Although several risk factors for anxiety-related phenotypes have been identified, few studies have quantified the weight of each one of these factors.

The aim of this study is to quantify contributions for anxiety variance in adolescents.

Ninety-seven adolescents from a community cross-sectional study and their parents were assessed with a structured DSM-IV diagnostic interview, several psychiatric and environmental rating scales and with a nutritional evaluation. The whole evaluation protocol generates 31 constructs previously associated with anxiety-related phenotypes in the literature. These constructs were organized in a hierarchical block analysis as follows:

1. (1) socio-demographic and early life environment;

2. (2) family environment and maternal psychiatric symptomatology and temperament;

3. (3) stressfull life-events and maltreatment;

4. (4) physical health and nutrition. The Screen for Children and Adolescent Related Anxiety Disorders (SCARED) was used as outcome.

The results of our hierarchical model accounts for only 24% of the total SCARED score. Hierarchically, block one and two did not explain significantly the SCARED variability (r2 = 0.078; p = 0.011 and r2=0.106; p = 0.064/pincremental = 0.090). After the inclusion of block 3, the hierarchical model achieved statistical significance (r2 = 0.231; p = 0.001/pincremental = 0.001), with little additional explanation after the inclusion of block 4 (r2 = 0.243; p = 0.001/pincremental = 0.242).

Even after an extensive exploratory analysis we were able to explain only about one fourth of anxiety symptoms variability in adolescents. Our findings support the notion that we still have a distant comprehension of the factors that lead to anxiety in adolescents and that the major determinants of anxiety disorders in adolescence are still missing.