Stress reactivity and vulnerability to depressed mood in college students

https://doi.org/10.1016/S0191-8869(03)00152-1Get rights and content

Abstract

Two studies reported here found that in response to common, minor stressors, stress reactivity (defined as mean stress per stressor) was a stronger predictor than total stress of depressed mood in traditional and nontraditional college men and women. A prospective study found individual reactivity scores varied over time, but relationships between stress and depressed mood held across four monthly assessments. Stress reactivity also accounted for more incremental variance in depressed mood than total stress after controlling for previous depressed mood. When students in the cross-sectional study were classified into reactivity groups, scores for depressed mood increased steadily for students in the very low through high reactivity groups, as did percentages of students with depressed mood scores that might indicate depression in normal populations. This study also found that stress reactivity was more strongly correlated than total stress with neuroticism and its facets (or traits) of depression, anxiety, and vulnerability to stress in the five-factor model of personality. Taken together, these studies suggest that elevated stress reactivity to minor stressors may indicate diminished ability to cope with everyday challenges and may predict increased vulnerability to depressed mood in a normal population.

Introduction

The assumption that stress accumulates over time to cause or trigger episodes of disorder (Rabkin & Struening, 1976) has guided research in stress and health for over three decades. It was a guiding principle in the pioneering development of the Social Readjustment Rating Scale (Holmes & Rahe, 1967) and subsequent life events inventories, which measure cumulative stress from exposure to minor and major events requiring adaptive responses. Sufficient stress to trigger disorder may accumulate from exposure to one or more very stressful events or a greater number of less stressful events. Many studies have supported this model by demonstrating reliable, but generally modest correlations between cumulative life events stress and negative physical and psychological health outcomes (Johnson & Sarason, 1979).

More recent efforts to assess stress have often employed “hassles” inventories, which measure stress from commonly occurring, minor stressors. There are theoretical and practical reasons for this approach. Many researchers found that major stressors occurred too infrequently to account for most of the stress that people experienced. Furthermore, many researchers study stress and outcomes in convenience samples such as college students, who generally experience few major stressors and little serious disorder, but frequent minor stressors, negative mood states, and deficits in performance. Hassles inventories are also based on the assumption that stress accumulates to cause or trigger negative outcomes. This approach has been successful and many studies found cumulative stress from minor stressors was a stronger predictor of physical and psychological disorder than stress from major life events, even when the same studies employed both measures of stress (DeLongis et al., 1982, Kanner et al., 1981, Monroe, 1983).

Hassles and life events inventories differ primarily in the nature of events listed. Life events inventories include items that cover a broader range of potential stress and required adaptation. Although the cognitive-transactional model of stress (Lazarus & Folkman, 1984) posits that individuals may vary greatly in appraising the stressfulness of any event, life events inventories include some items that most people might rate as slightly stressful and some items that most people might rate as extremely stressful. Individuals could achieve high cumulative stress scores by experiencing one or more very stressful events or a greater number of less stressful events. In contrast, and consistent with their name, hassles inventories do not contain any items that most people would consider especially stressful. They were developed in response to the hypothesis that the cumulative effect of frequent exposure to everyday stressors could trigger disorder. Yet, some hassles inventories allow respondents to rate minor stressors as extremely stressful and some respondents do so. Brantley and Jones (1989) suggested that individuals who rate minor stressors as very stressful may be dispositionally or temporarily more vulnerable to stressors and less able to cope. They might be expected to accumulate stress more quickly and be more likely to experience negative affective states than individuals who rate minor stressors as less stressful. It is also possible, however, that such individuals by virtue of their vulnerabilities to stress, may experience greater negative affect even if they do not accumulate more total stress.

Felsten (2002) tested this hypothesis by measuring mean stress per stressor (stress reactivity) and total stress in response to minor stressors, and found stress reactivity was the stronger predictor of depressed mood in college women. Reactivity was also moderately correlated with neuroticism, a stable dimension of personality associated with ineffective coping, vulnerability to stress, poor adjustment, and negative affect (Costa & McCrae, 1992). Based on this association, the author suggested that greater stress reactivity in response to minor stressors might be a stable marker for vulnerability to stress and a predictor of negative outcomes. The study was limited in that it included only women, did not evaluate associations between stress reactivity and personality in detail, and provided no actual test of the stability of stress reactivity. The present report addressed those limitations by reevaluating archival data, some of which have been previously published (Felsten, 1996a, Felsten, 1996b). These archival data contain measures appropriate for addressing the issues described above, but were not previously used in that manner.

Study 1 was a reanalysis of some of the published data from a cross-sectional study (Felsten, 1996a) used to evaluate expressive and neurotic hostility in the context of the five-factor model of personality (Costa and McCrae, 1992, McCrae and John, 1992). The reanalysis permitted evaluation of how strongly total stress and stress reactivity predicted depressed mood in men and women and also allowed a more detailed evaluation of personality correlates of stress reactivity. Data from the hostility inventory were not used in the reanalysis.

Section snippets

Participants

Participants included 100 women and 63 men who were undergraduate students at a small campus of a public university in the Midwest. Mean age was 23.6 years (S.D.=6.0) for men and 25.9 years (S.D.=9.0) for women. Almost all students were White and many were nontraditional students who worked for pay and/or were married. Participants provided informed consent.

NEO PI-R

The revised NEO Personality Inventory (Costa & McCrae, 1992) is a 240-item self-report questionnaire used to assess the five major domains

Results

Descriptive statistics appear in Table 1. Men and women did not differ on most variables, but women scored higher than men on stress reactivity, t(161)=2.6, P<0.05, neuroticism, t(161)=2.5, P<0.05, and agreeableness, t(161)=3.4, P<0.01. Women scored higher than men on three neuroticism facets (anxiety, depression, and vulnerability to stress) and four agreeableness facets (straightforwardness, altruism, compliance, and modesty). Although most scores for depressed mood were low, 18.4% of

Participants

Forty undergraduate students at a small, private college in the Northeast provided complete data for this reanalysis. Participants included 17 men, 21 women, and two students who did not indicate gender. All students attended school full-time; most were 18–21 years of age, White, and from middle and upper-middle class families. All students provided informed consent.

Materials and procedure

Students completed the Daily Stress Inventory (Brantley & Jones, 1989) and the Beck Depression Inventory (Beck & Steer, 1987) four

Results

Mean scores for the monthly measures of stress and depressed mood in men and women appear in Table 4. Independent samples t-tests found no gender difference for any measure and data were combined for further analyses. Repeated measures analysis of variance (using the Greenhouse-Geisser epsilon to correct for violations of sphericity) found overall differences across assessments in number of stressors F(3, 117)=30.1, P<0.001, total stress F(3, 117)=17.0, P<0.001), and depressed mood F(3,

Discussion

The cross-sectional and prospective studies reported here found that in response to common, minor stressors, stress reactivity was a stronger predictor than total stress of depressed mood in college men and women. This extended an earlier finding that stress reactivity to minor stressors was the better predictor of depressed mood in college women (Felsten, 2002). As in the earlier report, these studies found depressed mood was more strongly correlated with stress reactivity than with total

References (26)

  • P.T Costa et al.

    NEO PI-R professional manual

    (1992)
  • A DeLongis et al.

    Relationship of daily hassles, uplifts, and major life events to health status

    Health Psychology

    (1982)
  • G Felsten

    Five-factor analysis of Buss–Durkee Hostility Inventory neurotic hostility and expressive hostility factorsimplications for health psychology

    Journal of Personality Assessment

    (1996)
  • Cited by (0)

    View full text