Is there a health inequality in gambling related harms? A systematic review

Background Here we present a systematic review of the existing research into gambling harms, in order to determine whether there are differences in the presentation of these across demographic groups such as age, gender, culture, and socioeconomic status, or gambling behaviour categories such as risk severity and participation frequency. Primary and secondary outcome measures Inclusion criteria were: 1) focus on gambling harms; 2) focus on harms to the gambler rather than affected others; 3) discussion of specific listed harms and not just harms in general terms. Exclusion criteria were: 1) research of non-human subjects; 2) not written in English; 3) not an empirical study; 4) not available as a full article. Methods We conducted a systematic search using the Web of Science and Scopus databases in August 2020. Assessment of quality took place using Standard Quality Assessment Criteria. Results A total of 59 studies published between 1994 and 2020 met the inclusion criteria. These were categorised into thematic groups for comparison and discussion. There were replicated differences found in groups defined by age, socioeconomic status, education level, ethnicity and culture, risk severity, and gambling behaviours. Conclusion Harms appear to be dependent on specific social, demographic and environmental conditions that suggests there is a health inequality in gambling related harms. Further investigation is required to develop standardised measurement tools and to understand confounding variables and co-morbidities. With a robust understanding of harms distribution in the population, Primary Care Workers will be better equipped to identify those who are at risk, or who are showing signs of Gambling Disorder, and to target prevention and intervention programmes appropriately. Supplementary Information The online version contains supplementary material available at 10.1186/s12889-021-10337-3.

Migrants experienced similar gambling harms to the general population including selling possessions, relationship breakdown, mental health problems, drug use or drug selling, homelessness, domestic violence, sex work and suicide. Participants felt harms were exacerbated by factors including lack of a 'safety net', difficulty accessing informal support. Sub-Saharan African migrants (particularly men) felt that if they lost money they would feel a loss of community status, other migrants felt that they could not easily undo financial hardship as additional employment would have family consequences. Mild financial, psychological and work/study harms were broadly distributed across all PPGM groups. Mild financial and psych harms were common. Health, relationship and social deviance harms were mostly restricted to highest risk category (over 50% of reports). Mild harms were the most prevalent and were more common in gamblers not experiencing impaired control.
Severe financial harms were more likely in those displaying control issues. In psych harms more than 50% of all harms were reported in the lower risk categories, and more than 50% of severe harms were in the high risk group. Work/study and social deviance harms had low prevalence overall, health harms and relationship harms both had around 50% of all harm from the high risk group. Hiding debt, stealing, hidden bills leading to further debt. Use of online gambling reduced shame but the double life led to stress and emotional deterioration, mood swings, irritability, relationship issues, arguments, self-perception changes (low-esteem increased shame).
One individual showed changing perception of reality as he stole from work but did not process that this was theft. Verbal abuse was most common from young male gamblers towards their parents. Older males were found to coerce money from their wives.

Gender
Physical abuse was equally common between genders

Culture
Where traditional gender roles were present a mother gambling was seen to lead to child neglect, followed by father leaving work to care for the children and therefore causing financial hardships. The gambler was seen as self-centred and family relationships became strained.

Risk Severity
Among problem gamblers 93.8% indicated at least sometimes betting more than they could afford to lose, 92.9% reported betting larger amounts to get the same thrill and 96.2% reported loss chasing. A significantly higher proportion of moderate risk gamblers (18%) experienced family arguments than low risk gamblers (0.9% Counsellors noted that cultural acceptance was high for gambling and so those with a problem had a strong support network. However when discussing harms isolation from community was a key factor, with family and relationship difficulties. Harms included financial and personal distress, debt, cut off utilities, crime, loss of employment, homelessness, lack of food and no support. Non-Indigenous counsellors highlighted credit card debt as a key harm in the community, using one card to pay off another and getting into a cycle of loss chasing. Gamblers hid their losses due to shame, low self-esteem and guilt and these were highlighted as reasons help wasn't sought out. In contrast a win would lead to high popularity and community power.
Family issues were described as missing key events, neglecting children, lying and arguments, violence and breakups. These relationship issues sometimes led to social isolation. Within Indigenous communities' solitary gamblers were viewed as poor role models and were believed to have lost their cultural values. Community gamblers were seen as those with positive consequences and strong family groups.
Hubert and Griffiths Older participants scored higher on problem gambling and risk behaviour but there was no increase in harms.

Gender
Males scored significantly higher on psychological, social and financial harms but also higher for problem gambling and risk behaviour.

SES
Family life and parents' education level had no significant effect on harms. Lower achievers in school reported higher psychological harms. There was no significant differences in harms across school types

Risk Severity
There was a weak significant correlation between success in school and risk score. Gambling related thoughts of self-harm were more frequent in males. This contrasts with non-gambling thoughts of self-harm. Actual acts of self-harm were not related to gender.

SES
Gambling related thoughts of self-harm were more frequent in the unemployed but not linked to marital status. Gambling related acts of self-harm were also associated with unemployment, but not marriage status. Thoughts of harm also related to parents gambling.

