Part II. Socioeconomic Disparities in Mental Health ...

6 downloads 0 Views 485KB Size Report
Cannon's (1914) "fight-flight" model and Selye's (1936) refinements of it within the GAS ...... Rand D. Conger and Martha A. Rueter, Iowa State University at Ames ...... often associated with socially devalued roles and statuses (Hughes, 1945).

Citation: Ganster, Daniel C. (2002). The stressful workplace: Mental and physical health and the problem of prevention. In A. Maney and J Ramos (Eds.), Socioeconomic conditions, stress, and mental disorders: Toward a new synthesis of research and public policy. Washington, D.C.: Mental Health Statistical Improvement Program, National Institute of Mental Health. Retrieved on December 9, 2014 from http://archive.org/web/ https://web.archive.org/web/20021014140435/http:// www.mhsip.org/nimhdoc/socioeconmh_home.htm.

Part II.

Socioeconomic Disparities in Mental Health and Mental Disorder Section B. Stressful Economic Contexts Linked to Job Holding, Job Loss and Job Seeking In a Globalizing Society

Chapter 5. The Stressful Workplace, Mental and Physical Health, and the Problem of Prevention Daniel C. Ganster, University of Arkansas For many people, work occupies more time and attention than any other sphere of their life. In many ways, work is central to an individual's self-concept, and success or failure in the workplace has profound effects on material well being and social status. For this reason, scholars have examined conditions in the workplace as well as the nature of work itself to find explanations for mental and physical well being. Much of this research has been under the general rubric of work stress, and work stress literature forms the basis of this article. My objectives are to (a) summarize recent studies that have attempted to discover important work experiences, or job demands, that play an etiologic role in mental and physical health, (b) discuss the impact of demographic changes in the workforce, (c) review the success of interventions that address these demands, and (d) enumerate the prospects for conducting prevention research as well as the important constraints that investigators in this area face. Throughout this paper, I will focus on variables associated with conditions of work and the work place itself, and thus my discussion concerns persons who have jobs. I will not address the related concerns of unemployment and job loss except when they create stresses, such as job insecurity and role overload, for those still employed. Although I will not discuss the effects of unemployment itself, continuing trends in corporate downsizing and restructuring have potentially great effects on jobholders and thus will form a part of this review.

2

Theoretical Rationale for Studying Work Conditions A logical starting point is a discussion of the underlying theoretical rationale for assuming that demands in the workplace might be important in determining mental and physical health. While many theoretical perspectives would point the investigator to the workplace to seek out conditions that could affect well being, this research has been guided by the "stressful job demands" perspective, whose roots can be found in Selye's General Adaptation Syndrome (GAS; Selye, 1976). This perspective assumes that various environmental demands on individuals require them to adapt, and such adaptation takes energy. Demands that exceed a person’s ability to adapt, because of either severity or duration, lead to a breakdown of the mental and physical system. The GAS perspective provides a general underlying rationale, but it does not give a specific road map of particular demands that are important and generalizable across different jobs and organizational settings. The researcher needs a content theory that specifies the nature of the demands that are likely to matter most. The most influential theory in recent years is the job decision latitude, or demands-control, model (Karasek, 1979). In its basic form, the model specifies two broad constructs that can vary independently in the work environment. Job demands are defined as psychological stressors such as requirements for working fast and hard, having a great deal to do, not having enough time, and having conflicting demands. These are psychological demands and not physical ones. A fast and hectic work pace may impose physical requirements that lead to fatigue, but the psychological effects of this workload (e.g., the anxiety associated with the need to maintain the work pace and the associated consequences of failing to complete the work) cause the stress-related outcomes predicted by the model. Job decision latitude comprises two components: the worker's authority to make decisions on the job (decision authority), and the variety of skills used by the worker on the job (skill discretion). Operationally these two components are combined into one measure of decision latitude or control. The first hypothesis of the model is that "strain," which is a stressful condition that leads to mental and physical health problems, occurs when jobs are simultaneously high in demand and low in control. High demands produce a state of arousal in the worker that would normally be reflected in such responses as increased heart rate or adrenaline excretion. When there is a constraint on the responses of the worker, as would occur under conditions of low control, the

3 arousal cannot be appropriately channeled into coping response and thus produces an even great physiological reaction for a longer time. The second hypothesis is that positive outcomes (motivation, learning, and healthful regeneration) occur when an individual occupies an "active" job–one that has both a high level of psychological demands and a high level of control. The demands-control model almost completely dominates the occupational epidemiology literature concerned with work stress (Kristensen, 1989) and it has formed the basis for numerous studies in the organizational literature. Evidence for the buffering hypothesis has been mixed, but overall, control appears to be an important factor regarding employee well being (Ganster, 1989). The job decision latitude model evolved from the activation theory work done by Frankenhaeuser and her colleagues in Sweden (Frankenhaeuser, 1979; Frankenhaeuser & Gardell, 1976). Based on the activation theory proposition that performance, mental efficiency, and well being are maximized at some intermediate level of physiological activation, their studies focus on working conditions that are likely to produce either underload or overload. The primary independent variables of interest in their lab and field studies include work pace, attentional demands, control over pacing and other working conditions, and underutilization of skills. Dienstbier (1989) advanced this theoretical perspective and proposed a theory of physiological toughening, which has formed the basis for theories about the effects of long-term exposures to uncontrollable psychosocial demands in the workplace (Schaubroeck & Ganster, 1993). Dienstbier (1989) starts by articulating a more complex picture of physiological arousal and how it affects well being. Investigators often regard arousal as being unidimensional. Cannon's (1914) "fight-flight" model and Selye's (1936) refinements of it within the GAS suggest a single form of nonspecific response to stressors. Mason (1968), however, emphasized the difference between adrenal-cortical and adrenal-medullary arousal systems in terms of what they indicate about the nature of the stress response. •

The adrenal-cortical system is activated when the pituitary gland releases

adrenocorticotropin into the blood, which in turn stimulates the adrenal cortex to release glucocorticoids such as cortisol •

The adrenal-medullary arousal system refers to a pathway originating with

the hypothalamus and ending in the stimulation of the adrenal medulla, which releases the peripheral catecholamines (PCs) adrenaline and noradrenaline.

4

A healthy individual experiences low baseline levels of arousal. When faced with a challenging demand, there is a high PC arousal response (change from baseline) and a relatively small adrenal-cortical arousal response. At the termination of the demand, both types of arousal return relatively quickly to baseline levels. (Figure 1) Thus, high adrenal-medullary activity suggests a high level of active coping with a challenging demand. This activity does not presage later health problems or even subjective distress. In fact, adrenal-medullary arousal is positively associated with various facets of immune system functioning (Anisman & LaPierre, 1982). For example, natural killer cell activity has been enhanced by infusion of noradrenaline in rats (Solomon, Kay, & Morley, 1986). As described by Dienstbier (1989), PCs are major components of the sympathetic nervous system. Arousal of these stress hormones is often critical for responding adaptively to environmental demands. Cardiovascular arousal (e.g., heart rate, blood pressure responses) tends to correspond with PC arousal, as do other peripheral sympathetic nervous system responses (e.g., skin temperature response). On the other hand, cortisol elevations are associated with passive coping (e.g., avoidance) when an individual has difficulty responding to demands. Cortisol responses are associated with fear, anxiety, and depression (Dienstbier, 1989), and Antoni (1987) suggests that chronic cortisol elevation can lead to suppression of immunological functioning and to mental health problems such as depression. Drawing on a wide body of human and animal research, Dienstbier (1989) described the role of conditioning by chronic and intermittent stress exposures as a precursor to adrenal-cortical and adrenal-medullary responses to challenge. Intermittent, controllable exposures to challenging demands (e.g., aerobic exercise, cold temperature) promote adrenal regeneration (i.e., "toughening") of the individual. Although the complex micromediational mechanisms associated with this phenomenon are beyond the scope of this paper, this toughening presentation describes a regenerative process that is represented by (a) lower resting base rates on both the adrenalcorticol and adrenal-medullary systems, (b) more adrenal-medullary (e.g., cardiovascular and PC) responsivity to challenge, (c) less adrenal-corticol responsivity to challenge, and (d) rapid recovery to baseline levels of both systems following removal of the challenge. This pattern

5 corresponds to that illustrated in Figure 1.

6 The PC arousal components of this process, as they occur in humans, were illustrated in three studies of aerobic trainees reported by Dienstbier, LaGuardia, Barnes, Tharp, and Schmidt (1987), who found that aerobic conditioning strengthened the PC arousal response to mentally challenging demands. That is, intermittent exposure to controllable stressors (aerobic exercise) led to a toughening process. Dienstbier's (1989) theory of physiological toughening poses intriguing implications for how psychological demands in the workplace affect the health and effectiveness of workers. Whereas Dienstbier concentrated mostly on the toughening effects of intermittent and controllable demands, this theory could also explain the effects of exposure to demands that were chronic and uncontrollable, precisely the kinds of stressors highlighted by the job decision latitude model. An "untoughening" process (Ganster, 1991) was thus proposed in which chronic exposure to uncontrollable demands leads to a conditioning of sympathetic nervous system responses to demands in general. Individuals with such exposure develop a reduced capacity to cope physiologically with subsequent challenging demands (Figure 2). Relative to a healthy individual, a chronically stressed person shows higher resting base rates of both arousal systems. At the onset of a demand a chronically stressed individual shows a small PC arousal response and a large adrenal-cortical response. At the termination of the demand, both arousal systems show a poor rate of recovery to baseline levels. This reduced capacity has several important implications. First, degeneration in responsivity can impede an individual's ability to perform challenging tasks effectively and can lead to an increase in workplace hazards. Second, this diminished capacity reduces the individual's ability to cope actively with stressful demands. The inability to cope effectively with work stressors is theorized to determine whether the exposures result in health decrements over time. In sum, these processes suggest an interactionist model in which chronic and uncontrollable work environment demands not only lead to poor health, but also diminish adaptive physiological responses. Thus, over time, work stressors become more harmful to an individual as the capacity

7 to respond to them is reduced.

Consider the case of an air traffic controller. A crisis involving multiple aircraft on a collision course presents a stressful challenge that requires a rapid and strong physiological response from the adrenal-medullary system. This acute stressful demand differs from the chronic demands that the controller might experience from changing work shifts, infrequent breaks, and continually overloading demands. Individuals chronically stressed in this way are predicted to become untoughened, losing their ability to muster the strong adrenal-medullary response needed to

8 effectively cope with the acute challenges that occasionally arise. Maladaptive conditioning depletes the individual's coping capacity and makes stressful demands even more taxing and damaging. There is encouraging empirical support for the underlying toughening model proposed by Dienstbier (1989). The model’s ability to explain the effects of work stressors, however, is only beginning to be evaluated (Schaubroeck & Ganster, 1993).

Empirical Support for the Health-Impairing Effects of Work Demands

The empirical literature testing the links between work demands and mental and physical health is voluminous and is reviewed and compiled on a regular basis (see, for example, Ganster & Schaubroeck, 1991). I will only make a summary review of this vast array of empirical work in this paper and provide some representative illustrations. Controversies in this literature have been concerned less with theoretical debate than with methodological issues that make the interpretation of reported findings difficult. Kasl (1978, 1986) has been a convincing critic of this literature over the years, articulating the methodological criteria that must be satisfied to reach conclusions about whether occupational experiences (e.g., high levels of workload) are causal agents in mental or physical disorders. Ganster & Schaubroeck (1991) applied these standards in a critical review of the occupational stress research literature and reached several general conclusions. First, there are occupational differences in mortality and morbidity that are not easily explained by other risk factors. However, it is difficult to totally disentangle the causal effects of occupational characteristics, per se, from such factors as membership in a social class and the health-related behaviors that accompany such membership. Second, evidence from both naturalistic observations involving repeated within-subject comparisons and a few field experiments suggests that certain types of demands produce elevated levels of physiological response and emotional distress. Such demands include responsibility for tasks whose successful performance has significant consequences for the job incumbent and others, high levels of mental demands that require short-term memory and vigilance, and complex interactions with other people. Third, personal control in the workplace seems both to have a salutary main effect on well-being and, as predicted by the job decision latitude model, to help buffer the deleterious

9 effects of other demands. Finally, although the buffering effects of social support are still not clear, emotionally and instrumentally supportive relationships at work, especially with one's supervisor, seem to have a significant positive impact on well being. Having made these broad conclusions about the evidence, I will illustrate the research strategies that my colleagues and I have used, with a discussion of their strengths and weaknesses. One way to discover health-impairing effects of job experiences is to examine occupational differences in health outcomes. A number of large-scale studies have found impressive differences among occupations in morbidity and mortality (see Ganster, Duffy, & Hurrell, 1995, and Ganster & Schaubroeck, 1991, for reviews). Beyond occupational differences, investigators have attempted to link with exposures to specific occupational characteristics health outcomes. For example, Ganster et al. (1995) derived measures of occupational psychosocial demands and control from job analysis data and related them to measures of mental and physical health with a sample of 2,779 workers from 49 different occupations. Workers in occupations characterized by high levels of complexity and high levels of control experienced fewer symptoms of ill health and took fewer sick days. These relationships persisted even after controlling for gender, age, and exposure to nonpsychosocial conditions such as physical hazards. Even though measures of health outcomes were obtained from individual workers, the complexity and control measures were derived from occupational data. Thus, individual variation in exposure to job conditions within occupations, although probably considerable, was ignored and likely attenuated the findings. A second strategy is to achieve a finer grained analysis of job demands by focusing at a level of analysis below the occupation itself. The most frequent level is that of the individual job incumbent, and researchers almost always measure job demands at this level by obtaining the self-reports of the workers. An obvious advantage of this approach is that the investigator does not ignore the large differences in exposures across individuals working in the same occupational category. A much discussed disadvantage, however, is that it becomes very difficult to separate the characteristics of the work from those of the workers themselves. Problems stemming from response consistency effects, attribution effects, and even certain individual traits such as negative affectivity (Schaubroeck, Ganster, & Fox, 1992) make relationships observed in such studies ambiguous concerning the causal effects of job demands (see Ganster & Schaubroeck, 1991, for a review). My colleagues and I have recently attempted to combine the advantages of

10 working at the individual or job level of analysis while avoiding the problems associated with measuring job demands with just the reports of job incumbents. A recent study illustrates this strategy. Fox, Dwyer, and Ganster (1993) studied the relationships between job demands and employee control on the one hand and a variety of mental and physiological indicators of well being on the other with a group of nurses in a private hospital. Studying individuals in just one occupation has both advantages and disadvantages. As noted earlier, major limitations of occupational-level studies are the coarse nature of job demand assessments and the potential confounding of occupational differences in demands with differences in socioeconomic status. One can attempt to statistically control for the latter confounds (Ganster et al., 1995), but limiting the investigation to one occupation in which all respondents share virtually the same socioeconomic status provides even better control. The challenge with sampling individuals from just one occupation, however, is to obtain enough variance in job demands to allow meaningful tests of relationships. The nursing occupation represents an attractive setting in this regard because even though nurses working in one facility all share similar socioeconomic status, they face large disparities in the types of demands with which they must contend. Fox et al. (1993), asked 136 registered nurses to report on their job demands, such as work overload and conflict, as well as the personal control they had in performing their job duties. In addition, we obtained objective indicators of these demands by measuring patient loads, patient contact hours as a percentage of total work time, and the number of patient deaths each nurse witnessed. As expected, we found large differences in these demands across the different nurse specialties. Our outcome variables consisted of subjective well being measures as well as measures of blood pressure and cortisol levels obtained both during work shifts and at home during nonwork hours. Our intent was to test the job decision latitude model, so we examined the interactions between demands and control as well as their main effects. Several findings from this study are worth noting. First, the objective measures of job demands were significantly related to blood pressure and cortisol levels, but their effects were not buffered by employee control. Subjective appraisals of demands, however, showed few deleterious effects of their own but significantly interacted with employee control. The physiological outcomes were best explained by these interactions between subjective work demands and employee control beliefs. Greater workloads significantly lowered the job

11 satisfaction and raised the blood pressure and cortisol level of nurses with low levels of control, but showed no relationship with these outcomes for nurses with high personal control. Finally, interactions between demands and control explained physiological reactions recorded not only during the work shift but also at home, indicating that there are significant carryover effects that persist even when exposure to these demands ceases. The carryover effects found in the Fox et al. (1993) study are a reminder of the harmful effects of job demands: the immediate strains on the mental and physiological systems last longer than the exposure itself. (Most studies do not provide any direct assessment of this process.) Our study of nurses demonstrated that there were consistent after-work physiological elevations in blood pressure and cortisol that were related to job demands and control. However, the toughening theory of Dienstbier (1989) discussed earlier and the untoughening extension of it predict that chronic exposures to uncontrollable demands alter how people respond to subsequent challenges. The nursing data suggest that chronically stressed individuals maintain higher average levels of physiological arousal, as the model predicts, but they do not tell us whether such individuals also have a reduced capacity to muster these adaptive responses when they are needed to cope with an acute stressor. To test this conditioning hypothesis, the investigator needs to relate chronic job exposures to the individual's arousal response to a specific challenging demand in a controlled setting. A colleague and I attempted to directly test this conditioning, or untoughening effect, by sampling 568 workers from a variety of occupations who varied in their known levels of certain job demands (Schaubroeck & Ganster, 1993). From this sample, we selected only the 390 individuals who had been on the same job for at least two years. We arbitrarily chose this time period as one that would represent a fairly long-term exposure to the demands of a single occupation, a period presumably long enough for the hypothesized degenerative conditioning process to develop. We focused on demands concerning complex cognitive functions and demands relating to frequent interactions with demanding people. Previous literature indicates that such demands are experienced as highly stressful and uncontrollable. We tested our untoughening model in two ways. First, the model predicts that chronically stressed individuals will have consistently higher levels of arousal during the workday elicited by their high job demands. This elevated state of arousal, maintained for a long period, produces a depletion effect that leads to a diminished capacity to respond to acute challenges. We thus sampled blood

