BRIEF REPORT Stressful Life-Events and Fatigue in a Nonclinical Sample Rhonda F. Brown, PhD, and Einar B. Thorsteinsson, PhD Abstract: There is a lack of systematic examination of the relationship between different stressor features and fatigue. Thus, we examined which features of stressors (e.g. frequency, duration, severity, valence, type) were most strongly related to fatigue levels. Eighty-nine participants completed a short fatigue questionnaire and a comprehensive stress interview. High fatigue levels were found to be most strongly related to the number of acute interpersonal stressors and chronic difficulty stressors (6 months duration) and psychological distress, but no other stressor measure. Thus, acute and chronic stressor frequency counts might be the best measures used in future studies assessing the relationship between stressors and fatigue. Key Words: Fatigue, stressful life-events, interpersonal stressors, psychological distress. (J Nerv Ment Dis 2009;197: 707–710) F atigue is defined as a pervasive sense of tiredness or lack of energy that is not related exclusively to exertion. Fatigue severity may range from mild and brief states to severe and enduring clinical states that are characterized by fatigue, pain, and neurocognitive symptoms (Loblay et al., 2002). Fatigue is common in nonclinical populations such as community samples (LaChapelle and Finlayson, 1998; Loge et al., 1998), healthy working adults, (Åkerstedt et al., 2004; Bu ¨ltmann et al., 2002) and university students (Brown and Schutte, 2006; Thorsteinsson and Brown, 2009). For example, Loge et al. (1998) report that in a large community sample, 22% of participants reported substantial fatigue, with 11% fulfilling criteria for chronic fatigue caseness. We assessed fatigue in a nonclinical (i.e. university student) sample in this study. Our earlier research suggests that these samples include participants with nonclinical fatigue (i.e. related to stress/ distress and poor sleep quality) and possible clinical fatigue (i.e. related to stress and physical illness symptoms) (Liffman et al., 2009; Thorsteinsson and Brown, 2009). Thus, such studies may permit an increased focus on early fatigue experiences, which may or may not progress to a later diagnosed fatigue disorder. The literature suggests that there is a degree of overlap in the probable causal antecedents of these different fatigue experiences. Some authors suggest that certain cognitions, psychological states, and behaviors may interact with physical factors to determine how an individual experiences the fatigue; but possible underlying phys- iologic and psychosocial causal pathways leading to fatigue have rarely been advanced (Sharpe and Wilks, 2002) and need to be examined more comprehensively. Increased clarity on this matter may facilitate a better understanding of the complex processes involved in fatigue causation in healthy adults. Stressors (e.g. stressful life-events SLE, work stress) have consistently been identified as cross-sectional or longitudinal pre- dictors of fatigue in nonclinical (Åkerstedt et al., 2002; Åkerstedt et al., 2004; Van der Ploeg and Kleber, 2003) and clinical samples (e.g. chronic fatigue syndrome and multiple sclerosis patients) (Hatcher and House, 2003). Some studies indicate that particular aspects of work (e.g. hectic, high demands, dissatisfaction with organiza- tional support) are related to higher fatigue levels in healthy working adults (Åkerstedt et al., 2004; Van der Ploeg and Kleber, 2003). Stressor frequency is reported to be related to worse fatigue in chronic fatigue syndrome patients (Hatcher and House, 2003) and a failure to recover from acute infectious mononucleosis (a common cause of fatigue) (Buchwald et al., 2000). In addition, catastrophic stressors (e.g. hurricane, terrorist attack) are reportedly related to fatigue, but the results appear contradictory (Heim et al., 2004; Lutgendorf et al., 1995). However, much of the literature has focused on clinical fatigue and so there is little available research to help guide studies on healthy adults. Moreover, no studies have systematically compared which features of these stressors (e.g. duration, frequency, severity, valence, type) are most strongly re- lated to fatigue. Such an evaluation may indicate which stressor measure(s) are best employed in future studies examining the stressor-fatigue relationship. Much of the stress-fatigue literature has employed outdated checklist approaches to assess stressors (e.g. Holmes and Rahe, 1967) or used subjective stressor measures (e.g. perceived stress) that are likely to be biased inasmuch as they tend to incorporate an individual’s assessment of stress and psychological distress (Brown and Harris, 1978). In addition, few approaches currently permit the systematic comparison of different stressor measures other than the Bedford College Life Events and Difficulties Schedule (LEDS). In this study, we used the LEDS to generate semi-structured interview data and blinded objective ratings of stressor exposure, rather than evaluating individuals’ subjective impressions of the stress (Brown and Harris, 1989). A range of stressor measures were compared using the LEDS: stressor frequency, duration (i.e. acute, chronic), severity (i.e. num- ber of highly emotionally-threatening or goal-frustrating events), valence (i.e. positive, negative), and type (e.g. interpersonal, work stressors). Thus, in accordance with the limited available literature it was expected that: total stressor frequency count (i.e. acute events and chronic difficulties) and number of work stressors would be most strongly related to fatigue, in this nonclinical sample. METHOD Participants This study was conducted with full human research ethics committee approval. Participants were recruited via an advertise- ment placed on an online forum for young and mature-aged first- year university students enrolled at an Australian university. Stu- dents received course credit for their participation, but had a large variety of studies to choose from. They were eligible to participate in the study if they were over 18 years of age. They were excluded if they reported prior inpatient treatment of a psychiatric disorder; since recent distressing symptoms were expected to obscure the primary focus of the stress interview. A total of 112 adults were approached regarding participation but only 91 completed the interview and returned the questionnaire School of Behavioural, Cognitive and Social Sciences, University of New England, NSW, Australia. Send reprint requests to Einar B. Thorsteinsson, PhD, School of Behavioural, Cognitive and Social Sciences, University of New England, NSW 2351, Australia. E-mail: ethorste@une.edu.au. Copyright © 2009 by Lippincott Williams & Wilkins ISSN: 0022-3018/09/19709-0707 DOI: 10.1097/NMD.0b013e3181b3af36 The Journal of Nervous and Mental Disease • Volume 197, Number 9, September 2009 www.jonmd.com | 707