With this document you can:

This box is not visible in the printed version.

Gambling behaviour and harm among people from different ethnic groups

This report examines how gambling participation, gambling risk, adverse consequences and use of support services vary among people from different ethnic groups.

Published: 27 October 2026

Last updated: 27 October 2026

This version was printed or saved on: 27 October 2026

Online version: https://www.gamblingcommission.gov.uk/report/gambling-behaviour-and-harm-among-people-from-different-ethnic-groups

Executive summary

This report examines how gambling participation, gambling risk, adverse consequences and use of support services vary among people from different ethnic groups. It draws on pooled data from the 2023, 2024 and 2025 Gambling Survey for Great Britain (GSGB). Pooling 3 years of data provided a sufficiently large sample to examine participants from White, Mixed or Multiple ethnic, Asian, Black and Other ethnic backgrounds separately.

The research builds on our previous report, in which we found higher rates of adverse consequences among participants from ethnic minority backgrounds, including at lower Problem Gambling Severity Index (PGSI) scores. However, limited sample sizes meant that participants from different ethnic minority backgrounds had to be considered together. This report provides a more detailed picture by examining individual ethnic groups.

The analysis compared:

Key findings

Overall gambling participation was lower among participants from ethnic minority backgrounds than those from White backgrounds. However, this difference was largely driven by lower levels of lottery draw participation among ethnic minority groups. Across most non-lottery draw gambling activities, participants from ethnic minority backgrounds reported higher levels of engagement than those from White backgrounds. Those from ethnic minority backgrounds also reported gambling on a wider range of activities, despite gambling less overall.

Participants from ethnic minority backgrounds had higher PGSI scores than those from White backgrounds, including those with a PGSI score of 8 or more. Estimates varied from 1.9% for those from White backgrounds to 7.7% for those from Mixed and/or Multiple ethnic backgrounds. These differences persisted after accounting for demographic and socio-economic characteristics and gambling participation.

Participants from ethnic minority backgrounds also reported substantially higher rates of adverse consequences from gambling, often 3 to 4 times higher than those from White backgrounds. As with PGSI scores, this difference persisted for most ethnic groups once demographic, socio-economic and wider gambling factors were taken into account. The exception was those from Black backgrounds where the excess harm was mostly explained by demographic or gambling factors.

Among those who gambled in the past 4 weeks, participants from ethnic minority backgrounds were more likely than those from White backgrounds to report adverse gambling-related consequences, even when they had similar PGSI scores. This disparity was most pronounced at lower levels of gambling risk and narrowed substantially as PGSI scores increased. For example, among participants with a PGSI score of 0, at least one adverse consequence was reported by approximately 2.7% of those from White backgrounds, compared with around 7% to 10% of participants from ethnic minority backgrounds. However, among those with PGSI scores of 8 or more, the prevalence of adverse consequences was similarly high across all ethnic groups, ranging from 92% to 100%.

Use of gambling support services was also higher among those from ethnic minority backgrounds but remained well below the proportion reporting elevated PGSI scores or adverse consequences, suggesting that support-seeking has not kept pace with need.

This report highlights the importance of disaggregating the data on gambling participation by gambling activity. Only looking at headline figures of participation masks wider and varied patterns of gambling among people from different ethnic backgrounds.

This report references information that can be found in the ethnic group topic report findings excel files that accompany this report.

Introduction

Prior studies have shown that people from different ethnic minority backgrounds experience greater severity of gambling problems. For example, the British Gambling Prevalence Survey in 2010 (opens in new tab) found those from Black backgrounds were more likely to experience problem gambling despite gambling less overall.

More recently, the Gambling Commission's report Exploring demographic differences in adverse consequences from gambling found increased harm among participants from ethnic minority backgrounds, even at low levels of Problem Gambling Severity Index (PGSI). However, as with much prior research in this area, low base sizes meant participants from ethnic minority backgrounds had to be grouped within a single category, likely masking meaningful differences between groups. The consistent methodology used by the Gambling Survey for Great Britain (GSGB), and the large number of people interviewed each year, now allows data to be pooled across Years 1 to 3 and explore gambling behaviours among individual ethnic groups for the first time, delivering on this planned future work.

This report combines data from the GSGB 2023, 2024 and 2025 to examine: 

Definitions

Past 12 months gambling

Participants were provided with a list of 20 types of gambling legally available in Great Britain that are detailed in the Appendix, including private betting or gambling, and asked which of these activities they had participated in, in the past 12 months. Participants who answered yes to participating in one or more gambling activity are described in this report as participating in gambling in the past 12 months.

Past 4-week gambling

This is not captured by a single survey item. Instead, it is derived from a series of activity-specific frequency questions (covering lottery tickets, scratchcards and/or instant win games, betting, bingo, casino games, and fruit and/or slot machines), each asked only of those who reported gambling on that activity in the past 12 months, with response options ranging from "Not in the past 4 weeks" to "Every day". Football pools, private betting, and another form of gambling were instead asked as a simple yes or no for the past 4 weeks, giving 22 past 4-week activities in total (see Appendix). Past 4-week participation on each activity was derived by treating any response other than "Not in the past 4 weeks" as indicating participation. For this analysis participation in ‘Another form of gambling’ was not included, meaning participation in 21 past-4-week activities were included.

Full list of gambling activities

The list of gambling activities presented to participants was refined and tested during the development stages of the Gambling Survey for Great Britain (GSGB). The aim was to update the list used in previous surveys to better represent forms of commercial gambling currently available in Great Britain, and to better capture engagement in different online gambling formats. The Appendix shows the full list of gambling activities asked about for the past 12 months and past 4 weeks.

