Substance-related and behavioural addiction problems: Two surveys of Canadian adults (2015)

Objectives: To describe absolute and relative prevalence of 10 self-attributed substance-related and behavioural addiction problems among Canadian adults, to describe perceived prevalence of these problems in the general population, and to examine whether estimates varied by survey mode.
Methods: Sample 1 included 4000 adults recruited from an online research panel; Sample 2 included 2000 randomly selected adults who completed a computer-assisted telephone interview. Respondents in both samples were asked (1) whether or not they had experienced a problem in the preceding year with each of four substances (alcohol, tobacco, marijuana, and cocaine) and six behaviours (gambling, eating, shopping, sex, video gaming, and work), (2) whether they had ever personally knew someone exhibiting a problem with each behaviour, and (3) perceived prevalence of problems in the general population.
Results: About half of the respondents reported experiencing any past-year addiction problems. Tobacco and alcohol were the most common substance-related problems, while the most common behavioural problems were related to eating and work. Respondents consistently overestimated perceived population prevalence relative to self-attributed problems; however, the magnitude of overestimation was significantly greater among those who personally reported a problem with these behaviours. Online survey participants consistently reported higher self-attributed problem rates compared with CATI respondents, but rank-order correlations across self-, acquaintances-, and population-attributed prevalence estimates were very high in both samples.
Conclusions: Both survey modes provided accurate relative prevalence estimates, but further research should explore determinants of higher prevalence rates among online participants and respondents' consistent tendency to overestimate perceived population prevalence.