With the introduction of gambling-like features within video games (e.g., loot boxes) new forms of hybrid-gambling products have emerged, yet little is known about their relationship to gambling and problem gambling among those most likely to engage: young people. This article examines the relationship between the purchase of loot boxes, gambling behavior, and problem gambling among young people ages 16–24. Cross-sectional data were analyzed from wave 1 of the Emerging Adults Gambling Survey, an online survey of 3,549 people, aged 16–24. Data were weighted to reflect the age, sex, and regional profile of Great Britain. Measured included past-year purchase of loot boxes, engagement in 17 different forms of gambling (weekly, yearly, and weekly spend); and problem gambling status. Other covariates include impulsivity and sociodemographic status. Young adults who purchase loot boxes are more likely to be gamblers and experience problem gambling than others. In unadjusted regression models, the odds of problem gambling were 11.4 (95% confidence interval [CI] 7.6 to 16.9; p < 0.001) times higher among those who purchased loot boxes with their own money. This relationship attenuated but remained significant (odds ratio 4.5, 95% CI 2.6–7.9) when gambling participation, impulsivity, and sociodemographic factors were taken into account. The purchase of loot boxes was highly associated with problem gambling, the strength of this association being of similar magnitude to gambling online on casino games or slots. Young adults purchasing loot boxes within video games should be considered a high-risk group for the experience of gambling problems.

Keywords: gambling, young people, loot boxes, gaming

Recent years have seen an emerging trend of gambling-like features being embedded in different contexts.1 This is especially so within video games and is, arguably, best exemplified by the growth of loot boxes within video games.2,3 Loot boxes are items that may be bought for real-world money, but which contain randomized contents whose value is uncertain at the point of purchase.4 They are a popular form of microtransaction now included within video games to obtain money from players, upon which game developers are increasingly reliant as a revenue stream. Indeed, recent research has suggested that the majority of top-grossing mobile games on both Apple and Android devices now contain loot boxes.4

Similarly, a recent analysis of the desktop gaming platform Steam investigated the proportion of desktop play sessions that take place in games with loot boxes. It has suggested that more than 70 percent of desktop play sessions now take place in a game that is monetized through loot boxes.5

Loot boxes, along with other microtransaction processes, have been described as a “predatory practice,” which entraps people into repeated purchasing.2 It has been suggested that they are “psychologically akin” to gambling as individuals stake money on the uncertain outcome of a future event in the hope of receiving something of greater value.3 Some jurisdictions agree and have taken regulatory action: Belgium has banned the use of loot boxes within some video games stating they are a violation of gambling legislation; gambling authorities in the Netherlands have ruled that some loot boxes constitute unlicensed games of chance; and China has required that the odds of winning be displayed to consumers.

There is some evidence that consumers themselves view loot box purchase as a form of gambling. In two separate small-scale surveys in Canada, between 68 percent and 86 percent of participants agreed that loot boxes were a form of gambling and between 75 percent and 79 percent of participants agreed that opening a loot box felt like making a bet.6 In Great Britain, a recent study by the Royal Society for Public Health found that 79 percent of young people ages 11–24 thought that loot boxes were a highly addictive form of gambling.7

An emerging evidence base has demonstrated an association between the purchase of loot boxes and problem gambling, with these findings repeated across time and space despite studies using different methodologies.7–13 However, the majority of these studies simply look at the association between loot box purchasing and problem gambling and do not take into the broader gambling or gaming behaviors of these people. Gainsbury1 has noted the need for caution, arguing that observed relationships between loot box engagement and problem gambling may be explained by a confound: interest in both gambling and gaming.

Similar arguments occur within gambling studies, whereby it is postulated that the relationship between specific types of gambling and problem gambling are confounded by wider interest and engagement in gambling itself: the “involvement” hypothesis.14 Some studies have supported this, finding that once the depth and breadth of gambling engagement are taken into account, the relationship between specific gambling activities and problem gambling attenuates or is no longer significant.14,15 Others have found that the relationship persists for certain activities.16

Gambling involvement is just one of many possible confounds which may explain an association between loot box purchase and problem gambling. Other potential confounds include personality traits, such as impulsivity or sociodemographic/economic status. Impulsivity has been identified as having a strong relationship with problem gambling in many other studies.17 One study has suggested that impulsivity was related to both loot box purchasing and problem gambling. However, no investigation of whether impulsivity accounted for the associations between the two took place. In addition, shared features of the sociodemographic or economic profile of loot box purchasers and problem gamblers may also explain associations between loot box engagement and gambling. Being male, younger, from non-white ethnic groups, and unemployed are factors commonly associated with problem gambling.18

The aims of this study were to investigate