Preprints.org logo
Preprint
Article

This version is not peer-reviewed.

Psychometric Evaluation of the Bilingual Russian–Kazakh Problem Gambling Severity Index: Item-Level Response Patterns and Behavioral Factors Associated with Gambling-Related Harm

Submitted:

31 July 2026

Posted:

03 August 2026

You are already at the latest version

Abstract
Population-based evidence on gambling-related harm remains scarce in Central Asia, and Kazakhstan has lacked a nationally validated bilingual screening instrument. This study evaluated the psychometric properties of the bilingual (Kazakh–Russian) Problem Gambling Severity Index (PGSI), estimated the national distribution of gambling-related harm, and identified behavioral and socio-demographic correlates of elevated gambling risk. A stratified sample of 1,015 adults aged 18–60 years from all 17 regions and three major cities of Kazakhstan completed standardized questionnaires in Kazakh or Russian. The PGSI was assessed using categorical and continuous measures, examining internal consistency, construct validity, item-level response distributions, and multivariable logistic regression. The bilingual PGSI demonstrated good psychometric performance, including satisfactory internal consistency (Cronbach’s α = 0.846) and a unidimensional factor structure explaining 45.9% of variance. Overall, 27.1% of respondents met the screening threshold for elevated gambling risk (PGSI ≥ 8). Elevated scores primarily reflected cumulative moderate problems across domains rather than frequent severe responses. Gambling expenditure and frequency remained the strongest independent correlates of elevated risk, while demographic characteristics showed limited associations. These findings provide the first nationally representative evidence on gambling-related harm in Kazakhstan and support the bilingual PGSI for population screening, surveillance, and targeted prevention strategies.
Keywords: 
Kazakhstan
;  
Problem Gambling Severity
;  
gambling
;  
behavioral disorder
Subject: 
Social Sciences  -   Psychology

1. Introduction

Gambling disorder is recognized in both the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) and the International Classification of Diseases (ICD-11) as a behavioral addiction characterized by persistent gambling despite harmful consequences and impaired control over gambling behavior. Beyond its clinical manifestations, gambling-related harm encompasses a broader continuum of adverse financial, psychological, interpersonal, and occupational consequences that affect not only individuals but also their families and communities (Kristensen et al., 2024; Tulloch et al., 2023). Contemporary public health research increasingly emphasizes the identification of gambling-related harm across the full spectrum of risk rather than focusing exclusively on clinically diagnosed gambling disorder (Price et al., 2021).
International epidemiological studies indicate that gambling-related harm varies substantially across populations. Recent reviews estimate the annual prevalence of gambling disorder among adults to range between approximately 1.1% and 3.5%, while broader categories of problem gambling have been reported in 2.3–12.6% of the adult population depending on assessment methods and national context (Moreira et al., 2024; Hodgins et al., 2022). These differences reflect not only genuine variation in gambling behaviour but also methodological diversity, differences in gambling environments, and the availability of validated screening instruments. Reliable population-based screening is therefore essential for estimating gambling-related harm, monitoring trends, and informing prevention strategies.
The rapid expansion of online gambling, sports betting, and digital gambling platforms has further increased the importance of population surveillance. Digital gambling environments provide continuous access, reduce traditional barriers to participation, and have been associated with increased gambling involvement and harm (Watanapongvanich et al., 2020). At the same time, gambling-related problems frequently remain under-recognized because of stigma, delayed help-seeking, and limited routine screening within healthcare and social services (Quigley, 2022; Jääskeläinen & Kuusisto, 2025; Jones et al., 2025). As a result, prevalence estimates based solely on clinical diagnoses are likely to underestimate the actual burden of gambling-related harm.
Kazakhstan represents one of the least studied gambling environments in Central Asia. Despite the rapid development of legal gambling opportunities and online betting markets, nationally representative epidemiological evidence remains unavailable. Official health statistics report only a small number of clinically diagnosed gambling disorder cases despite a population exceeding 20 million, suggesting substantial under-identification rather than low occurrence (Ministry of Health, 2025). Moreover, no standardized national screening system has been implemented to estimate gambling-related harm in the general population, limiting the evidence available for prevention planning and public health policy.
A further limitation concerns the absence of a validated bilingual screening instrument suitable for Kazakhstan's Russian- and Kazakh-speaking population. The Problem Gambling Severity Index (PGSI) is among the most widely used instruments for measuring gambling-related harm in community samples (Ferris & Wynne, 2001). Unlike diagnostic interviews, the PGSI conceptualizes gambling-related harm as a continuum and has demonstrated good reliability and validity across diverse cultural settings. Nevertheless, its psychometric performance has not previously been evaluated in Kazakhstan, and no evidence exists regarding the response patterns of individual PGSI items within this population.
The present study addresses these gaps by providing the first psychometric evaluation of the bilingual Russian–Kazakh PGSI in a nationally representative sample of adults in Kazakhstan. In addition to examining the reliability and construct validity of the instrument, this study describes item-level response patterns, estimates the national distribution of gambling-related harm, and identifies behavioral and sociodemographic factors independently associated with elevated gambling-related risk. Collectively, these findings provide an empirical foundation for future epidemiological surveillance, cross-cultural research, and evidence-informed prevention strategies targeting gambling-related harm in Kazakhstan.

