Cybervictimization and psychological burden among internet users aged 15 years and older: online harassment, fraud, and co-victimization

Background Digital victimization is heterogeneous, yet population studies rarely distinguish interpersonal cybervictimization from fraud or determine whether co-occurrence adds psychological burden. Methods We analyzed public-use data from 17,403 Internet users aged ≥15 years in a Canadian population victimization survey. Five-year indicators defined four groups: neither exposure, fraud only, cyberstalking/cyberbullying only, and both; fraud delivery mode was unavailable. The primary outcome was fair/poor self-rated mental health; secondary outcomes were recent medication use to aid sleep, calmness, or depression and current life satisfaction (0–10). Weighted modified Poisson and linear models used all 500 survey bootstrap replicate weights. Core models adjusted for sociodemographic characteristics, religious affiliation, and Internet use; extended models also included conventional victimization, offline stalking, and discrimination. Results After extended adjustment, fair/poor mental health was more prevalent in the cyber-only group [prevalence ratio (PR) 1.30, 95% confidence interval (CI) 1.05–1.62]. The both-exposure estimate was elevated but imprecise (PR 1.38, 95% CI 0.99–1.94), and the fraud-only CI included the null (PR 1.12, 95% CI 0.93–1.36). All exposed groups had lower life satisfaction; medication associations were weaker. Interaction estimates did not establish synergy. Discussion The added adversity variables may have multiple temporal roles, so attenuation between models is not a mediation estimate. The released age variable combines ages 15–34, precluding adolescent-specific estimates. Coordinated assessment is a practice implication that requires evaluation.