Date of Award


Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PhD)





First Advisor

Dr Darren Morton

Second Advisor

Dr Barry Gane


111712 Health Promotion, 111714 Mental Health, 111799 Public Health and Health Services not elsewhere classified


Adolescent health is a major worldwide concern and is central to a number of current global health challenges. There is growing awareness that the factors influencing adolescent health are multi-dimensional and encompass a broad network of interrelated determinants. Although studies have explored the influence of selected determinants on adolescent health, few have explored the relative importance of the various determinants and the relationships that exist between them. Using Structural Equation Modelling (SEM), this thesis empirically investigates the complex network of factors that concomitantly predict health behaviours and health outcomes in adolescents. The thesis centres on data collected from a comprehensive health and lifestyle survey administered to 1734 adolescents (mean age = 14.5 ± 1.6 years; 54% males) attending 21 Seventh-day Adventist (Adventist) schools in Australia. The research findings are presented as a coherent series of five studies for publication in peer reviewed journals.

Study 1 focused on the self-rated health (SRH) status of the adolescents and investigated the association of a multitude of determinants including personal demographics; background factors such as childhood family dynamics (CFD) and adverse childhood experiences (ACEs); and selected health behaviours and health measures. The findings of Study 1 confirmed the complex relationship between these determinants and highlighted their respective relationships. Particularly noteworthy in this study was the relative high degree of association of mental health, body mass index (BMI) and ACEs on SRH. The findings from Study 1 formed the basis for Studies 2 to 5 which more extensively explored particular aspects of the model developed in Study 1.

Extending upon the observation in Study 1 of the significant influence of mental health status on the SRH of the adolescents, Study 2 examined the predictors of mental health in greater detail. The model developed in Study 2 through SEM analyses revealed that CFD was the strongest predictor of the adolescents’ mental health status, followed by having a sense of meaning and purpose, perceived social rejection and school academic performance. Multi-group analysis for gender found significant differences. The mental health of males was more greatly affected by physical activity, whereas the mental health of females was more greatly affected by sleep duration.

Another significant finding of Study 1 was the influence of BMI, a measure of overweight and obesity, on SRH and this formed the focus of Study 3. Compared to national norms, lower rates of overweight and obesity but higher rates of underweight were observed in the study, which was explained by the unique characteristics of the school system the study cohort was drawn from. BMI was lower among the young members of the cohort, as well as those who more regularly ate breakfast, consumed less soft drink and had a regular exercise program.

In Study 1, alcohol consumption was not significantly associated with self-rated health; however, a low rate of alcohol consumption was observed among the study cohort. Study 4, therefore, aimed to better understand the factors influencing alcohol consumption among this unique cohort. The findings of Study 4 indicated that adolescents’ intentions to consume alcohol were the greatest predictor of alcohol consumption. Consistent with the Theory of Reasoned Action, intentions to consume alcohol was influenced by their attitudes towards alcohol consumption and subjective norms surrounding alcohol consumption.

Finally, Study 5 focused on the unique faith-based aspect of the cohort to ascertain the influence of religious affiliation on the various health behaviours and health outcomes examined in Studies 1–4. The study found that adolescents who identified themselves as Adventist (57% of the total cohort) reported significantly better health behaviours than the other Christian and non-religious adolescents. This was especially significant among older adolescents (16–18 years). However, these better health behaviours did not translate to improved health status.

The findings from this thesis provide for the first time, a comprehensive picture of the complex network of factors associated with the health outcomes of adolescents attending Adventist schools in Australia. The findings support the need for age and gender appropriate multi-component interventions and prevention initiatives to promote positive health outcomes in adolescents, with an emphasis on prioritising not only modifiable health behaviours but upstream factors such as CFD and ACEs.


Used by permission: the author.

Staff and Students of Avondale College may access a print copy of this thesis from Avondale College Library ().