Clinical Correlates of Sports Betting: A Systematic Review

Sports betting is becoming increasingly widespread, and a growing number of individuals, both adolescents and adults, participate in this type of gambling. The main aim of this systematic review was to assess correlates of sports betting (sociodemographic features, gambling-related variables, co-occurring psychopathologies, and personality tendencies) through a systematic review conducted following the PRISMA guidelines. Relevant studies were identified via searches of NCBI/PubMed and APA PsycInfo databases. Individuals from the general population and/or with a clinical diagnosis of gambling disorder (GD) were included, irrespective of gender and age. In addition, the studies needed to have administered at least one clinical interview/psychometric instrument to assess the presence of problematic gambling/GD, contain at least one group of participants with sports betting, and directly analyze the association between sports betting and any of the following features: sociodemographics, gambling-related variables, co-occurring psychopathologies, and/or personality tendencies. Fifty-four articles were included. Multiple sociodemographic variables have been studied in relation to sports betting. In general, males with high impulsivity have greater tendencies for sports betting. The co-occurrence of certain pathologies, especially substance use or other addictive disorders, was also suggested. Most studies were cross-sectional, assessed participants using self-administered instruments, recruited samples using non-probability online panels, included small samples, had unbalanced samples, and included samples from only one country. Impulsive males may be particularly prone to sports gambling and related problems. Future research should examine prevention strategies that may help prevent the development of sport-betting-related GD and other addictive behaviors in vulnerable individuals.


Introduction
Over the last decade, the availability and accessibility of gambling has been expanding and evolving due to technological and sociological developments (Chóliz, 2016;Jiménez-Murcia et al., 2014). Consequently, consumers of gambling products have changed the way they behave and interact with these products, with a greater number of adults reporting gambling from home from their internet-connected devices (Lejoyeux, 2012). Online gambling has drastically increased the accessibility of gambling and as a result, the potential frequency of gambling and risk to experience symptoms of problem gambling Lejoyeux, 2012). Of all the available forms of online gambling, sports betting is one of the most widely endorsed forms (Jiménez-Murcia et al., 2014;Mestre-Bach et al., 2022).
The popularity of online sports betting (OSB) has been continuing to increase in part due to its legalization in multiple jurisdictions globally . In 2018, sports betting was the most popular form of online gambling in Europe (European Gaming & Betting Association, 2019). With increasing legalization, sports betting establishments have proliferated, where other addictive substances such as alcohol could be consumed (Li et al., 2020). In recent years, novel features have been incorporated into sports-betting experiences, such as cash-out features, additional live in-play betting and request-a-bet options, instant deposits, and micro-betting (Lopez-Gonzalez et al., 2019;Winters & Derevensky, 2019). Nevertheless, the fact remains that sports gambling is often done through illegal bookmakers (Morgan Stanley, 2014) which has resulted in some researchers requesting legislators to develop measures for better regulation of the sports betting market (Hing et al., 2015a).
In addition to broadening availability, some studies have reported on the potential influences of advertising on the development of gambling problems as a result of online gambling including OSB (Bouguettaya et al., 2020;Newall et al., 2019). Despite the role of advertising on increasing the risk of gambling problems associated with OSB, individuals that develop problems with gambling tend to have certain predispositions that increase this risk (Chóliz, 2008;Hing et al., 2015a;Russell et al., 2019aRussell et al., , 2019b. Based on the Diagnostic and Statistical Manual Fifth Edition, gambling disorder (GD) is an addictive condition characterized by persistent and recurrent problematic gambling behavior that generates clinically significant levels of distress and impairments in functioning (American Psychiatric Association, 2013). Although the DSM-5 disorder does not specify types of gambling activities, there is evidence suggesting that certain forms of gambling may be associated with greater risks for the development of GD (Lutri et al., 2018;Stevens & Young, 2010;Williams et al., 2021). Due to sports betting in both online and offline forms increasing in popularity, it is essential to identify demographic and clinical characteristics associated with the risk for the development of GD associated with sports betting.
It is known that individuals engaging in OSB gambling represent a particularly vulnerable group, with a higher proportion of individuals who are single, younger, and of lower socioeconomic status, report an earlier onset of gambling participation, endorse higher rates of substance use disorders, report greater psychological distress and have distinct personality profiles (i.e., higher impulsivity, reward dependence, and novelty seeking) (Estévez et al., 2017;Granero et al., 2020). The evidence suggesting that OSB involves gambling participation among younger individuals is of particular concern as adolescents and emerging adults appear at increased risk of gambling problems (Sarabia et al., 2014).
Other relevant factors that may result in maintaining sports betting behaviors and increase the risk for problem gambling are cognitive biases (e.g., illusion of control).
Cognitive biases may result in a reduction in the perception of risk and possible long-term effects as a result of gambling, and increase the perception that personal skills contribute significantly to gambling outcomes (Chóliz, 2010). However, cognitive biases specific to OSB remain understudied with inconsistent findings having been reported in their role for differentiating between problematic and non-problematic sports betting (Huberfeld et al., 2013).
Although there are increasing studies investigating the relationship between sports betting and problem/disordered gambling, many investigations conceptualize sports betting as one of the many forms of online gambling, rather than as an entity in and of itself. Moreover, few studies have examined the demographic and clinical characteristics of individuals reporting engagement in sports betting and whether certain characteristics differentiate between sports betting with or without problem/disordered gambling. The association between sports betting and GD has rarely been examined, particularly longitudinally, with no systematic review having been published on this topic. Despite the growing number of individuals experiencing clinically relevant problems with sports betting (Mestre-Bach et al., 2022), to the authors' knowledge, no previous systematic review has been conducted examining the clinical characteristics of people engaging in sports betting. This marks a significant gap in the field that should be addressed as such information could facilitate the development of more effective prevention and treatment interventions (Winters & Derevensky, 2019). Taken together, the research question this systematic review aims to answer is: what are the clinical correlates (i.e., sociodemographic features, gambling-related variables, co-occurring psychopathology and personality tendencies) of people engaging in sports betting?