2. Literature Review

Gambling-related harm is now viewed as a multidimensional continuum encompassing financial, psychological, interpersonal, occupational, and health-related consequences that emerge across varying levels of gambling involvement rather than solely among individuals meeting diagnostic criteria (Langham et al., 2016; Browne et al., 2017). This public health perspective has fundamentally altered approaches to gambling research and prevention. Rather than considering gambling disorder exclusively as an individual psychiatric condition, researchers increasingly emphasize the cumulative burden of gambling-related harm across populations. Browne et al. (2017) demonstrated that although severe gambling disorder accounts for considerable individual suffering, a substantial proportion of gambling-related harm occurs among individuals experiencing low- and moderate-risk gambling because these groups represent a much larger proportion of the population. Consequently, public health frameworks advocate shifting attention from treatment alone toward prevention, early identification, and systematic surveillance (Kristensen et al., 2024).
International epidemiological evidence demonstrates considerable variability in gambling-related harm across countries. Recent systematic reviews estimate the annual prevalence of gambling disorder among adults to range between approximately 1.1% and 3.5% (Moreira et al., 2024), whereas broader categories of problem gambling have been reported in 2.3–12.6% of adult populations depending on assessment methodology, gambling availability, and national context (Hodgins et al., 2022). Such variability reflects not only genuine differences in gambling behavior but also differences in legislative frameworks, gambling availability, cultural attitudes, and the psychometric performance of screening instruments, highlighting the need for culturally appropriate and psychometrically validated measures capable of identifying gambling-related harm across the full spectrum of severity rather than solely among clinically diagnosed cases.
Beyond prevalence estimates, gambling-related harm constitutes a significant public health concern owing to its consistent associations with psychiatric comorbidity, financial hardship, social dysfunction, reduced quality of life, and increased mortality (Karlsson & Håkansson, 2018). Moreover, the rapid expansion of online gambling, sports betting, esports wagering, and mobile gambling platforms has intensified these concerns by increasing both the accessibility and frequency of gambling opportunities (Granero et al., 2020; López-González et al., 2018; Moreira et al., 2023; Watanapongvanich et al., 2020).
Despite growing recognition of gambling-related harm, its early identification remains challenging because gambling often develops within sociocultural contexts that discourage disclosure. Stigma is among the most important barriers, as gambling problems are frequently perceived as signs of personal irresponsibility or moral failure rather than manifestations of an addictive disorder, thereby reducing help-seeking behavior (Hing et al., 2016). Self-stigma further reinforces these barriers through feelings of shame, guilt, and anticipated social rejection, resulting in concealment of gambling behaviors until financial, interpersonal, or psychological consequences become severe (Quigley, 2022). These processes contribute not only to delayed treatment seeking but also to systematic underestimation of gambling-related harm within epidemiological studies relying primarily on healthcare utilization or clinical diagnosis.
The normalization of gambling within contemporary leisure environments further complicates early recognition. Gambling has become increasingly embedded in sports, online gaming, esports, and digital media, blurring the distinction between recreational participation and harmful involvement (Watanapongvanich et al., 2020). As a result, gambling is often perceived as a socially acceptable activity rather than a potential health concern, particularly in online environments where participation occurs privately and is less visible than in traditional gambling venues.
Sociocultural barriers are further reinforced by institutional limitations. Jääskeläinen and Kuusisto (2025) argue that gambling-related problems remain insufficiently addressed within routine healthcare and social services because standardized screening procedures are rarely implemented outside specialized addiction settings. Similarly, Jones et al. (2025) emphasize that delayed recognition frequently reflects systemic deficiencies in professional education and assessment practices rather than the absence of clinically significant symptoms. Consequently, gambling-related harm represents not only an individual behavioral phenomenon but also a condition whose visibility is shaped by social attitudes, institutional practices, and the availability of sensitive assessment strategies.
These observations have important implications for gambling assessment. Contemporary research increasingly recognizes that delayed identification reflects not only stigma and limited access to care but also conceptual and psychometric limitations of existing assessment approaches. Gambling-related harm develops along a continuum of behavioral, cognitive, emotional, and motivational processes rather than as a discrete condition (Molander et al., 2021). However, many diagnostic frameworks continue to rely on categorical thresholds, reducing sensitivity to early or subclinical manifestations of gambling-related risk. Although categorical diagnoses remain essential for clinical decision-making, they are less effective for identifying emerging patterns of harm before diagnostic criteria are met, thereby limiting opportunities for early intervention.
The multidimensional nature of gambling-related harm is reflected in the psychological mechanisms underlying gambling persistence. Beyond monetary reinforcement, gambling behavior is sustained by cognitive distortions, emotional regulation, motivational processes, and reward expectations, including loss chasing, attentional preoccupation, illusion of control, and distorted outcome expectancies (Hing & Russell, 2017; Molander et al., 2021). Experimental evidence further demonstrates that structural characteristics of gambling products, such as near-miss effects and intermittent reinforcement, promote continued gambling despite objective losses (Çakıcı et al., 2021). Consequently, psychological vulnerability may emerge before behavioral indicators, including gambling frequency or expenditure, reach diagnostic thresholds.
These findings indicate that behavioral indicators alone provide an incomplete representation of gambling-related harm. Emotional vulnerability and cognitive distortions have been shown to improve the identification of gambling-related problems beyond behavioral measures (Nigro et al., 2021), supporting recommendations to integrate behavioral, cognitive, emotional, and motivational dimensions into contemporary screening approaches (Jääskeläinen & Kuusisto, 2025; Jones et al., 2025). The limitations of conventional assessment are further highlighted by emerging gambling-related activities – including speculative investing, esports betting, skin gambling, and loot boxes – which share important psychological characteristics with conventional gambling disorder despite frequently remaining outside traditional diagnostic classifications (Lee et al., 2023; Moreau et al., 2016). Collectively, this evidence suggests that multidimensional screening instruments are better suited than categorical diagnostic thresholds to identify gambling-related harm across its full continuum.
Conceptual limitations in gambling assessment are further compounded by institutional factors affecting diagnostic capacity. Gambling disorder remains under-recognized because routine screening is rarely incorporated into healthcare services, while limited professional training contributes to inconsistent assessment and delayed recognition (Jones et al., 2025). As a result, gambling-related problems are often attributed to other psychosocial or psychiatric conditions without systematic evaluation of gambling behavior. Diagnostic complexity is further increased by high rates of psychiatric comorbidity, particularly with substance use, mood, and personality disorders, which increase the risk of diagnostic overshadowing (Mundt & Baranyi, 2020; Nigro et al., 2021). Similar patterns have been reported in psychiatric and forensic settings, where gambling behavior is infrequently assessed despite its contribution to functional impairment, offending behavior, and relapse risk (Corbeil et al., 2023; Hing et al., 2015). Together, these findings suggest that under-detection reflects not only limitations of screening instruments but also broader institutional barriers to routine identification of gambling-related harm.
These international findings are particularly relevant for Kazakhstan, where gambling research remains limited despite the rapid expansion of legal gambling, online betting, and digital gambling platforms. Between 1990 and 2020, no peer-reviewed studies authored by researchers affiliated with Kazakhstani institutions and indexed in major international databases examined gambling addiction as a primary research focus. Existing studies have largely addressed legal regulation or public policy rather than behavioral epidemiology or clinical assessment. Preliminary research has reported cognitive distortions and financial overextension among gambling participants (Prilutskaya & Kuliev, 2016), emphasized gambling as an emerging public health issue (Buribayev & Khamzina, 2025), explored psychological correlates of gambling behavior (Tapalova et al., 2025), and examined legislative and criminological aspects of gambling expansion (Berdaliyeva et al., 2023). However, no nationally representative study has assessed gambling-related harm using internationally recognized screening instruments, and no psychometric validation of a bilingual gambling screening instrument has been conducted in Kazakhstan. Within this context, standardized screening instruments provide an essential methodological foundation for population-based research. Although structured diagnostic interviews remain the clinical reference standard, screening instruments enable efficient identification of varying levels of gambling-related risk and facilitate comparisons across populations, particularly in countries where routine clinical surveillance and epidemiological evidence remain limited.
Among available screening instruments, the Problem Gambling Severity Index (PGSI) is one of the most extensively validated measures of gambling-related harm in community samples (Ferris & Wynne, 2001). Unlike diagnostic interviews, the PGSI conceptualizes gambling severity as a continuum, consistent with contemporary public health approaches emphasizing early identification of harmful gambling (Browne et al., 2017). Psychometric studies have demonstrated good reliability and validity across diverse cultural settings, with translated versions showing comparable factorial structures and psychometric performance (Currie et al., 2013; Casu et al., 2023; Molander et al., 2021; Caler et al., 2017; Loo et al., 2011). Recent evidence further highlights the importance of rigorous cross-cultural adaptation, particularly in multilingual populations (Bastiani et al., 2025; Chinawa et al., 2023). These characteristics make the PGSI particularly suitable for Kazakhstan, where no validated bilingual gambling screening instrument has previously been available.
The increasing prevalence of online betting, mobile gambling, esports wagering, and other digital gambling formats further reinforces the need for screening instruments capable of detecting gambling-related harm across a continuum of severity (Granero et al., 2020; López-González et al., 2018; Moreira et al., 2023). Despite extensive international application of the PGSI, no psychometric evaluation of a bilingual Russian–Kazakh version has been conducted, and no nationally representative study has assessed gambling-related harm in Kazakhstan using a standardized screening instrument. Moreover, little is known about item-level response patterns or the behavioural factors associated with elevated gambling-related risk in the Kazakhstani population. Accordingly, the present study aimed to evaluate the psychometric properties of the bilingual Russian–Kazakh PGSI, examine item-level response patterns, estimate the national distribution of gambling-related harm, and identify behavioral and sociodemographic factors associated with elevated gambling-related risk in a nationally representative sample of adults in Kazakhstan.

