BACKGROUND: Different gambling formats are associated with varying rates of gambling harms. Some studies have suggested that this relationship is driven by wider gambling involvement and thus argued that how much people gamble is a more useful predictor of gambling harms than engagement in specific activities. This assumption was tested using data from two cross-sectional surveys, the Gambling Survey for Great Britain (GSGB), 2023 and the Health Survey for England (HSE), 2015-2018.

METHODS: Binary logistic regressions examined the relationship between past year engagement in different gambling formats and Problem Gambling Severity Index (PGSI) scores of 8 or more. Negative binomial regressions examined past year engagement in each gambling format with PGSI scores overall. All models were run with and without controls for the number of other gambling activities undertaken and the demographic/socio-economic profile of participants. For the GSGB, frequency of gambling on other gambling formats was also included.

RESULTS: Results from both surveys showed that controls for wider gambling involvement moderated the relationship between gambling format and PGSI scores. However, engagement in gambling activities like online casinos and/or slots (GSGB, online casino gambling incidence risk ratio (IRR): 2.33, 95 % CI: 1.78-3.04; HSE online casino, slots and bingo IRR: 5.93, 95 % CI:4.45-7.90) and Electronic Gambling Machines (EGMs) (GSGB: IRR = 1.81, 95 % CI: 1.43-2.28; HSE IRR = 3.94, 95 % CI: 3.02-5.16) were independently associated elevated PGSI scores even when wider gambling involvement was taken into account.

CONCLUSIONS: Commensurate with known theories about product risk and wider research evidence, results suggest there may be an impact on PGSI scores from engaging in online casino and/or slots and EGMs that is independent of an individual's broader gambling involvement. Evidence from this study suggests that some forms of gambling may be more closely associated with problem gambling. Regulatory policy should focus on these forms.