12 pressure at rest, but in the middle of the workshift. As expected, we found that workers chronically exposed to high job demands also showed higher resting levels of blood pressure. Second, the model predicts that chronically stressed workers will have less vigorous adrenalmedullary and cardiovascular arousal reactions to a challenging demand. We attempted to test this hypothesis by measuring the PCs, (adrenaline and noradrenaline) before and in the middle of a workshift. By examining the change in catecholamine levels during the workday, we found that chronically stressed workers demonstrated reduced PC responsivity when they encountered the demands of their job. Thus, these data lend some support to our untoughening theory. However, this test has a major limitation because the catecholamine responsivity was not assessed under the same conditions for each subject. In fact, samples were taken while subjects worked on their various jobs. Therefore, differences in responsivity might be a function of a maladaptive conditioning process, as we theorize, or might be caused by the naturally occurring differences in demands that the individuals faced during their workday. To conduct a more controlled test of this hypothesis, we tried to combine the advantages of field and lab research approaches. It was important that we assess each subject's responsivity to the same novel challenge under the same conditions. To accomplish this, we brought a motor home equipped as a mobile physiological testing lab to each work site. Each worker was brought to this lab and was exposed to two different challenge tasks. First, we administered the Stanford Research Institute version of the Structured Interview for the Type A Behavior Pattern (Dembroski, MacDougall, & Lushene, 1979). This protocol is basically a stress interview that exposes the respondent to persistent and stressful interactions with another person and thus represents an interpersonal challenge task. Next we administered the Stroop Color Work Conflict Task (Stroop, 1935) as a source of perceptual and cognitive challenge. Throughout these tasks, we took multiple readings of heart rate, blood pressure, and skin temperature to calculate each individual's responsivity to the demands and his or her speed of recovery when the demands were terminated. In this way, we were able to test whether chronically stressed workers would indeed show less vigorous sympathetic arousal to a novel challenging task and a slower recovery to baseline. We observed just these effects, after controlling for such factors as gender, education, age, body mass, Type A, and caffeine and nicotine consumption. The more that individuals had been exposed to chronic uncontrollable demands in their occupation, the less vigorous were their cardiovascular arousal and sympathetic nervous system responses to each of

13 the challenge tasks. Likewise, even though their adaptive arousal responses were lower and resting baseline levels higher, chronically stressed workers took more time to recover to baseline levels. In summary, these data strongly suggest that chronic exposures to uncontrollable demands can not only produce chronically elevated levels of arousal, which itself can cause damage, but can also impair an individual's ability to successfully cope with acute psychological demands. The overall empirical evidence that certain job demands can lead to mental and physical health problems is compelling, despite the limitations inherent in most individual studies. Among the most important of these demands are cognitively and perceptually overloading tasks, high intensity interactions with other people, and lack of control. On the positive side, high control and social support in the workplace (Ganster, Fusilier, & Mayes, 1986) appear to have positive effects that at times might even render other demands less injurious.

Implications of Recent Trends in the Workforce and in Organizations

Many observers of the American scene have identified significant changes in the demands on organizations to respond to global competitive pressures and the change in corporate cultures that this has wrought. Similarly, there are many predictions about how technology, especially the computer, will permanently alter the nature of work. Finally, many scholars have offered opinions about how the changing demographics of the workforce itself will change the nature of how work is organized, scheduled, and compensated. Currently the popular movements in management involve total quality management, restructuring, rightsizing, and reengineering-- all of which have implications for employee well being. I believe it was Mark Twain who noted that making predictions was risky–especially about the future. The world of work changes rapidly, making forecasts in this realm especially risky, but it seems that there are at least two trends that are highly relevant. First, one cannot dispute the changed demographics of the workforce, especially the prevalence of single working parents and families in which both parents work. In 1980, both parents worked in slightly more than half of families. In 1992, both parents worked in 70 percent of families with young children (Hayghe, 1994). These significant structural and functional changes in American families have not been accompanied by equally dramatic shifts in corporate

14 policies. The large number of dual career couples and single parents increases the number of workers who are stressed by the often-competing demands of work and home. Interest thus emerged in the 1980s in the stress of what is usually called work-family conflict (see Frone, Russell, & Cooper, 1992; Greenhaus & Parasuraman, 1987). There is growing evidence that the strain of balancing work and family responsibilities can lead to job dissatisfaction (Bachrach, Bamberger, & Conley, 1991), depression (Burden & Googins, 1987; Frone et al., 1992), and coronary heart disease (Haynes, Eaker, & Feinleib, 1984). Although these stressors may not reflect job demands, per se, the focus of this article, they do have implications for organizational practices that could either exacerbate or alleviate these strains (Thomas & Ganster, 1995). The other trend that shows little sign of abating is the change in corporate cultures and practices that affect job security. The 1980s saw the explosion of corporate mergers and acquisitions, often hostile and financed with high-risk bond issues. Many of these mergers and acquisitions were intended to capitalize on short-term gains in security prices rather than to improve operating efficiency. Consequently, many of the combinations were unstable and led to large layoffs. Such practices seem to have waned, but other forces have had the same effects on job security. In particular, global competition and its pressure on efficiency, especially with respect to labor productivity, still exert pressure for downsizing. Thus, job insecurity is likely to remain a major issue. The threat of layoff is a significant stressor and layoffs affect the attitudes and behaviors of those left behind-- the survivors (Brockner, et al., 1994). Despite the significance of layoffs, however, the prevention implications are not clear. Unlike for other stressors, such as certain job demands and work-family conflict, the literature does not suggest a prevention strategy. One can suggest that organizations adopt lifetime employment policies. But even the Japanese, for whom lifetime employment was a deeply held value in large corporations, have had difficulty maintaining this practice. Effective prevention is more likely to come from social policy decisions that emphasize preparing the workforce to cope with multiple job changes.

The Effectiveness of Prevention Approaches in the Workplace

Prevention attempts in the workplace can be categorized into those that are designed to help workers better cope with their job demands and those whose intent is to reduce or eliminate the

15 source of the stress itself. I will just briefly summarize the work that has been done in the first category and focus more attention on the programs that target the workplace conditions that give rise to health problems.

Approaches That Attempt to Change Individuals Many studies have evaluated the impact of stress management training programs conducted in the workplace. The content of these programs varies greatly but usually includes some attention to cognitive reappraisal, relaxation, exercise, and biofeedback (Murphy, 1988). Employee assistance programs also often provide counseling or referrals to employees showing stress-related symptoms. Evaluation studies suggest several general conclusions. First, welldesigned programs of sufficient duration can produce relatively short-term improvements in symptoms of psychological distress, including anxiety and depression. They also have demonstrated some effectiveness in reducing muscle tension and stress hormone levels, but there has been too little long-term follow-up to conclude that the successful interventions have a permanent impact. Most evidence suggests the need for periodic booster sessions, but these are rarely undertaken. Second, it is not clear that such programs are especially useful for preventing morbidity related to long-term exposures to uncontrollable job demands. My colleagues and I have argued for some time that perhaps the best use of such programs is to supplement organizational change efforts (Ganster, Mayes, Sime, & Tharp, 1982), but that the programs should not be considered the primary strategy for reducing mental and physical health problems in the workplace. Physical fitness levels can buffer some of the effects of work stressors. For example, Csanadi (1981) reported that psychological disorders were correlated with self-reports of work stressors, but only among subjects (law enforcement officers) who did not regularly engage in aerobic fitness activity. Dienstbier et al. (1987) also suggest that aerobic exercise might be effective in physiologically toughening individuals and making them more adaptive to acute mental demands. In view of the encouraging evidence supporting this physiological toughening theory, exercise intervention programs deserve more evaluation research. Especially needed is investigation of whether aerobically fit workers respond differently to work demands.

16

Approaches That Attempt to Change Work Conditions The research literature on work stress suggests several approaches for prevention, but evaluation research on work site change programs is still sparse. Especially rare are randomized experiments that test the effects of interventions targeting specific sources of distress and that employ meaningful outcome measures. The evidence would support experimentation with the following types of interventions: (a) job redesign interventions that increase worker control over important work processes; (b) job redesign interventions that reduce repetitiveness and sustained vigilance requirements; (c) training programs that help supervisors and work teams develop meaningful social support; and (d) organizational policies and practices that reduce work-family conflict, such as flexible scheduling and dependent care. In recent years, my colleagues and I have directed effort at evaluating some of these approaches. In one experiment, Schaubroeck, Ganster, Sime, and Dittman (1993) evaluated a two-stage intervention program that attempted to reduce high levels of role ambiguity. The target population was a group of middle managers and their subordinates in the business services division of a large university. The responsibilities of this division consisted of all purchasing and inventory management for the university, including food services. A new director had recently been appointed and the division was undergoing what turned out to be the penultimate reorganization in a series that started several years earlier. The last reorganization occurred soon after the study was completed. The reorganizations had created considerable turbulence in individual roles and relationships among the various departments of the division. Reported stress levels were high, and many employees reported psychosomatic disorders and frequent sick days, with most being attributed to conditions at work. In a two-and-one-half-year study, we conducted two surveys, had countless meetings with staff and managers, and finally implemented a role clarification intervention that began with charting the responsibilities of the managers and their departments and ended with individual role negotiation sessions between each manager and his or her subordinates. We randomly assigned groups of subordinates to treatment and waitcontrol conditions and evaluated employee stress symptoms several months after implementation. Although the intervention significantly reduced employee perceptions of role ambiguity, it seemed to have little impact on employee health symptoms or sick days.

17 In another study, Thomas and Ganster (1995) evaluated the effects of various familysupportive organizational practices on the health of nurses and other health care professionals. In this study, we surveyed 398 individuals who had children ages 16 years or younger at home and who worked in one of 45 different acute care facilities in Nebraska. From the individuals, we collected data on work-family conflict and stress, job satisfaction, depression, somatic health complaints, and blood cholesterol, and diastolic blood pressure. From organizational officials, we collected information on family-supportive practices and policies that each facility supported. These reports were cross- validated with the reports of the individual respondents. Familysupportive practices included five information and referral services, eight different dependent care services or resources, flexible work scheduling, and the presence of socially supportive supervisors. We found that such supportive practices, especially flexible scheduling and socially supportive supervisors, were associated with lower levels of work-family conflict and health problems, including lower blood cholesterol. These findings suggest that organizations can take specific steps to increase employees' control over family responsibilities. This control, in turn, appears to help employees better manage the often conflicting demands of work and family life. Again, this study did not involve the experimental manipulation of these practices but relied on naturally occurring variation across a sample of 45 organizations. Although our findings cannot support causal interpretations about the benefit of these practices, they certainly provide strong encouragement for organizations to consider their implementation. Particularly encouraging was that the family-supportive practices that were most strongly correlated with health outcomes (flexible scheduling and socially supportive supervisors) are fairly simple and inexpensive approaches. Although the stress research literature suggests that the approaches discussed above should help reduce health-impairing work conditions, there have been few successful efforts to implement and evaluate them. The lack of such experimentation is not attributed to lack of theory or basic research, but is more a function of difficult obstacles facing the researcher.

18

Obstacles to Intervention Research Although the evidence rather convincingly demonstrates that certain job demands are associated with a range of mental and physical health problems, the lack of experimental research generates two main inferential limitations. First, it is difficult to assert that job demands that consistently correlate with health status are in fact the important causal agents. Only controlled experimentation would allow that high level of confidence. Second, we have almost no evidence that efforts to change job demands will actually produce improvements in health. One can reasonably argue, then, that the first research priority should be to evaluate interventions that may reduce or eliminate exposures to job demands that the existing research suggests are toxic agents. Although this is not the first such call for experimentation (see Ganster et al., 1982; Ganster & Schaubroeck, 1991), researchers seem to have been reluctant to respond. Seven significant challenges may be discouraging hopeful intervention researcher. Identifying an At-Risk Working Population For intervention research to be successful, the working population being studied must show a high level of stress-related symptoms that are plausibly caused by working conditions. It helps if the population has a known high exposure to certain job demands that the literature indicates are linked to such health complaints. This discussion centers on intervention research, which targets problematic settings and seeks to make them more healthful, rather than on prevention research, which attempts to prevent the development of mental and physical health problems in the first place. For interventions to demonstrate a significant and practically meaningful effect in a time period that is short enough to maintain the integrity of an experimental design, one must start with a population that already shows severe problems. Finding such populations and establishing collaborative agreements with their employers is difficult. Establishing Collaborative Agreements Many, perhaps most, organizations in both the private and public sectors are reluctant to allow any sort of experimentation, especially when it concerns health and safety. There are probably several reasons for this hesitancy. For example, managers are fearful that interventions to render the workplace more healthful will involve high implementation costs and potential indirect costs stemming from reduced productivity. Managers are generally charged with

19 maintaining productivity and reducing costs. Any intervention that does not promise help in either area is thus undesirable. In fact, organizations spend enormous sums redesigning work processes, often with the help of expensive consultants. The issue is not just cost, then, but rather cost and benefit. Managers must be convinced that there will be measurable benefits from either improved productivity or reduced costs. Promising better health and well being for their employees does not seem to be an attractive enough inducement for many managers. A second hindrance to intervention research, and stress research generally, is that investigators are studying factors that presumably cause mental and physical health problems. More than one potential corporate collaborator has expressed fear about legal liability if researchers demonstrate that job conditions are making their employees sick. I do not dismiss such fears as totally unfounded, but I do think one can make some effective counterarguments. For one, the intervention is likely to be targeting practices that are common in the industry and that are generally seen as legal and acceptable. Thus, employers can argue that they were not engaging in practices that they should have known were dangerous. Moreover, a strong argument in favor of the company is that employers were making a sincere effort to improve the health of their employees. Implementing an intervention program should be convincing evidence that they were making a good faith effort to create a better workplace. Despite these arguments, however, getting past the organization's legal department often proves tough. Conducting Pilot Research In most intervention studies, the investigator must conduct a pilot study that produces the data with which to diagnose stress-related problems. This means collecting data about exposure to stressful job demands as well as employee health outcomes. These data must then show that the job demand exposures are related to the health outcomes. If the stressor measures are uncorrelated with the dependent variables of interest, it is difficult to make a convincing case to change the stressors through an intervention. A pilot study also provides the needed pretest measures both for the job demands that the intervention will attempt to affect and for the outcome variables. The difficulty with conducting such pilot studies is that they are often rather large and require funding support to be done successfully. Thus, the researcher must convince a funding agency either that the pilot study is worthy of support in its own right or that it will produce the data needed to design an intervention. This task can be difficult for the investigator. On one hand reviewers may see the pilot study as pedestrian if it is to be a project by itself. If it

20 is a first step in a multiyear grant, on the other hand, reviewers might think there are just too many unknowns to recommend a project for funding. The investigator is often in the position of having to argue the plausibility of an intervention on theoretical grounds and having to convince reviewers that it can be done in the proposed setting, even though the details of the intervention cannot be specified in the proposal. Choosing an Appropriate Focus for the Intervention The investigator must choose a site in which the most likely stressor, the one that an intervention will be designed to change, is really amenable to change. The investigator might seek settings, for example, in which employees suffer from low levels of personal control, a factor that much theoretical and empirical research supports as linked to well being. Meaningful approaches to augment employee control, however, might not exist in some of these settings. For example, the researcher might gain entry to a manufacturing facility in which employees experience low control because of machine-pacing requirements, but the job redesign necessary to provide self-pacing would require far more investment than an organization could make, at least in the short term. One must target job demands with an established etiologic role in employee well being and with technology or resources are available to make meaningful changes in them. Designing an Effective Intervention Designing an intervention to reduce the presence of certain job demands can involve a complex organizational change. Augmenting worker control, for example, can require a significant redesign of work processes, reporting relationships, and responsibilities. Such changes are not decided on and implemented overnight, but take considerable planning. Moreover, participants may require extensive training for the implementation to be effective. The attempt that my colleagues and I made to reduce stress through role clarification illustrates the difficulty of this step (Schaubroeck et al., 1993). After more than a year of pilot research and months of meeting with managers and staff groups, the researchers, staff, and managers decided that role clarity should be the highest priority. It was difficult enough getting to this step, and we still had to decide how to actually change the role ambiguity throughout the organization. The organizational structure consisted of six area managers who in turn directed the work of their staff. Although the target population was the staff group, we could not improve their role clarity until ambiguities and conflicts about responsibilities were resolved at the

21 manager level. We decided to use a method called responsibility charting to clarify roles and responsibilities among the managers and the areas they supervised. Responsibility charting is a fairly straightforward method, but no one on the research team had extensive experience with it. Fortunately, the collaborating organization could afford, and was willing, to hire a consultant. The responsibility charting procedure was completed over several weeks. We then trained each manager in how to clearly communicate the roles and responsibilities to their staff. This communication process actually involved a role negotiation procedure that differed from the way that many of the managers had previously communicated with their staff. We thought it necessary for the consultant to assist each of these dyads in the role negotiation process. Furthermore, we videotaped these sessions for later assessment. In short, identifying the stressor to be targeted took more than a year, and designing and implementing the intervention took another year, even with the assistance of an external consultant. Implementing a Randomized Design All the output of the previous steps results in only an elaborate case study that yields little interpretable evaluation data unless one can use a randomized design. Selection effects (Cook & Campbell, 1979) are usually the most pressing threat to internal validity in the field because managers often want to focus the intervention on those employees that need it most. The investigator must convince the organizational members that a randomized design is necessary to learn how effective the intervention really is. Implementing a randomized design and maintaining it in the field is a challenge. Unlike in laboratory experiments, subjects assigned to experimental conditions in the field do not always remain in the conditions to which they were initially assigned. Moreover, they often interact with each other extensively, which creates the threat of diffusion of the treatment to control group subjects or their resentful demoralization because they are not getting what is often perceived as a desirable job change. Combating these threats requires extensive communication with all subjects involved in the study. It also requires the investigator to be flexible and to make extensive contingency plans. Solving the Dependent Variable Problem In large-scale epidemiological studies, researchers can examine disease endpoints, hospital admissions, deaths, and other outcomes that represent the cumulative effects of long-term exposures to stressful job demands. For the field experimenter, time is the problem. Experimental groups can be maintained for only so long. People quit, retire, and change jobs,

22 and a randomized design often involves the use of a wait-control group. That is, to maintain the cooperation of the control group subjects over the course of the experiment, the investigator promises them treatment after the data have been assessed. People are not willing to wait indefinitely. Although investigators would like to see the effects of their interventions over the ensuing years, they simply cannot wait that long. Once the control group receives the intervention, of course, the researcher can continue to monitor outcomes for both groups but no longer has an internally valid design that supports causal inferences. Investigators need outcome variables whose underlying causal lags with the targeted stressor can reasonably be assumed to match the duration of the randomized experiment. On one hand, this constraint effectively rules out deaths from cardiovascular disease and probably hospital admissions for clinical depression as viable outcome variables. On the other hand, many observers view short-term outcomes involving employee attitudes and morale as relatively unimportant. Although I would not dismiss these criteria, the investigator also needs to examine more objective indicators. Measures of stress hormone production (e.g., cortisol) and markers of immune system functioning are useful outcomes, as are resting blood pressure and physiological reactivity to a standardized stressor (Schaubroeck & Ganster, 1993). Evidence indicates that such outcomes are responsive to interventions over several weeks or months, and they are epidemiologically meaningful. They can now also be obtained with relatively noninvasive techniques. The main problem is expense. Physiological reactivity measures are expensive in both subject and experimenter time, while biochemical measures require expensive assays. The investigator should also try to obtain as many measures relating to productivity and organizational expense as possible. It is probably unrealistic to expect large gains in labor productivity, but most stress reduction interventions should help organizational members be more effective. Although our stress outcome measures showed disappointing improvement in the role clarification study (Schaubroeck et al., 1993), managers later told us that they felt the organization was functioning much more smoothly and efficiently because of the intervention. Organizational archival measures, especially those concerning employee sick days, accidents, and health care costs, are important outcomes that have a large impact on the bottom line. Organizations must be convinced that stress-related interventions will have significant bottom line effects before they will wholeheartedly adopt them, and well-conceived interventions will achieve this goal.