Study Selection
The methodology employed in this review adheres to principles of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Moher, 2009). Relevant studies were identified via searches of NCBI/PubMed and APA PsycInfo databases using the following search terms: "sports bet*" OR "sports wager*" OR "sports gambl*" OR "fantasy sport*" OR "daily fantasy". The final search was conducted on May 28, 2021. No date range limits were applied to avoid selection bias. Only published or in-press empirical studies in peer-reviewed journals written in English, Spanish and French were considered for inclusion. In addition, only studies with an observational or descriptive design (e.g., cross-sectional, longitudinal, case-control) and a quantitative methodology were considered eligible for inclusion. Articles with no abstract, as well as publications that were not full articles, that had a qualitative design or that had the following specific formats (i.e., literature reviews, books, dissertations, case reports or series, editorials, clinical practice guidelines, commentaries, and gray literature) were excluded. Finally, studies evaluating treatment interventions were also excluded.
The present systematic review was performed on the basis of the following eligibility criteria: (1) human samples: general population (including both athletes and non-athletes) and/or individuals with a clinical diagnosis of GD, irrespective of sex and age; (2) administration of at least one clinical interview/psychometric instrument to assess the presence of problematic gambling/GD, (3) involvement of at least one group of participants with sports betting, and (4) direct analysis of associations between sports betting and any of these features: (A) sociodemographic characteristics (e.g., sex, age), (B) gambling-related variables (e.g., GD severity, gambling frequency, gambling-related cognitions, maximum bets), (C) co-occurring psychopathology, and/or (D) personality features (e.g., impulsivity).
Once duplicate results were excluded, all abstracts were screened for inclusion and exclusion criteria. Most studies excluded in this first step either investigated behaviors other than SB, had a dependent variable other than clinical correlates or were published in a language other than English or Spanish. For all studies identified for inclusion, a full text version was retrieved, and all studies were reviewed with regard to their quality and eligibility for the review. In case of exclusion of studies, reasons were documented. Included studies were read thoroughly, and the former defined measures were extracted and included in tables.

Study Assessment and Data Extraction
A two-step process was performed to select the final articles included in the present systematic review. First, two reviewers (EVM and BMM) screened the titles and abstracts of all potential studies individually. For the second level of the screening, articles identified for full review were further screened according to the eligibility criteria by two separate authors (EVM and BMM). Differences in rating between both reviewers were resolved through consensus, with the assistance of a third reviewer (GMB). The complete screening process was conducted using Covidence, a software based on the PRISMA standard and recommended by Cochrane Reviews (Kellermeyer et al., 2018).
A total of 442 records was retrieved from the literature search in the selected databases. After removing 111 duplicates, 155 of the 331 remaining articles were excluded taking title and abstract into account. The remaining 176 records were screened at a full-text level. From the 176 articles screened, 54 were ultimately included in the systematic review. Reasons for exclusion at the full-text screening stage are included in Fig. 1. Data were extracted including the full references of articles, main aims, study designs, sample characteristics and sizes, descriptions of methodology, and results.

Risk of Bias Assessment
The risk bias of the included studies was performed using four items of the Effective Public Health Practice Project Quality Assessment Tool (EPHPP): selection bias, study design, data collection methods and appropriate statistical testing; additionally, a global rating was calculated (Thomas et al., 2004). The EPHPP is a guide to systematically appraise study quality in seven ambits: selection bias, study design, confounders, blinding, data collection methods, withdrawal and dropouts, and intervention integrity and analyses. The tool can be used to evaluate the study quality of observational, cross-sectional, before and after, and randomized controlled trial studies and has good content and construct validity and adequate test-retest reliability (Armijo-Olivo, Stiles, Hagen, Biondo, & Cummings, 2012;Thomas et al., 2004). Being that all the studies had cross-sectional designs, the items specific to intervention studies were excluded (i.e., confounders, blinding, withdrawals and dropouts, and intervention integrity). The study design criteria were also slightly modified to reflect the fact that most studies were cross-sectional, with differentiations based on sample size in both general population (Strong: N > 1250; Moderate: N ranging from 630 to 1249; Weak N < 630) and clinical samples (Strong: N > 100; Moderate: N ranging from 1 3 30 to 99; Weak N < 30) being included. Two authors (JR and LM) performed the quality assessment separately for each study, and discrepancies were resolved.

Ethics
The present article is a systematic review of the literature, and no studies with human participants or animals were performed by any of the authors to conduct this work.

