Associations between emotional reactivity to stress and adolescent substance use: Differences by sex and valence

Abstract

Although stress is often related to substance use, it remains unclear whether substance use is related to individual differences in how adolescents respond to stress. Therefore the present study examined associations between substance use and daily emotional reactivity to stress within a year across adolescence. Adolescents (N = 330; Mage = 16.40, SD = 0.74 at study entry; n = 186 female; n = 138 Latine; n = 101 European American; n = 72 Asian American; n = 19 identifying as another ethnicity including African American and Middle Eastern) completed a longitudinal study, including three assessments between the 10th grade and 3-years post-high school. At each assessment, participants reported frequency of alcohol and cannabis use and the number of substances they had ever used. They also completed 15 daily checklists, in which they reported the number of daily arguments and their daily emotion. Multilevel models suggested that more frequent alcohol and cannabis use were related to attenuated positive emotional reactivity to daily stress (i.e., smaller declines in positive emotion on days when they experienced more arguments) for both male and female adolescents. Associations for negative emotional reactivity to stress varied by sex; more frequent alcohol use and use of more substances in one's lifetime were related to greater anxious emotional reactivity to stress among female adolescents, whereas more frequent alcohol and cannabis use and higher lifetime substance use were related to attenuated depressive emotional reactivity to stress among male adolescents. Taken together, substance use was related to emotional reactivity to daily stress within the same year during adolescence, although associations differed by valence and adolescent sex.

Keywords: adolescence; daily diary; drug use; emotion response; interpersonal stress.

FIGURE 1

Substance Use as a function of Age. All ps < 0.001. Frequency of alcohol and cannabis use over the past year were on a 10-point scale in which 1 represented no use, and lifetime substance use was on a scale of 0–7. CI, confidence interval.

FIGURE 2

Positive emotion as a function of daily stressors and frequency of alcohol use (a) and frequency of cannabis use (b). Associations controlled for age, ethnicity, parents’ education, and previous day’s emotion. Frequency of alcohol use, and frequency of cannabis use are continuous variables, and associations were probed at approximately one standard deviation below the mean, the mean, and one standard deviation above the mean. Simple slope coefficients are presented next to the legend. Please note that it is common for confidence intervals to overlap despite significantly different slopes (Schenker & Gentleman, 2001). Moderation indicates that the associations between daily arguments and emotion significantly differ, not that the predicted values of emotion are necessarily different at all levels of daily arguments. Participants generally showed a significant degree of emotional reactivity (i.e., lower positive emotion on days when they had more arguments than average), and assessments of regions of significance using the Johnson–Neyman technique revealed that this association was weaker for individuals with higher levels of substance use and no longer significant for individuals who had used alcohol 2 days per week and who had used cannabis one day week. CI, confidence interval; SD, standard deviation. *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 3

Anxious emotion as a function of daily stressors and frequency of alcohol use (a) and lifetime substance use (b) in female adolescents. Associations controlled for age, ethnicity, parents’ education, and previous day’s emotion. Frequency of alcohol use and lifetime substance use are continuous variables, and associations were probed at values approximately one SD below the mean, the mean, and one SD above the mean. Simple slopes are presented next to the legend. Please note that it is common for confidence intervals to overlap despite significantly different slopes (Schenker & Gentleman, 2001). Moderation indicates that the associations between daily arguments and emotion significantly differ, not that the predicted values of emotion are necessarily different at all levels of daily arguments. Female participants generally showed a significant degree of emotional reactivity (i.e., higher anxious emotion on days when they had more arguments than average), and assessments of regions of significance using the Johnson–Neyman technique revealed that this association significant at all levels of substance use, although stronger associations emerged for female individuals with higher levels of substance use. CI, confidence interval; SD, standard deviation. *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 4

Depressive emotion as a function of daily stressors and frequency of alcohol use (a), frequency of cannabis use (b), and lifetime substance use (c) in male adolescents. Associations controlled for age, ethnicity, parents’ education, and previous day’s emotion. Frequency of alcohol use, frequency of cannabis use, and lifetime substance use are continuous variables, and associations were probed at values approximately one SD below the mean, the mean, and one SD above the mean. Simple slopes are presented next to the legend. Please note that it is common for confidence intervals to overlap despite significantly different slopes (Schenker & Gentleman, 2001). Moderation indicates that the associations between daily arguments and emotion significantly differ, not that the predicted values of emotion are necessarily different at all levels of daily arguments. Male participants showed a significant degree of emotional reactivity (i.e., higher depressive emotion on days when they had more arguments than average), and assessments of regions of significance using the Johnson–Neyman technique revealed that this association was weaker for male individuals with higher levels of substance use and no longer significant for male individuals who had used alcohol once per week or more, who had used cannabis 2–3 days per month or more, and who had used four or more substances. CI, confidence interval; SD, standard deviation. *p < 0.05; **p < 0.01; ***p < 0.001.