Modeling Early Gambling Behavior Using Indicators from Online Lottery Gambling Tracking Data: Longitudinal Analysis.

摘要: Background: Individuals who gamble online may be at risk of gambling excessively, but internet also provides a unique opportunity to monitor behavior in real environments which allow intervention for those encounter difficulties. Objective: The objective this study was model the early trajectories individuals play lottery. Methods: Anonymized gambling‐related records initial 6 months 1152 clients French national lottery created their accounts between September 2015 and February 2016 were analyzed using two-step approach that combined growth mixture modeling latent class analysis. analysis based upon indicators activity (money wagered number days) problems (breadth involvement chasing). Profiles described probabilities following identified four indicators, several covariates (age, gender, deposits, type play, net losses, voluntary self-exclusion, Playscan classification—a responsible tool each player with assessment: green low risk, orange medium red high risk). Net classification used as external verification problem gambling. Results: We 5 distinct profiles gambling. Classes 1 (56.8%), 2 (14.8%) 3 (13.9%) characterized by values markers They displayed did not use self-exclusion measure, classified predominantly tags (range 90%-98%). Class 4 (9.7%) activity, played higher breadth game types 1-6), had zero few chasing episodes. losses (66%) or (25%) measure. (4.8%) very 1-17), episodes 0-5). experienced highest proportion (32%) (39%) within system represented only present. Conclusions: 1, 2, considered represent recreational representative players future problems. stood out terms much involvement, presence behavior. classes benefit from preventive measures.