• Adolescent;
  • educational status;
  • effect modifiers;
  • gambling;
  • personality;
  • proportional hazards models;
  • residence characteristics;
  • social class


Aims  To determine if impulsivity and socio-economic status (SES) interact to influence gambling onset in youth.

Design  Longitudinal study of grade 7 students followed for 8 years.

Setting  Montréal, Canada.

Participants  A total of 628 adult students aged 12.6 years on average at cohort inception.

Measurements  Impulsivity and SES (parent education, area deprivation) were collected during secondary school. Age of gambling onset was collected retrospectively when participants were aged 20.3 years. Cox proportional hazards regression was used to model the association between time to first report of gambling and interaction terms for each of impulsivity and parent education, and impulsivity and area deprivation accounting for sex and ethnicity.

Findings  Median (interquartile range) age of gambling onset was 17.0 (4.0) years. Impulsivity independently increased the risk of gambling onset among participants with no university-educated parent [hazard ratio (HR) 1.3; 95% confidence interval 1.1–1.5] and those living in highly deprived areas (HR 1.7; 1.5–2.0). Impulsivity was not associated with gambling onset among high SES youth. Among participants with high impulsivity, risks were elevated for those with no university-educated parent relative to one or more university-educated parent (HR 1.7; 1.1–2.7), and for participants living in deprived relative to advantaged areas (HR 5.0; 2.6–9.6). SES was not associated with gambling onset among participants with low impulsivity.

Conclusions  Impulsivity is a risk factor for gambling onset among low but not high SES youth, and low SES influences gambling onset primarily among impulsive youth. Gambling prevention programmes may need to consider potential interaction between impulsivity and SES.