Study Characteristics
Descriptions of the included studies and their findings are summarized in Table 1. Fiftyfour articles were included in the systematic review. Australia was the country with the most published research on sports betting. All studies included in the systematic review TCI-R People who gambling online and not on sports were more likely to be single, be younger, have higher monthly incomes, have begun gambling earlier and have the shortest durations of GD compared to people who bet on sports and gambled offline. People who bet on sports online made higher maximum bets. The online groups had higher cumulative debts compared to the offline group. People who bet on sports online obtained higher scores in persistence (industriousness, determination, ambition, perfectionism) compared to people who bet on online but not on sports and people who bet offline. As for substance use and abuse, no statistically significant differences were identified (Gainsbury et al.,  TCI-R The prevalence of people seeking treatment for online sport betting appears to be an increasingly common reason for consultation from 2005 (0.3%) to 2019 (16.1%). Of people seeking treatment for online sport betting, 49.8% reported substance use and 71.2% a secondary behavioral addiction. Two latent clusters were identified. Cluster 1 (n = 247, 76.5%) grouped patients who were more severly affected by online-sport-betting behaviors. These patients were characterized by being non-married, having lower socioeconomic level, using substances (alcohol, tobacco and other), being of younger age, having earlier onset of gambling, experiencing higher debts due to online sport betting, having higher psychopathological distress (based on overall SCL-90R scores), and demonstrating more dysfunctional personality profiles (higher novelty seeking, lower levels of self-directedness and cooperativeness). Cluster 2 (n = 76, 23.5%) grouped patients who were less affected by online-sport-betting behaviors. These patients were mostly married (or living with a stable partner), with higher socioeconomic levels, older age and later onset of the gambling activity, less substance use, and more functional psychopathological and personality profiles (Håkansson,  Past-year and pastmonth participation in different types of gambling In the total sample, 23% had low-risk gambling, 15% moderate-risk, 10% problem gambling, with a greater percentage of women reporting problem gambling (15% vs 6%). Having moderate-risk or problem gambling was more likely among people who bet on sports. Comparing people who bet on sports in the past month versus to past year, the former group was more likely to report other forms of gambling.
The group was also more likely to report past indebtedness and greater severity of gambling problems (moderate-risk and problem gambling of 18% and 13%, respectively)

K6
Felt need to seek treatment for alcohol or drug problems Overall, 54% of the sample reported live sports betting and 60% reported non-live sports betting. In the total sample, 23% had low-risk gambling, 19% had moderate risk, and 13% had problem gambling. Moreover, 4% of those reporting only live sports betting had problem gambling, and an additional 21% were in the moderate-risk range.
A total of 12% of the sample reported a history of over-indebtedness and 8% expected over-indebtedness, this being reported by only 4% of people who bet on live sports. Problem gambling was associated with psychological distress, recent online casino gambling, and recent combined online casino gambling and live sports betting (OR = 5.12), but not live-sports betting alone. History of over-indebtedness was associated with recent combined online casino gambling and live sports betting (OR = 1.57), but not live-sports betting alone Help-seeking in relation to gambling problems Comparing groups with non-problematic online sports betting to problematic online sports betting, individuals with problematic online sports betting were more likely to be gamble on sports offline and were more likely to consider themselves semi-professional gamblers. They were also more likely to use illicit drugs while gambling and reported greater psychological distress. Potential risk factors for online sports betting were being male (B = 1 . Compared to people with problematic EGM gambling, those with problematic online sports betting were younger, wer more educated, and engaged in fewer forms of gambling. Compared to people with problematic online race betting, those with problematic online sports betting were younger and less likely to have been born in Australia People who bet on sports appearing at a higher risk of problem gambling were those who were young, male, single, educated, and employed full-time (or were full-time students). People who bet on sports appearing at risk of problem gambling were found to have greater frequency and expenditure on sports betting, greater diversity of gambling involvement and more impulsive responses to betting opportunities, including in-play live action betting (as opposed to betting before a game). Normative influences (i.e., higher subjective norms) from media advertising and from significant others were also associated with greater problem-gambling risk In the general population that purchased sports-lottery tickets, four clusters were identified. Cluster I, people who gambled casually (45%), had the lowest score on each dimensions of the SAPG, with no negative impact from gambling or any psychological symptoms from problem gambling. Cluster II, people who gambled in an escalating fashion (28%), had higher scores on all dimensions of the SAPG than those in Cluster I, especially on the compulsive disorder dimension. This cluster was considered a group whose gambling might escalate in the future. Cluster III, people who gambled in an at-risk fashion (11%), had significant social and financial problems that may need social intervention to address and resolve concerns associated with gambling. Cluster IV, people who gambled in a compulsive fashion (10%), had higher scores on the compulsive disorder and over expectation dimensions of the SAPG. These players may need psychosocial intervention to address and resolve concerns associated with gambling Among those individuals who bet on sports, no differences were identified based on age or gender in terms of advertising impact on the individual. People experiencing more severe gambling problems also reported more knowledge of bookmakers' brands, more similarity to the main story characters in sports-betting advertisements and a higher perceived influence of advertising on their behavior in 2012, 37.6% reported they considered participating in a fantasy league with fees and prize money as gambling. Male participation in free fantasy sports leagues was 6-7 times more likely compared to females. Among male students at risk for or reporting a gambling problem (1.9%), more than half engaged in free fantasy leagues and approximately half also played in fee-based fantasy leagues. Almost half of female student-athletes with at-risk/probable-problem gambling (0.1%) reported playing free fantasy leagues, and a quarter reported playing fee-based fantasy leagues. Approximately half of individuals with non-problem gambling reported playing fantasy sports for free and less than a quarter reported playing fantasy sports for monetary gain. In the females with at-risk/probable-problem gambling, there was a 35.9% increase in free fantasy sports playing, and a 23.2% increase of fee-based fantasy sports playing. Most males with at-risk/probable-problem gambling (40.9%) were engaged in between two and five fantasy leagues. Males (65.5%) and females (66.7%) with at-risk/probable-problem gambling did not consider fantasy sports a gambling activity

