Copyright 2001 by the Educational Publishing Foundation 1076-8998/01/$5.00 DOI: 10.1037//1076-8998.6.4.348
Journal of Occupational Health Psvcholo. :()01. Vol. 6. No. 4. .148-360
Chronic and Acute Stressors Among Military Personnel: Do Coping Styles Buffer Their Negative Impact on Health? Aria L. Day and Holly A. Livingstone Saint Mary's University This study examined the moderating impact of positive and negative coping styles on the relationship of acute and chronic job stressors with self-reported health symptoms of 521 military personnel. The number of acute work-related events was associated with a high frequency of self-reported symptoms. Similarly, role ambiguity, overload, and lack of job stimulation were associated with increased symptoms. Only the negative coping styles (i.e., venting of negative emotions and denial/disengagement) were uniquely associated with symptoms. Only 5 of the 13 proposed Stressor X Coping interactions were significant, and they all involved the negative coping styles. That is, instead of alleviating the negative outcomes of work stressors, these coping styles were associated with high strain, regardless of the amount of stressor, and, in some cases, these coping styles exacerbated the negative effect of the stressor on the strain outcomes.
Job-related strain appears to be pervasive throughout organizations, and it has been associated with a number of negative individual and organizational outcomes (Dollard & Winefield, 1998; Jones, Flynn, & Kelloway, 1995; Ulleberg & Rundmo, 1997). This issue has received increased attention as workplaces introduce new and increased stressors caused by mergers, increased workloads, organizational restructuring, workplace innovations, and changes in job responsibilities (Balkin, Tremblay, & Westerman, 2001; Sauter & Murphy, 1995). Moreover, certain workers, such as police officers, firefighters, and military personnel, may be subject to extreme and dangerous situations in the course of their jobs (Anshel, 2000; Anshel, Robertson, & Caputi, 1997). Failure to cope effectively with workplace stressors can lead to negative physical and psychosocial outcomes (Folkman, Lazarus, Gruen, DeLongis, 1986; Lazarus, 1995; Quick, Murphy, Hurrell, & Orman, 1992). How do employees cope with these ongoing events, increased responsibilities, and extreme circumstances? The present study examines the ability of individual coping styles to reduce and buffer the
Aria L. Day and Holly A. Livingstone, Department of Psychology, Saint Mary's University, Halifax, Nova Scotia, Canada. A previous version of this article was presented in April 2000 at the annual meeting of the Society for Industrial and Organizational Psychology, New Orleans, Louisiana. Correspondence concerning this article should be addressed to Aria L. Day, Department of Psychology, Saint Mary's University. Halifax, Nova Scotia B3H 3C3, Canada. Electronic mail may be sent to
[email protected].
negative impact of these acute and chronic work stressors.
Stressors According to Hurrell, Nelson, and Simmons (1998), job stressors are defined as the work-related environmental conditions that have the potential to decrease the health and well-being of workers. These stressors may be classified as being either chronic or acute (Carayon, 1995). Chronic stressors are characterized by "their continuing problematic nature and extended duration" (Gottlieb, 1997, p. 5) and their high likelihood of recurrence (Barling, 1990). That is, they are long-term stressors that people experience on a daily basis (Carayon, 1995). Conversely, acute stressors are "extreme and unusual external [stimuli that are] perceived as threatening" (Anshel, 2000, p. 376). They are of a short duration, have a specific time of onset, and have a low likelihood of recurrence (Barling, 1990). Acute stressors may be similar in nature to traumatic stressors, which are defined as traumatic events or significant catastrophes that have a sudden onset, are temporary in nature, and tend to create negative emotional consequences (Carayon, 1995; Evans & Coman, 1993; Peterson, Prout, & Schwartz, 1991). Strain may result from particularly traumatic or stressful job-related events that involve an unexpected onset. Such events are very specific and identifiable in nature and may result in immediate distress (e.g., Eden, 1990; Newton, 1989). Acute stressors at work can include extreme experiences, such as exposure to dangerous conditions (e.g., violence; An348
STRESS AND COPING
shel, 2000; Anshel et al., 1997; Kleber & van der Velden, 1996), changes in company structure and the social network of an organization, environmental catastrophes (Kleber & van der Velden, 1996), technological changes (Amick & Celentano, 1991; Sauter & Murphy, 1995), and job transfer or loss (Anshel et al., 1997; Frone & McFarlin, 1989). In the past, researchers examining nonwork streams of psychology (e.g., aging, clinical, and health psychology) have tended to focus on acute stressors or on episodic life events and their effects (e.g., Holmes & Rahe, 1967). Only recently have these researchers started to examine chronic stressors (Gottlieb, 1997). Conversely, in the work literature, researchers have tended to neglect acute stressors and have focused primarily on chronic stressors and their effects (Anshel et al., 1997; Eden, 1990; Evans & Coman, 1993; Kleber & van der Velden, 1996). Because theories of work stress have generally assumed that job stressors are chronic (and have emphasized exposure to long-term stressors experienced in the workplace), these existing theories of job stress may not be appropriate for the study of acute job stressors (Kleber & van der Velden, 1996). Few studies have examined both acute and chronic stressors. Avison and Turner (1988) found that chronic