International Journal of Preventive Medicine (Int J Prev Med)2008-78024220130202146152ENDepartment of Bio‑statistics and Epidemiology, School of Health, Isfahan University of Medical Sciences, Isfahan, IranDepartment of Bio‑statistics and Epidemiology, School of Health, Isfahan University of Medical Sciences, Isfahan, IranDepartment of Pediatrics, Child Growth and Development Research Center, and Faculty of Medicine, Isfahan University of Medical Sciences, Isfahan, IranDepartment of Pediatrics, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Bureau of Population, Family, and School Health, Ministry of Health and Medical Education, Tehran, IranBureau ofPopulation, Family, and School Health, Ministry of Health and Medical Education, Tehran, IranBureau ofPopulation, Family, and School Health, Ministry of Health and Medical Education, Tehran, IranEndocrinology and Metabolism Research Institute, Tehran University of Medical Sciences, Tehran, IranBureau of Health and Fitness, Ministry of Education, Tehran, IranBureau ofPopulation, Family, and School Health, Ministry of Health and Medical Education, Tehran, Iran20130202<p><strong>Objective: </strong>To evaluate predictive factors of adolescents&rsquo; appraisal of their health.</p> <p><strong>Methods: </strong>The nationwide study, entitled &ldquo;Childhood and Adolescence Surveillance and Prevention of Adult Non‑communicable Diseases (CASPIAN) study,&rdquo; was conducted in 2010 among Iranian school students, aged 10‑18. In addition to demographic factors and physical examination, variables as family structure, nutrition habits, physical activity, smoking, hygienic habits, violence, school attachment, family smoking, and family history of chronic diseases were assessed. The dependent variable is the self‑rated health (SRH) and it was measured by 12 items, which had already been combined through latent class analysis. We had taken a dichotomous variable, i.e. the higher values indicate better SRH. The dependent variable was regressed on all predictors by generalized additive models.</p> <p><strong>Results: </strong>75% of adolescents had a good SRH. The linear and smooth effects of independent variables on SRH were observed. Among all the variables, physical activity had a positive linear effect on SRH (&beta; = 0.08, <em>P </em>value = 0.003). Smoking, violence, and family history of disease associated to SRH non‑linearly (<em>P </em>value &lt; 0.05). Family smoking (&beta; = &ndash;0.01) and hygienic habits (&beta; = 0.27) related to SRH both linearly and non‑linearly.</p> <p><strong>Conclusions: </strong>Physical health and high risk behavior, either of linear or non‑linear effect, are factors, which seem to shape the adolescents&rsquo; perception of health.</p> <strong>Keywords: </strong>Adolescents, health status, health status indicator, logistic models, non‑parametric statistics, school, self‑reporthttp://ijpm.mui.ac.ir/index.php/ijpm/article/view/1050http://ijpm.mui.ac.ir/index.php/ijpm/article/download/1050/798