• Open Access

Young adults' gambling and its association with mental health and substance use problems

Authors


Correspondence to: Dr Mohammad R. Hayatbakhsh, University of Queensland, School of Population Health, Herston Road, Herston, Queensland 4006; e-mail: m.hayatbakhsh@uq.edu.au

Abstract

Objective: To examine the socio-demographic characteristics of young adults’ gambling and its association with mental health and substance use behaviour.

Methods: The study is based on 3,512 young adults (1648 males) for whom data from the Mater-University of Queensland Study of Pregnancy (MUSP) were available on self-report gambling, gambling expenditure, Achenbach's Young Adult Self Report and substance use at the 21-year follow-up of the MUSP. The participants’ age ranged between 18.2 and 23.6 (mean = 20.6, standard deviation = 0.8) years.

Results: Two-fifths of the young adults reported gambling. Males reported more money spent on gambling and were significantly more likely to be at risk of problem gambling. Gambling and problem gambling were significantly more common in less-educated individuals, those who had higher income or those who had a paid job. Individuals who reported gambling were more likely to smoke cigarettes, drink more than a glass of alcohol per day, use illicit drugs, or exhibit high levels of externalising behaviour than non-gamblers.

Conclusions: The findings confirm the high prevalence of gambling and gambling expenditure in young adults. Individuals who are involved in gambling are more likely to report cigarette smoking, alcohol consumption, and use of illicit drugs. There is a need for further research to explore the mechanisms of association between gambling behaviour and individuals’ mental health and substance use.

Implications: Substance abuse and mental health services are recommended to consider co-morbid gambling problems in treatment-seeking patients.

Recent decades have witnessed a significant increase in the prevalence of gambling and gambling problems in young adults.1–4 In Australia, between 1991/92 and 2004/05, annual gambling expenditure (player losses) rose from $7.3 billion to $16.9 billion in real terms.5 Although a lot of research has been reported on the prevalence and correlates of gambling in adults, there is a shortage of evidence about the characteristics of adolescent and young adult gamblers. Research has shown that onset of gambling in adolescence and early adulthood is associated with greater gambling involvement in adulthood.6 The objectives of this study are based on the need to increase our understanding of gambling behaviour and its association with psychopathology and substance use disorders in a population of young adults. Such information may help key stakeholders, including those in the gambling health services and government agencies.

Research indicates that between 70% and 90% of adults have gambled at some time in their lives.7,8 These rates are similar to those reported for adolescents,9 suggesting that gambling behaviour begins relatively early in life.10 It is believed that the relaxation of gambling legalisation in most countries has been associated with an increase in gambling activity as well as problem gambling.8,11 The Australian Federal Government Productivity Commission8 exhaustively reviewed national and international research, and estimated that around 2.1% of the adult Australian population have moderate to severe problems with their gambling. In a critical analysis of prevalence rates of problem gambling across different developed countries, including Australia, Walker and Dickerson12 indicate current problem gambling involves 1–2% of the adult community. More recent studies in Australia have indicated that less than 1% of the adult population is affected by problem gambling.1,13 The difference in the prevalence estimates in these studies could be due to the definition and measurement of problem gambling or the time at which problem gambling was estimated.

International studies have shown that gambling behaviour is more common among males, those with a lower level of education,1,7,12–14 and people who have a paid job and higher income.1 Some research has found that the prevalence of gambling and problem gambling are greater in unmarried individuals,1,15 while others suggested the opposite.7

The adverse effects of uncontrolled gambling on individuals, their families and entire social systems are considered of growing concern to the community and public health.16–18 One of the strongest relationships consistently found in the literature associates problem gambling with some of the most common mental health disorders, such as depression, anxiety and substance use disorders.14,19–22 Gambling problems correlate with depression, anxiety and suicide in both adolescents and adults.23–31 The Queensland Household Gambling Survey (QHGS) in 2006–07 found that approximately 72% of problem gamblers report having felt seriously depressed in the previous year, with 39% having been under a doctor's care for stress-related issues.1 Alcohol and drug abuse are perhaps the best-documented co-morbid diagnoses for problem gamblers.20,23,24,32 Among a large national sample from the United States, in a 2001–02 survey, it was found that three-quarters of pathological gamblers also had an alcohol use disorder, 38% had a drug use disorder and 60% were affected by nicotine dependence.19 The QHGS also found that while only 22% of recreational gamblers report smoking cigarettes, more than 60% of people in the problem-gambling group were smokers.1 Problem gamblers have consistently been found to be substantially more likely to exhibit high levels of antisocial behaviours and use illicit drugs than non-problem gamblers.33,34 However, most of these findings have been derived from studies based on adults.

