The definition and measurement of social costs has been debated in gambling literature. Predominant framings of gambling have depicted it as a leisure activity without significant costs other than those caused directly by problem gambling. This view has been recently challenged with a public health perspective, adopting a wider definition of social costs that span beyond issues related to treatment and crime.

Definitional debates have resulted in highly heterogenous approaches to calculating the social costs of gambling. The aim of this review is, first, to assess the kind of costs that have been included in existing estimates; second, to compare overall estimates of social costs; and third, to compare methodological differences.

We conducted a scoping review following the Joanna Briggs Institute guidelines for scoping reviews, using the PRISMA-SCR/P tool. The literature search included both peer reviewed publications and grey literature and was conducted in June 2023 and repeated in March 2024. After screening, 26 references were included in the review.

Estimates in the included studies contained direct costs, such as crime and law enforcement, financial counselling, and treatment-related costs. Indirect cost included items such as health and social care, unemployment, and financial cost of divorce. Intangible costs included, for example, quantifications of suicide, emotional distress, and relationship problems. The number of included cost items varied across studies from 5 to 32. We identified seven different methodological approaches to calculating costs. Total estimates using different approaches varied greatly, ranging from 6 to 324,000 million int$ total, with an average of 3,980 int$ per adult and median 449 int$ per adult.

We conclude that the high variation in available total estimates is due to differences in number of cost items and methodological approaches. The results highlight the need for consistent international guidelines for calculating the total social cost of gambling.