23

Conclusions There is compelling theoretical and empirical support for conducting intervention research in the workplace. The obstacles that investigators face, however, are formidable. Intervention research teams require a mix of skills, including evaluation research, stress and health measurement, and organizational change expertise. Such research can be done, but it requires some flexibility on the part of investigators, and especially on the part of funding agencies. Agencies such as the National Institute of Mental Health (NIMH) and the National Institute for Occupational Safety and Health (NIOSH) should consider special funding initiatives for organizational intervention research that can accommodate the uncertainties inherent in multistage intervention research programs. Programs such as NIMH's Prevention Intervention Research Centers are useful in this regard, but they can fund research at only a limited number of sites. Another example is the National Science Foundation’s Transformation to Quality Organizations program. Although its emphasis is on general organizational effectiveness rather than on employee health and well being, the National Science Foundation has established a somewhat flexible set of criteria for funding research that assesses interventions in this area. The potential for improving worker health by redesigning work, increasing social support, and instituting family-supportive policies is great. Funding research that evaluates such interventions should be granted a high priority. The great advantage of conducting such intervention research is that successful interventions could demonstrate benefits for both employees and their employing organizations. Because interventions have so much potential to improve organizational effectiveness as well as employee well being, their subsequent diffusion is much more likely, even without the assistance of governmental research funding. Organizational decision makers, however, must be convinced with data from well-conducted studies that prove the feasibility of designing interventions and that show that such interventions improve the health of the organization and its members.

24

References Anisman, H., & LaPierre, Y. (1982). Neurochemical aspects of stress and depression: Formulations and caveats. In R. W. Neufeld (Ed.), Psychological stress and psychopathology (pp. 179–217). New York: McGraw-Hill. Antoni, M. H. (1987). Neuroendocrine influences on psychoimmunology and neoplasia: A review, Psychology and Health, 1, 3–24. Bacharach, S. B., Bamberger, P., & Conley, S. (1991). Work-home conflict among nurses and engineers: Mediating the impact of role stress on burnout and satisfaction at work, Journal of Organizational Behavior, 12, 39–53. Brockner, J., Konovsky, M., Cooper-Schneider, R., Folger, R., Martin, C., & Bies, R. J. (1994). Interactive effects of procedural justice and outcome negativity on victims and survivors of job loss, Academy of Management Journal, 37, 397–409. Burden, D. S.,& Googins, B. (1987). Boston University balancing job and homelife study. Boston: Boston University School of Social Work. Cannon, W. B. (1914). The emergency function of the adrenal medulla in pain and the major emotions, American Journal of Physiology, 33, 356–372. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation. Chicago: Rand McNally. Csanadi, S. B. (1981). Physical activity and stressor-strain relationships in law enforcement. Unpublished doctoral dissertation, University of South Florida, Tampa. Dembroski, T. M., MacDougall, J. M., & Lushene, R. (1979). Interpersonal interaction and cardiovascular response in Type A subjects, Journal of Human Stress, 5, 28–36. Dienstbier, R. A. (1989). Arousal and physiological toughness: Implications for mental and physical health, Psychological Review, 96, 84–100. Dienstbier, R. A., LaGuardia, R. L., Barnes, M., Tharp, G.,& Schmidt, R. (1987). Catecholamine training effects from exercise programs: A bridge to exercise-temperament relationships, Motivation and Emotion, 11, 269–295. Fox, M., Dwyer, D., & Ganster, D. C. (1993). Effects of stressful job demands and control on physiological and attitudinal outcomes in a hospital setting, Academy of Management Journal, 36, 289–318.

25 Frankenhaeuser, M. (1979). Psychoneuroendocrine approaches to the study of emotion as related to stress and coping. In H. E. Howe, Jr. and R. A. Dienstbier (Eds.), Nebraska Symposium on Motivation, 1978: Human Emotion (pp. 123–161). Lincoln: University of Nebraska. Frankenhaeuser, M., & Gardell, B. (1976). Underload and overload in working life: Outline of a multidisciplinary approach, Journal of Human Stress, 2, 35–46. Frone, M. R., Russell, M., & Cooper, M. L. (1992). Antecedents and outcomes of work-family conflict: Testing a model of the work-family interface, Journal of Applied Psychology, 77, 65–78. Ganster, D. C. (1989). Worker control and well being: A review of research in the workplace. In S. L. Sauter, J. J. Hurrell, and C. L. Cooper (Eds.), Job control and worker health (pp. 3–24). Chichester, England: Wiley. Ganster, D. C. (1991, April). Neuroendocrine approaches to studying job design. Paper presented at the annual meeting of the Society for Industrial and Organizational Psychology, St. Louis, MO. Ganster, D. C., Duffy, M. K., & Hurrell, J. J., Jr. (1995, August). The role of objective occupational demands in determining worker stress and well being. Paper presented at the annual meeting of the Academy of Management, Vancouver, British Columbia, Canada. Ganster, D. C., Fusilier, M. R., & Mayes, B. T. (1986). Role of social support in the experience of stress at work, Journal of Applied Psychology, 71, 102–110. Ganster, D. C., Mayes, B. T, Sime, W. E., & Tharp, G. D. (1982). Managing organizational stress: A field experiment, Journal of Applied Psychology, 67, 533–542. Ganster, D.,& Schaubroeck, J. (1991). Work stress and employee health, Journal of Management, 17, 235–271. Greenhaus, J. H., & Parasuraman, S. (1987). A work-nonwork interactive perspective of stress and its consequences. In J. M. Ivancevich & D. C. Ganster (Eds.), Job stress: From theory to suggestion (pp. 37–60). New York: Haworth. Hayghe, H. V. (1994). Are women leaving the labor force? Monthly Labor Review, 117, (7), 37–39. Haynes, S. G., Eaker, E. D. & Feinleib, M. (1984). The effects of employment, family and job stress on coronary heart disease patterns in women. In E. B. Gold (Ed.), The changing risk of disease in women: An epidemiological approach (pp. 37–48). Lexington, MA: Heath. Karasek, R. (1979). Job demands, job decision latitude, and mental strain: Implications for job redesign, Administrative Science Quarterly, 24, 285–306.

26

Kasl, S. (1978). Epidemiological contributions to the study of work stress. In C. L. Cooper & R. Payne (Eds.), Stress at work (pp. 3–48). Chichester, England: Wiley. Kasl, S. (1986). Stress and disease in the workplace: A methodological commentary on the accumulated evidence. In M. F. Cataldo & T. J. Coates (Eds.), Health and industry: A behavioral medicine perspective (pp. 52–85). New York: Wiley. Kristensen, T. S. (1989). Cardiovascular diseases and the work environment: A critical review of the epidemiological literature on nonchemical factors, Scandinavian Journal of Work, Environment, and Health, 15, 165-179. Mason, J. W. (1968). Organization of psychoendocrine mechanisms, Psychosomatic Medicine, 30, 565–808. Murphy, L. R. (1988). Workplace interventions for stress reduction and prevention. In C. L. Cooper & R. Payne (Eds.), Causes and consequences of stress at work (pp. 88–114). Chichester, England: Wiley. Schaubroeck, J.,& Ganster, D. C. (1993). Chronic demands and responsivity to challenge, Journal of Applied Psychology, 78, 73–85. Schaubroeck, J., Ganster, D. C.,& Fox, M. L. (1992). Dispositional affect and work–related Stress, Journal of Applied Psychology, 77, 322–335. Schaubroeck, J., Ganster, D. C., Sime, W.,& Dittman, D. (1993). A field experiment testing supervisory role clarification, Personnel Psychology, 46, 1–25. Selye, H. (1936). A syndrome produced by diverse noxious agents, Nature, 138, 32. Selye, H. (1976). The stress of life (2nd ed.). New York: McGraw-Hill. Solomon, G. S., Kay, N., & Morley, J. E. (1986). Endorphins: A link between personality, stress, emotions, immunity, and disease? In N. P. Plotnikoff, R. E. Faith, A. J. Murgo, & R. A. Good (Eds.), (Enkephalins and endorphins: Stress and the immune system pp. 129–144). New York: Plenum. Stroop, J. R. (1935). Studies of interference in serial verbal reactions, Journal of Experimental Psychology, 18, 643–662. Thomas, L. T.,& Ganster, D. C. (1995). The impact of family-supportive work variables on work-family conflict and strain: A control perspective, Journal of Applied Psychology, 80, 6–15.

1

Chapter 6. Economic Pressure in Rural Families: Couple Interactions that Reduce Risk for Emotional Distress and Marital Instability1 Rand D. Conger and Martha A. Rueter, Iowa State University at Ames Glen H. Elder, Jr. University of North Carolina at Chapel Hill

As early as the Depression years of the 1930s, scholars proposed that specific qualities of marital relations play a major role in conditioning the impact of unemployment and economic hardship on individual and family well-being (Angell, 1965; Cavan & Ranck, 1938; Komarovsky, 1940). Research findings from that era consistently indicated that strong marital ties reduce the adverse impact of economic problems, helping both couples and individuals weather more successfully the storm of job and income loss. Despite the uniform results, however, this research was limited by methodological weaknesses, a trait-like approach to the concept of marital bonds, and a lack of specificity regarding important dimensions of marital interactions. Methodological weaknesses in Depression-era studies are particularly evident in both their measurement and sampling strategies, which reflected convenience as much as known populations. Because little attention was given to the study of well-identified population groups, it is difficult to assess the generality of the findings even during the 1930s. An important limitation in this earlier work was its failure to study the influence of economic stress on rural couples, a limitation directly addressed in this paper. In addition, the approach to measurement was often qualitative, and little attention was given to issues of reliability and validity. Moreover, the measures often involved 1

The work reported in this chapter was supported by grants from the National Institute of Mental Health (MH48165, MH51361, MH43270.

2 broad categories of action that were conceptually rich, but operationally difficult to identify (e.g., Angell’s 1965 concept of family integration). This ambiguity and subjectivity in measurement make it difficult to replicate exactly the suggestive findings from these studies. A second, and especially important, limitation in the Depression-era research studies concerns their conceptualization of marital relations. Too often these early studies treated marriage as a “stable, intrinsic characteristic of the family" (Liem & Liem, 1990, p. 201). That is, marriages were described in global terms as either strong and supportive or not. From this perspective, the strength of a couple’s marital bonds was considered relatively immutable, even in the face of significant economic challenges. This view led to a static conception of the relationship between economic stress and marriage, a perspective not supported by more recent empirical findings. Since these early studies, research in the area has explored the dynamic nature of the relationship between financial troubles and the course and consequences of marital relations. Liker and Elder (1983), for example, found that couples who experienced major income loss during the Depression years also experienced elevations in marital tension and instability. Their research suggests that economic problems often led to increased conflicts over finances, which in turn lead to decreased marital quality. Interestingly, Liker and Elder also found that economic difficulties posed less of a threat to couples with happier marriages. They reported that, although couples with stronger marriages were not entirely protected, they seemed to experience fewer of the adverse consequences of economic decline than less satisfied couples. In an independent confirmation of these findings, Liem and Liem (1990) reported that high levels of preunemployment marital satisfaction protected husbands against the kinds of personal distress exacerbated by job loss. They found that, in general, husbands’ psychological distress from job loss led to subsequent declines in marital quality. However, men who were happily married before job loss were more resilient in terms of both emotional problems and disruptions in their marriages. These more contemporary findings then suggest the presence of a reciprocal process. A supportive marriage can help each spouse cope more successfully with financial problems. However, even a strong marital relationship will likely suffer under the weight of serious and continuing deprivation. Studies of these processes illuminate the dynamic role that marriage can play in providing the social support needed to cope effectively with economic stress. Although the view of a static

3 relation between hardship and marriage has been discarded, contemporary research evidence remains too global to clarify fully the processes of interest. For example, the Liem and Liem (1990) report that marital satisfaction protects against some of the psychological costs of unemployment reveals little about the interactional qualities in the marriage that actually provide such protection. Similarly, the Liker and Elder (1983) concept of marital tension is too global to improve understanding of the specific features of marriage that may decrease risk or increase vulnerability to economic problems. And as with earlier research in this area, contemporary studies have failed to examine links between economic stress and marital functioning in rural couples. The next generation of research on these issues must identify the particular acts of spouses that reduce the risks associated with economic difficulties and promote resilient marriages. As Pearlin and McCall (1990, p. 39) have suggested in their study of marital response to work stress, “The social and interactional character of support have either been ignored altogether or largely taken for granted.” In this report we contribute to the next generation of research by identifying some of the specific transactions between spouses that increase their individual and joint resilience to family financial difficulties. Through this effort we hope to both add to the theoretical understanding of family stress processes and enable the identification of specific interactional mechanisms that might be targeted in prevention or treatment programs designed to aid families under economic pressure. The present study also examines these issues with a sample of rural couples, thus extending the previous focus on urban families.

4

The Dynamic Relationship between Economic Pressure and Marital Instability In this report, we first present a model outlining a series of postulated processes through which family economic problems and the initial quality of the marriage are expected to affect each spouse's psychological distress and, ultimately, changes in marital instability (Figure 1). This model describes a chain of events that begins with two major difficulties in family life-- economic problems and a threatened marriage-- and then postulates a series of emotional and interactional events ensuing from these threats that may lead to further deterioration in marital relations. Specific dimensions of marital interaction are introduced at selected points in the overall model to determine the degree to which they might protect against the negative influence of economic stress on individual and family well-being. That is, these hypothesized protective influences are expected to inhibit the downward spiral depicted in the model. A Process Model of Economic Pressure and Marital Instability As shown in Figure 1, two avenues lead to emotional distress (i.e., depression, anxiety, and hostility) among husbands and wives. First, consistent with earlier research (Conger et al., 1992; Conger et al., 1994; Kessler, Turner, & House, 1988; Liem & Liem, 1990), the model proposes that economic pressure or strain will increase emotional distress. The distress construct refers to the dysphoria, worry, and irritability that frequently follow from hardship experiences (Conger & Elder, 1994). Economic pressure refers to the daily financial problems created by low income , such as the inability to pay bills, meet basic material needs, and avoid painful cutbacks in expenditures (see Conger, Ge, Elder, Lorenz, & Simons, 1994). In accordance with the Family Stress Model of economic influence (Conger & Elder, 1994), we expected that these economic problems would create a psychologically painful, aversive state for adults. Earlier research has shown that such psychologically aversive experiences are linked to feelings of both anger and despondency (Berkowitz, 1989).