Number of activities in the past 4 weeks

The number of discrete gambling activities undertaken in the past 4 weeks was calculated for those who had gambled in this period, based on whether a participant reported spending money on each activity. For the purposes of this count only, National Lottery draws (online and in-person) were combined and counted as a single activity, as were other charity lottery draws (online and in-person), giving a maximum possible count of 20 activities, aligning with the 20 activities captured by the past-12-months gambling variable. Elsewhere in this report, online and in-person National Lottery draws are analysed separately. Participants who had not gambled in the past 4 weeks were coded as not having taken part in any activities.

Frequency of gambling

Everyone who had gambled on a particular activity in the past 12 months was then asked to report how often, if at all, they had gambled on this activity in the past 4 weeks. Using this information, frequency of gambling was computed for each participant based on their reported highest frequency of gambling to any activity in the past 4 weeks. For example, if a participant reported purchasing National Lottery tickets once a week but bet on sports events a few times a week, they were categorised as gambling a few times a week. Response categories were: every day, a few times a week, about once a week, about once a fortnight, about once in the past 4 weeks,and less often than this (that is those who had gambled in the past 12 months, but not in the past 4 weeks).

Ethnic Group

Ethnicity is organised into 5 groups, aligning with Census ethnic group categorisation: participants from White, Mixed and/or Multiple ethnic, Asian, Black and Other ethnic backgrounds. These are derived from the following question:

Choose one option that best describes your ethnic group or background:

Problem Gambling Severity Index (PGSI)

The PGSI consists of nine questions and is asked of everyone who had gambled in the past 12 months, capturing the current experience of each of these items. Answer options were ‘Almost always’, ‘Most of the time’, ‘Sometimes’, and ‘Never’. Responses to the nine questions are summed and a score ranging between 0 and 27 is computed. Scores are grouped into the following categories:

PGSI score 0

Representing a person who gambles (including heavily) but who does not report experiencing any of the nine symptoms or adverse consequences asked about. In population prevalence analysis, participants who had not gambled in the past 12 months are also given a PGSI score of 0.

PGSI score 1 to 2

Representing low risk gambling by which a person is unlikely to have experienced any adverse consequences from gambling but may be at risk if they are heavily involved in gambling.

PGSI score 3 to 7

Representing moderate risk gambling by which a person may or may not have experienced adverse consequences from gambling but may be at risk if they are heavily involved in gambling

PGSI score 8 or higher

Representing problem gambling by which a person will have experienced adverse consequences from their gambling and may have lost control of their behaviour. Involvement in gambling can be at any level but is likely to be heavy.

Adverse consequences from own gambling

The GSGB includes a broader range of adverse consequences from gambling. These aim to add a greater level of insight into the adverse consequences of gambling that are experienced over and above the PGSI. These questions, asked of those who had gambled in the past 12 months, were split into 2 types:

  1. Those categorised as severe adverse consequences, where any experience of them is highly likely to be harmful. This includes relationship breakdown, losing something of significant financial value, violence or abuse and crime. For these questions participants were directed to answer ‘yes’ or ‘no’ to having experienced an adverse consequence.

  2. Those categorised as other potential adverse consequences, which are more likely to be harmful if experienced often or may be harmful depending on the individual’s specific circumstances. This includes spending less on everyday items, increased use of credit or savings to gamble, experience of conflict within relationships, feeling isolated, lying about the extent of gambling and poor work performance or work absences. For these questions participants were directed to answer ‘never’, ‘a little’ or ‘a lot’ to having experienced an adverse consequence.

Overall harm

A new variable was derived from the items on adverse consequences. Responses to each potential adverse consequence were summed giving a score of 3 if the potential adverse consequence happened very often, a score of 2 if fairly often and a score of 1 for occasionally. The potential adverse consequence items are as follows.

How often in the last 12 months has gambling led you to:

  1. reduce or cut back spending on everyday items such as food, bills, and clothing
  2. use savings or increase use of credit, such as credit cards, overdrafts, and loans
  3. experience conflict or arguments with friends, family, or work colleagues
  4. feel isolated from other people, left out, or feel completely alone
  5. lie to family, or others, to hide the extent of gambling
  6. be absent or perform poorly at work or study
  7. feel guilty about the way you gamble or what happens when you gamble
  8. borrow money or sell anything to get money to gamble
  9. experience any health problems, including stress or anxiety.

Responses to the questions on severe adverse consequences were given a score of 3 if they had occurred.

In the last 12 months:

This broadly replicates the DSM-IV convention of assigning higher weights to more serious items. This gives a variable with a maximum score of 39 (a maximum of 27 from the 9 potential adverse consequence items, each scored 0 to 3, plus a maximum of 12 from the 4 severe adverse consequence items, each scored 0 or 3). When these questions were first developed the Gambling Commission advised against presenting an aggregate prevalence figure using these combined data. Aligning with this approach, in this report we do not present a single total score for this derived variable. Instead, the derived variable is used only within statistical modelling as a continuous measure of relative harm severity, without assigning arbitrary threshold cut offs.

Health consequences

This is a binary variable derived from 2 items of the PGSI. In the past 12 months:

Response options include ‘Almost always’, ‘Most of the time’, ‘Sometimes’ and ‘Never’. Participants who selected any response other than 'Never' were categorised as having experienced health consequences.

Relationship consequences

This is a binary variable derived from 3 potential adverse consequences. Participants were asked how often, in the past 12 months, gambling had caused them to:

Response options include: ‘Very often’, ‘Fairly often’, ‘Occasionally’, ‘Never’. Participants who selected any response other than 'Never' were categorised as having experienced relationship consequences.