3. Methods

3.1. Study Design

This study employed a national cross-sectional, population-based survey to assess gambling-related harm in the adult population of Kazakhstan using the Problem Gambling Severity Index (PGSI). The primary objective was to estimate the distribution of screening-defined gambling risk in a nationally representative sample while evaluating the psychometric performance of the bilingual (Kazakh–Russian) version of the PGSI as a population-level screening instrument. A secondary objective was to identify demographic, socioeconomic, and gambling-related characteristics associated with elevated gambling-related risk.
The study was designed within a public health surveillance framework, recognizing that standardized screening instruments provide epidemiological evidence on the distribution of gambling-related harm rather than clinical diagnoses. Accordingly, the findings are interpreted as estimates of screening-defined gambling risk and should not be considered equivalent to the prevalence of clinically diagnosed gambling disorder. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations for reporting cross-sectional research whenever applicable.

3.2. Measurement Instrument

Gambling-related harm was assessed using the Problem Gambling Severity Index (PGSI), a nine-item screening instrument developed as part of the Canadian Problem Gambling Index (Ferris & Wynne, 2001). The PGSI measures behavioral and psychosocial indicators of gambling-related harm, including betting beyond financial means, chasing losses, borrowing money to gamble, perceived loss of control, and gambling-related guilt. Each item is rated on a four-point Likert scale ranging from 0 (“Never”) to 3 (“Almost always”), producing a total score between 0 and 27. Consistent with international recommendations, total scores were categorized as no-risk (0), low-risk (1–2), moderate-risk (3–7), and problem gambling (≥8). Because the PGSI conceptualizes gambling-related harm along a continuum rather than as a dichotomous disorder, it is particularly suitable for population-based screening and for detecting emerging or subclinical patterns of harm (James et al., 2016; Nower et al., 2013).

3.3. Translation and Cross-Cultural Adaptation

The Kazakh and Russian versions of the PGSI were developed following internationally accepted guidelines for the cross-cultural adaptation of patient-reported outcome measures (Beaton et al., 2000; Wild et al., 2005). The adaptation process included two independent forward translations, synthesis of the translated versions, two independent blind back-translations, and review by an expert committee to ensure semantic, conceptual, and cultural equivalence. Particular attention was given to culturally sensitive concepts, including gambling terminology, financial affordability, and household-related expressions, to maximize conceptual equivalence across both language versions. Cognitive interviews involving 25 respondents for each language version were subsequently conducted to evaluate clarity, comprehensibility, and cultural relevance. Feedback obtained during pilot testing informed minor wording refinements before implementation in the national survey.

3.4. Participants and Sampling

The study included 1,015 participants, comprising 551 men (54.3%) and 464 women (45.7%) (Table 1).
The age distribution was relatively balanced across the three younger age groups. Participants aged 22–35 years represented the largest proportion of the sample (30.1%), followed by those aged 18–21 years (29.8%) and 36–45 years (29.7%), whereas respondents aged 46–60 years accounted for 10.4% of the study population.
Regarding marital status, 49.1% of participants were married and 42.8% were single. Divorced respondents constituted 5.1%, widowed respondents 0.9%, and participants reporting other marital statuses 2.2%.
Students represented the largest occupational group (20.2%), followed by respondents employed in trade and services (16.3%), industry and construction (10.8%), and education and science (10.0%). Participants working in business and management accounted for 7.4%, those employed in transport and logistics for 6.5%, while 6.3% of respondents were unemployed. The remaining 22.5% were distributed across other occupational categories.