5

6

Second, contemporary research indicates that problematic marital relationships increase emotional distress among husbands and wives (Gotlib & McCabe, 1990). In the present study, marital instability refers to a series of thoughts or actions by spouses that frequently lead to actual separation or divorce (e.g., discussing the idea of obtaining a divorce or consulting with an attorney about a divorce). Thus, the theoretical model shows pathways of influence between Time 1 (the first year of the study) economic pressure and Time 2 (one year later) emotional distress as well as between Time 1 marital instability and Time 2 emotional distress. This sequencing of measures in time assures that the hypothesized causal variables actually occurred before the predicted emotional responses. Figure 1 also illustrates our expectation that the psychological difficulties brought on by economic pressure and marital instability increase the risk for damaging conflict and hostility in the marital relationship. The model proposes that the anger, worry, and sadness exacerbated by economic problems will often find expression through marital interactions marked by argument, derogation, and insensitivity. Indeed, previous studies have shown that economic problems are indirectly associated with marital conflict through emotional distress (e.g., Conger et al., 1992; Liem & Liem, 1990), consistent with the proposed model. The hypothesized association between emotional distress and marital conflict also is consistent with empirical evidence showing that psychological problems can lead to both angry interactions and also to withdrawal in close relationships (Berkowitz, 1989; Conger et al., 1994; Downey & Coyne, 1990; Gotlib & McCabe, 1990). In addition to causing emotional distress, marital instability was expected to influence marital conflict directly and indirectly. As shown in the Figure 1, the indirect relationship occurs through marital instability's influence on emotional distress, which in turn affects marital conflict. In addition, because unstable marriages are more likely to be marked by conflictual spousal interactions than stable marriages are (e.g., Conger et al., 1990), the model proposes a direct association between marital instability (Time 1) and marital conflict (Time 2). That is, the consideration of divorce will increase the likelihood of irritable interchanges between spouses. Next, the theoretical model proposes that couples with unstable marriages at time 1 will likely experience continuing problems over the course of the study. Net of the influence of a couple's

7 initial level of marital instability, however, we also expected time 2 marital conflict to lead to increased risk for time 3 marital instability. Consistent with earlier studies of marital distress (e.g., Conger et al., 1990), we proposed that high levels of spousal conflict would be associated with reports of relatively greater marital instability from year one to year three. Finally, the model hypothesizes that economic pressure will be indirectly related to time 3 marital problems through its impact on emotional distress and associated couple conflicts. It is important to note that much of the earlier work supportive of the model was done with urban populations. One might assume that the degree of stress produced by economic hardship would not be as great in rural families because of the presumably more benign and supportive environments found in rural areas. Because of the enormous economic upheavals in rural America (e.g., Lasley, 1994), however, we hypothesized the reverse. That is, we expected that the rural couples in the present study would suffer the negative consequences of economic stress depicted in the theoretical model. We also predicted that they would benefit from the protective mechanisms described in the following discussion. Marital Interactions and Couple Resilience Thus far the theoretical model builds upon earlier work by suggesting a dynamic process through which earlier economic pressure increases individual distress and exacerbates marital problems. The major concern of the present study, however, is the identification of points in these hypothesized processes at which particular patterns of marital interaction might protect against the damaging influence of economic pressure on each spouse and on their relationship. Theoretical insights from multiple sources led to the selection of two potentially important links in the model for the operation of protective influences: (a) marital empathy, lilnking economic pressure and emotional distress, and (b) solution generation, linking marital conflict and marital instability. Marital empathy. Theoretical insights provided by Weiss (1990) and by Pearlin and McCall (1990) played an important role in our selection of spousal behaviors that might reduce the iadverse impact of economic pressure. Weiss (1990) suggested that when outside threats such as economic pressure exist, reassurance of worth by a spouse might be the most protective form of support. Similarly, Pearlin and McCall (1990) proposed careful listening and expressions of affection are effective forms of marital support during stressful times. Both sets of investigators suggested that

8 emotional reassurance would be more protective against work-related stresses, such as insufficient income, than an emphasis on advice giving. Drawing on these ideas we developed a concept named marital empathy. Marital empathy refers to the tendency of each spouse (a) to listen to the other’s cares and concerns, (b) to maintain a cooperative and helpful posture in relation to expressed concerns, and (c) to indicate sensitivity to the partner’s point of view. Couples rated highly on this interaction style also express a willingness to make changes in their own behavior to help meet the needs of the spouse. We expected that this pattern of couple interaction would be especially important for families experiencing significant economic pressure; many cooperative forms of adjustment are required to meet the demands of difficult financial conditions. Specifically, we proposed that couples high on this interactional attribute would experience less economically related emotional distress than those demonstrating low marital empathy. Furthermore, we hypothesized that advice giving, in the form of proposing numerous solutions to current difficulties, would do little to alleviate the emotional distress produced by economic pressure. Generating solutions. The next point of concern in the processes shown in Figure 1 involves the connection between Time 2 marital conflict and Time 3 marital instability. Assuming that empathy does not entirely eliminate the connection between economic pressure and distress and that initially unstable marriages will continue to experience greater conflict, the question of interest turns to strategies that couples might use to reduce the adverse consequences of such conflict for the marriage. To help answer that question we turned to the work of Levinger and Huston (1990), who noted that conflict occurs in all marriages at some time and that it is the means by which couples resolve disputes that distinguishes more successful from less satisfying relationships. In many instances couples engage in conflict as a contest that results in a winner and a loser, or they withdraw from disagreements, leaving the underlying dispute as a nagging and unresolved issue in the marriage. Contrary to these destructive styles of dealing with conflict, Levinger and Huston (1990, p. 51) proposed that “collaborating to resolve a problem jointly is the principal way to go beyond a zero-sum conception of an interpersonal conflict.” They suggested that such collaborative problem solving provides the most effective means for resolving conflicts in a mutually satisfying manner.

9 Drawing on the work of Levinger and Huston (1990), we developed a concept called solution generation. Couples who do well at solution generation work effectively together to develop or identify realistic and nonexploitative solutions to conflicts they experience. We proposed that couples demonstrating this interactional quality, compared to those who rated low on this attribute, would report smaller increases in marital instability as a result of the normal conflicts and disagreements in married life. Additionally, we expected that marital empathy would not moderate the relationship between marital conflict and marital instability. That is, we proposed that internal family stressors such as marital conflict require more than an understanding response to reduce their negative influence on family relationships. Without some degree of resolution these stressors will be a continuing threat to family well-being. The following analyses empirically test both our basic process model and the hypothesized protective marital interactions linked to specific portions of the model.

Method Sample The married couples in the study were mothers and fathers participating in an investigation of rural family economic stress. At Year 1 of the study (1989), each family included two parents (average length of marriage, 17.9 years), a seventh-grade adolescent, and another sibling. The age of the sibling varied, but all were within 4 years of the seventh grader’s age All of the families lived in small towns or the countryside: 34% lived on a afarm, 12% lived outside a town but not on a farm, and 54% lived in a town with a population no greater than 6,500. The average family size was 4.95 members, and family median income for 1988 was $33,399. When first interviewed in 1989, the sample consisted of 451 white, lower-middle and middle class families. During the next three years, 44 families withdrew from the study and as many as 36 families were deleted from some of the following analyses because of missing data. Participants were interviewed in three successive years-- 1989, 1990, and 1991. The retention rate for each year of data collection was about 95 %. Families who withdrew from the study were, in most respects, not significantly different from those who remained in the study. In terms of education, however, husbands who dropped out averaged 12.74 years of education, whereas those who remained in the study averaged 13.58 years. This difference was statistically significant (p < .05). Wives who

10 dropped out averaged 12.87 years of education, not significantly different from the 13.48 years of education for wives who stayed in the study. Other longitudinal studies of families with adolescents also have reported greater attrition of less educated parents (e.g., Flanagan & Eccles, 1993). Procedures The families in the study were recruited from all 34 public and private schools with a seventhgrade class in selected communities of 6,500 people or less in eight counties. After receiving a letter explaining the research project, families were contacted by telephone and asked to participate. About 78% of the eligible families agreed to take part in the study. Each participating family member was paid approximately $10 per hour of participation. In each year of the study, families were visited twice in their own home. During the first visit, each of the four family members completed a set of questionnaires focusing on, family member characteristics and patterns of family interaction. The second visit occurred within 2 weeks of the first. A trained interviewer began the second visit by asking both parents and the two children to complete checklists designed to identify current family disagreements or conflicts. Potential areas of disagreement between parents and children included household chores, homework, and transportation. Potential areas of disagreement between the marital couple included money, relatives, and drug or alcohol use. After completing the checklists, family members were asked to gather around a table for the videotaping of four different structured interaction tasks. For the first interaction task (Task 1), family members were given a set of cards containing questions to discuss about their family life (e.g., activities they do together) that were designed to elicit family interactions. Family members were asked to review the cards in sequence and one at a time. As in all four tasks, the interviewer explained the task procedures, completed a practice card with the family, and checked the video recording equipment before leaving the room for another part of the house. Each family spent a total of 30 minutes discussing the Task 1 cards. The problem solving task (Task 2) began shortly thereafter and lasted 15 minutes. During this task, parents and children discussed and attempted to resolve three problems identified on the previously completed questionnaires. Families were asked to first discuss the problem that created the most difficulties and disagreements between parents and adolescents. They were told to go on to the second and third problems only after resolving the first one. Data for the measures of problem solving behavior were collected while the family addressed the first problem.

11 The third task involved only the siblings and lasted 15 minutes. The two children discussed such topics as things they do together, school activities, and their plans for the future. During the sibling interaction task, the parents completed questionnaires in another room. For the fourth task, the husband and wife spent 25 minutes discussing several topics, including the history and status of their relationship, areas of agreement or disagreement, and their plans for the future. Meanwhile, the two children worked on questionnaires in another room. The analyses reported here used observational data collected from the second and fourth interaction tasks. Survey data from the study's first, second, and third years were also used in these analyses. The video taped family interactions were coded by trained observers who used the Iowa Family Interaction Rating Scales (Melby et al., 1990) to rate styles of family interaction of theoretical interest. Before they began coding videos, all observers received 200 hours (20 hours per week for 10 weeks) of training and passed extensive written and viewing reliability tests. They then attended at least two coder training sessions each week to ensure their continued reliability. To test interobserver reliability, we randomly assigned 25 % of all videotaped tasks to be coded by a second, independent coder. The primary and secondary codings were then compared using intraclass correlations (Suen & Ary, 1989). An observer rating manual with a complete description of all coding procedures and all task procedures, as well as definitions for all rating scales, is available from the first author. Measures Previous research suggests that using just one reporter to measure study variables can produce strong associations among theoretical constructs that are, at least in part, a function of that reporter's dispositional characteristics (Bank, Dishion, Skinner, & Patterson, 1989; Lorenz, Conger, Simons, Whitbeck, and Elder, 1991). Indeed, high correlations have been found between a person's selfperception or mood state and his or her perceptions of others, particularly of close family members (Brewin, Firth-Cozens, Furnham, & McManus, 1992; Gara et al., 1993). These strong associations seem to reflect traits of negative or positive affectivity as much or more than linkages among separate empirical realities (Watson & Pennebaker, 1989). To reduce this method variance problem in the present study, we gathered information from several agents, using multiple methods of data collection. A description of each measure follows.

12

Economic pressure. Economic pressure encompasses the daily irritations and difficulties created by economic hardship, such as the inability to pay one’s bills, a lack of resources to finance economic necessities, and the need to continually reduce expenditures (see Conger et al., 1992, 1994). This construct is consistent with but also expands on the concept of economic strain (Conger & Elder, 1994). We used it as our primary exogenous variable reflecting family economic conditions because several studies have shown that presumably objective indicators of economic hardship, such as low income and job loss, affect individual and family functioning primarily through the actions or experiences included in the economic pressure construct (Broman, Hamilton, & Hoffman, 1990; Conger et al., 1994; Kessler et al., 1988). That is, because other indicators of hardship increase economic pressure, they affect the well-being of family members; therefore, including both the pressure construct and other measures of economic conditions in the model would be redundant and complicate later statistical estimation of the proposed processes. Economic pressure was measured in the study's first year. Self-reports from both parents were used to develop the three indicators for this construct. For the first indicator, husbands and wives independently responded to three items about madking ends meet. Both spouses were asked if they had enough money to meet their expenses (1 = strongly agree, 5 = strongly disagree), if they had difficulty paying monthly bills (1 = no difficulty at all, 5 = a great deal of difficulty), and if they had any money left over at the end of the month (1 = more than enough money left over, 4 = not enough money to make ends meet). Each spouse's responses were standardized and summed. The correlation between the spouse reports on the summed indexes was substantial (r = .64); therefore the two indexes were combined to produce a single indicator for the construct (α = .86). The survey items for the second indicator of economic pressure asked if the husband and wife believed their family had enough income to meet their basic material needs. Each spouse was asked whether they had the money needed to purchase clothing, household items, a home, a car, food, medical care, and recreational activities (1 = strongly agree, 5 = strongly disagree). The summed husbands' and wives' indexes were strongly correlated (r = .54) and were combined to form a second indicator for the economic pressure construct (α = .91). The final indicator of economic pressure, economic adjustments, measured the extent to which families had made cutbacks in expenditures during the past year in response to financial difficulties.

13 Husbands and wives independently responded (0 = no, 1 = yes) to a list of 16 possible adjustments, including giving up medical or household insurance, reducing household expenditures, and changing vacation plans. The husbands' and wives' indexes were significantly correlated and were combined into a single indicator (α = .89). For instances in which both spouses responded in the affirmative for a specific item, the index increased only by 1, resulting in a possible range of 0 to 16 for the economic adjustments measure. Marital instability. We measured marital instability during the study's first and third years using two indicators. The first indicator was developed from the husbands' responses to five selfreport survey items, and the second indicator was developed from the wives’ responses to the same five items. In 1989, both spouses used a 4-point scale (1 = never, 2 = yes, prior to the last three years, 3 = yes, within the last three years, 4 = yes, within the last three months) to report if they had thought their marriage might be in trouble, had thought of getting a divorce or separation, had discussed divorce or separation with a close friend, had ever seriously suggested the idea of divorce, or had talked about consulting a lawyer regarding divorce or separation (Booth, Johnson, & Edwards, 1983). During year 3 the possible responses were not in the past year, within the past year, within the last six months, and within the past three months. This altered response set allowed us to predict changes in marital instability from Year 1 to Year 3. Scale items were summed to produce a husband's and wife's report of marital instability at Time 1 (α = .82 for husbands, α = .85 for wives) and Time 3 (α = .82 for husbands, α = .87 for wives). These separate reports were used as indicators for the marital instability construct. Emotional distress. We measured the husbands' and wives’ levels of emotional distress during the study's second year using self-reports, spouse reports, and observer ratings. Each husband and wife completed the depression, anxiety, and hostility subscales of the Symptom Checklist-90Revised (Derogatis, 1983). For both husbands and wives, responses to these three subscales were highly intercorrelated. The self-report indicator of emotional distress, therefore, was formed by combining responses to the three subscales (α = .92 for husbands, α = .92 for wives). For the second indicator for the construct we asked each spouse four questions about the partner's level of emotional distress. These survey items were rated on a five-point scale (1 = strongly disagree, 5 = strongly agree). Each spouse was asked whether his or her partner is a happy person (reverse coded), is always sad or depressed, is usually tense and irritable, and is a worrier. Summing these

14 items produced the wife's report of her husband's distress (α = .79), and the husband's report of his wife's distress (α = .77). Three observational codes from task four (the marital interaction task) were summed to create the third indicator of emotional distress. Each observed behavior (both verbal and nonverbal) was rated on a 5-point scale (1= the behavior is not at all characteristic of the individual, 5 = the behavior is very characteristic of the individual). In the first code observers rated the extent to which each spouse appeared to be sad or unhappy. Observers rated each spouse's display of anxiety or agitation in the second code. Finally, a rating of each spouse's level of contentment or satisfaction with self and his or her surroundings was reverse coded to produce a measure of lack of positive affect. The three codes were summed to form a single measure of observed emotional distress for the wives (α = .69) and for the husbands (α = .69). Interobserver agreement for the rating scales used in this measure averaged .71 for wives and .67 for husbands. Marital conflict. We defined marital conflict as the extent to which the couple engaged in disagreements or conflictual interactions. Both parents and both children reported on the amount of marital conflict during the study's second year. The husband and wife reports were derived from their responses to the marital problems checklist. Using this checklist, both spouses reported how much they argued or disagreed about 16 different issues (0 = not at all, 4 = all the time). Potential areas of conflict included work, relatives, vacations, and money. Each spouse's responses were summed to form the two indicators of self-reported marital conflict (α = .85, husbands; α = .84, wives). The two children in the study provided the information for the third indicator of marital conflict. Each child responded to three questionnaire items. First, the children independently rated how often their parents argued or disagreed with each other (1 = never, 5 =often), how often their parents argued specifically about money (1 = never, 5 = always), and whether their parents had argued more than usual with each other during the past year (1 = no, 2 = yes). Each set of responses was standardized and summed. After determining that the two children's measures were highly correlated, we combined them to form a 6-item scale (α = .67). Nine observational codes assessed during the marital interaction task (Task 4) were summed to form the final indicator of marital conflict. For each of these codes, observers rated verbal and nonverbal behavior using a 5-point scale (1= the behavior is not at all characteristic of the

15 individual, 5 = the behavior is very characteristic of the individual). The first code, hostility, measured the extent to which each spouse directed hostile, angry, critical, disapproving , or rejecting behavior toward his or her partner. Observers also coded the degree to which each spouse used verbal or nonverbal threats or bullying to gain compliance from his or her partner. A third code measured the extent to which an individual was impatient, self-centered, or unwilling to comply with his or her partner's wishes. Observers rated each spouse on these three codes, producing a total of six ratings. For the next three codes observers rated the couple as a unit (producing a total of three ratings). Transactional conflict measured the extent to which the couple engaged in reciprocal exchanges of hostility that became progressively more negative. This measure differed from the hostility code in that it assessed intensifying conflict in the marriage, rather than individual displays of anger and hostility. A high score on the eighth observational measure, relationship quality (reverse coded), reflected a marital relationship that was brittle and conflictual. Finally, with the silence/pause code, observers rated the extent to which the couple engaged in protracted silences that were tense and uncomfortable. These nine observational ratings were summed to form the observer's report of marital conflict (α = .89). Interobserver reliability for the individual ratings used in the scale averaEged .76. Marital empathy. Marital empathy included behaviors indicative of cooperation and helpfulness between the spouses, sensitivity to one another's situation, and a willingness to modify one's own behavior to meet the other's needs. This construct was measured during the study's first year using observational ratings from Tasks 2 (the family problem solving task) and Task 4 (the marital interaction task). Thus, couples showing a high level of marital empathy exhibited empathetic behavior toward one another while confronting a problem as well as during a general marital discussion. Task 2 and Task 4observers rated the wife's empathy toward her husband and the husband's empathy toward his wife, and these four ratings were summed. Interobserver reliability for the individual ratings included in this measure averaged .65 (α = .55). Solution generation. Solution generation, a couple’s demonstrated ability to identify or develop a number of useful solutions to a problem, was measured during the study's first year using reports from three informants. First, each spouse rated his or her partner's ability to generate solutions to problems. Using a 7-point scale (1 = always, 7 = never) spouses were asked how often,

16 when confronted with a problem, their partner "has good ideas about how to solve the problem" (reverse coded) and how often their partner "has poor ideas about how to solve the problem." Four observational codes also were used to measure solution generation. Task 2 observers (the problem solving task) rated both the husband and the wife on the number of problem solutions each identified (1 = no solutions identified, 5 = 4 or more solutions identified). Observers also rated each spouse on the quality of his or her solutions (1 = no solutions identified, 5 = at least one solution identified that is reasonable, nonexploitative, realistic, achievable, and specifically stated). Interobserver reliability for these codes averaged .76. To form a single measure of solution generation, the wives’, husbands', and observers' ratings were all standardized and summed (α = .62).

