Factor structure of health and oral health-related behaviors among adolescents in Arusha, northern Tanzania

Abstract

Objective. This study aimed to evaluate the factor structure of health and oral health-related behaviors and it's invariance across gender and to identify factors associated with behavioral patterns. Materials and methods . A cross-sectional study included 2412 students attending 20 secondary schools in Arusha. Self-administered questionnaires were completed at school. Results . Principal component analysis of seven single health and oral health-related behaviors (tooth brushing, hand wash after latrine, hand wash before eating, using soap, intake of sugared mineral water, intake of fast foods and intake of sweets) suggested two factors labeled hygiene behavior and snacking. Confirmatory factor analyses, CFA, provided acceptable fit for the hypothesized two-factor model; CFI = 0.97. Multiple group CFA across gender showed no statistically significant difference in fit between unconstrained and constrained models (p = 0.203). Logistic regression revealed ORs for hygiene behaviors of 1.5, 0.5, 1.5, 1.5 and 0.6 if being a girl, current smoker, reporting good relationship with school, access to hygiene facilities and bad life satisfaction, respectively. ORs for snacking were 1.3, 1.4, 0.4 and 0.5 if female, in the least poor household quartile, low family socio-economic status and high perceived control, respectively. Conclusion . The two factors suggest that behaviors within each might be approached jointly in health promoting programs. A positive relationship with school and access to hygiene facilities might play a role in health promotion. Provision of healthy snacks and improved perceived behavioral control regarding sugar avoidance might restrict snacking during school hours.