3.5. Quantitative Procedures and Statistical Analysis

Data were collected using a structured questionnaire administered in Kazakh and Russian. The questionnaire included three domains: sociodemographic characteristics, gambling behaviour, and gambling-related harm assessed using the nine-item Problem Gambling Severity Index (PGSI). Additional items assessed gambling frequency, duration, preferred gambling modalities, gambling expenditure, perceived interpersonal consequences, attempts to reduce gambling, and exposure to gambling advertising.
Data quality was evaluated by examining missing values, duplicate records, logical consistency, questionnaire completeness, and denominator accuracy. PGSI scores were calculated according to the original scoring guidelines and classified into four categories: no-risk (0), low-risk (1–2), moderate-risk (3–7), and problem gambling (≥8). Descriptive statistics were generated for the overall sample and for respondents who completed all PGSI items.
Continuous variables are presented as means and standard deviations or medians and interquartile ranges, whereas categorical variables are reported as frequencies and percentages. In addition to total PGSI scores, response distributions for individual PGSI items were examined to characterize patterns contributing to gambling-related harm. Floor and ceiling effects were also assessed.
The psychometric properties of the PGSI were evaluated using Cronbach’s alpha, corrected item–total correlations, the Kaiser–Meyer–Olkin measure, Bartlett's test of sphericity, and exploratory factor analysis.
Associations between PGSI categories and participant characteristics were examined using Pearson's chi-square tests and Cramér’s V. Multivariable binary logistic regression was performed to identify factors independently associated with problem gambling (PGSI ≥8), with results reported as adjusted odds ratios (aORs) and 95% confidence intervals.
Sensitivity analyses were conducted using an alternative threshold for gambling-related harm (PGSI ≥3) and by repeating the analyses among respondents who completed the PGSI assessment. Statistical significance was set at p < .05. All analyses were performed using IBM SPSS Statistics Version 29 (IBM Corp., Armonk, NY, USA).

3.6. Ethics Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of Zhetysu University (Protocol No. 2025-04/K). Written informed consent was obtained from all participants in the quantitative component. Participation was voluntary and respondents were informed of their right to withdraw at any stage without consequence. Personally identifiable information was not collected or retained, and all data were anonymized and stored on encrypted password-protected drives accessible only to the core research team.

4. Results

4.1. Psychometric Properties of the Bilingual PGSI

The bilingual PGSI demonstrated good internal consistency (Cronbach’s α = 0.846) (Table 1). Corrected item–total correlations ranged from 0.36 to 0.69, and deletion of individual items did not materially improve the overall reliability coefficient.
The Kaiser–Meyer–Olkin measure indicated excellent sampling adequacy (KMO = 0.881), while Bartlett's test of sphericity confirmed the suitability of the correlation matrix for factor analysis (χ2(36) = 1465.46, p <0.001). The first principal component explained 45.9% of the total variance, supporting an essentially one-factor structure of the instrument. Factor loadings ranged from 0.37 to 0.77, with the highest loading observed for the item assessing gambling-related financial problems and the lowest loading for the item assessing gambling-related guilt (Table 2).

4.2. Sample Characteristics and Distribution of PGSI Categories

A total of 1,015 participants were included in the final analysis. According to the conventional PGSI classification, 574 participants (56.6%) were classified as PGSI 0, 30 (3.0%) as low-risk gamblers (PGSI 1–2), 136 (13.4%) as moderate-risk gamblers (PGSI 3–7), and 275 (27.1%) met the threshold for problem gambling (PGSI ≥8) (Table 3, Table 4).

4.3. Distribution of Gambling-Related Harms

Item-level analyses demonstrated variability in endorsement frequencies across the nine PGSI indicators (Table 4). For each item, the largest proportion of positive responses was observed within the response category “Sometimes”, whereas endorsement of “Most of the time” and particularly “Almost always” was substantially less frequent.
The highest frequencies of positive responses were observed for behavioral indicators reflecting repeated gambling involvement, whereas financial and interpersonal consequences were endorsed less frequently but remained consistently represented across the study population. Responses indicating the highest severity (“Almost always”) accounted for only a small proportion of endorsements across all PGSI items.

4.4. Bivariate Associations with Problem Gambling

Bivariate analyses identified statistically significant associations between problem gambling (PGSI ≥8) and several demographic and gambling-related characteristics (Table 5).
Sex was significantly associated with problem gambling (χ2=23.20, df=1, p<0.001, Cramér’s V=0.151). Occupational status also demonstrated a statistically significant association (χ2=29.88, df=13, p =0.005, Cramér’s V = 0.172). In contrast, neither age group (χ2=6.30, df=3, p =0.098) nor marital status (χ2= 7.29, df=6, p =0.295) was significantly associated with PGSI ≥8.
Among gambling-related variables, the strongest associations were observed for gambling frequency (χ2 = 91.61, df =4, p <0.001, Cramér’s V = 0.440) and the proportion of monthly income spent on gambling (χ2 = 86.57, df =3, p <0.001, Cramér’s V=0.427). Duration of gambling involvement was also significantly associated with problem gambling (χ2 = 13.53, df =4, p =0.009), although the corresponding effect size was comparatively small (Cramér’s V=0.169).

4.5. Multivariable Logistic Regression

A multivariable logistic regression model was constructed to identify independent predictors of problem gambling (Table 6).
After simultaneous adjustment for all variables included in the model, the proportion of monthly income spent on gambling remained the strongest independent predictor of problem gambling (adjusted OR = 2.80, 95% CI: 1.90–4.12, p <0.001). Gambling frequency also remained independently associated with PGSI ≥8 (adjusted OR = 2.13, 95% CI: 1.62–2.80, p <0.001).
By contrast, sex was no longer significantly associated with problem gambling after adjustment (adjusted OR=1.16, 95% CI: 0.70–1.90, p =0.565). Likewise, none of the age categories demonstrated statistically significant independent associations with PGSI ≥8. Duration of gambling involvement also failed to retain statistical significance in the adjusted model (adjusted OR = 1.14, 95% CI: 0.93–1.41, p =0.212).
Overall, the multivariable analysis demonstrated that variables describing gambling behavior and financial involvement remained independently associated with problem gambling after adjustment for demographic characteristics, whereas demographic variables did not retain statistical significance.

5. Discussion

5.1. Psychometric Performance of the Bilingual PGSI

The bilingual Russian–Kazakh version of the PGSI demonstrated satisfactory psychometric performance. Good internal consistency (Cronbach's α = 0.846), excellent sampling adequacy (KMO = 0.881), a significant Bartlett's test, and a unidimensional factor structure support the reliability and construct validity of the instrument. These findings indicate that the bilingual PGSI is suitable for epidemiological research and screening in the Kazakhstani context.