Results The results are generally supportive of the process model derived from the research literature and illustrated in Figure 1. All of the postulated direct paths were statistically significant as were factor loadings for the indicators of the various constructs. Moreover, the overall goodness-of-fit index was quite robust. The findings suggest that two major family stressors, economic pressure and marital instability, are associated with high levels of emotional distress for both husbands and wives. (See Conger & Elder, 1994) These emotional problems, in turn, lead to conflicts in the marriage (see also Liem & Liem, 1990). The results replicate earlier findings regarding the association between psychological distress and problems in marital and other family relations (Conger et al., 1990, 1994; Coyne & Downey, 1991). Presumably, emotions such as anger and depression surface in the marriage through instances of irritable conflicts and withdrawal from intimacy. As expected, distress-related marital conflicts led to a subsequent increase in marital instability. Also as predicted, the influence of economic pressure on later marital instability was entirely indirect through the proposed mediating processes, personal distress amd marital conflict. Moreover, just as the eventual influence of economic stress on marital instability was indirect, so was its association with marital conflict. We conclude that economic hardship has its most

17 significant impact on marriage through its exacerbation of wives’ and husbands’ emotional problems. But are there specific marital resources that can inhibit this destructive set of processes? We predicted that specific types of interactional skills in marriage would be protective at different points in the postulated set of causal relationships. The theoretical ideas of Weiss (1990) and Pearlin and McCall (1990) suggested that affection, sensitivity, and personal affirmation by marital partners are most effective in reducing the negative impact of work-related threats on personal well-being, whereas advice giving has little protective influence. Drawing on these perspectives, we proposed that marital empathy would be especially likely to act as a buffer against the emotional costs of economic pressure for both husbands and wives. The measure of empathy involved the observed tendency of both spouses to demonstrate concern, caring, sensitivity, and helpfulness to one another during two videotaped interaction tasks. As expected, couples who rated high on this interactional skill compared to those who were not were significantly less likely to suffer emotionally as a result of family economic difficulties. Also as expected, the ability to generate solutions to problems did not alleviate emotional distress brought on by economic concerns. Theoretically, this finding provides important new information regarding specific properties of marriages that are protective in the face of economic stress. The results tell us what couples actually do that has a protective influence, rather than simply indicating that those who are happy with their marriages are more resilient in the face of economic difficulties than those who are not. Presumably it is these actions that account both for their satisfaction with their relationships and for the buffering effect of happy marriages. In fact, as we have defined marital empathy, it involves a set of behaviors quite similar to a range of positive acts found to be associated with marital satisfaction (Gottman, 1993). Data regarding specific social behaviors provides the type of information needed to assist in the development of interventions for stressed couples. The behaviors included within the empathy construct can be taught to and applied by couples who, as a result, should increase their degree of resilience to such outside threats as economic stress. If the source of stress for couples results from difficulties within the family itself, however, other interactional skills are needed to reduce their adverse impact. That is, concern and understanding should provide affirmation that benefits a spouse when stressors from the outside pose a threat to well-being. When one’s partner is the

18 source of stress, however, resolution of the problematic behavior is required rather than empathy alone. Drawing on the work of Levinger and Huston (1990), we proposed that couples with strong problem-solving skills would be most able to respond effectively to marital conflict, reducing its impact on later marital instability. As noted, we also predicted that marital empathy would do little to moderate the relationship between marital conflict and marital instability. The findings were consistent with these hypotheses. Couples who demonstrated the ability to generate realistic and nonexploitative solutions to their conflicts and disagreements were less likely to suffer instability in their marriages as a result of such conflicts than less capable couples. On the other hand, a couple’s ability to respond with empathy did little to protect their marriage from the damaging effect of marital conflict. These findings suggest that, when faced with an internal family stressor, couples need to do more than provide sensitivity and concern. They also need to be able to negotiate, bargain, and reach agreement on realistic solutions to these internal family matters. This finding sharpens our knowledge regarding specific social support processes and also provides a potential tool for assisting couples, almost all of whom will experience periods of conflict in their relationships. We expect solution-generating skills to be teachable and useful in programs designed to help couples develop strategies for promoting more satisfying relationships.

Implications The research reported here moves in the direction of a next generation of research on family stress processes in general and economic stress in particular. Such research should go beyond earlier work that has dealt primarily with perceptions of satisfaction and closeness in family relationships to a focus on specific interactional qualities and individual characteristics that either promote resilience or increase vulnerability to difficult life conditions. Such research will increase theoretical knowledge and also lead to more effective programs for assisting families during stressful periods in their lives. Interventions involving such marital interactions, of course, represent only a sample of potential strategies for reducing the adverse consequences of economic hardship. The possibilities for helping families range from social structural- , such as the provision of alternative means for

19 augmenting income, to individual therapy for the alleviation of personal distress. Although income generating solutions are ideal, other methods of assisting families with financial difficulties are often required as well.. Another important dimension of the current work involves the extension of economic stress research to rural couples and marriages. These findings demonstrated that the rural Midwest does not fit the stereotype of benign, untroubled country living (see Davidson, 1990 for a description of the harsh realities in contemporary rural America). Moreover, the economic stress effects found among rural families were particularly robust. For example, Liker and Elder (1983) reported that the Depression-era men that they studied became emotionally distressed because of economic problems only if they had a prior history of emotional instability. In contrast the rural men in the present study were at risk for stress-induced emotional problems, regardless of their history of psychological difficulties (i.e., economic pressure had a main effect on men’s emotional distress). Moreover, whereas Liker and Elder (1983) reported no direct effect of economic problems on the emotional lives of women, we found strong associations between these two variables in the present analyses. Thus, not only do rural couples experience the same degree of disruption in their lives as a result of economic hardship as urban couples do, but they may even be more traumatized by financial difficulties. The present study, of course, requires replication with both urban and rural families. Moreover, these findings need confirmation with other ethnic groups and with other family forms, such as childless married partners. Also important will be extension of this work to other family relationships. For example, will empathy and problem-solving skills have similar influences for parent/child or extended family relationships? The model of economic stress and marital instability tested here also needs to be compared with other possible models regarding the same process. The current results remain tentative until additional work is done to address these issues. Nonetheless, we can conclude that the data are consistent with both the hypothesized process model of economic pressure and marital instability and with the protective mechanisms hypothesized to moderate specific linkages in the model. This study attempted to shed new light both on the processes through which economic stress affects marital relations and on particular characteristics of interactions between spouses that might reduce the cumulative harm produced by such processes. The initial impetus for the research

20 derived from the Depression-era studies of the 1930s, which produced suggestive evidence that spouses with strong marriages were less likely to suffer either personally or as a family when jobs or income were lost (Angell, 1965; Cavan & Ranck, 1938; Komarovsky, 1940). This research, however, had several conceptual and methodological weaknesses including limitations in measurement and sampling and a static rather than dynamic view of marriage. Many of the problems in these early studies have been addressed by more contemporary findings that have shown that marital quality both influences and is affected by the economic stress process (Liem & Liem, 1990; Liker & Elder, 1983). This work has taken the field beyond the earlier, trait-like conception of marital quality.

21

References

Aneshensel, C. S. (1992). Social stress: Theory and research, Annual Review of Sociology, 18, 15-38. Angell, R. C. (1965). The family encounters the depression. Gloucester, MA: Charles Scribner. Bank, L., Dishion, T. J., Skinner, M., & Patterson, G. R. (1989). The glop problem in structural equation modeling. In G. R. Patterson (Ed.), Family social interaction, (pp. 247-280). Hillsdale, NJ: Erlbaum. Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and reformulation, Psychological Bulletin, 106, 59-73. Booth, A., Johnson, D., & Edwards, J. N. (1983). Measuring marital stability,. Journal of Marriage and the Family, 38, 387-394. Brewin, C. R., Firth-Cozens, J., Furnham, A., & McManus, C. (1992). Self-criticism in adulthood and recalled childhood experience, Journal of Abnormal Psychology, 101, 561-566. Broman, C. L., Hamilton, V. L., & Hoffman, W. S. (1990). Unemployment and its effects on families: Evidence from a plant closing study, American Journal of Community Psychology, 18, 643-659. Cavan, R. S., & Ranck, K. M. (1938). The family and the depression. Chicago: University of Chicago Press. Conger, R. D., & Elder, G. H., Jr. (1994). Families in troubled times: Adapting to change in rural America. New York: Aldine. Conger, R. D., Conger, K. J., Elder, G. H., Jr., Lorenz, F., Simons, R., & Whitbeck, L. (1992). A family process model of economic hardship and influences on adjustment of early adolescent boys, Child Development, 63, 526-541. Conger, R. D., Ge, X., Elder, G. H., Jr., Lorenz, F. O., & Simons, R. L. (1994). Economic stress, coercive family process, and developmental problems of adolescents, Child Development, 65, 541-561. Conger, R., Elder, G. H., Jr., Lorenz, F., Conger, K., Simons, R., Whitbeck, L., Huck, S., & Melby, J. (1990). Linking economic hardship to marital quality and instability, Journal of Marriage and the Family, 52, 643-656.

22

Coyne, J. C., & Downey, G. (1991). Social factors and psychopathology: Stress, social support, and coping processes. In M R. Rosenzweig & L. W. Porter (Eds.), Annual review of psychology Vol. 42, (pp. 401-425). Palo Alto, CA: Annual Reviews, Inc. Davidson, O. G. (1990). Broken heartland: The rise of America’s rural ghetto. New York: The Free Press. Derogatis, L. R. (1983). SCL-90-R administration, scoring, and procedures manual-II. Towson, MD: Clinical Psychometric Research. Downey, G., & Coyne, J. C. (1990). Children of depressed parents: An integrative review, Psychological Bulletin, 108, 50-76. Flanagan, C. A., & Eccles, J. S. (1993). Changes in parents' work status and adolescents' adjustment at school, Child Development, 64, 246-257. Gara, M. A., Woolfolk, R. L., Cohen, B. D., Goldston, R. B., Allen, L. A., & Novalany, J. (1993). Perception of self and other in major depression, Journal of Abnormal Psychology, 102, 93100. Gotlib, I. H., & McCabe, S. B. (1990). Marriage and psychopathology. In F. D. Fincham & T. N. Bradbury (Eds.), The psychology of marriage: Basic issues & applications (pp. 226-257). New York: Guilford Press. Gottman, J. M. (1993). A theory of marital dissolution and stability, Journal of Family Psychology, 7, 57-75. Hammen, C. (1991). Depression runs in families: The social context of risk and resilience in children of depressed mothers. New York: Springer-Verlag. Joreskog, K. G., & Sorbom, D. (1989). LISREL VII: User's reference guide. (1st ed.). Mooresville, IN: Scientific Software. Kessler, R. C., Turner, J., & House, J. S. (1988). Effects of unemployment on health in a community survey: Main, modifying, and mediating effects, Journal of Social Issues, 44, 6985. Komarovsky, M. (1940). The unemployed man and his family. New York: Dryden Press. Lasley, P. (1994). Rural economic and social trends. In R. Conger & G. H. Elder, Jr. (Eds.), Families in troubled times: Adapting to change in rural America. New York: Aldine de Gruyter. Levinger, G., & Huston, T. L. (1990). The social psychology of marriage In F. D. Fincham & T. N. Bradbury (Eds.), The psychology of marriage: Basic issues & applications (19-58). New York: Guilford Press.

23

Liem, J. H., & Liem, G. R. (1990). Understanding the individual and family effects of unemployment. In J. Eckenrode & S. Gore (Eds.), Stress between work and family (pp.175204). New York: Plenum Press. Liker, J. K., & Elder, G. H., Jr. (1983). Economic hardship and marital relations in the 1930s, American Sociological Review, 48, 343-359. Lorenz, F. O., Conger, R. D., Simons, R. L., Whitbeck, L. B., & Elder, G. H., Jr. (1991). Economic pressure and marital quality: An illustration of the method variance problem in the causal modeling of family processes, Journal of Marriage and the Family, 53, 375-388. Melby, J., Conger, R., Book, R., Rueter, M., Lucy, L., Repinski, D., Ahrens, K., Black, D., Brown, D., Huck, S., Mutchler, L., Rogers, S., Ross, J., & Stavros, T. (1990). The Iowa family interaction rating scales. Ames, IA: Iowa Youth and Families Project. Pearlin, L. I. (1989). The sociological study of stress, Journal of Health and Social Behavior, 30, 241-256. Pearlin, L. I., & McCall, M. E. (1990). Occupational stress and marital support: A description of microproceses. In J. Eckenrode & S. Gore (Eds.), Stress between work and family (pp. 39-60). New York: Plenum Press. Suen, H. K., & Ary, D. (1989). Analyzing quantitative behavior observation data. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Turner, R. J., Wheaton, B., & Lloyd, D. A. (1995). The epidemiology of social stress, American Sociological Review, 60, 104-125. Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress and distress: Exploring the central role of negative affectivity, Psychological Review, 96, 234-354. Weiss, R. S. (1990). Bringing work stress home. In J. Eckenrode & S. Gore (Eds.), Stress between work and family (17-37). New York: Plenum Press.

1

Chapter 7. The Job Seeker Role as Resource: Achieving Reemployment and Enhancing Mental Health Richard H. Price, Amiram D. Vinokur, and Daniel S. Friedland Institute for Social Research, University of Michigan

In the modern economy, job loss is pervasive. As economic changes trigger workplace shutdowns and reductions in the workforce, many workers who would never have thought themselves vulnerable in the past are losing jobs (Price, 1990). In 1994, approximately 8 million people were unemployed in the United States. Of those 8 million people, between 50% and 65% were unemployed because they lost their jobs (U.S. Department of Labor, 1995). Given the pervasiveness of the phenomenon, we must understand both the impact of job loss on health and mental health and the mechanisms through which that impact occurs. The transition from unemployment to reemployment is a turning point in an individual’s life (Elder, Gimbel, & Ivie, 1991; Pickles & Rutter, 1991). An unsuccessful transition can close opportunities and place a person on a downward trajectory in terms of social mobility. Successfully negotiating the transition can relieve a number of the stressors of unemployment. In some cases, finding a new job may provide new opportunities and place the individual on a

1

2 trajectory of upward mobility. Successfully negotiating the transition may also have an amplifying effect; success breeds more confidence. Job loss is a stressful life event that influences physical and mental health (Pearlin, 1989; Pearlin, Lieberman, Menaghan, & Mullan, 1981). As a primary stressor, job loss can lead to an array of secondary stressors, most notably economic hardship. The physical and mental health consequences of job loss depend not only upon the job loss itself, but also on the number and strength of secondary stressors, such as increased debt and family conflict, triggered by the event. This formulation has been particularly useful in distinguishing life events as primary stressors from their various secondary sequelae. However, most accounts of the relationship between stressors such as job loss and health do not sufficiently emphasize the goal directed orientation of individuals as they cope with stressful events (Thoits, 1994). Recent developments in role theory (Baker & Faulkner, 1991; Callero, 1994) complement the view of the individual as an active agent in the coping process. These accounts view roles as resources and elaborate on our conception of job loss as a role loss with consequent losses of both material and personal psychological resources. This view supplements the stress process perspective in important ways, by both specifying the resources at stake for the job loser and emphasizing that the job seeker is an active agent, striving to claim a new job role in society. In this paper, we review the impact of job loss on physical and mental health, emphasizing the mechanisms by which job loss leads to deleterious outcomes. We then describe the transitional role of job seeker, and examine the challenges faced by job seekers. The JOBS program, a preventive intervention developed by the Michigan Prevention Research Center to help job seekers meet the challenges of finding a new job, is then described. We discuss results of preventive trials that document its impact on reemployment and mental health. We view these preventive trials as experiments in life course development, which can yield dividends in both improved well being and scientific understanding.