Notwithstanding considerable evidence in the existing literature about substantial risk gambling and gambling problems, most studies have focused on adult gamblers.3,35 There remains a need for further investigation into this problem among young people.2,36 This study involves the merging of specifically collected data relating to gambling behaviour with an existing prospective data set describing the early life course of a population sample in Brisbane. The main contribution of this study is to describe socio-demographic characteristics of gambling in a cohort of young people who are at risk of gambling and substance use problems. It aims to identify mental health problems and substance use disorders associated with young adults’ gambling behaviour.

Methods and materials

Study sample

The data for this study have been taken from the Mater-University of Queensland Study of Pregnancy (MUSP). The Mater Misericordiae Mothers’ Hospital is one of two major obstetric units in Brisbane, Australia. The project involves a 21-year longitudinal investigation that began in 1981. Pregnant women attending their first clinic visit (at approximately 18 weeks gestation) at the Mater Hospital were invited to participate in the study.37,38 Over three years (between 1981 and 1984), 8,556 consecutive pregnant women were invited to join the study and 8,458 agreed to participate (phase 1). Of these, 7,223 gave birth to a live singleton infant and it is this group of mothers and offspring that constitutes the MUSP birth cohort sample. The original cohort were from south-east Queensland (mainly Brisbane, the Gold Coast and the Sunshine Coast) and appeared to be representative of public obstetrical patients but differed from private patients in a number of characteristics.38

In the 21-year survey of the MUSP, between June 2001 and December 2004, 3,512 young adults (1,648 males) responded to questions about gambling involvement. The Canadian Problem Gambling Index (CPGI)39 questionnaire was attached to the main questionnaire from mid-July 2003, resulting in data for 969 young adults (428 male). Data collection was mainly via face-to-face interview. Persons living outside Brisbane or who were unable to make an appointment for a face-to-face interview completed a mailed questionnaire and did not do the physical assessment. Informed consent was obtained from all participants.

Measurements

Young adults’ prevalence of gambling and problem gambling, socio-demographic information, substance use and psycho-behavioural characteristics were assessed at the 21-year phase of the MUSP.

Gambling and problem gambling

Prevalence of gambling among young adults was measured at the 21-year follow-up of the MUSP by asking participants ‘Do you spend money on gambling (e.g. buy lottery tickets, play the pokies, go to the casino, bet on horses, dogs, etc)?’ According to their response, young adults were divided into two groups: ‘non-gambler’, and ‘gambler’. The second question asked about the amount of money young adults spent on gambling per week. The range of answers varied from zero to $500 per week. Subsequently, participants were divided into four groups: no money spent (60%), $1 to $6 (20%), $7 to $34 (15%), and $35 or more (5%) per week.

The self-administered CPGI39 was used to establish the prevalence of problem gambling in young adults. The CPGI is a 31-item questionnaire which measures problems which correspond to DSM-IV criteria for pathological gambling40 and the South Oaks Gambling Screen (SOGS)41 and is, therefore, considered an appropriate measure of problem gambling for use in the general population. The CPGI has three main sections: gambling involvement, problem gambling assessment and correlates of problem gambling (including familial history of gambling). It yields five categories of gambling behaviours: no gambling, non-problem gambling, low-risk gambling, moderate-risk gambling and high-risk (problem) gambling. Initial studies indicate that the CPGI demonstrates good reliability and validity.39

Within the CPGI, nine items make up a sub-scale known as the Problem Gambling Severity Index (PGSI). The PGSI distinguishes four gambler sub-types: non-problem, low risk, moderate risk and problem. The non-problem group is further divided into gamblers and non-gamblers as these sub-types are believed to display different characteristics. Tabulation of the nine items is as follows: a score of 1 for each response of ‘sometimes,’ a score of 2 for each response of ‘often’ and a score of 3 for each ‘always’ response. Based on this scoring procedure, a respondent's index can range from 0 to 27 and the cut-off points for each gambler sub-type are as follows: 0 = non-problem gambler; 1–2 = low risk gambler; 3–7 = moderate risk gambler; and 8 or higher = problem gambler.