DSM-5 criteria
Overall, 11.5% reported participating in fantasy sports in the past year with 43.5% of these people playing for money. Males were significantly more likely to play fantasy sports, both for no money and for money. Among those who played fantasy sports without money, 14.9% endorsed ≥ 1 DSM-5 criteria and the rate was higher for female players (26.7%) than male players (11.8%). Among those played fantasy sports for money, 26.9% endorsed ≥ 1 DSM-5 criteria and the rate was higher for male players (27.8%) than female players (16.7%). Fantasy participants (regardless of whether they played for money) were over five times more likely to endorse ≥ 1 DSM-5 GD criteria, and those who participated in fantasy sports for money were significantly more likely to experience one or more gambling-related problems. Male fantasy participants (regardless of whether they played for money) were almost two times more likely to endorse ≥ 1 DSM-5 GD criteria, and males who participated in fantasy sports for money were nearly three times more likely to experience one or more gambling-related problems. Female fantasy participants (regardless of whether they played for money) were almost three times more likely to endorse ≥ 1 DSM-V GD criteria (Martin et  Fantasy sports players were more likely to report gambling compared to non-players. Being male was the only significant statistical predictor of gambling-related problems, and fantasy sports players were more likely than non-players to experience gambling-related problems. Being male was the only significant statistical predictor of fantasy sports playing, and students with gambling-related problems were more likely to play fantasy sports than those without gambling problems. Fantasy players with gambling-related problems were significantly more likely to gamble on fantasy sports than those without problems Participants who paid an entry fee/deposit to play fantasy sports gambled more frequently than those who did not (p < .001). Those who played fantasy sports gambled more frequently than those who did not play fantasy sports (p < .001), and individuals who played DFS gambled more frequently than those who only played season-long fantasy (p < .01). Among the total sample, 4.9% endorsed one or more DSM-5 GD criteria; people who gambled who played fantasy sports of any kind endorsed more criteria than those who did not play fantasy sports (p < .001), and people who gambled who played DFS endorsed more criteria than those who only played season-long fantasy (p < . (CSB) had current problem gambling (PGSI score 8), compared to 6.7% of those who did not use CSB products.
PGSI scores were significantly higher amongst people using CSBs (p < 0.001). People using CSBs had higher illusions of control (p = 0.031), experienced more gambling-related harms (p < .001), and had higher scores in the consumption screen for problem gambling (p < 0.001) Past year engagement G-SAS IGT, The colour-shapeshifting task, BIS Most people who bet on sports reported gambling-related behavioral problems in the mild (n = 23) or moderate range (n = 11), with only one participant falling into the extreme and severe symptom category on the G-SAS. Between people who bet on sports and those who did not, no significant differences were reported in the number of chosen disadvantageous decks nor the outcomes on the IGT. Regarding cognitive flexibility, error rates differed significantly between groups (p = .02), where people who bet on sports (13%) committed more errors than those who did not (8% and III) Gambling experiences Sports wagering appears to be a frequent activity among men.
Rates of betting on one's own team decreased over time, with the highest rates being in 2008 (p < .001). However, no significant changes were reported for betting on another college team. Participation in fantasy leagues involving entry fees and prize money significantly increased from 17% in 2008 to 20% in 2016. Females reported lower rates of gambling participation compared to men, and the rates of sports wagering decrease over time. In 2016, 39% of men and 20% of women reported believing that sports wagering was acceptable as long as they were wagering on a sport in which they do not participate. Similarly, in 2016, 49% of men and 31% of women perceived sports wagering as a harmless pastime and 306 (26.7%) as having problem gambling. Potential risk factors for having moderate-risk or problematic sports betting included: a higher monthly sports betting expenditure, higher gambling urges, alcohol-use-related concerns, and lower self-control. The most significant statistical predictors of moderate-risk or problematic sports betting included: being motivated by money, experiencing greater gambling urges, having greater erroneous cognitions, alcohol-userelated concerns, and lower self-control. Those at higher risk of sports-betting-related problems were younger, spoke a language other than English as their main language, and had lower self-control 36.8% reported placing at least 1% of their sports bets on micro-events in the previous year. Those who reported betting on micro-events scored higher on impulsivity and 77.8% were classified as having problem gambling in contrast to 28.7% of those who did not bet on micro-events. People who bet on micro-events were younger, were more likely to be single, had a higher level of education, engaged in more forms of gambling, bet on sports more frequently, bet on a greater number of different sports, placed more bets per sports-betting day, placed more of their bets via telephone, perceived