Incidence and predictive factors of Internet addiction among Chinese secondary school students in Hong Kong: a longitudinal study.

<bold>Purpose: </bold>Internet use has global influences on all aspects of life and has become a growing concern. Cross-sectional studies on Internet addiction (IA) have been reported but causality is often unclear. More longitudinal studies are warranted.<bold>Methods: </bold>We investigated incidence and predictors of IA conversion among secondary school students. A 12-month longitudinal study was conducted among Hong Kong Chinese Secondary 1-4 students (N = 8286). Using the 26-item Chen Internet Addiction Scale (CIAS; cut-off >63), non-IA cases were identified at baseline. Conversion to IA during the follow-up period was detected, with incidence and predictors derived using multi-level models.<bold>Results: </bold>Prevalence of IA was 16.0% at baseline and incidence of IA was 11.81 per 100 person-years (13.74 for males and 9.78 for females). Risk background factors were male sex, higher school forms, and living with only one parent, while protective background factors were having a mother/father with university education. Adjusted for all background factors, higher baseline CIAS score (ORa = 1.07), longer hours spent online for entertainment and social communication (ORa = 1.92 and 1.63 respectively), and Health Belief Model (HBM) constructs (except perceived severity of IA and perceived self-efficacy to reduce use) were significant predictors of conversion to IA (ORa = 1.07-1.45).<bold>Conclusions: </bold>Prevalence and incidence of IA conversion were high and need attention. Interventions should take into account risk predictors identified, such as those of the HBM, and time management skills should be enhanced. Screening is warranted to identify those at high risk (e.g. high CIAS score) and provide them with primary and secondary interventions. [ABSTRACT FROM AUTHOR]