Socio-demographic factors

The participants’ educational level was categorised as did not complete high school, completed high school, completed tertiary, including college (e.g. business, trade, secretarial, teachers) and TAFE, and university degree. A separate question asked about education; options were: no, full-time and part-time. Marital status was either married/de-facto relationship or unmarried (including single, divorced and separated). Level of income was divided into three groups: low income (up to 25th percentile), middle income (between 25th and 75th percentiles) and high income (highest 25 percentiles). The participants were grouped into a dichotomous variable: paid job and no paid job, according to whether they had a paid job at the time the survey was conducted

Substance use

The extent of smoking by young adults was assessed via the average number of cigarettes smoked per day during the week preceding the survey (non-smokers, fewer than 10 cigarettes per day and 10 or more cigarettes per day). Regarding alcohol consumption, the participants were divided into three groups: no alcohol use, up to one drink (glass) per day, and more than one drink per day.

Consumption of illicit drugs was assessed from a self-report questionnaire. The illicit drugs under study included cannabis, amphetamines (amphetamine and ecstasy), heroin, cocaine, inhalants and hallucinogens. Young adults were asked two separate questions regarding consumption of cannabis. The first question was ‘in the last month how often did you use cannabis, marijuana, pot, etc?’ Options included: have never used, use every day, use every few days, used once or so and not used in last month. In the analysis, young adults were grouped into three categories: never used, occasional users (including ‘once or so’ and ‘not in the last month’), and frequent users (including ‘every day’ and ‘every few days’). The participants’ use of other illicit drugs during the last 12 months was categorised as ‘never used’ and ‘ever used’.

Psycho-behavioural factors

In the present study, the young adult's symptoms of problem behaviours during the last six months were measured using the Young Adult Self-Report (YASR), version of the Child Behaviour Checklist (CBCL).42 The YASR is a questionnaire for individuals aged 18–30 years which contains 114 problem items that can be scored on eight syndromes, including externalising behaviours (such as delinquency and aggression), internalising behaviours (such as withdrawal behaviours, anxiety and depression), symptoms of social problems, attention problems and thought problems.43,44. For the purpose of this study, YASR internalising and externalising behaviours were selected as measures of the young adult's mental health. For both variables we used the 90th percentile as a cut-off, above which were considered the cases with high levels of externalising and internalising behaviours.45

Statistical analysis

Frequencies and cross-tabulations were used to describe gambling practices, gambling expenditure and sub-types of gambling in young adults. We also explored socio-demographic characteristics of young adults who participated in the survey. Chi-square analyses were used to test cross-sectional associations between gambling practice and various mental health and substance use factors. Logistic regression models examined the association of gambling behaviour and other factors (expressed in odds ratio) that were significantly associated with gambling.

Of 969 young adults who had data available on the CPGI, 41.4% gambled at 21 years; of those 30.1% were non-problem gamblers (NPG), 6.3% were categorised as low-risk gamblers, 3.8% as moderate-risk gamblers, and just over 1.0% met the criteria for problem gambling (or high-risk gamblers). Due to the small number of participants, the low, moderate and high-risk problem gamblers were combined into one group. This did not substantially change the pattern of results. In this study, the combined low, moderate and high risk gamblers are called ‘at risk and problem’ gamblers (ARPG). This group constituted 11.3% of the cohort of young adults.

Results

Of 3,512 (1,648 male and 1,864 female) young adults, 76.9% had a paid job (44.5% full-time), while 23.1% reported no paid job. Nearly 23.0% of the participants were studying full-time and another 12.0% were in part-time study. The overall prevalence of young adult gambling and a demographic profile of gamblers and non-gamblers are outlined in Table 1. Some 40.1 % (44.2% males and 36.6% females) of the participants reported gambling. Male participants reported higher amounts of money spent on gambling. In addition, male participants were significantly more likely to be NPG and ARPG compared with females.