Resource consequences

This is a binary variable derived from 3 of the potential consequences items and 1 PGSI item. Participants were asked how often, in the past 12 months, gambling had caused them to:

Response options were ‘Never’, ‘Occasionally’, ‘Fairly often’ and ‘Very often’.

The following PGSI item also captures resource consequences. In the last 12 months:

Have you borrowed money or sold anything to get money to gamble?

Response options were 'Almost always', 'Most of the time', 'Sometimes' and 'Never'.

Participants who selected any response other than 'Never' were categorised as having experienced resource consequences.

Use of support services

Participants were asked whether in the last 12 months, their own gambling led them to seek help, support or information online, in-person or by telephone from:

Response options were Yes or No.  

Methods

This section references information that can be found in the ethnic group topic report findings excel files that accompany this report.

Bivariate analyses

Bivariate analyses were used to look at how gambling behaviours, Problem Gambling Severity Index (PGSI) scores, adverse consequences and help seeking vary among people from different ethnic backgrounds. The gambling behaviours evaluated included overall participation, the total number of activities and participation in specific activity types. Both potential and severe adverse consequences were assessed. To account for the complex survey design of the Gambling Survey for Great Britain (GSGB), all bivariate analyses used weighted cross-tabulations, with chi-squared tests used to assess statistical significance between groups.

Overall participation rates can be influenced by the popularity of lottery draws and mask wider variations in gambling participation among sub-groups. As such, additional analyses were performed excluding participants who had only taken part in lottery draw-based activities, to provide a clearer picture of the variation in gambling behaviour.

Multivariable analyses

Negative binomial models were used to explore the reasons why PGSI scores and adverse consequences vary by ethnic background. All models accounted for the complex survey design.

PGSI score

To examine PGSI scores among participants who gambled in the past 4 weeks, 3 sequential models were tested:

Model 1 (unadjusted): Compares PGSI scores across ethnic groups using those from White backgrounds as the reference baseline (as shown in Table 6 in the ethnic group topic report findings).

Model 2 (demographic and socio-economic adjustment): Controls for sex, age, area-level deprivation, income, employment status and level of education. Statistically significant results indicate that differences in PGSI score across ethnic groups persist regardless of demographic and socio-economic factors (as shown in Table 7 in the ethnic group topic report findings).

Model 3 (participation in gambling activities): This repeats model 2 but also includes whether a participant reported participating in any of 21 gambling activities in the past 4 weeks. Statistically significant results indicate that differences in harm scores persist regardless of demographic, socio-economic and both the volume and specific types of gambling activities reported across the 21 gambling activities (as shown in Table 8 in the ethnic group topic report findings).

Adverse consequences

Potential and severe adverse consequences were combined into a single harm score to capture the overall severity and breadth of gambling-related harm in one outcome measure (see definitions for a full description). The 3 sequential models tested were:

Model A (unadjusted): This compares the harm scores across ethnic groups, relative to those from White backgrounds (as shown in Table 10 in the ethnic group topic report findings).

Model B (demographic and socio-economic adjustment): Controls for sex, age, area-level deprivation, income, employment status and level of education. Statistically significant results indicate that differences in harm score across ethnic groups persist regardless of demographic and socio-economic factors (as shown in Table 11 in the ethnic group topic report findings).

Model C (participation in gambling activities): This repeats model B but also includes whether a participant reported participating in any of 21 gambling activities in the past four weeks. Statistically significant results indicate that differences in harm scores persist regardless of demographic, socio-economic and the number/type of activities an individual engages in (as shown in Table 12 in the ethnic group topic report findings).

Interpreting model results

Model results are reported as Incidence Rate Ratios (IRRs) relative to participants from White backgrounds (the baseline reference group):

Each IRR is presented with a 95% Confidence Interval (CI) to indicate estimation precision. If the confidence interval does not overlap 1, the difference between that ethnic group and the reference baseline is statistically significant.

Sample approach

Because overall gambling participation varies considerably by ethnic group, PGSI scores and adverse consequences were examined among those who had gambled in the past 4 weeks. This ensures a consistent and comparable baseline population across analyses.

Use of support services was examined among those who had gambled in the past 12 months, as this is the only timeframe for which this measure was collected.

Results

How does gambling behaviour vary among people from different ethnic backgrounds? 

This section references information that can be found in the ethnic group topic report findings excel files that accompany this report.

Gambling participation by ethnic group

Gambling participation varied significantly across ethnic groups. Gambling participation in the past 4 weeks was highest among participants from White backgrounds (50.3%) and lowest among those from Other ethnic backgrounds (28.6%), with rates of 44.1% among those from Mixed and/or Multiple ethnic backgrounds, 36.1% among those from Black backgrounds and 31.0% among those from Asian backgrounds as shown in Figure 1. This pattern was broadly similar for men and women, with male participation exceeding female participation within most ethnic groups, except for those from Mixed and/or Multiple ethnic backgrounds where participation was similar among men (43.9%) and women (44.3%) (as shown in Table 1 in the ethnic group topic report findings).

However, a different pattern was observed when looking at gambling on things other than lottery draws alone. Past 4-week participation in non-lottery draw activities was highest among those from Mixed and/or Multiple ethnic backgrounds (30.9%) and the difference between participation rates for White backgrounds and other ethnic groups was not as pronounced as shown in Figure 1.