Gambling Behaviour
Thoughts of harm were associated with years gambling Gambling for mood modification and money was associated with thoughts of self-harm. Gambling for enjoyment was negatively associated. Non-gambling related self-harm was also significantly related to gambling for mood modification and inversely with gambling for enjoyment. Harmonious passion (someone choosing to gamble) was related to positive emotions and thoughts, and negatively related to feeling judged or guilty. However after engaging in gambling it no longer related to guilt. Harmonious passion was not related to anxiety. Obsessive passion (feeling compelled to gamble) was strongly associated with guilt, anxiety, and negative emotions. Obsession was negatively related to feeling in control and having fun.

Game Choice
Casino users reported more positive and negative outcomes overall than lottery players. The prevalence of gambling within prison was higher than a general population but prevalence of gambling before incarceration was significantly lower. High rate mid-20s peak offenders were 5.3x more likely to chase losses and experience harms. Rate chronic persisters were 3.7x more likely. There was no significant link between gambling and drug use, but occasional gamblers were more likely to abstain from alcohol and drugs.
Risk Severity Nearly 2/3 who scored highly on the PGSI abstained from drugs and alcohol.
McCarthy et al. Increasing school year was associated with increasing socioemotional harms and increased gambling. Harms increased with age more acutely in males.

Gender
Women were less likely to feel bad about gambling. Harms increased with age was more acute in males.

SES
Higher family affluence was associated with more harms but also more gambling. Feeling less school connectedness/ belonging led to more harms and also more gambling.

Culture
Non-white British students experienced more harms. White female students were less likely to feel bad about gambling.
Mihaylova, Kairouz, and Nadeau (2013 Online poker gamblers were more likely to gamble at least weekly, spend more and have higher annual debt as a result of poker specifically.

Online vs. Offline
Online poker players had a greater risk of alcohol dependency. Cannabis use did not differ between online and offline gamblers but online poker users were more likely to have used other illicit drugs over the past year. Academic performance did not differ between online and offline gamblers bur online poker players were more likely to report family issues, studying issues and financial issues. Age Children understood the slim chance of winning and that continuing to play meant you would lose "until you're poor." Children gambled with pocket money or verbal agreements and so financial harms were not experienced. Children didn't seem to perceive harms other than peer pressure and financial loss, though other studies cited show exposure as a child led to increased alcoholism and cigarette use. Children were found to be more likely to experience harm later in life due to believing sport bets were skill based and gambling as a child with minimal harms due to family protection.

Gender
Reported harms differed very little between genders. However when not examining frequent gamblers males reported more financial issues than females.

Gambling Behaviour
Considerably more frequent gamblers reported harms than occasional gamblers. The most common harm for frequent gamblers was guilt or shame, followed by relationship issues and daily routine disruptions. Daily gamblers were significantly more likely to report all harms.
Secondary Data Analysis PGSI Frequency of Play

Ministry of Social Affairs and Health Helsinki
Age Guilt was not related to age. Younger ages were more likely to report harms.

Gender
Men had a higher prevalence of harm but gambled more and spent more. Increased spends, betting above means and guilt were most prevalent harms. When controlling for frequency and spends gender was no longer related to harms.

Risk Severity
The most harm reported were found in lowmedium risk scorers.

Gambling Behaviour
Higher frequency of play and higher spends caused more harms. Harm risk increased when weekly spends exceeded E21.
Raisamo et al. Females aged 18-24 experienced more harms but also reported an increase in occasional gambling. Gambling more than intended was most common for all age groups.

Gender
Gambling prevalence was higher in males however females 18-24 reported experiencing more harms. Loss of control (gambling more than intended) and guilt were the most common harms for both genders. The portion of women who lost control, felt guilty, and hid betting slips increased while harms such as lying about wins and loss chasing decreased in males. Age Students spent pocket money or wages from part-time jobs on gambling. There was no significant difference between self-report and peer-reported signs of a gambling problem suggesting adolescents may be less aware of what to look for.

Gender
Males were more likely to be preoccupied with thoughts of gambling, feel the need to increase spending, and experience arguments.

Risk Severity
Highest scoring individuals on the SOGS were more likely to miss school, sell personal property, commit illegal acts, and try other risks like smoking and drugs. Legal drugs and ecstasy use was the same for all risk categories. Migrants were less likely to gamble than non-migrants but were found to be more likely to experience harms. There was minimal evidence of specific harms migrants experience and whether these differed from non-migrants. However they found that Spanish migrants tended to spend over 300 euros daily, claim losses as wins, and chase losses. Australian migrants experienced financial harm, shame, relationship issues, suicide, mental health issues, isolation and prostitution. Probable pathological gamblers reported the highest degree of harms in all categories, including longest time spent online generally and for gambling specifically. As risk scores increased so did physical, mental health, social relationship and academic harms.

Online vs. Offline
Online gambling was associated with binge drinking but not smoking. Around 60% of participants who were online gamblers rated as high risk (problem gambler) and as scores increased so did physical, mental health, social relationship and academic harms.