that their peers bet more frequently on sports, reported a lower frequency of exposure to gambling-related advertising, and used promotions more frequently. In a regression model, statistical predictors of betting on micro-events were: betting on a higher number of different sports, placing more bets per sports-betting day, placing more of their bets by telephone, and having problem gambling. Placing a higher proportion of bets on microevents was significantly associated with speaking English as a main language at home, having problem gambling, placing a lower proportion of bets via the internet, and placing one's last impulse bet more recently. Significant statistical predictors of placing a higher proportion of sports bets on micro-events included having problem gambling, having more accounts with different operators and placing a lower proportion of bets via the internet People with moderate-risk and problem gambling reported significantly higher sports-betting expenditures than those with non-problem gambling. Compared to people with nonproblem gambling, those with problem gambling placed bets on more days per month and higher proportions of bets during the match and on micro-events within the match. People with problem gambling reported that their peers bet a significantly higher amount of money. People with low-risk, moderate-risk and problem gambling reported using a higher number of different types of promotions and had higher impulsiveness scores compared to people with non-problem gambling Three subgroups of people with problem gambling were identified: behaviorally conditioned (not fitting in the other groups), emotionally vulnerable (reporting at least one episode of anxiety or depression before having a gambling problem) and antisocial/impulsivity (reporting antisocial personality disorder or high novelty-seeking). The antisocial/impulsivity gambling group preferentially chose semi-skillful gambling such as sports gambling, whereas the emotionally vulnerable group was more likely to gamble on games of chance. The antisocial/impulsivity group had a greater proportion of co-occurring disorders: 53.5% had a history of mood disorders, 36.4% a history of anxiety disorders, 50.5% a history of addictive disorders other than gambling, and 9.1% a history of a psychotic condition. Sports betting was preferred by 12.4% of the antisocial/impulsivity group, versus 8.3% of the emotionally vulnerable group and 10.8% of the behaviorally conditioned group (Wang et  to bet on sports was statistically predicted by attitude (p < 0.001) and subjective norms (p < 0.001). Actual sportsbetting behaviors were predicted by intention (p < 0.001) and perceived behavioral control (PBC) (p = 0.02). Both attitudes and subjective norms, directly and indirectly, influenced college students' sports-betting behaviors (p < 0.001).
In the lower-risk gambling group, sport gambling intention was mostly statistically predicted by attitude (p < 0.001), followed by subjective norms (p < 0.001). Within this group, sports betting was statistically predicted by intention (p < 0.001) and PBC (p = 0.016). In the higher-risk gambling group, sports-betting intention was statistically predicted by subjective norms (p < 0.001) and attitudes (p = 0.004), but not by PBC. Finally, the problem-gambling group's sportsbetting behavior was statistically predicted by both intention (p < 0.001) and PBC (p = 0.02). For both groups, the indirect effects of attitude and subjective norms via intention on behavior were significant. The role of attitude toward sports gambling was more influential in predicting the gambling behaviors of the lower-risk gambling group (p = 0.04) of gambling on at least one activity during the initial COVID-19 lockdown. 5.4% of males betting on sports and 3.9% of females betting on sports experienced problem gambling (PGSI score ≥ 8) during the initial COVID-19 lockdown. A further 10.7% of men and 8.3% of women experienced moderate-risk gambling (PGSI score of 3-7).
Men were more likely to experience problem gambling if they were younger (under 35 compared with over 55 years) (AOR = 5.24), had lower rather than higher wellbeing scores (AOR = 2.23), started a new form of gambling during lockdown (AOR = 2.50), or had changed their level of spending on gambling. For women, age (being under 35 years), lower rates of wellbeing, shielding status, and increases in gambling frequency were all associated with moderate-risk or problem gambling of gambling behaviors Former athletes had the highest frequency of involvement in sports gambling (70.0%), followed by current athletes (40.0%) and non-athletes (20.0%). In athletes (current and former), there was a relationship between the sport that they played and the sport on which they gambled (p < .001). In general, former athletes were more likely to participate in skill-based forms of gambling such as sports gambling and poker, whereas non-athletes were more likely to engage in forms of gambling that are based predominantly on chance had a cross-sectional design. Sample sizes ranged from 60 to 20,587. One study included only females (McCarthy et al., 2018), three studies included only males (Barrera-Algarín & Vázquez-Fernández, 2021;DiCicco-Bloom & Romer, 2012;Estévez et al., 2017), and the remaining 50 studies included both males and females. The percentage of males and females included in the last group of studies varied considerably, with the percentage of females being lower than that of males in most cases, with the exception of eight studies (Baggio et al., 2018;Hing & Haw, 2009).