Table 1.  Gambling and young adult socio-demographic characteristics.
VariablesNGambling Gambling expenditure (dollars per week)Problem Gambling Severity Index
  NoYesp%None<7.07.0–34.935.0+p% NGNPGARPGp%
  1. Note: % p value derived from Chi-square tests; NG non-gambler; NPG no-problem gambler; ARPG at risk and problem gambler; NS non-significant.

Gender   <0.001    <0.001 %%%<0.001
Male1,63755.844.2 55.816.220.97.1 42650.934.314.8 
Female1,87563.436.6 63.421.912.81.9 54364.626.98.5 
Education completed   <.001    <0.001    NS
Incomplete high school71552.947.1 52.914.323.69.2 18353.029.517.5 
Completed high school1,87361.538.5 61.420.614.73.3 49260.830.19.2 
Tertiary education78162.137.9 62.119.515.72.7 23759.130.410.6 
University14362.237.8 62.225.99.12.8 5756.131.612.3 
Paid job   <0.001    <0.001    <0.01
Yes2,70257.842.2 57.820.017.64.6 76757.232.510.3 
No81066.933.1 66.916.713.03.4 20263.921.314.9 
Income   <0.001    <0.001    <0.01
Low97670.030.0 70.017.710.81.5 21969.421.09.6 
Middle1,76358.941.1 58.919.717.44.0 46257.832.010.2 
High77349.350.7 49.320.022.08.7 28851.734.014.2 
Marital status   NS    <0.05    NS
Married/de facto57257.742.3 57.720.119.13.1 26654.935.010.1 
Single/separate2,76060.539.5 60.519.015.84.7 70360.028.311.7 

Less-educated individuals were significantly more likely to gamble than respondents holding tertiary and university degrees. Those who didn't complete high school were also likely to spend greater amounts of money on gambling than those who had higher education. Middle and high incomes were associated with greater gambling expenditure. Having a paid job and a higher level of income were associated with both gambling and at risk or problem gambling. Young adults who did not have a paid job were less likely to be NPG and more likely to be ARPG. By contrast, those with high incomes were more likely to gamble and to be ARPG than lower income earners; the highest proportion of ARPG was observed among individuals who reported higher income compared to others. Young adults who were married or living in de facto relationships were only slightly less likely to spend money on gambling activities than those who were separated or single.

The association between young adults’ gambling behaviour and substance use and mental health is presented in Table 2. Gambling was associated with cigarette smoking, alcohol consumption, use of illicit drugs and externalising behaviour. Individuals who gambled were more likely to smoke cigarettes, drink more than one glass of alcohol per day, use cannabis or other illicit drugs and exhibit a higher level of externalising behaviour. Similarly, gambling expenditure was associated with the use of legal and illegal substances and externalising behaviour. Report of greater gambling expenditure was associated with increased likelihood of cigarette smoking, alcohol consumption, use of cannabis and other illicit drugs and more symptoms of externalising behaviour.

Table 2.  Association of young adults' gambling, substance use and mental health.
VariablesGamblingGambling expenditure ($ per week)Problem gambling severity index
 NoYesp%None<7.07.0–34.935.0+p%NGNPGARPGp%
 (2,103)(1,409) (2,103)(676)(581)(152) (568)(292)(109) 
 %% %%%% %%% 
  1. Notes: % p value derived from chi-square tests; NG non-gambler; NPG no-problem gambler; ARPG at risk and problem gambler; NS non-significant.

Cigarette smoking (per day)  <0.001    <0.001   <0.001
Non-smoker69.255.6 69.265.748.538.2 69.762.039.5 
<10 per day16.119.4 16.117.921.318.4 17.117.119.3 
10+ per day14.725.0 14.716.430.143.4 13.220.941.3 
Alcohol consumption  <0.001    <0.001   <0.001
Abstainer11.34.1 11.33.44.55.3 11.15.15.5 
≤1 drink per day59.247.3 59.259.839.222.4 59.950.035.8 
>1 drink per day29.548.7 29.536.856.372.4 29.144.958.7 
Cannabis ever used  <0.001    <0.001   <0.001
No55.042.9 55.047.940.828.3 55.151.027.5 
Yes45.057.1 45.052.159.271.7 44.949.072.5 
Pattern of current cannabis use  <0.001    <0.001   <0.001
No use55.042.9 55.047.940.828.3 55.151.027.5 
Occasional use35.341.2 35.340.441.543.4 35.738.445.9 
Frequent use9.716.0 9.711.717.728.3 9.210.626.6 
Use of other illicit drugs  <0.001    <0.001   <0.001
No77.168.9 77.174.766.851.3 74.872.654.1 
Yes22.931.1 22.925.333.248.7 25.227.445.9 
Internalising  NS    NS   NS
Normal90.189.9 90.189.190.990.1 89.690.884.4 
Top 10%9.910.1 99.910.99.19.9 10.49.215.6 
Externalising  <0.001    <0.001   <0.001
Normal92.888.0 92.889.589.277.0 94.089.074.3 
Top 10%7.212.0 7.210.510.823.0 6.011.025.7 