nonwork stressors resulted in more psychological distress than did acute nonwork stressors because chronic stressors "represent unresolved, continuing difficulties for the individual" (p. 261). However, Eden (1990) conducted a longitudinal study of 29 university students, computer workers, and faculty and found that an acute job stressor (i.e., closure of the university computer system) resulted in higher levels of psychological and physiological strain than did chronic job stressors. However, the level of chronic job stressors was not measured directly, and the status quo was taken as a measure of chronic stressors. Compas, Connor, Osowiecki, and Welch (1997) suggested that researchers must integrate and compare the research on chronic and acute stress to examine these issues more closely. However, not all individuals will assess these chronic and acute stressors as being stressful. According to Lazarus and Folkman (1984), when confronted with acute and chronic job stressors, an individual will first engage in primary appraisal. If the individual views the situation as being negative (i.e., stressful), she or he will engage in secondary appraisal, in which the individual assesses how she or he is able to cope with the situation. If coping methods are unavailable, ineffective, or impractical, then
349
several psychological, behavioral, and physical strain outcomes can occur (Lazarus & Folkman, 1984). Coping Coping refers to "cognitive and behavioral efforts to master, reduce, or tolerate the internal and/or external demands that are created by a stressful event" (Folkman, 1984, p. 843). Researchers have typically divided coping into two styles: emotion focused and problem focused (e.g., Lazarus & Folkman, 1984). Emotion-focused coping involves regulating the emotional response to the problem and may include avoiding, minimizing, and distancing oneself from the problem (Lazarus & Folkman, 1984). Problemfocused coping is directed at managing or altering the problem that is causing the distress and may include defining the problem, generating alternative solutions, determining the costs and benefits of these solutions, and acting to solve the problem (Lazarus & Folkman, 1984). In a critique of coping measures, Carver, Scheier, and Weintraub (1989) suggested that there may be other forms of coping efforts that have not been addressed by previous researchers. Carver et al. (1989) developed items for their COPE Inventory to measure different aspects of problem-focused coping (e.g., active coping and planning), emotion-focused coping (e.g., positive reinterpretation and growth), and coping through behavioral and mental disengagement (e.g., using drugs and alcohol to cope, giving up, sleeping, or working more to avoid the problem). Some forms of coping have been linked to various positive psychological health outcomes (Kirkaldy, Cooper, & Brown, 1995). Coping that involves social support and cognitive strategies has been associated with lower perceived strain (Decker & Borgen, 1993). Problem-focused coping (i.e., creating a plan of action and focusing on the problem) has been associated with increased job satisfaction (Bhagat, Allie, & Ford, 1995), increased health (Parkes, 1990), and decreased depression, anxiety, and somatization (Greenglass & Burke, 1991). Emotional-focused coping (i.e., positive appraisal and acceptance of the problem) has also been associated with less depression, anxiety, and somatization (Greenglass & Burke, 1991). In their study of 180 police recruits, Violanti (1992) found that both problem-focused coping (i.e., "planful problem solving") and emotion-focused coping (i.e., "distancing oneself from the problem") were associated with decreased psychological distress. However, Violanti also found that other types of emotional-based coping, such as positive reap-
.'50
DAY AND LIVINGSTONE
praisal and accepting responsibility, were not associated with distress. Moreover, Bhagat et al. (1995) found that emotion-focused coping (i.e., positive appraisal and acceptance of the problem) was positively associated with physical and psychological strain. These discrepancies in findings may be because there are many coping scales with different factor structures and scale definitions (Latack & Havlovic, 1992), and no single clear factor structure has emerged across a number of studies (e.g., Carver et al., 1989). Moreover, both positive and negative coping styles exist within the same measures of problemand emotion-focused coping (see, e.g., Carver et al., 1989). For example, although emotion-focused coping may involve a positive reappraisal of the situation, it may also involve venting of negative emotions. Harcourt, Rumsey, and Amber (1999) noted that the inconsistency in coping research warrants further investigation. Despite the discrepancy in findings with problemand emotion-focused coping, some forms of coping are consistently associated with negative outcomes. Mental and behavioral disengagement have been shown to be associated with increased perceived stress (e.g., Griffith, Steptoe, & Cropley, 1999). Begley (1998) found that coping through increased alcohol and drug use was associated with increased somatic complaints and decreased psychological well-being. Similarly, in their study of 2,638 teachers, principals, and directors, Cooper and Kelly (1993) found that respondents who reported that they smoked, drank, and used medication as coping mechanisms tended to perceive greater psychological distress. Coping behaviors that involve denying or avoiding the problem have also been associated with higher self-reported psychological distress (Tyler & Cushway, 1995; Violanti, 1992). Finally, in a nonwork context, Bar-Tal and Spitzer (1994) found that venting of emotions was associated with increased distress.