3

Impacts of Job Loss on Health and Mental Health: Empirical Findings Findings from the Great Depression to the present, based on cross-sectional, longitudinal, and prospective designs, have documented the psychological and social costs of job loss for the unemployed person, for members of the person’s family, and for the family as a whole (for a recent review, see Dew, Penkower, & Bromet, 1991). Although some people may lose their jobs because of mental health problems, several studies have demonstrated that job loss produces mental health problems which, in fact, extend beyond any prior problems (Dooley, Catalano, & Wilson, 1994; Kessler, Turner, & House, 1987). Job loss has adverse effects on the job seeker’s social and psychological functioning (Vinokur, Caplan, & Williams, 1987). It leads to increases in depressive symptoms (Catalano, 1991; Catalano & Dooley, 1977; Kessler, Turner & House, 1988; 1989), increased anxiety (Catalano, 1991), decreased subjective perceptions of competence (Warr, Jackson, & Banks, 1988), and decreased self-esteem (Jackson & Warr, 1984). Job loss is also associated with increased risk of suicide attempts (Platt & Kreitman, 1985), increased risk of alcohol abuse (Catalano, Dooley, Wilson, & Hough, 1993), and increased propensity for violent behavior (Catalano, Dooley, Novaco, & Wilson, 1993). Job loss also affects members of the job seeker’s family (Dew et al., 1991; Elder & Caspi, 1988). The job seeker’s increased propensity for aggressive, even violent, behavior often manifests itself in the context of the family. Positive correlations have been found between job loss and both spousal abuse (Windschuttle, 1980) and child abuse (Gil, 1970; Parke & Collmer, 1975). Research also indicates that the wives of job losers have a higher prevalence of psychiatric disorders than wives of people who remain employed (Bebbington, Hurry, Tennant, Stuart & Wing, 1981). Finally, job loss has been linked to marital and family dissolution (Liem & Liem, 1988). Not all job loss is harmful, however. While the majority of research on job loss has documented its deleterious effects, job loss may, in some instances, actually be beneficial for an

4 individual’s physical and mental health (Wheaton, 1990). The meaning and impact of a job loss depend, in part, on the history of the job role for the individual. For example, many jobs and organizational contexts are stressful (Ganster & Schaubroeck, 1991; Kahn & Byosiere, 1992). When individuals lose jobs that expose them to chronic stressors, the loss may actually relieve stress and benefit health.

Job Loss as Role Loss: Impact on Material and Personal Resources We propose that job loss is linked to physical and mental health risk through role loss and its consequences. Roles afford access to both material and personal resources in society (Baker & Faulkner, 1991; Callero, 1994). Involuntary job loss entails not only loss of income and access to the resources that income allows, but also a loss of status, occupational identity, and feeling of control over events. These losses of material and personal resources and the stressors associated with them trigger distress and mental health problems ( Jahoda, 1981); Warr & Jackson, 1984) Of course, the degree of stress depends importantly on both the salience of the work role (Thoits, 1991) and the degree to which the individual was dependent on the lost resources.

Job Loss and Economic Hardship A number of studies have identified economic hardship as a key mediating influence between job loss and depressive symptomatology (Kessler, Turner & House, 1987; Price, van Ryn, & Vinokur, 1992; Vinokur, Price, & Schul, 1995). Furthermore, substantial evidence supports the relationship between economic hardship and health outcomes more generally (Umberson, Wortman, & Kessler, 1992). In the case of job loss, the resulting economic hardship may have effects on mental and physical health through at least three pathways. First, economic hardship may deprive individuals and their families of resources needed to sustain physical health and general well being. Second, economic hardship can produce a cascade of stressful events with both

5 immediate and delayed effects on health and well being. Third, the stresses of economic hardship can influence family dynamics, which in turn affect mental health. Economic Deprivation. Economic hardship may affect physical health and general well being through negative effects on nutrition and loss of access to health care. Poor nutrition has its own radiating set of consequences on well-being: Poor nutrition can make both children and adults more susceptible to physical illness (Beasley, 1991), and malnourished children show lower levels of school achievement (Pollitt, 1994), increasing the likelihood of continued downward social mobility in later generations. Because health insurance in the United States is frequently directly tied to paid employment, one of the most immediate consequences of job loss is diminished access to health care. Price (1990) has observed that families will often reallocate limited benefits among family members. For example, a family may seek treatment for children while neglecting acute conditions among adults. Individuals may fail to seek either preventive services or care for acute and chronic conditions. In these circumstances, acute conditions can become chronic, and chronic conditions may deteriorate still further (Price, 1990). These health effects, of course, also influence the capacity of unemployed persons to seek new employment that might reverse the flow of negative health and unemployment effects. Cascade of Stressful Events Economic hardships produce a cascade of stressful economic life events that challenge the coping capacities of families and individuals. In the short term, economic hardship forces people to worry about facets of life that previously had been taken for granted. Inability to meet payments for housing may lead to foreclosure or eviction. Loss of an automobile means the loss of not only family transportation, but also a key resource for an effective job search. Economic hardship can also have delayed effects. People may cope with their financial difficulties by using credit cards or installment purchase plans, or by drawing heavily from savings. These approaches to coping with financial strain can have long term negative consequences. Drawing heavily from savings and taking on additional debt create a spiral of financial economic problems that will continue even after employment is regained. These long-

6 term financial problems may be even greater for older individuals who lose their pensions along with their jobs (Price, 1990). Reverberation of Economic Stressors throughout the Family Still another set of mechanisms implicating economic hardship and family dynamics mediates the relationship between unemployment status and poor mental health. Several studies suggest that the distress displayed by job losers affects the well being of their spouses (Liem & Liem, 1988; Penkower, Bromet, & Dew, 1988) as well as their children (Elder & Caspi, 1988; Justice & Duncan, 1977; Steinberg, Catalano & Dooley, 1981). Recent results reported by Vinokur, Price and Caplan (1996) reveal that couple dynamics influence the mental health of the unemployed person. Their analyses suggest that economic hardship increases depressive symptoms in both job losers and their spouses. Depressed spouses or partners then withdraw social support from the unemployed person and increase undermining behaviors. Both reduced support and increased undermining by the spouse then increase the depression and reduce the marital satisfaction of the unemployed person. Economic hardship associated with job loss in a family has direct effects on the spouse or partner, which in turn erodes that person’s capacity to support the job loser, with predictable effects on the job loser’s mental health.

Influence of Job Loss on Personal Identity and Mastery Roles are used to construct the self (Callero, 1992; 1994; Turner, 1978). Losing the central roles of worker and provider is a major challenge to a person’s identity. Ezzy (1993) argues that job loss is a form of status passage that directly disrupts an individual’s attempt to sustain a consistent and positive self-image and therefore increases the risk of mental health problems. Furthermore, because life domains are interrelated, role loss in one domain has radiating effects

7 on other domains. Thus, loss of an occupational role also presents identity challenges to the individual in the role of friend, spouse and parent. Job Loss and Erosion of Personal Identity and Friendship Ties Some jobs are low in status, but few are as stigmatized as unemployment. Unemployment status represents a form of “spoiled identity” (Goffman, 1963); job losers will often construct an alternative work identity such as “consultant” or “student” to avoid the erosion of self-esteem and demoralization often associated with socially devalued roles and statuses (Hughes, 1945). Job loss may also present a challenge to identity and self esteem by altering an individual’s network of friends and social support. The loss of a job may result in the loss of a primary source of contact with friends (Bolton & Oatley, 1987). Since friendships often arise and are maintained by proximity (Whyte, 1956), the bonds of friendship are more difficult to maintain when people are no longer employed by the same organization. Over time, the frequency of contact with friends from the previous job decreases (Atkinson, Liem, & Liem, 1983), and this loss of friendship networks can erode mental health. Kessler, Turner, and House (1988) found that being integrated into an affiliative network reduced the impact of unemployment on anxiety, depression, somatization and physical illness of job losers. Their findings underline the importance of supportive friendships for identity and well-being. Job-Loss and Family Relations Finally, one’s identity, sense of mastery and competence are normally sustained in valued and sanctioned social roles as provider, spouse and parent within the family (Thoits, 1991). Job loss disrupts these roles and the sense of personal identity and mastery they provide. This role disruption takes a number of different forms. Job loss introduces new and pressing agendas into the family, including coping with financial hardship and mobilizing to find reemployment, that can disrupt previously stable household role allocations and relationships (Conger, et al., 1990; Menaghan, 1991). When role reallocation involves shifts in authority and status in the family, the resulting shift in power dynamics can lead to conflicts that threaten the short-term stability of the couple relationship (Atkinson, Liem, & Liem, 1986). Such a re-alignment can undermine the selfconfidence of the job seeker and the partner in coping with job loss individually and as a couple

8 (Howe, Caplan, Foster, Lockshin, & McGrath, 1995; Vinokur & Caplan, 1987). It can also reduce the likelihood that the couple will develop workable solutions to the concrete demand of the crisis and, consequently, can increase the risk of depression for both partners (e.g., Kowalik & Gotlib, 1987; Vinokur, et al., 1996). This destabilizing process can result in a downward trajectory in the marital relationship that may end in separation and divorce (Gottman, 1993). Job loss and economic hardship also place strains on parent child relationships (Conger et al., 1990; Elder & Caspi, 1988). These strains not only undermine the parent’s sense of identity and mastery, but often increase the likelihood of parental irritability, conflicts between parent and child, harsh punishment, and family violence and child abuse (Broman, Hamilton, & Hoffman, 1990). Summary Job loss is role loss. Paid work roles provide access to critical material resources that sustain life, health, and well being for individuals and families. Paid work roles also provide emotionally significant identity resources to individuals in their role as family provider and as a contributing member of the community. Job loss strips away access to these essential material and identity resources. The loss of these resources may then result in a series of short and longterm problems for individuals and families, including mental health problems and increased risk of divorce, family violence, and child abuse. A recent study by Turner (1995), using a national probability sample of unemployed persons, found that the negative health consequences of unemployment are determined by identity strains and economic hardship. However, he also suggests that those two classes of strains are differentially important for different groups of job losers. Job losers with high SES, especially those for whom the work role is most salient, may suffer more from the loss of identity than from the loss of material resources. Job losers with low SES, on the other hand, may suffer more from the increased financial strain (Turner, 1995).

The Transitional Role of Job Seeker As we have argued, job loss is a major role loss that casts the individual into a new stigmatized social status as an unemployed person (Ezzy, 1993). Status as an unemployed

9 person does not provide the individual with a coherent positive identity or with an alternative social role that guides action. Consequently, in addition to the humiliation, disappointment, feelings of betrayal, and financial pressures that accompany job loss, job losers also experience a sense of crisis. Since unresolved crises tend to have powerful negative effects on mental health (Turner & Avison, 1992), the resolution of this crisis by successfully assuming a more satisfactory role in the social structure is an important first step. Once a new role is assumed, the sense of crisis gives way to the struggle of enacting the new role successfully. Although individuals are motivated to actively solve the problems that face them (Thoits, 1994), their motivation becomes directed and translated into action only when they can mold a plan that will lead to their goal. Having a realistic achievable goal enhances one’s sense of mastery and further energizes actions. Being unemployed does not provide a vision of a realistic goal. Hence, the first turning point for job losers is to escape the transitory status as an unemployed person by defining themselves with a more satisfactory role. Some job losers define themselves as students and return to school; others define themselves as parents and resume or take on child-rearing responsibilities; and still others define themselves as retired. However, for the majority of job losers, to alleviate economic hardship and protect their identity, they must often claim their old occupational role by finding a new job. Successfully Enacting the Role of Job Seeker The task of finding and assuming a new job is not straightforward. Successfully coping with unemployment and obtaining a new job largely depends on an individual’s ability to assume the transitional role of job seeker. Defining themselves as job seekers is the first step toward the development of an articulated, concrete course of action that seems realistic and achievable. Assuming the role of job seeker is the turning point that transforms the psychological distress of job loss into a directed plan of action. . Successfully enacting any role, including that of job seeker, involves several skills. Although the transitional role of job seeker is generally recognized in society, job-seeking skills

10 are seldom formally taught. Most people learn job search skills informally through either selfhelp programs (Azrin & Beasalel, 1982) or self-help books (e.g., Bolles, 1995). Mastery of these job search skills, which are critical to landing a job, becomes a new challenge for the job seeker. What knowledge, skills, and attitudes allow job seekers to enact their role successfully? First job seekers have to recognize the set of marketable skills they possess. Second, they need to decide which occupational role they wish to claim and know how to find employers who desire employees with the occupational role and skills. Third, they need to know how to present themselves and communicate their skills so employers will recognize their value. And fourth, they need to be able to evaluate the merit of jobs that are offered to them. These skills form a repertoire of skilled social behaviors involved in the enactment of a social role: They involve merging and presenting components of one's identity (skills to be used in the workplace as well as personality traits) to complementary role players (i.e., employers) according to norms that are appropriate for socially constructed channels (job applications, resumes) and situations (networking, calls for job inquiries, interviews). One new perspective expands the traditional concept of social roles to include their dynamic aspects as claims to and uses of resources (Baker & Faulkner, 1991; Callero, 1994). This perspective suggests that the role of job seeker itself, skillfully and persistently enacted, is a vital resource in successfully claiming a role that will lead to reemployment. This new perspective underscores the importance for the job loser of assuming the job-seeking role. Through the psychological assumption of the role and its successful communication to others, the job seeker is able to mobilize various resources that are instrumental in pursuit of a new job. For example, presenting oneself as a job seeker allows the person to ask relatives, friends, and former coworkers and employers to provide leads and information on promising jobs, and for transportation to job interviews. Self-presentation as a job seeker also allows the person to legitimately ask a spouse for help in various tasks, including help with various household and family duties (e.g., child care), so that time is available for the job search. In fact, claiming this role may be a requirement of official employee assistance programs such as job retraining, career- planning seminars, and job search workshops.

11

Challenges in the Job Seeker Role Even within the role of job seeker, there is uncertainty associated with searching for a new job. This uncertainty has a number of elements, that add to the complexity and threat of the major tasks involved in enacting this role (Caplan, Vinokur, Price, & van Ryn, 1989). Job seekers may be unaware of which of their skills are most marketable and how those skills conform to possible occupations. In addition, in many cases the job seeker will not know where geographically or in which industrial sectors to look for a new job. Even if one has a clear sense of which skills are marketable and where to look, there remains the question of how to present oneself in the job search. For example, it is not enough to declare oneself a job seeker in general. A provisional occupational role must be claimed, both to direct ones own efforts and to mobilize and focus the aid of supportive friends and family. Finally, the job seeker will face numerous setbacks before a new job is won, including refusals, rejections and delays. If setbacks create discouragement and hopelessness, a sustained and effective job search is unlikely. Furthermore, these setbacks can trigger more serious problems for vulnerable individuals, including depression, drinking problems, and family conflict (Kessler et al., 1988; Price, 1992). A successful job-search program must meet several formidable challenges in teaching job seeking skills. The program must recruit individuals who are shaken by their recent job loss, uncertain about their financial and career futures, and frequently haunted by self-doubts about their marketability. The program must gain the trust of these often fragile candidates, help them realize the positive aspects of assuming the role of job seeker, determine their own marketable skills, identify provisional occupational roles, appraise the barriers facing them in their job search, develop their job-seeking skills, and increase their confidence in the job search process, support them in a sometimes unfamiliar job market, inoculate them against the inevitable setbacks, and finally, help them make career decisions that will launch them into the next phase of their working lives.

12

The Michigan JOBS Program as Training in Skillful Role Enactment The Michigan Prevention Research Center developed the JOBS program, a preventive intervention in which unemployed people learn how to be skillfull job seekers (Caplan, et al., 1989; Price & Vinokur, 1995; Vinokur, Price & Schul, 1995). The program consists of a job search workshop with co-trainers who facilitate group processes oriented to maximize active learning experiences. The program helps participants recognize and assume the role of job seeker as a positive transitional role that can facilitate reemployment. The JOBS program further helps the participants discover how to draw on and play out the personal and social resources that become accessible by enacting the job seeker role. More generally, the program (1) clarifies the cognitive and procedural tasks involved in the enactment of the job seeker role, (2) teaches effective ways to exercise the skills needed to perform those tasks, and (3) inoculates the participants against setbacks by using cognitive, motivational and behavioral techniques. Taken together, these elements bolster the motivational and coping resources of the program’s participants.

Clarifying the Job Seeker Script and Learning to be a Skillful Job Seeker. The JOBS program is designed to clarify the frequently unfamiliar job seeker script, starting with how to begin the job search and finishing with a focus on how to handle the outcomes. The program begins with self and market appraisal where the candidate identifies marketable skills, assesses market opportunities and identifies alternative job and career possibilities. This stage essentially allows job seekers to assess “possible selves” (Markus & Nurius, 1986) and examine their previous work and personal life to identify transferable skills that match new job or career options. In the second stage, participants begin to engage in the job environment by searching for potential employment opportunities. Job seekers learn that conventional sources of information,

13 such as want ads, are frequently of little use and that personal and social network contacts are much more valuable (Granovetter, 1974). After identifying potential employment opportunities, the job seeker must contact those employers and create a favorable impression. The key to skillful self-presentation is learning to “think like an employer” (Curran, 1992). A job seeker must communicate aspects of his/her background and skills that would legitimize the employer’s decision to hire them. Thinking like an employer provides the participants with the appropriate perspective so they can effectively target their presentation. This new perspective also helps participants to identify weaknesses in their qualifications and to reframe them as potential strengths. For example, older workers might argue strongly that their long career reflects a substantial depth of experience, not easily obtained elsewhere, rather than a lack of energy or other attributes the employer might associate with older workers. Participants also learn how to skillfully contact prospective employers (Granovetter, 1974). One tactic involves obtaining purely informational interviews to learn more about the kind of work the organization does and what kind of skills, qualifications, training and experiences are needed. The informational interview lets prospective employers observe the skills and demeanor of a job seeker without feeling pressured to respond to direct queries about a job. Informational interviews also allow job seekers to communicate their assets and availability. In some cases, they may offer to do volunteer work or a limited scope project to demonstrate their skills and make themselves memorable to a prospective employer. Finally, a persistent and skilled job search may generate more than one job offer. Therefore, the management of job choice (Power & Aldag, 1985) is critical. Participants learn to evaluate the true quality of potential jobs to further their long-term economic interests and wellbeing and to manage relationships with prospective employers, even if a job offer is declined.