Figure 1: Gambling participation in the past 4 weeks, by ethnic group (percentage)

Figure 1: Gambling participation in the past 4 weeks, by ethnic group (percentage)

Figure 1: Gambling participation in the past 4 weeks, by ethnic group (percentage)
Ethnic group Gambling participation in the past 4 weeks Gambling participation in the past 4 weeks excluding lottery draw only players
White 50.3 28.4
Mixed and/or Multiple 44.1 30.9
Asian 31.0 19.1
Black 36.1 24.8
Other 28.6 21.9

Base: All productive individuals excluding participants who had only played National Lottery draw-based or other charity lottery-draw based games. Participants within each ethnic group (unweighted): White groups = 43,809; Mixed and/or Multiple ethnic groups = 996; Asian groups = 2,818; Black groups = 862; Other ethnic groups = 337.

Gambling participation: number of activities

Despite being less likely to gamble overall, participants from Mixed and/or Multiple ethnic backgrounds were more likely to gamble on three or more activities (18.5%) than all the other groups. This was followed by those from Black backgrounds (14.9%) and those from White backgrounds (13.3%). Participants from Asian backgrounds were the least likely to take part in three or more gambling activities in the past four weeks, irrespective of whether lottery draws were included or excluded as shown in Figure 2 (and as shown in Table 2 in the ethnic group topic report findings).

Figure 2: Participants gambling on 3 or more activities in the past 4 weeks, by ethnic group (percentage)

Figure 2: Participants gambling on 3 or more activities in the past 4 weeks, by ethnic group (percentage)

Figure 2: Participants gambling on 3 or more activities in the past 4 weeks, by ethnic group (percentage)
Ethnic group Gambling participation on 3 or more activities in the past 4 weeks Gambling participation on 3 or more activities excluding lottery draw only players
White 13.3 13.5
Mixed and/or Multiple 18.5 18.8
Asian 9.3 9.4
Black 14.9 15.1
Other 11.2 11.3

Base: Productive individuals who gambled in the past 4 weeks excluding participants who had only played National Lottery draw-based or other charity lottery draw-based games. Participants within each ethnic group (unweighted): White groups = 44,413; Mixed and/or Multiple ethnic groups = 1,006; Asian groups = 2,840; Black groups = 875; Other ethnic groups = 340.

Gambling participation: individual activities

Table 3 in the ethnic group topic report findings shows whether participation in individual gambling activities varied by ethnic group. The majority of individual activities significantly varied by ethnic group (p-values from less than .001 to .03) (except other charity lottery in-person p=0.21, and fruit and/or slot machines in-person p=0.05).

A clear pattern emerged when comparing specific individual activities. For 2 activities, National Lottery draws played online, and other charity lottery draws played online, participants from all 4 ethnic minority backgrounds had lower participation than among those from White backgrounds (as shown in Table 3 in the ethnic group topic report findings).

For most other activities, participants from White backgrounds were less likely to participate than those from other groups. This was with the exception of fruit and/or slots games played online, bingo either online or in-person, betting on sports and/or racing events online and National Lottery scratchcards, where participation rates were lowest among those from Asian backgrounds as shown in Figure 3. In some cases, the gap was striking: betting on outcomes of events in-person was around five times more common among participants from Mixed and/or Multiple ethnic backgrounds than those from White backgrounds (4.4% versus 0.9%), and betting on outcomes of events online was around 4 times more common among participants from Black backgrounds than those from White backgrounds (5.6% versus 1.5%). The 2 activities that showed no meaningful differences between ethnic groups were other charity lottery draws played in-person, and fruit and/or slot machines played in-person. Participation rates for these activities were similar across all groups.

Figure 3: Participation in selected gambling activities in the past 4 weeks, by ethnic group (percentage)

Figure 3: Participation in selected gambling activities in the past 4 weeks, by ethnic group (percentage)

Figure 3: Participation in selected gambling activities in the past 4 weeks, by ethnic group (percentage)
Activity White groups Mixed and/or Multiple ethnic groups Asian groups Black groups Other ethnic groups
National Lottery scratchcards 11.7 14.7 8.6 11.2 9.4
Betting sports and/or racing (online) 9.0 10.0 4.9 11.5 6.5
Betting on outcomes (online) 1.5 4.8 2.7 5.6 4.9
Betting on outcomes (in-person) 0.9 4.4 2.2 4.1 3.7
Bingo (online) 2.3 4.7 1.6 3.3 5.0
Bingo (in-person) 3.4 4.3 1.8 2.9 3.4
Fruit and/or slot games (online) 3.2 6.6 2.9 3.3 4.9

Base: All productive individuals. Participants within each ethnic group (unweighted: White groups = 43,785; Mixed and/or Multiple ethnic groups = 994; Asian groups = 2,818; Black groups = 857; Other ethnic groups = 336.

How do Problem Gambling Severity Index (PGSI) scores vary by ethnic group?  

This section references information that can be found in the ethnic group topic report findings excel files that accompany this report.

PGSI scores varied significantly by ethnic group overall. Participants from each ethnic minority background had a higher rate of gambling risk than those from White backgrounds (PGSI scores greater than zero: 13.3%). Rates were highest among participants from Black (21.6%) and Mixed and/or Multiple ethnic backgrounds (21.5%), followed by those from other ethnic (18.6%) and Asian (17.6%) backgrounds. The same pattern was more pronounced for PGSI scores of 8 or more, with rates 3 to 4 times higher among those from ethnic minority groups than those from White backgrounds (as shown in Table 4 in the ethnic group topic report findings).