Table 2 also shows that heavy smokers and drinkers of more than one glass of alcohol per day were disproportionately over-represented in the ARPG group. A higher proportion of the ARPG group reported having used cannabis, in particular frequent use, and having ever used illicit drugs other than cannabis, than non-gamblers and no-problem gamblers. Finally, ARPG were more likely to manifest extreme levels of externalising behaviour. By contrast, there appears to be no association between young adults’ internalising and gambling problems.

Individuals who reported gambling or problem gambling were more likely to smoke cigarettes, use cannabis or use other illicit drugs (Table 3). There was a direct relationship between amount of money spent on gambling and frequency of substance use. For example, participants who spent between $7 and $35 per week on gambling were nearly three times more likely to smoke 10 or more cigarettes per day (OR = 2.9; 95% CI: 2.3–3.6), while for those who spent more than $35 the association was significantly stronger (OR = 5.3; 95% CI: 3.7–7.8). The data in Table 3 illustrate a similar pattern for the association of gambling and use of other legal and illegal substances. Further, it is noted that young adults who reported non-problem or problem gambling were statistically significantly at increased risk of reporting high levels of externalising behaviour.

Table 3.  Association of young adults’ gambling, substance use and mental health.
VariablesGamblingGambling expenditure
(dollars per week)
 Problem Gambling Severity
Index
 NoYesNone<7.07.0–34.935.0+NGNPGARPG
  OR (95% CI) OR (95% CI)OR (95% CI)OR (95% CI) OR (95% CI)OR (95% CI)
  1. Notes: NG non-gambler; NPG no-problem gambler; ARPG at risk and problem gambler

Cigarette smoking (per day)
Non-smokerRefRefRefRefRefRefRefRefRef
<10 per dayRef1.5 (1.3–1.8)Ref1.2 (0.9–1.5)1.9 (1.5–2.4)2.1 (1.3–3.3)Ref1.1 (0.8–1.7)2.0 (1.1–3.5)
10+ per dayRef2.1 (1.8–2.5)Ref1.2 (0.9–1.5)2.9 (2.3–3.6)5.3 (3.7–7.8)Ref1.8 (1.2–2.6)5.5 (3.4–8.9)
Alcohol consumption
AbstainerRefRefRefRefRefRefRefRefRef
≤1 drink per dayRef0.4 (0.3–0.6)Ref0.3 (0.2–0.5)0.6 (0.4–0.9)1.2 (0.6–1.7)Ref0.6 (0.3–1.0)0.8 (0.3–2.0)
>1 drink per dayRef2.1 (1.8–2.4)Ref1.2 (1.1–1.5)2.5 (2.4–3.5)6.5 (4.4–9.6)Ref1.8 (1.4–2.5)3.4 (2.2–5.2)
Cannabis ever used
NoRefRefRefRefRefRefRefRefRef
YesRef1.6 (1.4–1.9)Ref1.3 (1.1–1.6)1.8 (1.5–2.1)3.1 (2.2–4.5)Ref1.2 (0.9–1.6)3.2 (2.1–5.1)
Pattern of current cannabis use
No useRefRefRefRefRefRefRefRefRef
Occasional useRef1.5 (1.3–1.7)Ref1.3 (1.1–1.6)1.6 (1.3–1.9)2.4 (1.6–3.5)Ref1.2 (0.9–1.6)2.6 (1.6–4.2)
Frequent useRef2.1 (1.7–2.6)Ref1.4 (1.0–1.9)2.5 (1.9–3.3)5.7 (3.6–8.9)Ref1.3 (0.8–2.0)5.8 (3.2–10.5)
Use of other illicit drugs
NoRefRefRefRefRefRefRefRefRef
YesRef1.5 (1.3–1.8)Ref1.1 (0.9–1.4)1.7 (1.4–2.0)3.2 (2.3–4.5)Ref1.1 (0.8–1.5)2.5 (1.7–3.8)
Externalising
NormalRefRefRefRefRefRefRefRefRef
Top 10%Ref1.8 (1.4–2.2)Ref1.5 (1.1–2.0)1.6 (1.2–2.1)3.9 (2.6–5.8)Ref1.9 (1.2–3.2)5.4 (3.1–9.4)

