Gaming and gambling: French study finds striking differences in risk among teenagers 

Teenagers who engage in both digital games featuring gambling-like elements and real-money gambling are more frequently found in the highest category of gambling-related problems. This is the finding of a French study involving 5,000 young people aged 15 to 17. For Belgium, the study relates to a familiar issue for the Belgian Gaming Commission: the increasingly blurred line between video games and gambling. 

The researchers identified an association. Their findings do not prove that loot boxes, casino mechanics or other forms of digital gaming cause gambling problems. However, they highlight why research involving young people needs to look beyond traditional forms of gambling. 

The study examines six specific gaming categories 

The French research and prevention organisation ARPEJ conducted the ENJEU-Mineurs study with funding and expert support from the Autorité Nationale des Jeux (ANJ), France’s gambling regulator. The authors remained responsible for the conclusions. 

Participants completed an online questionnaire between August 29 and September 16, 2025. They were recruited through family panels, with parental consent. The researchers used quotas based on gender, age and region, and applied weighting based on population data. 

The study defines “hybrid digital games” as six categories: social casino games, free gambling demos, video games featuring casino activities, loot boxes, fantasy sports, and Web3 or play-to-earn products. 

The research therefore does not concern video games in general. Nor does this classification mean that all these products are legally considered gambling. The ANJ publication on hybrid games among minors uses the term to examine different types of games with gambling-like features together. 

Nearly four in ten teenagers played a hybrid game 

Among the teenagers surveyed, 39% reported participating in at least one form of hybrid gaming. However, the primary sources differ regarding the period covered by this percentage. 

The full ARPEJ report refers to participation during the three months preceding the study. The ANJ mentions twelve months in its summary. The percentage therefore cannot be presented as a clear annual figure without explaining this discrepancy. 

It is also important to use the correct reference group when discussing spending. Of the entire sample, 15.6% reported spending money on hybrid games. Among users of these games alone, the figure was 39.9%. 

These percentages describe the same spending behaviour across different groups. It would therefore be incorrect to state that only 15.6% of hybrid game users spent money, or that almost 40% of all teenagers surveyed paid to play. 

Highest category of gambling problems more common among those combining both activities 

Among hybrid game users, 58.8% also reported participating in real-money gambling. 

The researchers assessed gambling problems using the Canadian Adolescent Gambling Inventory, a measurement tool designed for young people. Among those who combined both activities, 26% fell into the highest severity category. Among teenagers who reported only real-money gambling, the figure was 3.9%. 

This is a substantial difference, but it requires careful wording. The percentages represent proportions within specific study groups, not clinical diagnoses of gambling addiction. Nor does the finding mean that 26% of all teenagers surveyed experienced severe gambling problems. 

Above all, the comparison raises a specific research question: which characteristics explain why the group combining both activities reports problems more frequently? The association alone does not provide an answer. 

Playing earlier does not prove a causal progression to gambling 

Among teenagers who combined hybrid games with real-money gambling, 67.4% recalled first playing free games featuring gambling-like elements. 

This percentage applies exclusively to the group engaging in both activities. It does not describe the experiences of two-thirds of all 5,000 participants. 

Furthermore, participants were asked to recall the sequence of events retrospectively. Memories may be incomplete. Even if the reported sequence is accurate, it does not prove that the earlier gaming experience caused subsequent gambling. 

The study did not follow teenagers over an extended period to establish who developed gambling problems after playing a particular game. Other characteristics, such as an existing interest in gambling or social influences, may also be associated with both activities. 

The full ARPEJ report on ENJEU-Mineurs also discusses the limitations of self-reported behaviour and online panels. Such panels may overrepresent excessive behaviour compared with the general population. 

Belgium already has a legal framework concerning loot boxes 

For Belgium, the role of paid random rewards is particularly relevant. In 2018, the Belgian Gaming Commission classified paid loot box systems as gambling. Products that meet the legal criteria for gambling cannot be offered without the required authorisation. 

Gambling Club previously examined how Belgium and Europe are addressing loot boxes. In this debate, distinguishing between legal classification and potential behavioural risks is essential. 

A product may have gambling-like features without automatically being subject to the same rules as an online casino. Conversely, placing a paid chance-based mechanism inside a video game does not automatically exempt it from gambling regulations. 

The French study does not change Belgian law. However, it provides additional information for the debate about which gaming features deserve attention and how minors encounter them. 

Advertising reaches young people through different channels 

Among hybrid game users, 68.9% said they had seen messages encouraging participation in real-money gambling. 

The channels reported included social media and online games. This does not mean that every message appeared within a game. Nor did the study establish which specific advertisement influenced each respondent’s behaviour. 

For Belgium, this connects with earlier questions about the distribution of loot box advertisements through Meta. Gambling Club reported on research into the visibility of such advertisements among Belgian users. 

Several questions are relevant for regulatory oversight: what is being offered, how is it being promoted, and who is seeing the message? However, advertising research and studies of gambling problems measure different things. Their findings should not be combined as though they constituted a single body of causal evidence. 

Parents and schools do not always see the same gaming environment 

For prevention efforts in Belgium, one useful lesson is the importance of recognising specific gaming mechanisms. A conversation about “gaming” in general may be too broad when the real issue concerns paid mystery boxes, virtual currencies or simulated casino activities. 

Gambling Club previously explored the role of parents and schools in addressing gambling among young people. That broader perspective is also relevant here: young people’s gaming environments involve much more than the moment they attempt to open a gambling account. 

The French percentages cannot be directly applied to Belgian teenagers. A Belgian assessment would require national data on participation, spending, exposure to advertising and the availability of these types of games. 

The study provides concrete questions for further investigation without predetermining what the answers will be in Belgium.