Stressors, Coping, and Strain In addition to the direct impact of coping on psychosocial outcomes, coping may moderate the relationship between work stressors and strain outcomes. Parkes (1990) suggested that only certain methods of coping would moderate the relationship between stressors and strain. She found that problem-focused coping moderated the relationship between chronic job demands and health. That is, individuals who used problem-focused coping tended to report lower levels of distress, regardless of the amount of job
demands and work support they experienced. However, individuals who did not tend to use problemfocused coping experienced more distress, especially when job demands were high or when work support was low. Bhagat et al. (1995) found that problemfocused coping moderated the relationship between chronic work stressors and life strains, including serious illness, emotional exhaustion, and feelings of depersonalization, whereas emotion-focused coping only moderated the relationship between chronic work stressors and depersonalization. Greenglass and Burke (1991) found that problem-focused coping tended to moderate the relationships of chronic work stressors with anxiety and depression, whereas emotion-focused coping (in terms of accepting daily life hassles) did not moderate the relationship of stress and any psychosocial outcomes. Similarly, coping by withdrawing or ignoring the problem did not moderate the relationship between job demands and health (Parkes, 1990). It is also feasible that coping only moderates the relationship between certain types of stressors and strain. No research has examined the moderating impact of coping on the relationship between acute work stressors and strain. However, in one study of nonwork acute stressors, none of the coping strategies (i.e., problem focused, positive appraisal, emotional discharge, and avoidance) moderated the relationship between acute nonwork stressors and psychological distress (Bar-Tal & Spitzer, 1994).
Overview and Hypotheses Researchers have tended to examine chronic or long-term job stressors and have directed little attention toward the acute stressors experienced by individuals in the workplace. Military personnel face many unique stressors in their work environment, among which are traumatic events related to their service duties. However, there has been little research examining the impact of these acute work stressors on the well-being of military members, and even fewer studies have examined the joint impact of acute and chronic stressors. Because the effects of coping on the stress-strain relationship may be dependent not only on the type of coping measure used (Parkes, 1994) but also on the type of stressor being examined, it is important to include both positive and negative coping methods and both acute and chronic stressors. Because people typically use more than one type of coping simultaneously (Carver & Scheier, 1994; Parkes, 1990), it is most appropriate to examine the joint impact
STRESS AND COPING
of all of the coping styles (i.e., to examine each coping style while controlling for the impact of the other coping styles). Therefore, the three main purposes of the present study were (a) to compare the relationship between the chronic and acute stressors experienced by military personnel and their reported strain (i.e., negative psychological, behavioral, and physiological symptoms); (b) to examine the direct relationship of positive and negative coping styles with health symptoms; and (c) to examine the moderating impact of coping styles on the stressor-strain relationship. On the basis of past research regarding stressors and strain (e.g., Jones et al., 1995), the following hypothesis was made: Hypothesis 1: There will be a direct positive relationship of both acute (i.e., number of negative events encountered at work) and chronic stressors (i.e., role overload, ambiguity, lack of stimulating work, and having responsibility for other people) experienced by military personnel with reported psychological, behavioral, and physiological complaints or symptoms.