When focusing only on people who gambled in the past 4 weeks, this pattern became more pronounced (as shown in Table 5 in the ethnic group topic report findings). Participants from White backgrounds were most likely to score PGSI 0 (76.7%), compared with 39.6% to 57.9% among participants from ethnic minority groups. Conversely, those from ethnic minority groups showed higher rates of PGSI 3 to 7 (8.7% to 13.1% versus 5.7%) and PGSI 8 or more (15% to 25.1% versus 3.4%), than those from White backgrounds as shown in Figure 4. This pattern held when focusing on participants who had gambled in the past 4 weeks on something other than lottery draws alone (as shown in Table 5 in the ethnic group topic report findings). The proportion of participants with a PGSI score greater than zero rose from 34.9% for White backgrounds to between around 53% to 61% for those from ethnic minority backgrounds. For PGSI scores of 8 or more, rates rose to 5.7% among those from White backgrounds to between 21% to 30% for those from ethnic minority backgrounds as shown in Figure 4.

Figure 4: Gambling Participation among those with a PGSI score of 8 or more, by ethnic group (percentage)

Figure 4: Gambling Participation among those with a PGSI score of 8 or more, by ethnic group (percentage)

Figure 4: Gambling Participation among those with a PGSI score of 8 or more, by ethnic group (percentage)
Ethnic group Gambling participation in the past 4 weeks among those with a PGSI score of 8 or more Gambling participation in the past 4 weeks among those with a PGSI score of 8 or more (excluding lottery draw only players)
White 3.4 5.7
Mixed and/or Multiple 16.3 21.6
Asian 15.2 21.1
Black 15.0 21.5
Other 25.1 29.7

Base: Productive individuals who gambled in the past 4 weeks excluding participants who had only played National Lottery draw-based or other charity lottery draw-based games. Participants within each ethnic group who gambled in the past 4 weeks (unweighted: White groups = 21,452; Mixed and/or Multiple ethnic groups = 454; Asian groups = 862; Black groups = 305; Other ethnic groups = 99). Participants within each ethnic group who gambled in the past 4 weeks excluding lottery draw only players (unweighted: White groups = 11,219; Mixed and/or Multiple ethnic groups = 295; Asian groups = 514; Black groups = 196; Other ethnic groups = 72.

What explains the differences in PGSI scores?

There are socio-economic and demographic differences between ethnic groups which might explain the pattern seen. The populations of ethnic minority groups are often younger, experience greater deprivation or engage in higher-risk gambling activities, all of which are risk factors for higher PGSI scores. To investigate this, multivariable models were used to control for these factors. Incidence Rate Ratios (IRRs) above 1 indicate a higher expected PGSI score relative to the reference group (participants from White backgrounds); for example, an IRR of 2 indicates an expected PGSI score twice that of the reference group.

The unadjusted model (model 1) mirrored the descriptive findings: participants from ethnic minority backgrounds had, on average, PGSI scores around 3 to 4.7 times higher than participants from White backgrounds (IRRs - Mixed and/or Multiple backgrounds: 3.17; Asian backgrounds: 3.20; Black backgrounds: 3.28; Other ethnic backgrounds: 4.66) (see the following Table 1 and Table 6 in the ethnic group topic report findings).Controlling for socio-economic and demographic characteristics, in model 2, reduced these IRRs, but they remained statistically significant, indicating that socio-economic and demographic characteristics did not fully account for differences in PGSI scores across ethnic groups (as shown in the following Table 1 and Table 7 in the ethnic group topic report findings).

Model 3 added in participation across 21 distinct gambling activities, the IRRs reduced further but remained statistically significant, indicating that the differences in PGSI scores between participants from ethnic minority backgrounds and participants from White backgrounds are not fully explained by this measure of gambling involvement (see the following Table 1 and Table 8 in the ethnic group topic report findings). These models show that the higher PGSI scores seen in participants from ethnic minority backgrounds were only partially explained by differences in socio-economic and demographic characteristics and participation across 21 distinct gambling activities.

Table 1: The unadjusted and adjusted model results for the association between ethnic group and PGSI score

Table 1: The unadjusted and adjusted model results for the association between ethnic group and PGSI score
Ethnic group Incidence rate ratio Lower 95 percent confidence interval Upper 95 percent confidence interval
Model 1: unadjusted
Mixed and/or Multiple ethnic groups 3.17 2.55 3.95
Asian groups 3.20 2.73 3.76
Black groups 3.28 2.45 4.40
Other ethnic groups 4.66 3.29 6.59
Model 2: Adjusted for socio-economic and demographic characteristics
Mixed and/or Multiple ethnic groups 2.06 1.62 2.62
Asian groups 2.28 1.94 2.68
Black groups 2.12 1.57 2.85
Other ethnic groups 3.57 2.24 5.69
Model 3: Adjusted for socio-economic, demographic characteristics and participation across 21 distinct gambling activities
Mixed and/or Multiple ethnic groups 1.75 1.33 2.30
Asian groups 2.17 1.84 2.55
Black groups 1.51 1.18 1.93
Other ethnic groups 2.43 1.56 3.76

How does the experience of severe adverse and other potential adverse consequences from gambling vary among those from different ethnic groups?

This section references information that can be found in the ethnic group topic report findings excel files that accompany this report.