Assessment of Sports Betting and GD
Most studies assessed past-year engagement in sports betting, online sports betting, or fantasy sports. Some studies assessed engagement in sports betting during average weeks or months. Other factors assessed included: sports gambling expenditures, frequency of sports lottery purchasing in a week, time commitments to daily sports lottery-related activities, types of sports lottery purchased, channels used to bet on sports, devices used, engagement with in-play betting, gambling activities participation (frequency, seasonal fantasy sports betting, daily fantasy sports betting, and sports betting in general), monthly sports-betting expenditures over the past year, frequency of sports-betting behaviors, proportion of sports bets in land-based venues, via the internet and via telephone calls, payment methods used for sports betting, and age of first sport betting. The psychometric instruments used to assess GD were heterogeneous and are described in Table 2. The most frequently used measures to assess for the presence of problem gambling and/or GD were PGSI (Ferris & Wynne, 2001), DSM-IV criteria (American Psychiatric Association, 2010) and DSM-5 criteria (American Psychiatric Association, 2013).

Risk Bias Assessment Results
As depicted in Table 3, most studies had global ratings corresponding to strong and moderate quality, almost in equal proportions (strong n = 24; moderate n = 25), and only 5 studies were rated as having a weak global quality.

Sports Betting and Sociodemographic Characteristics
Sports betting was highly associated with sex, with males being more likely to participate in sports-betting and fantasy-sports leagues (Håkansson et al., 2017;Marchica & Derevensky, 2016;Martin & Nelson, 2014;Richard et al., 2019). Differences have also been reported between males and females in the association between sports betting and substance use. While in the case of females, this association seems to be relatively weak, for males it appears to be particularly strong (Baggio et al., 2018). Regarding age, Marchica et al. (2017) observed that students aged 16-19 years were more likely to exhibit features of problem gambling when they engaged in sports-related gambling regularly, compared with younger adolescents aged 13-15 years.
Numerous attempts have also been made to define a phenotype related to sports betting by considering multiple sociodemographic variables together. Findings suggest that individuals engaging in sports betting are more likely to be male, single, and younger than non-sports-betting individuals (Cooper et al., 2021). They are also more likely to be college-educated, have full-time jobs with higher salaries, and report betting a higher percentage of their monthly incomes (Cooper et al., 2021). In a predictive model of sports betting delineated by DiCicco-Bloom and Romer (2012), non-Hispanic Black ethnicity was a significant statistical predictor of sports betting, as well as having a friend who gambles or approves of one's gambling. Jiménez-Murcia et al. (2021) reported that being male, younger, engaging in more frequent sports betting, and presenting higher impulsivity and lower self-directedness levels, among other measures, were factors increasing the risk of online sports betting. The highest frequency of sports betting was observed in males, with high social status and low education levels also being implicated (Jiménez-Murcia et al., 2021). Granero et al. (2020) identified two latent clusters of sports betting. Cluster 1 included individuals most involved in online sports betting. This group was characterized by being younger and unmarried, having a lower socioeconomic status, engaging in gambling activities early in life, reporting higher gambling-related debts, and reporting more substance-use disorders, worse psychopathology and dysfunctional personality traits. Cluster 2 included individuals less involved in online sports betting. These individuals were older, mostly married or with a stable partner, and were of higher socioeconomic status.

Problem and Disordered Gambling Severity
A significant association between sports betting and problem gambling was reported in most studies (Baggio et al., 2018;Lopez-Gonzalez et al., 2020;Nweze et al., 2020; Cooper et al. (2021) observed that, compared to non-sports-betting individuals, those betting on sports were more likely to report problem gambling, and this could be related to group differences in attitudes towards gambling, cognitive distortions, gambling motivations or the number of gambling activities performed. Similarly, Gainsbury et al. (2019) reported a positive correlation between a higher frequency of online sports betting and problem gambling. In this vein, Bodor et al. (2018) reported that participants with alcohol use disorder who reported a higher frequency of sports betting were more likely to exhibit at-risk/problem gambling or to have GD. In DiCicco-Bloom and Romer's study (2012), symptoms of GD including tolerance and worry were statistically predictive of sports betting, while withdrawal and subjective loss of control were not. Not all studies identified clear associations between sports betting and problem gambling. For example, Håkansson and Widinghoff (2020) highlighted associations between problem gambling and the combination of online casino gambling and live sports betting. However, this association was not significant for live-sports betting alone.
Some studies were indicative of a relationship between sports betting and GD severity, with sports betting being a statistical predictor of problem gambling Orlowski et al., 2020;Phillips et al., 2013;Russell et al., 2019aRussell et al., , 2019b. Others have reported that problem-gambling severity was the strongest statistical predictor of the frequency of sports betting (). Individuals participating in fantasy sports, including those playing fantasy sports for money, were more likely to report problematic gambling compared to those without such engagement (Lopez-Gonzalez et al., 2019;Martin et al., 2016Martin et al., , 2018, (Martin & Nelson, 2014). Regarding adolescents, the strongest statistical predictor of at-risk gambling for individuals aged 13-15 years was regular participation in daily fantasy sports (Marchica et al., 2017).

Other Gambling Activities and Gambling Frequency
Participation in sports, lottery, and slot-machine gambling have been found to be statistical predictors of sports betting (DiCicco-Bloom & Romer, 2012). Marchica et al. (2017) found that at-risk gambling was statistically predicted by all forms of sport-relevant gambling activities. Other gambling-related variables considered potential risk factors for online sports betting included a higher frequency of sports betting in general and a greater presence of negative attitudes towards gambling .
In Estévez et al.'s study (2017), online sports betting was associated with higher wagers compared to non-sports online betting and offline gambling. In this vein, Håkansson and Widinghoff (2020) reported that people who combined online casino gambling and live sports betting were more likely to report over-indebtedness, which did not occur in the case of live-sports betting alone. Gambling-related factors, such as betting on a greater number of different sports, more frequent exposure to promotions, and more positive attitudes towards them were statistical predictors of betting on micro-events within sporting events (Russell et al., 2019a(Russell et al., , 2019b or greater intended frequency of sports betting ().

Sports Betting and Psychopathology
In Gainsbury et al.'s study (2019), psychological distress was associated with a higher frequency of sports betting. Similarly,  noticed that problematic online sports betting, as compared to non-problematic online sports betting, was associated with more psychological/emotional distress. Likewise, the use of alcohol and other substances, as well as the presence of other behavioral addictions, have been reported in different studies, with percentages being substantial, ranging from 22.1% to 71.2% among individuals engaged in sports betting (Bodor et al., 2018;Granero et al., 2020;.

