Objective  To examine the prevalence of myopia and its related personal eye habits among junior high school students and to explore stage-specific myopia prevention behaviors of the students based on health action process approach (HAPA) and theory of planned behavior (TBA).

  Methods  Using stratified random cluster sampling, we recruited 4 894 students at 6 junior high schools in 5 prefectures of Guangdong province. An anonymous questionnaire survey was conducted among 2 289 students online and 2 570 students on-site during July 2020.

  Results  Of the 4 859 students with valid responses, 3 013 (62.0%) reported being myopic and 994 (20.5%) reported suffering from myopia during past 6 months. The students being female, in higher schooling grade, and with parental myopia reported a significantly higher myopia prevalence rate than other students (P < 0.001 for all). The students′ myopia prevalence differed significantly by sleeping time during school days, proper reading/writing time during school days or weekends, daily outdoor activity, outdoor activity in school days, outdoor activity with parents/friends during free days, posture while reading/writing, and rest time after continuous reading for 40 minutes (all P < 0.05). The results of multivariate logistic regression analysis revealed following risk factors of myopia for the students: being female, in higher schooling grade, with parental myopia, low frequency of outdoor activity for 2 hours per day, and low frequently of having a rest of 10 minutes after reading/writing for 40 minutes. Based on reported frequencies of having daily outdoor activity for at least 2 hours and having a rest of 10 minutes after reading/writing for 40 minutes, we classified 1 543 (31.8%), 1 433 (29.5%), and 1 883 (38.3%) of the students as at the pre-intention, intention, and action stage of developing eye habits for myopia prevention; high proportions of the boy students and the students in low schooling grade were assessed in the action stage. The students at the action stage reported a higher social cognition and better eye habits related to myopia prevention than the students at both pre-intention and intention stage (P < 0.01 for all) but no difference in risk cognition was observed among the students at different stages. The results of hierarchical regression analysis showed that for the students at pre-intention stage, the attitude towards and self-efficacy of action were major predictors for myopia prevention behaviors, jointly explaining 15% of total variance of behavior change; for the students at intention stage, risk perception, subjective norm, and action self-efficacy were major predictors, altogether explaining for 12% of total variance of behavior change; whereas, attitude towards action, subjective norm, action and coping self-efficacy, and behavioral planning and coping were major predictors, explaining 20% of total variance of behavior change, for the students at action stage.

  Conclusion  Among junior high school students, the prevalence of myopia varies with individual characteristic and eye habits; the influence of social cognition variables on myopia prevention behaviors is different among the students at various stages of developing eye habits for myopia prevention.