Some of the perceived discrepancies in results involving the positive and negative relationships that coping exhibits with strain outcomes may be due to the simultaneous measurement of positive and negative forms of coping. Therefore, in the present study, these two forms were measured separately. Hypothesis 2: Positive coping styles (i.e., problemfocused coping, positive emotion-focused coping, and social support) will be associated with decreased reported psychological, behavioral, and physiological complaints or symptoms, whereas negative coping styles (i.e., venting of negative emotions, denying, and disengaging from problems) will be associated with increased reported psychological, behavioral, and physiological complaints or symptoms.
Finally, there have been inconclusive findings regarding the moderating impact of coping. These discrepancies may be because past research failed to distinguish between negative and positive forms of coping. Moreover, no research has examined coping as a moderator between acute work stressors and strain outcomes. Hypothesis 3: Coping style will moderate the relationship between the stressors and reported psychological and physiological complaints or symptoms. More specifically, at low levels of the stressor, complaints should be low regardless of coping style. At high levels of the stressor, individuals who have negative coping styles (such as denial of problems, disengagement of problems, and venting of negative emotions) will tend to experience more negative outcomes than will individuals who do not have negative coping styles. More-
351
over, at high levels of the stressor, individuals with positive coping styles (such as problem solving, seeking support, and positive interpretation of problems) will report fewer symptoms than will individuals who do not have positive coping styles.
Because of the distinct nature of the acute and chronic stressors and the positive and negative coping styles, we made more specific hypotheses and predicted a total of 13 interactions. Problem-focused coping may be associated with lower strain only in situations that are within the control of the individual. Therefore, we expected that problem-focused coping will moderate only the relationship between strain and the ongoing job-based stressors of ambiguity and overload. Because the acute work events have already occurred and cannot be changed, we predicted that problem-focused coping will not have any impact on their outcomes. Conversely, social support, emotion-focused coping, denial/disengagement, and venting of negative emotions will moderate the relationship between the occurrence of acute work events and strain outcomes. We also predicted that only venting emotions and denial/disengagement from one's problems will moderate the relationship between lack of job stimulation and strain outcomes. If there is a lack of stimulation in an individual's job, denying/disengaging from the problem or venting negative emotions may be his or her sole means of address. However, these coping styles are anticipated to exacerbate the strain symptoms. Similarly, venting of emotions will also moderate the relationship between having responsibility for others at work and strain. Finally, we expected social support to moderate the relationship of ambiguity and overload with strain, because it is feasible that increased information (as a result of the support) may effectively reduce the amount of ambiguity and overload in the job.
Method Sample Questionnaires were mailed to a random sample of 1,000 military personnel who had been stratified based on gender, first official language, command, and rank. There were 620 respondents (454 men and 166 women), for a response rate of 62%. Usable data came from 521 respondents (379 men and 142 women) who had been involved in the military service full time for an average of 15.42 years (SD = 6.81). Only 20 of the respondents identified themselves as being either aboriginal or other visible minority. The mean age of the participants was 36 years (SD = 6.71 years).
