Perceptual & Motor Skills: Exercise & Sport 2014, 119, 3, 679-697. © Perceptual & Motor Skills 2014

AEROBIC EXERCISE, BALL SPORTS, DANCING, AND WEIGHT LIFTING AS MODERATORS OF THE RELATIONSHIP BETWEEN STRESS AND DEPRESSIVE SYMPTOMS: AN EXPLORATORY CROSS-SECTIONAL STUDY WITH SWISS UNIVERSITY STUDENTS1, 2, 3 MARKUS GERBER, SERGE BRAND, CATHERINE ELLIOT, EDITH HOLSBOER-TRACHSLER, AND UWE PÜHSE University of Basel Summary.—This exploratory study was designed to compare four types of exercise activities in Swiss university students. A sample of 201 medical students (136 women, 65 men; M age = 23.2 yr., SD = 2.4) and 250 exercise and health sciences students (144 women, 106 men; M age = 22.3 yr., SD = 2.2) participated in the study. They completed the Perceived Stress Scale, the Depression Scale, and the Office in Motion Questionnaire. Interaction effects between stress and exercise activities were analysed using hierarchical regression analyses, after controlling for age, sex, and academic discipline. Frequent participation in ball sports and dancing were associated with decreased depressive symptoms among students with elevated perceived stress, whereas no such relationship existed among their peers with lower perceived stress. No stress-moderating effect was found for aerobic exercise. Weight lifting was only associated with lower depressive symptoms among students with low perceived stress. The present findings suggest that, among Swiss university students, certain exercises may have better potential to moderate the relationship between perceived stress and depressive symptoms than others. Future research could analyze whether personalized exercise programs created to satisfy participants' individual needs are more beneficial for stress management.

Stress has become an increasingly important topic in public health debates (Leka, Griffiths, & Cox, 2003). Evidence suggests that stress is associated with an increased risk for both physical and mental disorders (Grant, Compas, Thurm, McMahon, & Gipson, 2004; Uchino, Smith, Holt-Lunstad, Campo, & Reblin, 2007). Moreover, the prevalence of stress and stress-related disorders is high in most industrialized countries (Seco, 2011). The “Stress in America” survey shows that stress has increased during the last decade and weighs heavily on younger Americans.3 High levels of perceived stress are also found among university students (Bayram & Bilgel, 2008; Egan & Moreno, 2011). A recent study showed that perceived Address correspondence to Markus Gerber, Department of Sport, Exercise and Health, University of Basel, Basel, Switzerland or e-mail ([email protected]). 2 All authors declare that they have no competing interest. This study was conducted with funding from the Freiwillige Akademische Gesellschaft Basel (FAG), Switzerland. The FAG did not influence data collection, data entry, data analyses, the interpretation of the data, or the writing and submission of the manuscript. The authors thank Klara Spalek, Daniela Beutler, David Fasler, and Patrick Winiger for data entry. 3 www.stressinamerica.org 1

DOI 10.2466/06.PMS.119c26z4

02-PMS_Gerber_140106.indd 679

ISSN 0031-5125

03/12/14 4:24 PM

680

M. GERBER, ET AL.

life stress is associated with symptoms of burnout among university students (Santen, Holt, Kemp, & Hemphill, 2010). These findings support that symptoms of burnout are associated with negative consequences such as dropping out of medical school (Dyrbye, Thomas, Power, Durning, Moutier, Massie, et al., 2010) or increasing the risk of suicidal ideation (Dyrbye, Thomas, Massie, Power, Eacker, Harper, et al., 2008). The term stress has been affiliated with several unique definitions interspersed throughout the scientific literature. Scholars like Cannon (1929) and Selye (1946) defined stress as a dependent variable or a physiological, psychological, or behavioral response to a stressor. Typically, researchers distinguish between reaction-oriented, stimulus-oriented, and cognitivetransactional stress models (Schwarzer, 2000). Modern stress research is rooted in a reaction-oriented stress model. In reaction-oriented models, stress is conceptualized as an unspecific reaction of a living organism to any form of demand (cp. Forcier, Stroud, Papandonatos, Hitsman, Reiches, Krishnamoorthy, et al., 2006; Hamer, Taylor, & Steptoe, 2006). Alternatively, stress is defined as an independent variable within stimulus-oriented stress models. Under this definition, stress is seen as an external factor that disturbs the internal homeostasis of an organism (Gerber, 2008). Critical life events have captured the attention of some scientists (Holmes & Rahe, 1967), but on the other hand some researchers have honed in on minor yet frequent stressors, which are commonly referred to as daily hassles (Hahn & Smith, 1999). Reaction- and stimulus-oriented models propose that the risk for various chronic diseases increases with elevated stress exposure (Schwarzer, 2000). Nevertheless, these models inadequately explain the individual differences in the perception of and reactivity to stress (Gerber, 2008). In an attempt to overcome this shortcoming, stress has also been defined as a cognitive transaction between the person (internal demands) and the environment (external demands) involving several subjective appraisals (Lazarus & Folkman, 1984). Cognitive-transactional models serve to explain why individuals often perceive the stressfulness of the same event in different ways. According to cognitive-transactional stress theorists, such discrepancies can be attributed to internal factors such as the subjective importance of an event, characteristics of the person, past experiences, and an individual's (perceived) resources to solve problems. Since the late 1970s, researchers have started to examine potential moderators of the stress-illness relationship. For example, Kobasa (1979) discovered that hardiness is a mindset that facilitates successful coping with stress. Nearly a decade later, Cohen and Wills (1985) demonstrated that perceived social support has the potential to prevent stress-related health impairments, whereas Antonovsky (1987) proposed a high sense of coherence to be a stress resilience resource. Meanwhile, a wide array of de-

02-PMS_Gerber_140106.indd 680

03/12/14 4:24 PM

EXERCISE, STRESS, DEPRESSIVE SYMPTOMS

681

mographic factors (e.g., sex, age, race/ethnicity) as well as psychological (e.g., self-schema, attributional style) and social resources (e.g., family and peer environment) have been labeled as protection against stress-related symptoms (for review, see Grant, Compas, Thurm, McMahon, Gipson, Campbell, et al., 2006; Luthar, Sawyer, & Brown, 2006). Scholars underscore that physical exercise also may moderate the relationship between stress and health (Gerber & Pühse, 2009; Stults-Kolehmainen, 2013). A literature review revealed a stress-moderating effect of frequent exercise participation in nearly half of all published articles (Gerber & Pühse, 2009). Disparities in health were of minor magnitude between exercisers and non-exercisers experiencing low stress levels; however, exercisers reported fewer health complaints than non-exercisers when exposed to elevated stress. Upon further analysis, participation in exercise moderated the relationship between stress and health regardless of participant age, stress conceptualization, exercise assessment, and type of health indicator (Gerber & Pühse, 2009; Fuchs & Klaperski, 2012). Nevertheless, the strength of the stress-moderating effects was relatively weak in most studies with levels of explained variance varying from