Of those who gambled in the past 4 weeks, experience of any severe adverse consequence, or potential adverse consequences relating to health, relationships or resources varied significantly by ethnic group overall. Participants from ethnic minority backgrounds reported higher rates of both potential and severe adverse consequences compared with those from White backgrounds. Potential health consequences were reported by 10.8% of those from White backgrounds who had gambled in the past 4 weeks, rising to 24.6% for those both from Mixed and/or Multiple ethnic backgrounds and Black backgrounds to 29% for those from Asian backgrounds, and 44.3% of those from Other ethnic backgrounds (as shown in Table 9 in the ethnic group topic report findings). This same pattern was observed for potential relationship and resource consequences with those from ethnic minority backgrounds experiencing consistently higher rates than those from White backgrounds (roughly 3 to 4 times as high) as shown in Figure 5. The prevalence of experiencing at least 1 severe adverse consequence was also substantially higher for those from ethnic minority backgrounds with 1.9% of those from White backgrounds reporting this, compared with 10.3% for those from Mixed and/or Multiple ethnic backgrounds, 11.7% for those from Asian backgrounds, 12.5% for those from Black backgrounds and 21.4% of those from Other ethnic backgrounds.

Figure 5: Reporting of adverse consequences among participants who gambled in the past 4 weeks, by ethnic group (percentage)

Figure 5: Reporting of adverse consequences among participants who gambled in the past four weeks, by ethnic group (percentage)

Figure 5: Reporting of adverse consequences among participants who gambled in the past 4 weeks, by ethnic group (percentage)
Consequence type White groups Mixed and/or Multiple ethnic groups Asian groups Black groups Other ethnic groups
Severe adverse consequence 1.9 10.3 11.7 12.5 21.4
Health consequence 10.8 24.6 29.0 24.6 44.3
Relationship consequence 7.9 24.9 26.5 24.5 36.4
Resource consequence 8.5 26.3 30.0 24.8 33.8

Base: Productive individuals who had gambled in the past 4 weeks. Participants within each ethnic group who gambled in the past 4 weeks (unweighted: White groups = 21,470; Mixed and/or Multiple ethnic groups = 454; Asian groups = 863; Black groups = 308; Other ethnic groups = 98).

This pattern was also observed when looking at those who had gambled on activities other than lottery draws in the past 4 weeks. Here, potential health, relationship and resource consequences affected around 1 in 3 participants from Mixed and/or Multiple ethnic backgrounds, Asian backgrounds, Black backgrounds and Other ethnic backgrounds, compared with around 1 in 6 of those from White backgrounds (as shown in Table 9 in the ethnic group topic report findings).

What explains the differences in adverse consequences?

A negative binomial model was used to further investigate the relationship between ethnicity and adverse consequences using the ‘overall harm’ score described in our methods section as the outcome variable. The unadjusted model (Model A; as shown in the following Table 2 and Table 10 in the ethnic group topic report findings) reflected the descriptive findings with participants from all ethnic minority backgrounds having significantly higher IRRs than those from White backgrounds, the baseline group. When socio-economic and demographic characteristics were included, the IRRs lowered but remained significantly higher than those from White backgrounds (Model B; as shown in the following Table 2 and Table 11 in the ethnic group topic report findings). A further adjusted model (Model C; as shown in the following Table 2 and Table 12 in the ethnic group topic report findings) controlling for participation across 21 distinct gambling activities attenuated (that is, lowered the IRRs further) but remained statistically significantly higher than participants from White backgrounds except for those from Black backgrounds. This indicates the increased rates of harm in those from Black backgrounds could be explained by differences in gambling participation, but this is not the case for those from Mixed and/or Multiple ethnic, Asian, and Other ethnic backgrounds.

Table 2: The unadjusted and adjusted model results for the association between ethnic group and ‘overall harm’ score.

Table 2: The unadjusted and adjusted model results for the association between ethnic group and ‘overall harm’ score
Ethnic group Incidence rate ratio Lower 95 percent confidence interval Upper 95 percent confidence interval
Model A: unadjusted
Mixed and/or Multiple ethnic groups 4.46 3.46 5.75
Asian groups 4.68 3.88 5.65
Black groups 4.22 2.84 6.28
Other ethnic groups 7.64 5.13 11.38
Model B: Adjusted for socio-economic and demographic characteristics
Mixed and/or Multiple ethnic groups 2.54 1.97 3.28
Asian groups 3.37 2.70 4.20
Black groups 2.77 1.76 4.36
Other ethnic groups 6.55 3.51 12.22
Model C: Adjusted for socio-economic, demographic characteristics and participation across 21 distinct gambling activities
Mixed and/or Multiple ethnic groups 1.92 1.43 2.58
Asian groups 2.81 2.32 3.40
Black groups 1.77 0.96 3.29
Other ethnic groups 4.72 1.92 11.61

Adverse consequences by Problem Gambling Severity Index (PGSI) category, by ethnic group

The previous sections have examined PGSI scores and adverse consequences by ethnic group separately, both showing elevated rates among those from ethnic minority groups. In this section we explore whether certain ethnic groups experience disproportionate potential adverse consequences at identical PGSI categories. To do this, we analysed the prevalence of experiencing at least one potential adverse consequence across PGSI categories. Among participants who had gambled in the past four weeks, experiencing an adverse consequence varied significantly by PGSI scores overall. The likelihood of experiencing an adverse consequence rose sharply with PGSI category for all ethnic groups, from around 3% at a PGSI score of 0 to over 90% at a score of 8 or more (as shown in Table 13 in the ethnic group topic report findings).

Within all PGSI categories, those from ethnic minority backgrounds consistently reported higher levels of experiencing at least one adverse consequence than participants from White backgrounds, with the largest differences observed at lower levels of gambling risk.

Looking at those with a PGSI score of 0 is informative here. These are people who have gambled in the past four weeks but not endorsed any of the PGSI items. However, 2.7% of those from White groups reported experiencing at least one adverse consequence whereas 10.6% of those from Asian groups or Other ethnic groups reported the same.

