Gambling Harms Statistically Mediate the Association between Risk-weighted Gambling Involvement and Wellbeing in a Population-representative Sample

The widely adopted functional definition of gambling-related harm implies a three-part pathway: gambling engagement gives rise to harmful consequences, which in turn reduce health and wellbeing. This pathway has not previously been tested in general-population data. A population-weighted Australian analytic sample (N=4,001) was used to test whether this structure holds empirically. A risk-weighted gambling involvement index was derived from form-specific frequencies, with greater weight given to forms most strongly associated with gambling problems. Weighted mediation models showed that gambling harms statistically accounted for the observed association between gambling involvement and reduced wellbeing: indirect effects through harms were consistently negative and robust to demographic adjustment, while direct effects of gambling involvement were essentially zero once harms were included. The pattern held for both an expanded 26-item harm composite and the shorter 10-item gambling harms screen (GHS-10), and was clearest for life satisfaction and mental wellbeing. These results provide the first cross-sectional pathway-based support for the structure implied by this definition in representative population data. As a corollary, they support the construct validity of direct harm screens, which behave as expected of measures occupying the harm position in this pathway. Population-level effect sizes were modest, reflecting the concentration of harm burden in a minority of gamblers, but the wellbeing deficit among those reporting harms was substantial.