Sports Betting and Personality Tendencies
Higher levels of impulsivity, sensation-seeking and positive and negative urgency have been reported among people betting on sports (Cooper et al., 2021). Among the multiple factors analyzed, impulsivity (and more specifically negative urgency) was the measure that most distinguished non-sports-betting and sports-betting groups. DiCicco-Bloom and Romer (2012) observed that one of the most significant statistical predictors of sports betting was sensation-seeking.
When analyzing impulsive in-play bets, Hing et al., (2018aHing et al., ( , 2018b reported that these types of wagers were statistically predicted by problem gambling and greater buying impulsiveness, among other factors. In this vein, Hing et al., (2018aHing et al., ( , 2018b observed that people who gambled on sports immediately before a match were characterized by higher levels of impulsivity, which is in concordance with the study by Jiménez-Murcia et al., (2021).
In reference to persistence as a personality feature, mixed results have been reported. Whereas in some studies it was reported that online sports betting was linked to higher levels of persistence, compared to non-sports online betting and offline gambling (Estévez et al., 2017), others have found that greater sports-betting frequency associated with lower persistence (Jiménez-Murcia et al., 2021). Finally, it has been suggested that, in the case of sports betting, other relevant statistical predictors of problem-gambling severity include lower self-directedness and greater psychopathological distress (Jiménez-Murcia et al., 2021).

Discussion
The increased participation in sports betting facilitated by online-gambling modalities has led to an increase in the number of sports-betting patients treated in behavioral addiction units (Mestre-Bach et al., 2022). A significant gap in the literature has involved conceptualizing OSB as simply another form of online gambling without attending to the potentially specific risks associated with online or offline forms of sports betting. The present systematic review was conducted to obtain clinically relevant information on sociodemographic characteristics, gambling-related variables, co-occurring psychopathology and personality features related to sports betting.
Most studies included in this systematic review were published within the last decade, underlining the novelty of research in this area. The number of studies reviewed is sufficient to establish sociodemographic and clinical profiles of people who bet on sports who may be at an increased risk of problem or disordered gambling. The present study has synthesized research findings revealing distinctive characteristics of individuals reporting sport betting who suffer from gambling problems.
Gambling problems related to sports betting are associated with being younger, male, single, and college-educated, and experiencing a faster evolution of GD which is often associated with higher levels of severity compared to other subtypes of individuals with gambling problems. We also found that sports betting with gambling problems to be associated with impulsivity sensation-seeking, psychological distress, frequent use of alcohol or other substances and other (non-GD) behavioral addictions. In this sense, sports betting is linked to specific personality profiles with certain biologically based predispositions (e.g., impulsivity, compulsivity and sensitivity to reward and punishment) that may promote vulnerability to addictions. Such an endophenotype may be particularly concerning as it is not only a risk factor for problematic engagement in sports betting, but also a factor that may maintain problematic behaviors across time.
Findings from this systematic review have important implications for prevention and treatment. As sports betting related to problem-gambling risk is linked to younger age, individuals within this subset of the population should be considered a target for preventative actions and programs. As younger individuals with GD are also more likely to drop out of treatment, bolstering the effectiveness of current preventative approaches to decrease the risk of individuals developing problems with gambling is important (Estevez et al., 2021;Ford & Håkansson, 2020;Granero et al., 2020;Jiménez-Murcia et al., 2015). Moreover, as sports-betting advertising has expanded as a form of marketing in sporting events (Bouguettaya et al., 2020;Newall et al., 2019), preventative messages included within these advertising campaigns could help increase awareness of the risks associated with sports betting and knowledge as to when it is important to seek out additional consultation or treatment.
Besides the homogeneity found in the description of people who bet on sports (i.e., male, younger age, single), a recent study identified sports betting with gambling problems to be associated with other distinct features (i.e., earlier onset of gambling activity, higher psychopathological distress, and more dysfunctional personality profile) (Granero et al., 2020). Since this study was conducted in Spain, and most of the studies included within this review were performed in Australia, it will be interesting to increase our efforts in differentiating features linked to sports betting based on cultural backgrounds. This may lead to the development of more specific and effective treatments that are tailored to the needs of people with sports-betting-related GD based on their geographic locations and cultural backgrounds.
Regarding personality, both impulsivity and sensation-seeking have been reported to be frequently present in this profile of individuals with gambling problems (Zilberman et al., 2018) and have been associated with a higher severity or disordered gambling among those meeting the diagnostic criteria (Hing et al., 2018a(Hing et al., , 2018b. This may be particularly worrying considering that the types of wagers becoming available via OSB are based on microevents within sporting events and include in-play bets, which may be particularly attractive to impulsive individuals. This tendency may lead people to spending more money or wagering more frequently than they had initially intended, resulting in various psychological or financial consequences. These findings suggest the importance of the inclusion of features within OSB websites or applications to deactivate certain forms of wagering based on individuals' preferences or predetermined level of risk. This may be an important aspect to consider in discussions involving regulators and companies providing sports-betting products to an ever-increasing population of people across the globe.