Active Learning and Inoculation Against Setbacks: Key Ingredients Active learning and inoculation against setbacks are key ingredients in the delivery of the JOBS program. These facets are essential for the skills learned in the program to transfer to the

14 actual job search. They provide participants with the motivational and coping resources to maintain a fosused and energized job search even after setbacks. Building Skills and Motivation through Active Learning. Active learning helps participants discover the resources that they possess and how to apply them most effectively to cope with their job search problems and meet their reemployment goals. The purpose of the active learning process is twofold: to maximize specific job seeking skills and to enhance both the specific job search self-efficacy and the global sense of mastery (Pearlin et al., 1981). Together, job search self-efficacy and the sense of mastery serve as motivational forces that sustain job search activities (Bandura, 1977). In the JOBS program, the group is used as the medium for implementing the active learning process. Participants practice the skills in situ during the course of the program and receive feedback from other participants and trainers to further refine their job seeking skills. For example, in learning how to perform in a job interview, participants first observe trainers modeling skilled interview behavior. Next, participants role-play the interview situation, with each person in the group serving as both interviewer and job seeker. Finally, participants offer each other advice and support to further improve tactics of self- presentation. This process exposes the participants to each other’s strengths and weaknesses and allows them to learn from and help each other in a socially supportive environment guided by trainers who provide constructive feedback. By actively engaging participants in their own learning, the program enhances feelings of mastery over the previously unfamiliar job seeker role. Bolstering Coping Resources through Inoculation against Setbacks Another key component in the coping repertoire of a job seeker is the inoculation against setbacks (Meichenbaum, 1985). Because even the most skillfully enacted roles will meet with periodic failure and rejection, the individual must anticipate setbacks in the job search and rehearse coping strategies. Participants are encouraged to diagnose possible barriers to their activities and to imagine possible rejections and failures in the job search process. They then generate and evaluate possible tactics to remove the barriers and avoid or deal with the setbacks. This process inoculates job seekers against setbacks by training them to rapidly and effectively mobilize effective coping strategies when setbacks loom ahead. The result is a repertoire of

15 responses that minimize demoralization and helplessness when the inevitable problems are encountered.

Impact of the JOBS Program The JOBS program has been studied in two large-scale, randomized, preventive trials initiated in Michigan during the past decade. In each trial, recently unemployed workers were recruited in Michigan Unemployment Offices. They were invited either to participate in the JOBS program or (in the control group) to receive a take home set of materials that contained the written content of the program. In both randomized trials, the JOBS program was conducted with groups of 15 to 20 unemployed participants using male-female trainer pairs. The program was conducted in five half-day sessions and has been described in a detailed implementation manual (Curran, 1992). In both trials, the JOBS program was delivered in community centers, school classrooms, and rented rooms in hotels or other settings that were geographically accessible to the participants. Detailed reports of both trials are available in Caplan et al. (1989) and Vinokur et al. (1995). The major findings of these trials are summarized below. Reemployment and Mental Health The results of the JOBS I trial clearly indicated that the participants of the JOBS program obtained new jobs sooner and achieved higher rates of reemployment (Caplan et al., 1989). Also, participants obtained higher paying jobs than their control group counterparts did. Over the two and a -half year follow-up, the economic advantages gained by the JOBS participants persisted and showed no indication of diminishing in comparison with the control group (Vinokur, van Ryn, Gramlich, & Price, 1991). Perhaps most important, participants in the JOBS program showed clear mental health benefits (Caplan et al., 1989; Vinokur, Price, & Caplan, 1991). JOBS participants showed significantly lower levels of psychological distress and depression when compared with their control group counterparts. The mental health benefits appeared to be connected to reemployment, and they persisted throughout the course of the initial JOBS trial. Furthermore, participation in the JOBS program appeared to have

16 psychological benefits beyond those associated with reemployment. Even participants who were unable to obtain reemployment initially continued to show higher levels of job search confidence and morale over time than did their control group counterparts, for whom job search confidence sharply eroded with continued unemployment (Vinokur, Price, & Caplan, 1991). These initial results indicate that being taught the skills needed to enact the job seeker role had a number of important consequences for the program participants, including more rapid reemployment at higher paying jobs and prevention of depression associated with prolonged unemployment. Beyond that, these initial results suggest that participation in the program increased participants’ confidence and sense of efficacy, even among those whose job search was more prolonged and who were not yet reemployed. Greater Gains for Those at Higher Risk Not all people experiencing involuntary job loss are at equal risk for serious physical and mental health problems. To identify those job losers most at risk for subsequent mental health problems, particularly depression, we conducted a set of risk analyses (Price et al., 1992 soc). These analyses indicated that participants at particularly high risk for depression were those with elevated levels of depressive symptomatology, high levels of economic hardship, and low levels of social assertiveness at baseline measurement. A reanalysis of the JOBS I trial that compared high- and low-risk participants in the JOBS program and their control group counterparts clearly indicated that the vast majority of mental health benefits of the JOBS program were for high risk participants (Price et al., 1992). Furthermore, these benefits appeared to occur for those high- risk participants who would have otherwise developed the most severe depressive symptoms had they not been exposed to the JOBS program. To provide more rigorous causal evidence for these findings about high- risk participants, the JOBS II trial was conducted as a prospective trial in which high-risk participants, as well as lower risk participants, were randomly assigned to the experimental (JOBS) and control conditions. Results of the JOBS II trial clearly replicated the reemployment and mental health findings of JOBS I. Participation in the JOBS program promoted higher levels of reemployment and mental health. Perhaps more important, however, JOBS II confirmed the benefits of JOBS for high-risk participants. The high-risk participants received the majority of benefits. While

17 both high- and low-risk participants experienced an increased sense of mastery after participating in the JOBS program, reemployment rates and mental health benefits were considerably higher for high- risk participants. High-risk people, left to their own devices, are most likely to lose confidence, remain unemployed, and periodically fall into episodes of depression. It is precisely these high- risk people, however, who also show the greatest preventive benefits from the skills and confidence gained through the JOBS program (Vinokur et al., 1995). The JOBS II trial also collected data from the spouses and significant others of job losers about the distress and social and emotional functioning of JOBS participants and controls. These data from a “natural rater” in the environment of job losers corroborate the self-reports of job losers themselves, providing external validity for the impact of the JOBS program. Benefits and Costs of the JOBS Program A benefit-cost analysis conducted by Vinokur, van Ryn, Gramlich, and Price (1991) indicated that the JOBS program not only produced substantially higher personal income for JOBS participants, but also generated higher state and federal tax revenues. These findings suggest that adoption of the program by government-based human services would pay for itself in less than a year. Some may fear that the JOBS program, in promoting reemployment for one group within society, is simply taking jobs from others, but there is reason to believe that such displacement effects are minimal. Labor economists Davidson and Woodbury (1993) have examined the issue of displacement effects of programs that promote reemployment among workers who lost their jobs and qualify for unemployment compensation. They found that the displacement effects of the programs were minimal and were offset by the overall improvement in the economy that results from reducing the duration of job vacancies. This accelerated job filling, in turn, leads to the creation of more job opportunities through increases in economic efficiency. Thus, if such programs succeed in reemploying people more quickly or in helping people obtain jobs more suitable for their skills, they enhance the efficiency of the labor market and ultimately contribute to economic growth.

18 Mechanisms of Action: Reduced Financial Strain and Increased Mastery The broad outlines of some of the main mechanisms of action of the JOBS program on mental health are beginning to emerge. Specifically, financial strain is the critical mediator between unemployment and poor mental health (Kessler et al., 1987; 1988; Vinokur, Caplan & Williams, 1987). Financial strain has also been found to be a mediator between stressful life events such as widowhood and negative mental health outcomes (Umberson et al., 1992). Other research has demonstrated that reemployment restored mental health to pre-unemployment levels by reducing economic hardship and financial strain (e.g., Kessler et al., 1988). The JOBS intervention had significant effects on reemployment rates, monthly income, and mental health. Therefore the JOBS program is most likely to help improve mental health by facilitating reemployment, which restores income and reduces economic hardship. Findings from the JOBS trials also suggest that the program facilitates reemployment by enhancing feelings of mastery and enabling participants to engage in a skilled and persistent job search. Our results suggest that an enhanced sense of mastery is a likely mediating mechanism between experience in the JOBS program and reemployment. Analysis of the JOBS II preventive trial by Vinokur, Price, & Schul (1995) indicated that respondents who participated in the JOBS program showed increased levels of mastery. Thus, a likely mechanism of action is that JOBS provides needed job seeking skills, inoculation against setbacks, and an increased sense of mastery that facilitates reemployment, decreases economic hardship, and improves mental health. The intervention is particularly successful for high-risk respondents who are less able to mobilize psychological resources for a successful job search.

Toward Tests of Generalizability and Implementation in the Community Both from a theoretical and a practical perspective, these initial results suggest the need for a second generation of preventive trials to examine generalizability across various populations and community contexts. For example, results of both JOBS trials demonstrated that high-risk

19 respondents (those with initially higher levels of depressive symptoms and economic hardship) benefit most from the intervention, suggesting that efficiency and cost- effectiveness might be increased by recruiting only high-risk participants. However, the effectiveness of the JOBS intervention appears to depend on the supportive participation of low-risk participants. Future effectiveness trials are needed to test this hypothesis. A second relevant example involves the role of spouses and significant others. Supportive and undermining behavior from an intimate partner is consequential for effective enactment of the search role and for the mental health of the couple (Howe, et al., 1995; Vinokur, et al., 1996). This suggests a second kind of trial, enlisting the cooperation and support of the spouse or significant other in the JOBS program itself. A set of preventive trials is under way to test this hypothesis and will allow replication of a new JOBS intervention for couples in Michigan and Maryland, states with quite different labor markets. Beyond questions of generalizability are questions of implementation of preventive interventions in the community. This third generation of prevention research poses questions about cultural differences in community receptivity and organizational readiness for the adoption of innovative prevention programs (Price & Lorion, 1988; Price & Vinokur, 1995). For example, we need to know a great deal more about the fit between preventive programs such as the JOBS program, thus far delivered in relatively ideal conditions, and existing agencies and communities with different resources, cultures, and organizational capabilities.

Conclusion: Preventive Trials as Experiments in Life Course Development Job loss is a psychosocial condition that can lead to deleterious emotional, behavioral, and physical effects. It entails the loss of a valued role in society that carries with it crucial material and personal resources. It often results in economic hardship and corrodes one’s sense of mastery, personal identity and close relationship. This assault on material and personal

20 resources in turn produces a broad array of impacts on physical and mental health, not only for the job loser, but often for family members as well. Job loss also represents a turning point in life. Left to their own devices, some job losers will successfully claim a new work role in society and reap all of the protective economic and psychological resources that a good job provides. Many others, overwhelmed by the cascade of negative financial events and an eroded sense of mastery, will suffer discouragement and depression and may become a burden or source of stress to their families. The transitional role of job seeker can facilitate coping with unemployment and the passage to reemployment. The job seeker role is challenging and requires persistent, skilled role performance under conditions in which setbacks and rejection are the rule rather than the exception. The JOBS preventive trials show that skillful and persistent use of the job seeker role can be taught, resulting in positive outcomes for individuals, families, and communities. Individuals who enact the job seeker role skillfully experience less psychological distress and depression and greater economic gains, and their family members exhibit more supportive relationships and lower likelihood of divorce and spousal abuse. Communities could experience economic growth and generate greater tax revenues. When turning points in life, such as job loss, are positively influenced under the controlled conditions of a preventive trial, there are scientific benefits as well. In the past, our understanding of the mechanisms and impact of social adversity on health and mental health has relied heavily on the results of social and epidemiological survey studies, sometimes supplemented by intensive qualitative research (Kessler, Price, & Wortman, 1985). Longitudinal studies, in particular epidemiological studies, have uncovered antecedents of physical and mental health and have confirmed the importance of personal and social resources, such as mastery and social support, in understanding the stress process and its sequelae. We believe that preventive trials such as the JOBS studies provide a different kind of opportunity to understand the antecedents, mechanisms, and outcomes of the stress process. Their special contribution is to provide an opportunity to actively and positively intervene in the lives of vulnerable persons under controlled conditions. Preventive trials such as the JOBS trials are experiments in life course development. Like other experiments, they afford an opportunity

21 for greater control over hypothesized antecedents, and therefore provide a firmer grasp of causal processes influencing developmental and health outcomes. In the case of the JOBS trials, for example, our experimental results, in conjunction with longitudinal analyses of mediational processes, have increased our confidence about the decisive role of economic hardship and mastery in influencing the occupational and health outcomes of job loss. Our understanding of the stress process for a number of adverse psychosocial conditions may have now reached the point where a number of such preventive experiments to influence the life course are possible. If so, we may be on the threshold of a new generation of studies on coping with adversity that will provide the dual benefit of scientific understanding and social improvement.

22

References

Atkinson, T., Liem, R., & Liem, J. (1986). The social costs of unemployment: Implications for social support, Journal of Health and Social Behavior, 27, 317-331. Azrin, N. H., & Beasalel, V. B. (1982). Finding a job. Berkeley, CA: Ten Speed Press. Baker, W. E., & Faulkner, R. R. (1991). Role as resource in the Hollywood film industry, American Journal of Sociology, 97, 279-309. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change, Psychological Review, 84, 191-215. Beasley, J. D. (1991). The betrayal of health: The impact of nutrition, environment, and lifestyle on illness in America. New York: Times Books. Bebbington, P., Hurry, J., Tennant, C., Stuart, E. Y., & Wing, J. K. (1981). Epidemiology of mental disorders in Camberwell, Psychological Medicine, 11, 561-579. Bolles, R. N. (1995). What color is your parachute? A practical manual for job-hunters & career-changers. Berkeley, CA: Ten Speed Press. Bolton, W., & Oatley, K. (1987). A longitudinal study of social support and depression in unemployed men, Psychological Medicine, 17, 453-460. Broman, C. L., Hamilton, V. L., & Hoffman, W. S. (1990). Unemployment and its effect on families: Evidence from a plant closing study, American Journal of Community Psychology, 18, 643-659. Callero, P. L. (1992). The meaning of self-in-role: A modified measure of role-identity, Social Forces, 71, 485-501. Callero, P. L. (1994). From role-playing to role-using: Understanding role as resource, Social Psychology Quarterly, 57, 228-243.

23 Caplan, R. D., Vinokur, A. D., Price, R. H., & van Ryn, M. (1989). Job seeking, reemployment, and mental health: A randomized field experiment in coping with job loss, Journal of Applied Psychology, 74, 759-769. Catalano, R. (1991). The health effects of economic insecurity, American Journal of Public Health, 81, 1148-1152. Catalano, R., & Dooley, D. (1977). Economic predictor of depressed mood and stressful life events in a metropolitan community, Journal of Health and Social Behavior, 18, 292-307. Catalano, R., Dooley, D., Novaco, R., & Wilson, G. (1993). Using ECA survey data to examine the effect of job layoffs on violent behavior, Hospital and Community Psychiatry, 44, 874879. Catalano, R., Dooley, D., Wilson, G., & Hough, R. (1993). Job loss and alcohol abuse: A test using data from the Epidemiologic Catchment Area project, Journal of Health and Social Behavior, 34, 215-225. Conger, R. D., Elder, G. H., Lorenz, F. O., Conger, K. J., Simons, R. L., Whitbeck, L. B., Huck, S., & Melby, J. N. (1990). Linking economic hardship to marital quality and instability, Journal of Marriage and the Family, 52, 643-656. Curran, J. (1992). JOBS: A manual for teaching people successful job search strategies. Ann Arbor: Michigan Prevention Research Center, Institute for Social Research, University of Michigan. Davidson, C., & Woodbury, S. A. (1993). The displacement effect of reemployment bonus programs, Journal of Labor Economics, 11, 575-605. Dew, M. A., Penkower, L., & Bromet, E. J. (1991). Effects of unemployment on mental health in the contemporary family, Behavior Modification, 15, 501-544. Dooley, D., Catalano, R., & Wilson, G. (1994). Depression and unemployment: Panel findings from the Epidemiologic Catchment Area study, American Journal of Community Psychology, 22, 745-765. Elder, G. H. Jr.,, & Caspi, A. (1988). Economic stress in lives: Developmental perspectives, Journal of Social Issues, 44, 25-45. Elder, G. H., Jr., Gimbel, C., & Ivie, R. (1991). Turning points in life: The case of military service and war, Military Psychology, 3, 215-231.

24 Ezzy, D. (1993). Unemployment and mental health: A critical review, Social Science & Medicine, 37, 41-52. Ganster, D. C., & Schaubroeck, J. (1991). Work stress and employee health. Special Issue: Yearly review of management, Journal of Management, 17, 235-271. Gil, D. G. (1970). Violence against children: Physical abuse in the United States. Cambridge, MA: Harvard University Press. Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewoods Cliff, NJ: Prentice-Hall. Gottman, J. M. (1993) A theory of marital dissolution and stability, Journal of Family Psychology, 7, 57-75. Granovetter, M. S. (1974). Getting a job: A study of contacts and careers. Cambridge, MA: Harvard University Press. Hasenfeld, Y. (1983). Human service organizations. Englewood Cliffs, NJ: Prentice-Hall. Howe, G. W., Caplan, R., Foster, D., Lockshin, M., & McGrath, C. (1995). When couples cope with job loss: A research strategy for developing preventive intervention. In L. R. Murphy, J. J. Hurrell, Jr., S. L. Sauter, & G. P. Keita (Eds.), Job stress interventions, (pp. 139-158). Washington, DC: American Psychological Association. Hughes, E. C. (1945). Dilemmas and contradictions of status, American Journal of Sociology, 1, 353-359. Jackson, P. R., & Warr, P. B. (1984). Unemployment and psychological ill-health: The moderating role of duration and age, Psychological Medicine, 14, 605-614. Jahoda, M. (1981). Work, employment and unemployment: Values, theories, and approaches in social research, American Psychology, 36, 184-191. Justice, B., & Duncan, D. F. (1977). Child abuse as a work-related problem, Corrective and Social Psychiatry and Journal of Behavior Technology, Methods and Therapy, 23, 53-55. Kahn R. L., & Byosiere, P. (1992). Stress in organizations. In M. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd ed., pp. 571-650). Palo alto, CA: Consulting Psychologists Press. Kessler, R., Price, R., & Wortman, C. (1985). Intervening processes in the relationship between unemployment and health, Psychological Medicine, 17, 949-961.