5.2. Response Pattern Across PGSI Items

One of the principal findings of this study was that 27.1% of respondents met the conventional threshold for problem gambling (PGSI≥8). Despite the relatively high proportion of respondents classified as problem gamblers (27.1%), endorsement of the highest response category (“Almost always”) remained uncommon across individual PGSI items. Instead, elevated PGSI scores were more frequently characterized by repeated endorsement of the response category “Sometimes” across multiple items.
First, prevalence estimates obtained using the PGSI vary substantially according to sampling strategy, recruitment procedures, and population characteristics (Dellis et al., 2014; López-González et al., 2018; Sundqvist & Wennberg, 2021). Second, the PGSI was developed as a population-based measure of gambling-related harm rather than a diagnostic instrument, and prevalence estimates are therefore influenced by the distribution of gambling-related consequences within the surveyed population (Delfabbro & King, 2017; Dowling et al., 2018). An additional consideration concerns the internal response structure of the PGSI. Previous psychometric studies have shown that elevated PGSI scores are not necessarily characterized by frequent endorsement of the highest response category for individual items. Instead, respondents may reach higher total scores through repeated endorsement of intermediate response categories across several items, reflecting the cumulative nature of gambling-related harm measured by the instrument (Sharp et al., 2012). The response pattern observed in the present study is consistent with these findings, as “Sometimes” was endorsed substantially more often than “Almost always” across most PGSI items despite the relatively high proportion of respondents classified as PGSI ≥8.
In the multivariable model, behavioral characteristics remained independently associated with problem gambling, whereas demographic variables did not. The proportion of monthly income spent on gambling was the strongest independent predictor, followed by gambling frequency. In contrast, sex, age, and gambling duration were no longer statistically significant after adjustment. These findings suggest that measures reflecting current gambling behavior and financial involvement may provide more informative indicators of gambling-related harm than demographic characteristics alone and may therefore be useful in screening and early identification strategies. Our findings are consistent with recent evidence indicating that behavioral indicators remain more robust predictors of gambling-related harm than demographic characteristics after multivariable adjustment (Dowling et al., 2021; Clune et al., 2024).

5.3. Limitations

This study has several limitations. First, it relied on self-reported data, which may be affected by recall bias and social desirability, particularly in stigmatized contexts where gambling behavior may be underreported. Second, the study employed the PGSI as a screening instrument rather than a diagnostic tool; thus, the results reflect screening-defined gambling risk, not a clinical diagnosis. This distinction is especially relevant in under-researched settings where formal diagnostic assessment is limited, and screening constitutes the primary form of surveillance. Finally, the study included only noninstitutionalized adults; populations in clinical or correctional environments were not sampled and may exhibit different risk patterns. Consequently, the findings should be interpreted as population-level risk estimates rather than representing the full spectrum of gambling-related harm.

6. Conclusions

This study provides the first comprehensive evaluation of the bilingual Russian–Kazakh version of the Problem Gambling Severity Index (PGSI) in a representative sample from Kazakhstan. The findings support the reliability and construct validity of the instrument, indicating that it is suitable for epidemiological research and population-based screening.
A substantial proportion of respondents met the conventional threshold for problem gambling (PGSI ≥8). Item-level analyses further showed that elevated PGSI scores were more commonly characterized by the accumulation of moderate gambling-related problems across multiple domains than by frequent endorsement of the most severe response category. In addition, behavioral indicators – particularly gambling expenditure and gambling frequency – remained independently associated with problem gambling after adjustment for demographic characteristics.
These findings extend the evidence supporting the use of the PGSI in culturally diverse populations and provide the first psychometric evidence for its application in Kazakhstan. They also highlight the value of behavioral and financial indicators for the early identification of gambling-related harm and provide an empirical basis for future epidemiological research, prevention strategies, and public health initiatives addressing gambling-related harm in Kazakhstan.

Author Contributions

Conceptualization, O.T. and N.Z..; methodology, O.T; software, S.T.; validation, N.S. and O.T.; formal analysis, N.S.; investigation, N.S.; resources, N.Z.; data curation, N.S.; writing—original draft preparation, N.S.; writing—review and editing, O.T.; visualization, S.T.; supervision, Y.B.; project administration, Y.B.; funding acquisition, Y.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Committee of Science of the Ministry of Science and Higher Education of the Republic of Kazakhstan (Grant No. BR24992927, “Integrative Study of Gambling Addiction in Kazakhstan and Multidisciplinary Strategies for Its Minimization”). The funding body had no role in the study design; collection, analysis, or interpretation of data; writing of the manuscript; or the decision to submit the article for publication.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Zhetysu University (Protocol No. 2025-04/K). Written informed consent was obtained from all participants in the quantitative component. Participation was voluntary and respondents were informed of their right to withdraw at any stage without consequence. Personally identifiable information was not collected or retained, and all data were anonymized and stored on encrypted password-protected drives accessible only to the core research team.

Data Availability Statement

The data that support the findings of this study contain sensitive participant information and are available from the corresponding author upon reasonable request. It is nonetheless subject to approval by the relevant ethics committee and must comply with data protection regulations.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Abbreviations

The following abbreviations are used in this manuscript:
PGSI Problem Gambling Severity Index
DSM Diagnostic and Statistical Manual of Mental Disorders
ICD International Classification of Diseases