Limitations and Future Research
The studies included in this review have noteworthy limitations. Most studies (1) were cross-sectional, meaning that causality could not be inferred and therefore longitudinal studies are needed to understand temporal relationship between the measured variables; (2) assessed participants using self-administered instruments, which may be subject to biases associated with social desirability or introspective differences; (3) recruited samples using non-probability online panels, which limits the generalizability of the results and could have promoted greater selection of people who bet online versus offline; (4) included small samples; (5) had unbalanced samples, with very low representation of women; and (6) used different instruments to assess each of the measured variables, with this heterogeneity limiting the generalizability of the results. Further, given the heterogeneity of the samples and their geographic locations, with different legislations in different countries, comparability between the studies is complicated and limited.
Limitations specific to the present systematic review should also be noted. First, only studies in English, French and Spanish were included, which could limit generalizability. Second, in some countries, there is considerable gray literature related to GD, as highlighted by other authors (King et al., 2020), which was excluded in the present systematic review due to the desired empirical rigor of the included articles. Third, although the present review was focused on specific sociodemographic and clinical/psychological variables associated with sports betting, there are other factors, including neurobiological, physiological, and sociocultural variables that may be associated with sports betting that were not included. These are hypothetically important variables that should be considered in future reviews.
Based on the findings included within the present systematic review, future investigations of the following topics should be considered. First, empirical studies focused on fantasy sports (both daily and league-based) and their possible associations with GD should be conducted. There was a paucity of such studies included in this review, and this is an important domain of sports betting worth studying. Second, longitudinal studies across different developmental periods (e.g., adolescence, emerging adulthood, adulthood) and involving vulnerable groups are needed to understand the trajectories or maintenance of sports betting and problem/disordered gambling across time while considering factors that may moderate or mediate such relationships.

Clinical Implications
Further exploration of the phenotypes related to sports betting, among both adolescents and adults, especially those with problem gambling, is clinically important. Data obtained regarding risk and maintenance factors associated with sports-gambling behaviors should help treatment-development efforts. Likewise, delving into the possible co-occurrence of problem gambling related to sports betting and other disorders may help to understand poor treatment outcomes in these individuals and promote more personalized interventions.
As described in this systematic review, the findings of the studies published to date suggest that people with sports-gambling-related GD have a specific sociodemographic, clinical and personality profile that differentiates them from others with GD. Thus, their youth, the rapid evolution of the problem, the cognitive distortions (minimizing the risk of sports betting and overestimating their skills and knowledge to obtain prizes), as well as the associated personality traits (high impulsivity and novelty seeking), suggest the need to implement tailor-made treatment programs. In addition, all the characteristics mentioned above are associated with higher likelihoods of dropout from treatment programs. Therefore, cognitive restructuring strategies, the implementation of healthy and alternative leisure activities and the use of new technologies (such as serious games) for the treatment of these underlying personality characteristics, but from a motivating and entertaining environment, may be particularly useful.

Conclusions
Sports betting is becoming increasingly widespread, and a growing number of individuals, both adolescents and adults, are involved in this form of gambling. The present systematic review included studies that explored the sociodemographic and clinical characteristics associated with sports betting. Many studies indicated associations between sports betting and problem gambling. Further, findings indicated that males with high levels of impulsivity reported more frequent engagement in sports betting and had high likelihoods of experiencing gambling problems. Associated psychiatric disorders, especially substance or behavioral addictions, have also been reported. Taken together, the synthesis of findings from this review are an initial step in the direction of identifying individuals at elevated risk for problems associated with sports betting, with further research investigating sports betting and fantasy sports being necessary as the social and technological context of sports betting continues to evolve.
Authors' Contribution SJM, MNP, and JLD designed the study. BMM, EVM and JR carried out the literature search and the screening process. LM and JR conducted data extraction. GMB with the support of the rest of the authors were in charge of writing the manuscript.
Funding Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This research was funded by International Center for Responsible Gaming (ICRG). Dr. Potenza's involvement was supported by the Connecticut Council on Problem Gambling. This work was additionally supported by a grant from the Ministerio de Ciencia e Innovación (grant PDI2021-124887OB-I00) and co-funded by FEDER funds/European Regional Development Fund (ERDF), a way to build Europe. The research was also funded by the Delegación del Gobierno para el Plan Nacional sobre Drogas (2021I031) and by AGAUR-Generalitat de Catalunya (2021-SGR-00824).We thank CERCA Programme/Generalitat de Catalunya for institutional support and Instituto de Salud Carlos III (ISCIII).

Data Availability
The data that support the findings of this study are available from the corresponding author upon request.

Declarations
Conflict of interest Dr. Potenza discloses that he has consulted for and advised Game DayData, Addiction Policy Forum, Baria-Tek, AXA, Idorsia, and Opiant Therapeutics; been involved in a patent application with Yale University and Novartis; received research support from the Mohegan Sun Casino and the Connecticut Council on Problem Gambling; consulted for or advised legal and gambling entities on issues related to impulse control and addictive behaviors; provided clinical care related to impulse-control and addictive behaviors; performed grant reviews; edited journals/journal sections; given academic lectures in grand rounds, CME events, and other clinical/scientific venues; and generated books or chapters for publishers of mental health texts. Dr. Derevensky has had a number of consultancy engagements from gambling operators and has provided webinars internationally on gambling and gaming disorders. He has also been the recipient of multiple government research grants. He has also worked as the Director of Research for the Florida Council on Compulsive Gambling and has provided expert testimony to government regulators internationally. Dr. Jimenez-Murcia received consultancy honoraria from Novo Nordisk. The rest of the authors declare no conflict of interest with the content of this manuscript. 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.

Ethics Approval
The present article is a systematic review of the literature and no studies with human participants or animals were performed by any of the authors to conduct this work.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.