Longitudinal investigations provide the best information about the etiological contributors to problem gambling. Many prior longitudinal studies have been conducted, including large-scale ones in Alberta (LLLP), Ontario (QLS), Sweden (Swelogs), Victoria, Australia (VGS), New Zealand (NZ-NGS), and Massachusetts (MAGIC). Collectively, these studies have provided a wealth of useful information. However, while instructive, a major limitation is that they all had very small numbers of problem gamblers (generally between 50 - 150), limiting the power of their analyses. Several of these studies also did not comprehensively assess all variables of potential etiological importance. An important finding deriving from these investigations is that the etiological predictors identified appear to be somewhat jurisdiction and time frame specific. In this regard, there have been no national Canadian longitudinal studies, and none conducted in recent years. The present study addresses these above issues with a national cohort of 10,119 Canadian gamblers that included 1,346 problem gamblers in the 2018 baseline as well as 2,710 with a mental health disorder and 778 with a substance use disorder. This presentation will report on the predictors of both concurrent problem gambling in the 2018 baseline as well as variables that most strongly predicted future problem gambling at the follow-up in 2019. It includes an analysis of the self-reported perceived causes of problem gambling among 1040 problem gamblers and the policy implications that derive from these collective findings.