Discussion

Gambling is an emerging public health problem with remarkably little known about the onset of gambling behaviour over the early life course. In this study, we examined the profile of young adult gambling, gambling expenditure and ARP gambling in Brisbane, Australia. We found that 40.8% of young adult respondents had participated in at least one gambling activity. Males were more likely (44.2%) than females (36.6%) to gamble. Those with higher incomes constituted a greater proportion of the gambling group and individuals with a higher level of education (tertiary education and university) were less likely to gamble relative to the less-educated groups. We also found that gambling activity was associated with young adults’ concurrent substance use. An increased quantity of cigarettes smoked and illicit drug use was related to greater risk of engagement in gambling activities. In addition, young adults who reported having gambled or had problem gambling were more likely to display externalising behaviour (including aggression and delinquency).

Our findings are largely consistent with previous research, including the QHGS.1 In addition, the data in this study confirms other research that suggested gambling behaviour begins relatively early in life9 and that current problem gambling involves 1–2% of the community.12 The present study also supports previous research that indicated an association between gambling problems on one side and psychopathology, including substance use disorder on the other.19,21,46 However, our data contradict the findings of those studies that suggest an association between anxiety and depression disorders and gambling problems. One possible reason for this discrepancy could be due to the different methods in assessment of mental health problems. Unlike McCormick and colleagues31 and Petry and colleagues,19 we used self-reported symptoms of internalising problems assessed by YASR.

Notwithstanding the existing literature on the association between gambling problems and psychopathology and substance use disorders, there remain questions that require further investigation: What is the cause-effect sequence? Does problem gambling precede or follow (or occur at the same time) as mental health and substance abuse? Due to simultaneous collection of data on both gambling and mental health problems, our study does not have the capacity to address the issue of cause and effect association, although it suggests that involvement in problem gambling is associated with greater legal and illegal substance use and problem behaviour.

Limitation

A few limitations should be taken into account when interpreting the findings of this study. First, the present study had a sizeable reduction in the sample between the first antenatal visit and the 21-year follow-up. Of the 7,223 participants who were included in the original cohort (1981–83), only 48.6% responded to the 21-year follow-up. Attrition in this study was more common among males and those from a lower socio-demographic background.38 Our observed associations could vary if the association between gambling, mental health and substance use in those lost to follow-up was higher or lower than in the study group. In addition, we only had data on CPGI for 969 young adults. This might have led to some of the associations between problem gambling, and mental health and substance use behaviour becoming statistically non-significant. Finally, this study was conducted between 2001 and 2004. Since then, access to the gambling facilities has increased and age of initiation to gambling has declined. However, there is no reason to believe that this study's findings cannot be inferred to the present time.

Conclusion

Some 40.1% of respondents were gamblers and 11.3% met the CPGI criteria for problem gambling with 4.8% of the sample being moderate or high risk problem gamblers. The data in this study points to high rates of gambling and gambling expenditure in young adults. With the increase in access to gambling services, it is clear that many young adults now engage in gambling behaviour and a substantial minority develop problem gambling.1 Individuals who are involved in gambling, and in particular those who spend more on gambling, are more likely to report cigarette smoking, alcohol consumption, use of illicit drugs and exhibit high levels of externalising behaviour. It is recommended that substance abuse and mental health services consider co-morbid gambling problems in treatment-seeking patients. There is a need for research to explore the mechanisms of association between gambling behaviour and individuals’ mental health and substance use.

Ancillary