References

  1. Bastiani, O.; Khouja, J.; Skatova, A.; Newall, P. The relationship between Problem Gambling Severity Index scores and suicidality: Results of a 9-year cohort study of young United Kingdom adults. Addiction 2025, 121(1), 196–207. [Google Scholar] [CrossRef] [PubMed]
  2. Beaton, D.E.; Bombardier, C.; Guillemin, E.; Ferraz, M.B. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine 2000, 25(24), 3186–3191. [Google Scholar] [CrossRef] [PubMed]
  3. Berdaliyeva, A. S.; Kim, A. I.; Seraliyeva, A. M.; Gassanov, A. A.; Dunentayev, M. V. Criminological measures to counteract corruption offences in the field of illegal gambling. Journal of Financial Crime 2023, 30(1), 4–23. [Google Scholar] [CrossRef]
  4. Brodeur, M.; Audette-Chapdelaine, S.; Savard, A.; Kairouz, S. Gambling and the covid-4. 19 pandemic in the province of Quebec (Canada): protocol for a mixed-methods study. BMJ Open 2021, 11(9), e048785. [Google Scholar] [CrossRef] [PubMed]
  5. Browne, M.; Rockloff, M. The dangers of conflating gambling-related harm with disordered gambling. Journal of Behavioral Addictions 2017, 6(3), 317–320. [Google Scholar] [CrossRef] [PubMed]
  6. Browne, M.; Greer, N.; Rawat, V.; Rockloff, M. A population-level metric for gambling-related harm. International Gambling Studies 2017, 17(2), 163–175. [Google Scholar] [CrossRef]
  7. Bureau of National Statistics. Бpaки и paзвoды в 2024 гoдy [Marriages and divorces in 2024]. Agency for Strategic Planning and Reforms of the Republic of Kazakhstan. 2024. Available online: https://stat.gov.kz/ru/news/braki-i-razvody-v-2024-godu.
  8. Buribayev, Y. A.; Zhienbayeva, N. B.; Khamzina, Z. A.; Yeshimbetova, S. Z.; Aknazarov, S. A.; Tapalova, O. B. Cognitive and emotional predictors of gambling disorder: From impulsivity to pathological dependence. Bulletin. Series: Psychology 2025, 82(1), 022. [Google Scholar] [CrossRef]
  9. Buribayev, Y.; Khamzina, Z. Gambling in Kazakhstan: Seeking a balance between economic benefits and combating gambling addiction. Journal of Gambling Issues 2025, 59. Available online: https://cdspress.ca/wp-content/uploads/2025/03/JGI-Dec-24-REVIEW-059.R1_Khamzina_Proof_FINAL.pdf.
  10. Çakıcı, M.; Sancar, N.; Buran, A.; Şahan, G.; Yılmaz, B. Development and validation of a near miss scale for assessing gambling tendency. Journal of Gambling Studies 2021, 38(3), 1045–1058. [Google Scholar] [CrossRef] [PubMed]
  11. Caler, K. R.; Garcia, J. R.; Nower, L. Problem gambling among ethnic and racial minority groups: A systematic review. Addictive Behaviors 2017, 72, 125–138. [Google Scholar] [CrossRef] [PubMed]
  12. Casu, M.; Belfiore, C.; Caponnetto, P. Rolling the dice: a comprehensive review of the new forms of gambling and psychological clinical recommendations. Psychiatry International 2023, 4(2), 105–125. [Google Scholar] [CrossRef]
  13. Chinawa, A.; Ossai, E.; Odinka, P.; Nduaguba, O.; Odinka, J.; Aronu, A.; …; Chinawa, J. Problem gambling among secondary school adolescents in Enugu, Nigeria. African Health Sciences 2023, 23(3). [Google Scholar] [CrossRef] [PubMed]
  14. Clune, S.; Ratnaike, D.; White, V.; Donaldson, A.; Randle, E.; O’Halloran, P.; Lewis, V. What is known about population level programs designed to address gambling-related harm: rapid review of the evidence. Harm Reduction Journal 2024, 21(1). [Google Scholar] [CrossRef] [PubMed]
  15. Coloma-Carmona, A.; Carballo, J.; Llinares, F.; Aguerri, J.; Griffiths, M. Development and validation of the trading disorder scale for assessing problematic trading behaviors. Journal of Behavioral Addictions 2025, 14(2), 941–958. [Google Scholar] [CrossRef] [PubMed]
  16. Corbeil, O.; Soulard, M.; Huot-Lavoie, M.; Béchard, L.; Fournier, É.; Brodeur, S.; …; Demers, M. Problem gambling among people with first-episode psychosis: protocol for a prospective multicenter cohort study. 2023. [Google Scholar] [CrossRef] [PubMed]
  17. Currie, S. R.; Hodgins, D. C.; Casey, D. M. Validity of the Problem Gambling Severity Index interpretive categories. Journal of Gambling Studies 2013, 29(2), 311–327. [Google Scholar] [CrossRef] [PubMed]
  18. Delfabbro, P.; King, D. L. Prevention paradox logic and problem gambling: Does low-risk gambling impose a greater burden of harm than high-risk gambling? Journal of Behavioral Addictions 2017, 6(2), 163–167. [Google Scholar] [CrossRef] [PubMed]
  19. Dellis, A.; Sharp, C.; Hofmeyr, A.; Schwardmann, P. M.; Spurrett, D.; Rousseau, J.; Ross, D. Criterion-related and construct validity of the Problem Gambling Severity Index in a sample of South African gamblers. South African Journal of Psychology 2014, 44(2), 243–257. [Google Scholar] [CrossRef]
  20. Dowling, N. A.; Merkouris, S.; Manning, V.; Volberg, R. A.; Lee, S.; Rodda, S. N.; Lubman, D. I. Screening for problem gambling within mental health services: a comparison of the classification accuracy of brief instruments. Addiction 2018, 113(6), 1088–1104. [Google Scholar] [CrossRef] [PubMed]
  21. Dowling, N. A.; Merkouris, S.; Rodda, S. N.; Smith, D.; Aarsman, S. R.; Lavis, T.; Lubman, D. I.; Austin, D.; Cunningham, J.; Battersby, M.; Eric, O. GamblingLess: A Randomised Trial Comparing Guided and Unguided Internet-Based Gambling Interventions. Journal of Clinical Medicine 2021, 10(11), 2224. [Google Scholar] [CrossRef] [PubMed]
  22. Ferris, J.; Wynne, H. The Canadian Problem Gambling Index. Final report. Canadian Consortium for Gambling research for the Canadian Centre on Substance Abuse. 2001. Available online: https://www.greo.ca/Modules/EvidenceCentre/files/Ferris%20et%20al(2001)The_Canadian_Problem_Gambling_Index.pdf.
  23. Granero, R.; Fernández-Aranda, F.; Lara-Huallipe, M.; Gómez-Peña, M.; Moragas, L.; Baenas, I.; …; Jiménez-Múrcia, S. Latent classes for the treatment outcomes in women with gambling disorder and buying/shopping disorder. Journal of Clinical Medicine 2022, 11(13), 3917. [Google Scholar] [CrossRef] [PubMed]
  24. Granero, R.; Jiménez-Múrcia, S.; Pino-Gutiérrez, A.; Mora, B.; Valenciano-Mendoza, E.; Baenas, I.; …; Fernández-Aranda, F. Gambling phenotypes in online sports betting. Frontiers in Psychiatry 2020, 11. [Google Scholar] [CrossRef] [PubMed]
