.1080/10826084.2026.2710339] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Gebauer L, LaBrie R, Shaffer HJ. Optimizing DSM‐IV‐TR classification accuracy: a brief biosocial screen for detecting current gambling disorders among gamblers in the general household population. Can J Psychiatry. 2010;55(2):82‐90. 10.1177/070674371005500204] [DOI] [PubMed] [Google Scholar]
- 25. Young MM, Hodgins DC, Currie SR, et al. Not too much, not too often, and not too many: the results of the first large‐scale, international project to develop lower‐risk gambling guidelines. Int J Mental Health Addict. 2024;22(1):666‐684. 10.1007/s11469-022-00896-w] [DOI] [Google Scholar]
- 26. Himelhoch SS, Miles‐McLean H, Medoff DR, et al. Evaluation of brief screens for gambling disorder in the substance use treatment setting. Am J Addict. 2015;24(5):460‐466. 10.1111/ajad.12241] [DOI] [PubMed] [Google Scholar]
- 27. Currie SR, Hodgins DC, Casey DM, et al. Deriving low‐risk gambling limits from longitudinal data collected in two independent Canadian studies. Addiction. 2017;112(11):2011‐2020. 10.1111/add.13909] [DOI] [PubMed] [Google Scholar]
- 28. Dowling NA, Greenwood CJ, Merkouris SS, et al. The identification of Australian low‐risk gambling limits: a comparison of gambling‐related harm measures. J Behav Addict. 2021;10(1):21‐34. 10.1556/2006.2021.00012] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Goodie AS. Associations between gambling games and gambling problems: whole games compared with temporal, skill characteristics, and other structural characteristics. Curr Addict Rep. 2015;2(3):249‐253. 10.1007/s40429-015-0068-5] [DOI] [Google Scholar]
- 30. Boldero JM, Bell RC. Chance‐ and skill‐based dimensions underlying young Australians' gambling activities and their relationships with gambling problems and other factors. Int Gambl Stud. 2012;12(2):145‐162. 10.1080/14459795.2011.643907] [DOI] [Google Scholar]
- 31. Welte JW, Barnes GM, Tidwell M‐CO, et al. Gambling and problem gambling in the United States: changes between 1999 and 2013. J Gambl Stud. 2015;31(3):695‐715. 10.1007/s10899-014-9471-4] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Goodie AS, Fortune EE. Measuring cognitive distortions in pathological gambling: review and meta‐analyses. Psychol Addict Behav. 2013;27(3):730‐743. 10.1037/a0031892] [DOI] [PubMed] [Google Scholar]
- 33. Petry NM. A comparison of treatment‐seeking pathological gamblers based on preferred gambling activity. Addiction. 2003;98(5):645‐655. 10.1046/j.1360-0443.2003.00336.x] [DOI] [PubMed] [Google Scholar]
- 34. Hing N, Smith M, Rockloff M, et al. How structural changes in online gambling are shaping the contemporary experiences and behaviours of online gamblers: an interview study. BMC Public Health. 2022;22(1):1620. 10.1186/s12889-022-14019-6] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. LaPlante DA, Nelson SE, LaBrie RA, et al. Disordered gambling, type of gambling and gambling involvement in the British Gambling Prevalence Survey 2007. Eur J Public Health. 2011;21(4):r. 10.1093/eurpub/ckp177] [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.