The same pattern occurred for those with PGSI scores of 1 to 2: 16.6% of participants from White backgrounds reported at least 1 adverse consequence, compared with 18.9% of participants from Mixed and/or Multiple ethnic backgrounds, 18.7% of those from Black backgrounds, 25.3% of those from Other ethnic backgrounds and 27.0% of those from Asian backgrounds. This is consistent with the Gambling Commission's earlier report Exploring demographic differences in adverse consequences from gambling, which found similar patterns of elevated harm among participants from ethnic minority groups at low PGSI scores.

Interestingly, at PGSI scores of 8 or more, this gap narrowed considerably, and rates were broadly similar across all ethnic groups (91.8% to 100%), with participants from White backgrounds (94.0%) sitting within this range (as shown in Figure 6).

Figure 6: Reporting of adverse consequences among those who gambled within the past 4-weeks by PGSI risk category and ethnic group (percentage)

Figure 6: Reporting of adverse consequences among those who gambled within the past 4-weeks by PGSI risk category and ethnic group (percentage)

Figure 6: Reporting of adverse consequences among those who gambled within the past 4-weeks by PGSI risk category and ethnic group (percentage))
PGSI category White groups Mixed and/or Multiple ethnic groups Asian groups Black groups Other ethnic groups
PGSI 0 2.7 7.2 10.6 7.0 10.6
PGSI 1 to 2 16.6 18.9 27.0 18.7 25.3
PGSI 3 to 7 51.5 61.4 59.1 60.7 61.2
PGSI score 8 or higher 94.0 100.0 93.5 91.8 100.0

Base: Productive individuals who had gambled in the past 4 weeks. Participants within each ethnic group who gambled in the past 4 weeks (unweighted: White groups = 21,436; Mixed and/or Multiple ethnic groups = 454; Asian groups = 861; Black groups = 305; Other ethnic groups = 99).

How does the use of support services vary by ethnic group?  

This section references information that can be found in the ethnic group topic report findings excel files that accompany this report.

The survey asked about the use of 4 different support services because of gambling (that is, mental health, foodbanks, relationship, gambling) in the past 12 months - the only timeframe for which this measure was collected. This covered obtaining help, support or information from these services. Use of these services varied by ethnic group. Overall, 3.4% of participants who had gambled in the past 12 months had sought any form of support, but this masked large differences between groups: 2.5% of participants from White backgrounds had sought any support, compared with 13.6% of those from Other ethnic backgrounds, 10.8% of those from Mixed and/or Multiple ethnic backgrounds, 8.7% of those from Black backgrounds and 8.0% of those from Asian backgrounds (as shown in Table 14 in the ethnic group topic report findings).

This pattern held across each individual type of support. Gambling support services specifically were accessed by 0.9% of participants from White backgrounds, compared with 5.8% of those from Other ethnic backgrounds, 5.1% of those from Mixed and/or Multiple ethnic backgrounds, 3.7% of those from Black backgrounds and 2.4% of those from Asian backgrounds. The most-used service differed by group: mental health services for participants from White backgrounds (1.2%), foodbanks for those from Asian backgrounds (5.1%), mental health services for those from Other ethnic backgrounds (8.9%) and those from Mixed and/or Multiple ethnic backgrounds (6.3%), and relationship support for those from Black backgrounds (4.8%) (as shown in Figure 7).

Figure 7: Reported use of support services among those who gambled in the past 12 months, by ethnic group (percentage)

Figure 7: Reported use of support services among those who gambled in the past 12 months, by ethnic group (percentage)

Figure 7: Reported use of support services among those who gambled in the past 12 months, by ethnic group (percentage)
Service type White groups Mixed and/or Multiple ethnic groups Asian groups Black groups Other ethnic groups
Any support 2.5 10.8 8.0 8.7 13.6
Mental health services 1.2 6.3 4.7 4.0 8.9
Foodbanks 1.1 6.2 5.1 4.4 8.0
Relationship support 1.0 5.3 3.4 4.8 9.1
Gambling support 0.9 5.1 2.4 3.7 5.8

Base: Productive individuals who gambled in the past 12 months. Participants within each ethnic group who gambled in the past 12 months (unweighted: White groups = 26,569; Mixed and/or Multiple ethnic groups = 588; Asian groups = 1,175; Black groups = 430; Other ethnic groups = 135).

Among participants with a Problem Gambling Severity Index (PGSI) score of 3 or more, service use was highest among participants from Mixed and/or Multiple ethnic and Other ethnic backgrounds, who sought support at roughly double the rate of participants from White backgrounds across most service types (any support: 41.4% and 42.1% respectively, versus 21.0% among participants from White backgrounds) as shown in Figure 8) (see Table 15 in the in the ethnic group topic report findings Excel file). Those from Asian and Black backgrounds showed a more modest gap, generally sitting between participants from White backgrounds and those from Mixed and/or Multiple and Other ethnic backgrounds, and in the case of gambling support specifically, the rate among participants from Asian backgrounds (8.9%) was similar to or slightly below that of participants from White backgrounds (10.0%). Even where service use was highest, it remained far from universal: around 76% of those with a PGSI score of three or more had not sought any support, regardless of ethnic group.