  25. Hing, N.; Russell, A. How anticipated and experienced stigma can contribute to self-stigma: the case of problem gambling. Frontiers in Psychology 2017, 08. [Google Scholar] [CrossRef] [PubMed]
  26. Hing, N.; Russell, A. Psychological factors, sociodemographic characteristics, and coping mechanisms associated with the self-stigma of problem gambling. Journal of Behavioral Addictions 2017, 6(3), 416–424. [Google Scholar] [CrossRef] [PubMed]
  27. Hing, N.; Russell, A.; Gainsbury, S. Unpacking the public stigma of problem gambling: the process of stigma creation and predictors of social distancing. Journal of Behavioral Addictions 2016, 5(3), 448–456. [Google Scholar] [CrossRef] [PubMed]
  28. Hing, N.; Russell, A.; Gainsbury, S.; Nuske, E. The public stigma of problem gambling: its nature and relative intensity compared to other health conditions. Journal of Gambling Studies 2015, 32(3), 847–864. [Google Scholar] [CrossRef] [PubMed]
  29. Hodgins, D. C.; Young, M. M.; Currie, S. R.; Abbott, M. W.; Billi, R.; Brunelle, N.; Nadeau, L. Lower-risk gambling limits: Linked analyses across eight countries. International Gambling Studies 2022, 23(2), 2143546. [Google Scholar] [CrossRef]
  30. Jääskeläinen, P.; Kuusisto, K. Problematic gambling in municipal social work in Tampere, Finland: social workers’ perceptions of service pathways before the casino opening and the health and social services reform. Nordic Studies on Alcohol and Drugs 2025, 42(3), 226–242. [Google Scholar] [CrossRef] [PubMed]
  31. James, R.; O’Malley, C.; Tunney, R. Sociodemographic predictors of latent class membership of problematic and disordered gamblers. Addictive Behaviors Reports 2016, 3, 61–69. [Google Scholar] [CrossRef] [PubMed]
  32. Jones, M.; Boering, P.; Patel, K.; Dymond, S. Gambling, suicide and mental health treatment utilisation in wales: case-control, whole-population-based study. Bjpsych Open 2025, 11(6). [Google Scholar] [CrossRef] [PubMed]
  33. Karlsson, A.; Håkansson, A. Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. Journal of Behavioral Addictions 2018, 7(4), 1091–1099. [Google Scholar] [CrossRef] [PubMed]
  34. Kim, H.; Leslie, R.; Stewart, S.; King, D.; Demetrovics, Z.; Andrade, A.; Hodgins, D. A scoping review of the association between loot boxes, esports, skin betting, and token wagering with gambling and video gaming behaviors. Journal of Behavioral Addictions 2023, 12(2), 309–351. [Google Scholar] [CrossRef] [PubMed]
  35. Kristensen, J. H.; Pallesen, S.; Bauer, J.; Leino, T.; Griffiths, M. D.; Erevik, E. K. Suicidality among individuals with gambling problems: A meta-analytic literature review. Psychological Bulletin 2024, 150(1), 82–106. [Google Scholar] [CrossRef] [PubMed]
  36. Langham, E.; Thorne, H.; Browne, M.; Donaldson, P.; Rose, J.; Rockloff, M. Understanding gambling related harm: A proposed definition, conceptual framework, and taxonomy of harms. BMC Public Health 2016, 16, 80. [Google Scholar] [CrossRef] [PubMed]
  37. Lee, U.; Lewis, L. E.; Mills, D. J. Association between gambling and financial trading: A systematic review. F1000Research 2023, 12, 111. [Google Scholar] [CrossRef]
  38. Loo, J. M. Y.; Oei, T. P. S.; Raylu, N. Psychometric evaluation of the Problem Gambling Severity Index among Chinese gamblers. Journal of Gambling Studies 2011, 27(3), 467–485. [Google Scholar] [CrossRef] [PubMed]
  39. López-González, H.; Estévez, A.; Griffiths, M. D. Internet-based structural characteristics of sports betting and problem gambling severity: Is there a relationship? International Journal of Mental Health and Addiction 2018, 16(1), 136–150. [Google Scholar] [CrossRef]
  40. López-González, H.; Estévez, A.; Griffiths, M. D. Spanish validation of the Problem Gambling Severity Index: A confirmatory factor analysis with sports bettors. Journal of Behavioral Addictions 2018, 7(3), 814–820. [Google Scholar] [CrossRef] [PubMed]
  41. Lorains, F.; Cowlishaw, S.; Thomas, S. Prevalence of comorbid disorders in problem and pathological gambling: systematic review and meta-analysis of population surveys. Addiction 2011, 106(3), 490–498. [Google Scholar] [CrossRef] [PubMed]
  42. Molander, O.; Wennberg, P.; Berman, A. The gambling disorders identification test (gdit): psychometric evaluation of a new comprehensive measure for gambling disorder and problem gambling. Assessment 2021, 30(1), 225–237. [Google Scholar] [CrossRef] [PubMed]
  43. Moreau, A.; Chabrol, H.; Chauchard, É. Psychopathology of online poker players: review of literature. Journal of Behavioral Addictions 2016, 5(2), 155–168. [Google Scholar] [CrossRef] [PubMed]
  44. Moreira, D.; Azeredo, A.; Dias, P. Risk factors for gambling disorder: a systematic review. Journal of Gambling Studies 2023, 39(2), 483–511. [Google Scholar] [CrossRef] [PubMed]
  45. Moreira, D.; Dias, P.; Azeredo, A.; Rodrigues, A.; Leite, Â. A systematic review on intervention treatment in pathological gambling. International Journal of Environmental Research and Public Health 2024, 21(3), 346. [Google Scholar] [CrossRef] [PubMed]
  46. Mundt, A.; Baranyi, G. The unhappy mental health triad: comorbid severe mental illnesses, personality disorders, and substance use disorders in prison populations. Frontiers in Psychiatry 2020, 11. [Google Scholar] [CrossRef] [PubMed]
  47. Nigro, G.; Matarazzo, O.; Ciccarelli, M.; Pizzini, B.; Sacco, M.; Cosenza, M. Positive illusions: the role of cognitive distortions related to gambling and temporal perspective in chasing behavior. Journal of Gambling Studies 2021, 38(3), 889–904. [Google Scholar] [CrossRef] [PubMed]
  48. Nower, L.; Martins, S. S.; Lin, K.-H.; Blanco, C. Subtypes of disordered gamblers: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Addiction 2013, 108(4), 789–798. [Google Scholar] [CrossRef] [PubMed]
  49. Potenza, M.; Mestre-Bach, G.; Demetrovics, Z.; Derevensky, J.; Fernández-Aranda, F.; Gainsbury, S.; …; Kraus, S. Online gambling in youth; 2024; pp. 597–603. [Google Scholar] [CrossRef]