352
DAY AND LIVINGSTONE
The questionnaire contained the following scales.1 Control variables. To control for the effects of extraneous variables, we asked respondents to indicate their age, gender, and the number and type of acute nonwork stressors. Respondents indicated whether they had been directly involved in 15 tragic or life-threatening incidents that were not related to their military service (e.g., witnessing abusive violence, being threatened with assault, seeing a friend killed, being physically assaulted, and being seriously injured). Each item was scored as either occurring or not occurring, and the mean number of events (across these 15 events) was calculated (possible scores range from 0 to 15). Thirty-six percent of the sample had experienced at least 1 of these 15 events Acute work events. To assess acute work-related stressors, we asked respondents to indicate whether they had been directly involved in 15 tragic or life-threatening incidents that were related to their military service (e.g., witnessing abusive violence, harming a person, being threatened with death, seeing a colleague killed, being held hostage, and being seriously injured). Events that pertained directly to military service were selected because specific measures of acute stressors (which are relevant to a particular occupation) should be more accurate in the measurement of acute job stressors (Eden, 1990). Each item was scored as either occurring or not occurring, and the mean number of events (across these 15 events) was calculated (possible scores range from 0 to 15). Thirty-two percent of the sample had experienced at least 1 of these 15 events. Role stressors. Four of the subscales from the Occupational Environment Scales (OES; Osipow & Spokane, 1983) were used to assess the kinds of chronic stressors that people experience in their work.2 Ten items assessed their lack of job stimulation and job fit (e.g., "I am bored with my work"; a = .85). Ten items assessed respondents work-role ambiguity ( e.g., "The priorities of my work are clear to me"; a = .74). Nine items were used to assess role overload (e.g., "At work, I am expected to do too many different tasks in too little time"; a = .88),3 and 10 items were used to assess role responsibilities (e.g., "I have on-the-job responsibility for the activities of others"; a = .70). Using a 5-point Likert scale (1 = rarely or never; 5 = most of the time), respondents indicated the degree to which each statement fit their personal circumstances. Coping. Twelve of the subscales from the COPE Scale (Carver et al., 1989) were used to assess how members of the military typically handle various problems and difficult situations that arise during service life. Problem-focused coping consisted of 12 items assessing active coping, planning, and suppression of competing activities (e.g., "I make a plan of action"; a = .86). Seeking social support consisted of 8 items that assessed seeking both emotional and instrumental support (e.g., "I try to get advice from someone about what to do"; a = .84). Positive emotion-focused coping comprised 8 items measuring positive reinterpretation/growth and acceptance (e.g., "I look for something good in what is happening"; a = .73). Disengagement and denial consisted of 16 items assessing denial, behavioral disengagement, mental disengagement, and disengagement through alcohol and drug use (e.g., "I pretend that things haven't really happened" or "I take prescribed drugs in order to think less about my concerns"; a = .79). Finally, venting of negative emotions comprised 4 items that assessed a focus on, and venting of, emotions (e.g., "I get
upset and let my feelings out"; a = .73).4 Respondents indicated on a 4-point Likert scale the extent to which they engaged in each of the various coping behaviors (1 = "/ usually don't do this at all"; 4 = "/ usually do this a lot"). Psychological, behavioral, and physical strain. The 20item Symptoms Checklist (Bartone, Ursano, Wright, & Ingraham, 1989) was used to assess the prevalence of psychological, behavioral, and physiological health symptoms and complaints (e.g., being jumpy/easily startled, increased smoking, and headaches). Members responded to each item based on a 4-point Likert scale ranging from never (1) to very often (4). In the present study, Cronbach's alpha for this scale was high (a = .88).
Results Correlations and Descriptive Statistics Correlations, means, and standard deviations of the study variables are presented in Table 1. Neither gender nor age was related to strain. However, older employees tended to have experienced more acute work events, responsibilities, overload, and more stimulating jobs. They also tended to use more problem-focused coping and less denial/disengagement. Women tended to have experienced fewer acute work events and less ambiguity on the job. Women also tended to seek social support and vent emotions to a greater degree than did men. Reported health symptoms were positively associated with the acute nonwork events, acute work events, and three of the four chronic work stressors. Differences were also noted among the relationships among the coping styles and health symptoms. Only the negative coping styles (i.e., venting of emotions and denial/disengagement) were associated with increased negative health symptoms. Problem-focused coping, positive emotion-focused coping, and seeking support were all unrelated to health symptoms.
Moderated Regression Analysis To assess the moderating impact of coping styles on the relationship between the acute and chronic 1 There were other scales included in the questionnaire that were not examined in the present study. 2 The role boundaries scale was not used because it was not directly relevant to our sample. 3 Although the scale originally consisted of 10 items, 1 item (i.e., "I am competent in what I do") was deleted because it does not assess overload (and it exhibited low item-total correlation). 4 Although Carver et al (1989) included venting of emotions with social support, venting was assessed separately to keep these two conceptually distinct constructs independent from each other.
353
STRESS AND COPING
2
~~ * *
2
!
fN CO
CO
§ * *
o
~ ON
* * * rj
oo" OO
* * m
* O O
OO
r^ Tf
*
p s t^p O4 O O O 1
*_
.
II in
1 o
iQ •*•
^
** ,—, ** **
^
OO
i
№
1
-c *-l
*
° °
' rr r i i ^ ** *** * *_ o —~ P8 r — rr ** ** * * * *
CO ^t OO -O ^
co
— 3 |o'l)|l
> S. 1. "b 1 * w
o.'S l-i oC4>Oa>
l-g* §•'6
, , £ s S a j >
