In this analysis, we examined the relationships between common sources of worry or stress, mental health symptoms and substance use in a population-based sample of young adolescents. In contrast to previous research that typically examined only 1–2 stressors in relation to a single outcome, our approach examined 6 sources of stress in relation to several mental health and substance use outcomes.
We found that adolescents reported worry or stress about a wide range of life events and difficulties. Two previous province-wide, school-based surveys of adolescents aged 12–18 years (n=1841 and n=2484, respectively) that used the same stress measure19, reported similar or lower proportions. Specifically, 22.5% and 20.3% (versus 21.3% in our study) reported romantic breakup stress; 5.7% and 5.8% (versus 18.7%) reported stress related to the divorce or separation of parents; 3.1% and 3.1% (versus 7.1%) reported stress related to a new family; 15.5% and 14.6% (versus 22.3%) reported stress related to relationship with father, 14.3% and 13.9% (versus 19.2%) reported stress related to relationship with mother; and 10.8% and 8.5% (versus 25.8%) reported stress related to a health problem. Our higher proportions may reflect that young adolescents find these life events and difficulties more stressful or worrisome than their older counterparts.
Consistent with previous reports [64
], worry and stress were reported more frequently by girls than boys. This is likely related to well-established sex differences in how girls and boys cope with stress [66
]. Girls tend to ruminate (i.e., passively and repetitively focus on their perceived failures) and use emotions to cope, both of which correlate highly with a “stressful reaction.” Boys are more problem-focused and use distraction to cope, both of which are unrelated to a “stressful reaction”.
It is notable that, with the exception of illicit drug use, romantic breakup stress was independently associated with all outcomes investigated. This finding is consistent with a growing literature suggesting that romantic involvement and breaking up in young adolescents is associated with depression, anxiety and conduct disorders [43
]. Romantic relationships at age 12–13 may be a particularly salient source of stress because of the tendency among younger adolescents to over attribute (i.e., “blame”) current mental symptoms to romantic relationships, rather than other potential concomitant stressors, such as problems with friends or schoolwork. The high emotional salience of romantic relationships in the adolescent mindset may reflect a lack of romantic experience and maturity, which may be associated with “catastrophic” reactions to relatively minor upsets in romantic relationships.
Similarly, young adolescents may also be particularly sensitive to stress related to other personal relationships. The finding that interpersonal stress was associated with depression is consistent with recent literature suggesting a longitudinal bi-directional association between depressive symptoms and interpersonal relationship problems [69
Family disruption stress was associated with depressive symptoms, which confirms previous reports and may reflect the effects of a detrimental familial environment where young adolescents spend the majority of their time [41
]. Parental or family problems may decrease the family attachment felt by adolescents, which could lead to an increase in involvement with substance-using peers [74
]. Cigarette and marijuana use have been associated with parental discord, separation, and divorce in the few extant studies in non-clinical adolescent populations [51
]. It is also possible that adolescents use cigarettes and marijuana for their mood-lifting and anxiolytic effects [80
] or as an avoidance or escape mechanism from family problems at home.
Not surprisingly, all three sources of personal stress (health, weight and schoolwork), which relate to physical or intellectual traits, were associated with depression symptoms. This supports the conceptual model of depression that posits self-blame or attribution of negative events (viewed as “personal failures”) as triggers of depression [82
]. Weight [83
], acne [86
] and school work [87
] (which adolescents may view as personal failures) have each been reported to be associated with anxiety and depression.
Stress or worry about health was inversely associated with binge-drinking, perhaps reflective that many young adolescents are aware of, and perhaps fear, the detrimental health effects of alcohol use.
It is interesting that several sources of stress relate to a variety of diverse mental health and substance use indicators. Possible mechanistic underpinnings may relate to demands which require adaptation, placed on young adolescents by these stressors. While most adults have developed positive coping styles to manage these demands (i.e., seeking advice, obtaining information, accepting social support, internal reflection), younger adolescents who experience frequent stress for the first time, may not have yet developed appropriate coping skills, and instead use immature or negative coping styles such as withdrawal, avoidance or distraction with substance use [6
]. There is some evidence to suggest that age 15 represents a turning point when youth begin to use more effective coping styles in response to stress [91
Study limitations include its cross-sectional design which limits causal inference. While data on the exposures and outcomes were collected at the same time, the reference time frames varied. Exposure to stress or worry was measured using a lifetime time frame. The outcomes were measured using a “past 7 days” time frame (depression symptoms), a “past 6 months” time frame (conduct disorder symptoms, binge drinking, cigarette smoking, marijuana use) or a “lifetime time frame” (illicit drug use). Different reference time frames may have obscured the estimates. Lifetime illicit drug use may have been under-reported. Our stress measure may have reflected a tendency to have a “stressful or worrisome reaction” or to perceive an event as worrisome or stressful rather than an actual experience of worry or stress. Lastly, given that the sample was one of convenience and the loss to follow-up, our findings may be prone to selection bias.