  50. Price, A.; Hilbrecht, M.; Billi, R. Charting a path towards a public health approach for gambling harm prevention. Journal of Public Health 2021, 29(1), 37–53. [Google Scholar] [CrossRef] [PubMed]
  51. Prilutskaya, M.V.; Kuliev, R.S. Gambling behavior in people at different levels of risk of pathological gambling. Neuroscience and Behavioral Physiology 2016, 46(6), 682–687. [Google Scholar] [CrossRef]
  52. Quigley, L. Gambling disorder and stigma: opportunities for treatment and prevention. Current Addiction Reports 2022, 9(4), 410–419. [Google Scholar] [CrossRef] [PubMed]
  53. Roberts, A.; Bowden-Jones, H.; Roberts, D.; Sharman, S. Should GPS routinely screen for gambling disorders? British Journal of General Practice 2019, 69(682), 226–227. [Google Scholar]
  54. Savolainen, I.; Vuorinen, I.; Sirola, A.; Oksanen, A. Gambling and gaming during covid-19: the role of mental health and social motives in gambling and gaming problems. Comprehensive Psychiatry 2022, 117, 152331. [Google Scholar] [CrossRef] [PubMed]
  55. Sevy, S.; Robinson, D.; Holloway, S.; Alvir, J.; Woerner, M.; Bilder, R.; Kane, J. Correlates of substance misuse in patients with first-episode schizophrenia and schizoaffective disorder. Acta Psychiatrica Scandinavica 2001, 104(5), 367–374. [Google Scholar] [CrossRef]
  56. Sharp, C.; Steinberg, L.; Yaroslavsky, I.; Hofmeyr, A.; Dellis, A.; Ross, D.; Kincaid, H. An Item Response Theory Analysis of the Problem Gambling Severity Index. Assessment 2012, 19(2), 167–175. [Google Scholar] [CrossRef] [PubMed]
  57. Sundqvist, K.; Wennberg, P. The Association Between Problem Gambling and Suicidal Ideations and Attempts: A Case Control Study in the General Swedish Population. Journal of Gambling Studies 2021, 38(2), 319–331. [Google Scholar] [CrossRef] [PubMed]
  58. Tulloch, C.; Browne, M.; Hing, N.; Rockloff, M.; Hilbrecht, M. Trajectories of wellbeing in people who live with gamblers experiencing a gambling problem: An 18-year longitudinal analysis of the Household, Income and Labour Dynamics in Australia (HILDA) survey. Plos One 2023, 18(1), e0281099. [Google Scholar] [CrossRef] [PubMed]
  59. Watanapongvanich, S.; Binnagan, P.; Putthinun, P.; Khan, M.; Kadoya, Y. Financial literacy and gambling behavior: evidence from Japan. Journal of Gambling Studies 2020, 37(2), 445–465. [Google Scholar] [CrossRef] [PubMed]
  60. Wild, D.; Grove, A.; Martin, M.; Eremenco, S.; McElroy, S.; Verjee-Lorenz, A.; Erikson, P. Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: Report of the ISPOR Task Force for Translation and Cultural Adaptation. Value in Health 2005, 8(2), 94–104. [Google Scholar] [CrossRef] [PubMed]
Table 1. Sociodemographic characteristics of the study sample (valid %).
Table 1. Sociodemographic characteristics of the study sample (valid %).
Variable Category n %
Gender Male 551 54.3
Female 464 45.7
Age group 18–21 302 29.8
22–35 306 30.1
36–45 301 29.7
46–60 106 10.4
Marital status Married 498 49.1
Single 434 42.8
Divorced 52 5.1
Widowed 9 0.9
Other 22 2.2
Occupation Student 205 20.2
Trade/Services 165 16.3
Industry/Construction 110 10.8
Education/Science 101 10.0
Business/Management 75 7.4
Transport/Logistics 66 6.5
Unemployed 64 6.3
Other occupations* 229 22.5
Income spent on gambling <10% 287 28.3
10–25% 128 12.6
26–50% 42 4.1
>50% 17 1.7
Table 2. Psychometric properties of the bilingual PGSI.
Table 2. Psychometric properties of the bilingual PGSI.
Indicator Value
Cronbach's α 0.846
KMO 0.881
Bartlett χ2 (36) 1465.46
p <0.001
Variance explained 45.9%
Table 3. Distribution of response categories across individual PGSI items in the study population (N = 1,015).
Table 3. Distribution of response categories across individual PGSI items in the study population (N = 1,015).
PGSI item Never n (%) Sometimes n (%) Most of the time n (%) Almost always n (%)
Bet more than you could really afford to lose 661 (65.1) 259 (25.5) 78 (7.7) 17 (1.7)
Needed to gamble with larger amounts to achieve the same excitement 674 (66.4) 253 (24.9) 69 (6.8) 19 (1.9)
Went back another day to try to win back money lost 624 (61.5) 230 (22.7) 100 (9.9) 61 (6.0)
Borrowed money or sold anything to gamble 722 (71.1) 214 (21.1) 60 (5.9) 19 (1.9)
Felt that you might have a gambling problem 729 (71.8) 202 (19.9) 60 (5.9) 24 (2.4)
Felt that gambling caused health problems, including stress or anxiety 752 (74.1) 181 (17.8) 63 (6.2) 19 (1.9)
People criticized your betting or told you that you had a gambling problem 680 (67.0) 211 (20.8) 90 (8.9) 34 (3.3)
Gambling caused financial problems for you or your household 723 (71.2) 192 (18.9) 74 (7.3) 26 (2.6)
Felt guilty about the way you gamble or about what happens when you gamble 686 (67.6) 224 (22.1) 67 (6.6) 38 (3.7)
Table 4. Distribution of PGSI categories in the study sample.
Table 4. Distribution of PGSI categories in the study sample.
PGSI category n %
No risk 574 56.6
Low risk 30 3.0
Moderate risk 136 13.4
Problem gambling 275 27.1
Table 5. Bivariate associations with problem gambling (PGSI ≥ 8).
Table 5. Bivariate associations with problem gambling (PGSI ≥ 8).
Predictor χ2 (df) p-value Cramér's V
Sex 23.20 (1) <0.001 0.151
Age group 6.30 (3) 0.098 0.079
Marital status 7.29 (6) 0.295 0.085
Occupation 29.88 (13) 0.005 0.172
Share of monthly income spent on gambling 86.57 (3) <0.001 0.427
Gambling frequency 91.61 (4) <0.001 0.440
Duration of gambling involvement 13.53 (4) 0.009 0.169
Table 6. Multivariable logistic regression predicting problem gambling (PGSI ≥ 8).
Table 6. Multivariable logistic regression predicting problem gambling (PGSI ≥ 8).
Predictor Adjusted OR 95% CI p-value
Male sex (vs. female) 1.16 0.70–1.90 0.565
Age 22–35 years 0.65 0.38–1.13 0.126
Age 36–45 years 0.79 0.44–1.42 0.430
Age 46–60 years 0.69 0.31–1.52 0.354
Share of income spent on gambling (per category increase) 2.80 1.90–4.12 <0.001
Gambling frequency (per category increase) 2.13 1.62–2.80 <0.001
Duration of gambling involvement (per category increase) 1.14 0.93–1.41 0.212
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.