Figure 8: Reported use of support services among those with PGSI score of 3 or more, who gambled in the past 12 months, by ethnic group (percentage)

Figure 8: Reported use of support services among those with PGSI score of three or more, who gambled in the past 12 months, by ethnic group (percentage)

Figure 8: Reported use of support services among those with PGSI score of 3 or more, who gambled in the past 12 months, by ethnic group (percentage)
Service type White groups Mixed and/or Multiple ethnic groups Asian groups Black groups Other ethnic groups
Any support 21.0 41.4 27.3 27.0 42.1
Mental health services 10.0 24.5 16.6 13.2 29.2
Foodbanks 10.9 24.6 16.7 12.9 26.1
Relationship support 8.0 22.5 13.1 17.3 31.2
Gambling support 10.0 23.2 8.9 14.4 20.8

Base: Productive individuals who gambled in the past 12 months with a PGSI score of 3 or more. Participants within each ethnic group who gambled in the past 12 months and have a PGSI score of 3 or more (unweighted: White groups = 1,725; Mixed and/or Multiple ethnic groups = 109; Asian groups = 242; Black groups = 91; Other ethnic groups = 32).

Summary

Key findings

Commensurate with prior research, this analysis showed that participants from White backgrounds were the most likely to gamble overall. However, analysis in this report shows that this difference is driven mainly by participation in lottery draws, with those from White backgrounds being substantially more likely to take part in lottery draws than other ethnic groups. When looking at gambling activities separately, participants from ethnic minority backgrounds had much higher participation in non-lottery draw activities compared to those from White backgrounds, in some cases, their rates of participation were up to 4 to 5 times higher than those from White groups. This pattern was broad rather than limited to 1 or 2 activities: 9 individual gambling activities showed higher participation among participants from all 4 ethnic minority backgrounds compared with those from White backgrounds, while only 2 showed lower participation. Those who gambled from ethnic minority groups also tended to take part in more forms of gambling activity than their counterparts from White backgrounds.

Given the increased propensity of those from ethnic minority groups to engage in a wider range of gambling activities, and in non-lottery draw activities, it was perhaps unsurprising that Problem Gambling Severity Index (PGSI) scores were also higher among these groups. These patterns held, and in some cases strengthened, when lottery draw-only participants were excluded, suggesting the findings were not simply a lottery draw effect. Regression models showed that participants from ethnic minority backgrounds had higher PGSI scores even when taking into account demographic, socio-economic characteristics and gambling participation. Whilst the strength of the relationship attenuated, these results suggest that some other, unmeasured, factor may be contributing to increased PGSI risk among people from different ethnic minority groups.

Commensurate with the patterns observed for PGSI scores, participants from ethnic minority backgrounds reported higher rates of potential adverse and severe adverse consequences, often 3 to 4 times higher than participants from White backgrounds. With the exception of those from Black backgrounds, regression models also showed this difference in gambling participation, demographic or socio-economic factors did not fully account for this. Furthermore, this gap was not consistent across all levels of risk: at low PGSI scores, participants from ethnic minority backgrounds reported substantially higher rates of experiencing at least 1 potential adverse consequence than White participants though this gap narrowed among those with PGSI scores of 8 or more, where experience of adverse consequences was the norm across all groups. Interestingly, around 7% to 10% of participants from ethnic minority groups who had gambled in the past 4 weeks and had a PGSI score of 0 reported experiencing at least 1 adverse consequence from gambling. This compares with 2.7% (about 1 in 37) for those from White backgrounds. This suggests participants from ethnic minority backgrounds may experience more harm at earlier stages of risk rather than uniformly across the whole spectrum. This is consistent with the Gambling Commission's earlier report, which similarly found a higher prevalence of harm among those from ethnic minority backgrounds at lower PGSI scores, with this gap narrowing at higher scores.

As would be expected given the findings in this report, there was higher support-seeking among participants from ethnic minority backgrounds. However, even among participants with a PGSI score of 3 or more roughly 3 in 4 had not sought any support, regardless of ethnic group, suggesting not everyone who needs help is seeking it.

Limitations

This report pooled together 3 years of the GSGB to get a large enough sample to look at the ethnic minority groups separately. Despite this, the sample size is relatively small, for example there was a total of 337 people from Other ethnic backgrounds compared to 43,809 people from White backgrounds. This impacts our confidence in the estimates as the sample could be more sensitive to outliers. The results of the modelling should be considered alongside the 95% confidence intervals which reflect the uncertainty associated with a smaller sample. Going forward, as more years of the GSGB are available, it would be ideal to repeat these analyses to check the results still stand with a larger sample.

When reporting on the adverse consequences we used a derived continuous variable in the models. This was done to get an understanding of the spectrum of harm (compared to using a binary variable) and was created following the scoring methodology used in the DSM-IV. However, we acknowledge that this derived variable assumes the potential adverse consequences are of equal consequence to each other (and the same for severe consequences), when this might not be the case. For example, having to reduce or cut back spending on everyday items such as food, bills, and clothing could be interpreted as less severe compared to having to use savings or increase use of credit, such as credit cards, overdrafts, and loans.

Implication for future research

This report highlights the importance of disaggregating the data on gambling participation by gambling activity. Only looking at headline figures of participation masks wider and varied patterns of gambling among people from different ethnic backgrounds. Whilst people from ethnic minority groups gamble less overall, those that do have an increased propensity to gambling on non-lottery draw based products and to gamble on a wider range of activities.

The models in this report could not fully explain the relationship between ethnic background and adverse consequences or PGSI scores. Since these models considered demographic and socio-economic characteristics, and gambling involvement, other unobserved factors must drive the higher levels of harm and elevated PGSI scores observed among those from ethnic minority backgrounds. Future research should focus on identifying the underlying mechanisms. This could involve qualitative research with those with lived experiences which could inform future data collection and analysis.

Appendix

Gambling Activities Listing

The full description of gambling activities shown to participants when reporting past 12 month participation are as follows:

The full description of gambling activities shown to participants when reporting past 4 week participation are as follows: