Compliance
PGSI (Problem Gambling Severity Index)
Definition
A nine-question screening instrument, scored from 0 to 27, used in population surveys to classify respondents as non-problem, low-risk, moderate-risk or problem gamblers. The standard measure behind prevalence statistics quoted by regulators, governments and the industry.
Key takeaways
- The PGSI is a nine-question screen scoring 0 to 27, banding respondents from non-problem to problem gambling at 8 or more.
- It is the source of almost every prevalence figure regulators and governments quote.
- Survey method changes the number; online panels and new methodologies produce higher rates, as Britain’s 2023 change showed.
- It measures risk, not a clinical diagnosis, and it drives policy on stake limits, advertising and levies.
Why it matters
When a regulator says that a certain percentage of adults are problem gamblers, the figure almost always comes from the PGSI. The index asks respondents how often in the past twelve months they have, among other things, bet more than they could afford to lose, needed to gamble with larger amounts to get the same feeling, gone back to try to win back money lost, borrowed money or sold anything to gamble, or felt guilty about their gambling; each answer scores from never (0) to almost always (3); and the total places the respondent in one of four bands, with 8 or more classed as problem gambling. It was developed in Canada in 2001 and is used in national surveys across Europe, Australia and North America.
The industry needs to understand the PGSI because prevalence numbers drive policy. A survey showing rising moderate-risk and problem gambling rates is the basis for stake limits, advertising restrictions and levies; a survey showing stable rates is the industry's defence. The method matters to the number: surveys of online panels produce higher rates than surveys of the general population, changes in survey method produce apparent changes in prevalence, and Britain's move to a new survey methodology in 2023 produced a headline problem-gambling rate several times the previous one, with the regulator cautioning against comparison and the industry disputing the design.
Operators also use PGSI-derived questions in their own customer interactions and research, and some responsible-gambling tools incorporate self-assessment based on it. It is a screen, not a diagnosis: the clinical definition of gambling disorder is separate, and a PGSI score indicates risk rather than a condition.
Sources
- Gambling Survey for Great Britain: annual report (Year 1, 2023) - Gambling Commission
Frequently asked questions
What PGSI score counts as problem gambling?
A score of 8 or more out of 27. Scores of 1 to 2 are low risk, 3 to 7 moderate risk, and 0 non-problem gambling, under the standard banding.
Why do problem gambling rates differ between surveys?
Because the method differs. Online panel surveys, telephone surveys and in-person surveys reach different people and produce different rates, and a change in method produces an apparent change in prevalence that is not a real one.
Do operators use the PGSI?
Indirectly. Some self-assessment tools and research use PGSI questions, and operators’ markers of harm draw on the same behaviours, but the index itself is a population survey instrument.