25

Kessler, R., Turner, J., & House, J. (1987). Intervening processes in the relationship between unemployment and health, Psychological Medicine, 17, 949-961. Kessler, R. C., Turner, J. B., & House, J. S. (1988). The effects of unemployment on health in a community survey: Main, modifying, and mediating effects, Journal of Social Issues, 44, 69-86. Kessler, R. C., Turner, J. B., & House, J. S. (1989). Unemployment, reemployment and emotional functioning in a community sample, American Sociological Review, 54, 648657. Kowalik, D. L., & Gotlib, I. H. (1987). Depression and marital interaction: Concordance between intent and perception of communication, Journal of Abnormal Psychology, 96, 127-134. Liem, R., & Liem, J. H. (1988). The psychological effects of unemployment on workers and their families, Journal of Social Issues, 44, 87-105. Markus, H., & Nurius. P. (1986). Possible selves, American Psychologist, 41, 954-969. Meichenbaum, D. (1985). Stress inoculation training: A clinical guidebook. New York: Pergamon. Menaghan, E. G. (1991). Work experiences and family interaction processes: The long reach of the job, Annual Review of Sociology, 17, 419-444. Parke, R., & Collmer, C. (1975). Review of child development research, Child Abuse: An Interdisciplinary Review. Chicago: University of Chicago Press. Pearlin, L. I. (1989). The sociological study of stress., Journal of Health and Social Behavior, 30, 241-256. Pearlin, L. I., Lieberman, A., Menaghan, E. G., & Mullan, J. T. (1981). The stress process, Journal of Health and Social Behavior, 22, 337-356. Penkower, L., Bromet, E., & Dew, M. (1988). Husbands’ layoff and wives’ mental health: A prospective analysis, Archive of General Psychiatry, 45, 994-1000.

Pickles, R., & Rutter, M. (1991). Statistical and conceptual models of “turning points” in developmental processes. In L. R. Bergman, G. Rudinger, and B. Törestad (Eds.), Problems

26 and methods in longitudinal research: Stability and change (pp. 133-165). Cambridge, England: Cambridge University Press. Platt, S., & Kreitman, N. (1985). Parasuicide and unemployment among men in Edinburgh, 1968-1982, Psychological Medicine, 15, 113-123. Pollitt, E. (1994). Poverty and child development: Relevance of research in developing countries to the United States. Special Issue: Children and poverty, Child Development, 65, 283-295. Power, D. J., & Aldag, R. J. (1985). Soelberg's job search and choice models: A clarification, review and critique. Academy of Management, 10, 48-58. Price, R. H. (1990). Strategies for managing plant closings and downsizing. In D. Fishman & C. Cherniss (Eds.), The human side of corporate competitiveness (pp. 127-151). Beverly Hills, CA: Sage Publications. Price, R. H. (1992). Psychosocial impact of job loss on individuals and families, Current Directions in Psychological Science, 1, 9-11. Price, R. H., & Lorion, R. P. (1989). Prevention programming as organizational reinvention: From research to implementation. In D. Schaffer, I. Phillips, & N. B. Enzer (Eds.),[M.M. Silverman and V. Anthony (Assoc. Eds.)] Prevention of mental disorders, alcohol and drug use in children and adolescents (pp. 97-123). (DHHS Publication No. ADM 89-1646). Washington D.C.: U.S. Government Printing Office. Price, R. H., van Ryn, M., & Vinokur, A. D. (1992). Impact of preventive job search intervention on the likelihood of depression among the unemployed, Journal of Health and Social Behavior, 33, 158-167. Price, R. H., & Vinokur, A. D. (1995). Supporting career transitions in a time of organizational downsizing: The Michigan JOBS Program. In M. London (Ed.), Employees, careers and job creation: Developing growth-oriented human resource strategies and programs. (pp. 191-209). San Francisco, CA: Jossey-Bass. Steinberg, L. D., Catalano, R., & Dooley, D. (1981). Economic antecedents of child abuse and neglect, Child Development, 52, 975-985. Thoits, P. A. (1991). On merging identity theory and stress research, Social Psychology Quarterly, 54, 101-112. Thoits, P. (1994). Stressors and problem-solving: The individual as psychological activist, Journal of Health and Social Behavior, 35, 143-159.

27

Turner, J. B. (1995). Economic context and the health effects of unemployment, Journal of Health and Social Behavior, 36, 213-229. Turner, R. (1978). Role and the person, American Journal of Sociology, 84, 1-23. Turner, R. J., & Avison, W. R. (1992). Innovations in the measurement of life stress: Crisis theory and the significance of event resolution, Journal of Health and Social Behavior, 33, 36-50. Umberson, D., Wortman, C., & Kessler, R. (1992). Widowhood and depression: Explaining long-term gender differences in vulnerability. Journal of Health and Social Behavior, 33, 10-24. U.S. Department of Labor (1995). [Tables.] Monthly Labor Review, 118(7), 98-99. Vinokur, A., & Caplan, R. D. (1987). Attitudes and social support: Determinants of job-seeking behavior and well-being among the unemployed, Journal of Applied Social Psychology, 17, 1007-1024. Vinokur, A., Caplan, R. D., & Williams, C. C. (1987). Attitudes and social support: Determinants of job-seeking behavior and well-being among the unemployed, Journal of Applied Social Psychology, 17, 1007-1024. Vinokur, A. D., Price, R. H., & Caplan, R. D. (1991). From field experiments to program implementation: Assessing the potential outcomes of an experimental intervention program for unemployed persons. American Journal of Community Psychology, 19, 543-562. Vinokur, A. D., Price, R. H., & Caplan, R. D. (1996). Hard times and hurtful partners: How fnancial strain affects depression and relationship satisfaction of unemployed persons and their spouses, Journal of Personality and Social Psychology, 71, 166-179. Vinokur, A. D., Price, R. H., & Schul, Y. (1995). Impact of the JOBS intervention on unemployed workers varying in risk for depression, American Journal of Community Psychology, 23, 39-74. Vinokur, A. D., van Ryn, M., Gramlich, E., & Price, R. H. (1991). Long-term follow-up and benefit-cost analysis of the JOBS program: A preventive intervention for the unemployed, Journal of Applied Psychology, 76, 213-219. Warr, P. B., Jackson, P. (1988). Men without jobs: Some correlates of age and length of unemployment. Journal of Occupational Psychology, 57: 77-85. Warr, P. B., Jackson, P., & Banks, M. (1988). Unemployment and mental health: Some British studies, Journal of Social Issues, 44, 47-68.

28

Wheaton, B. (1990). Life transitions, role histories, and mental health, American Sociological Review, 55, 209-223. Whyte, W. H., Jr. (1956). The organization man. New York: Simon & Schuster. Windschuttle, K. (1980). Unemployment: A social and political analysis of the economic crisis in Australia. Ringwood, Victoria: Penguin. clw C:/turnpaper.118

New Directions. Stressors and the World of Work Patrick C. Heggy Heggy Leadership Enterprises

The world of work is in a state of transformation. The modern global economy is shifting from a focus on “atoms” to a focus on “bits,” shifting from a focus on physical products to a focus on ideas and information. This transformation is far from a smoothly integrated, welldesigned process. It is a process composed of bumps and jolts and has created many casualties. Tomorrow’s workplace promises even more of this discontinuity and an even faster rate of change. Reengineering, downsizing, outsourcing, contingent workers, telework-- all are terms that reflect structural alterations in what work is and how it is to be done. These changes have enormous implications for the physical and mental health of Americans. The challenges of the transforming workplace are addressed in the preceding chapters, each highlighting a particular mental health concern stemming from the changing world of work: Rand Conger and Martha Reuter focus on the stressors and consequences of unemployment; Rick Price, Dan Friedland, and Amirim Vinokur on the transition from unemployment to reemployment; and Dan Ganster on key stressors associated with work. I will examine their perspectives and forecast other important changes for the coming decade.

Job Loss The stress caused by loss of employment has important health effects. For the foreseeable future, job loss will be an almost routine part of the modern economy. The lifelong, singlecompany employment model that existed earlier in the 20th century is gone. Employees are

struggling to cope with corporate uncertainties and are changing their expectations about jobs (“World trends and forecasts,” 1996). Global competition, fueled by information technologies, has created a ceaseless pressure on corporations to get more and more productivity from fewer employees. Many of these newly unemployed millions will come from upper and middle management, and they will have little hope of finding comparable positions (Rifkin, 1995). As Conger and Reuter explain, such circumstances create economic pressures that lead to emotional distress. Conflicts in finances can, in turn, lead to decreased marital quality stressors. Yet Conger and Reuter’s work indicates that this is a reciprocal relationship and that marital empathy can serve as a buffer to emotional distress. The positive implications of these marital skills for mental health may be offset by America’s rising divorce rate, however. More than half of all first marriages in the United States now end in divorce, depriving many couples of a potentially supportive relationship in an era of high workplace turmoil (Boyette and Boyette, 1995). The significance of Conger and Reuter’s work for future workers-- today’s children-- lies in teaching them the problem solving skills involved in generating solutions. The expansion of those skills might be expected to mitigate mental distress, as well as to enhance the buffering effect of marital stability and marital empathy.

Job Seeking In the future, new technologies will allow companies to eliminate or automate more jobs. These new technologies will also simplify and speed the process of finding workers for specific, immediate, transitory corporate needs. These two features will combine to thrust more people into the role of perpetual job seeker. Not only will more people be competing for fewer jobs, but many individuals will also confront the challenge of learning job-seeking skills. Rick Price, Dan Friedland, and Amiram Vinokur explore this stressor and describe how job loss can hinder a person’s success in re-entering the world of employment. Job loss lowers the individual’s status and threatens mental health by disrupting an established network of friendships and social supports. It demands that the unemployed individual claim a new role, that of job seeker. The transition to this new role can itself act as a stressor.

In the future, as corporate restructuring efforts combine with the expansion of technology, more people will find themselves passing through the role of job seeker. For some, such as contingent or temporary workers, the role of job seeker could become almost permanent. Contingent workers, contract consultants, or “temps,” currently constitute nearly 30 % of the American workforce. By the year 2000, that figure could climb as high as 50 percent (Coates, 1995). More and more people will experience periods of repeated or extended unemployment. When such a large percentage of the workforce is maintained in this state of role uncertainty, mental health concerns may become paramount. Price and Friedland identify several skills that the job seeker can use to mitigate the mental health consequences of job seeking stresses. These skills engage the job seeker in active learning, inoculate him or her against the likely setbacks that will routinely occur, and enhance a sense of personal mastery. The JOBS program described by Price and his colleagues treats the job seeker role as a distinct identity that requires distinct skills. The inclusion of role identity as a key element of stress management is a valuable insight. Today’s social definitions and expectations surrounding the world of work are likely to change in the future. As large numbers of workers become excluded from traditional concepts of what it means to have a job, the identities they create will change. Work and its attendant significance for personal status and self-esteem may be disaggregated by the highly educated but intermittently employed workers of the early 21st century. The majority of workers will be pushed through the roles of job loser, job seeker, and jobholder, and virtually no one will be immune to this sequence. Being unemployed may come to be seen as a natural, transitional period and not an indication of personal failure. An increasing number of workers can be expected to move from high-paying information economy jobs to unemployment. No position and no specific job skill will guarantee security and well-paid employment. In the information economy, even people with proficient skills as symbolic analysts or knowledge workers will find themselves on the outside of the formal economy on a regular basis. This situation is likely to divide workers into those who have well paid jobs (the Haves), those who have the skills to hold such jobs but can currently find no position (the Used to Haves), and those who lack complex symbolic analytic or knowledge, as well as access to the training necessary to attain those skills (the Can’t Haves).

Under these circumstances, attitudes toward paid employment will change. Being unemployed may come to be seen as a natural, transitional period and not an indication of personal failure. The unemployed in the Depression era believed they were at fault for their unemployment, but the workers of the early 21st century could have a postmodern sense of identity that de-stigmatizes unemployment (Levine, 1985). This phenomenon already has a name, downshifting, and is a choice of a growing number of America’s workers. More and more people are willingly trading paid employment time for a less stressful lifestyle (Nardone, 1995; Bowman, 1996). As this trend continues, it may help mitigate the stressors created by the frenetic pace of change in work and jobs. A post-industrial perspective is forming in which people will expect to make identity shifts. In fact, attempting to maintain a stable, coherent identity in an economy dominated by continuous change may be maladaptive and actually generate rather than relieve stress. Individuals who are too consistent in their identities and reactions are probably immured in inflexible environments. In the future, identity recreation will be adaptive, and identity constancy will be dysfunctional (Anderson, 1995). During the transition to the information age economy, people can be expected to continue to pursue the receding industrial jobs of the 20th Century. As they do so, programs such as the JOBS described by Price, Friedland, and Vinokur will provide them with valuable job-seeking skills. Nonetheless, researchers and mental health professionals must remain aware of the stress that can be created by attempting to apply Industrial Age tools to an information economy. Both, corporations and individuals will need to adjust continually to a transforming work environment (Collins & Perras, 1994). Therefore, a future-oriented perspective and futures studies should be an integral part of any training program. For example, how job-seeking skills are obtained will change. Information technologies, which will include on-line, video-based interactive expert systems, will augment learning of the face-to-face interactions in the JOBS program, such as market appraisal and the search for potential employment opportunities. Through these computer programs on job skills training, more of the unemployed will have access to the information they need to re-assert themselves in the workforce. One of the challenges is to ensure that those who can learn job search skills also have the skills to perform jobs in the global information economy of the early 21st Century. Fortunately,

some of the skills addressed by Price, Friedland, and Vinokur will have value for tomorrow’s workers. For example, the JOBS program’s emphasis on learning and personal mastery can provide tomorrow’s workers with key skills. Because jobs are changing so rapidly, learning particular skills is perhaps less important than learning how to learn and having confidence in oneself (Swobodar, 1996). .Price and Friedland thoughtfully address the resources needed to conduct such life course training. They identify the differing benefits received by high- and low-risk participants and the attributes of an organization ready to undertake such training. Hypothetically, the positive mental health benefits of the JOBS training program will apply in many cases. The sophisticated job search tools that will be available in the future may simplify the role of job searcher, eliminating many of the challenges that today’s job seekers face. For example, sophisticated electronic searching tools will be available to employers and job seekers will change the way “matches” are discovered. Search programs will be launched into electronic databases to retrieve resumes that meet established criteria. One already existing group, OnLine Solutions ([email protected]), helps individuals format their resumes specifically for Internet utilities. An employer seeking to add a worker to do long-range planning, for instance, might require 5 years of experience with a Fortune 500 company and publication credits in Harvard Business Review and Forbes. The computer search program would automatically retrieve all matches. These programs, also known as “knowbots” or Artificial Intelligent agents (AIgents), will learn from previous search protocols to automatically modify the search criteria in response to search failures or an overabundance of matches. The application of these tools will make possible nearly flawless and instantaneous matches for employee qualifications, interests, experience, and customer needs. The term “job search” will be replaced by the more collaborative concept of “job matching.” These types of automated handshaking programs may alter the skills required to regain employment. That is, they highlight the actual talents of the individual more than the secondary skills that are provided as part of the JOBS program, such as clarifying the job seeker script. Additionally, job seekers will be insulated from rejection because the rejections will be invisible. Rejections will simply be matches that are not made via software utilities, rather than face-to-face denials. These utilities will also advance the capacity of job seekers to align themselves with particular kinds of work. Because the employer-employee relationship will be less

geographically bound, broader pools of potential employers will become available. Therefore, job seekers will be able to define more precisely the type of work they enjoy without unduly constraining the number of potential employers. The employer-employee match will become more accurate. Information or knowledge workers will be able to seek jobs that best match their personal preferences. As a result, more employees will bring a higher level of personal commitment to their work, which implies greater engagement and stronger social support, factors that can offset workplace stressors. In the first decade of the 21st century, these technologies will be available but will not be ubiquitous and so other forms of validation will be needed. For those employers and job hunters without access to sophisticated expert systems or multimedia technologies, third party intermediation will be a likely alternative. This intermediation may come from neutral clearinghouses for employee skills and qualifications that will serve as reference checkers whose impartial seal of approval validates a job hunter’s credentials. Reference validation organizations might also participate in the matching process, taking on the current role played by headhunters in many professions. It is also likely that the dominant temporary agencies of the mid-1990s will take on that clearinghouse role. Doing so will involve an expansion of services that are already being offered -- employee skill screening. The difference is that in tomorrow’s fast-paced economy, the employer will simply log onto the temporary agency’s databases to select the individual who best matches the needs of the moment.

Job Holding Mental health challenges, of course, do not disappear once an individual enters the workforce. Dan Ganster’s chapter describes how the modern workplace exacerbates stressful job demands. The increased percentage of working single parents and families with both parents working heightens the work-family conflict, which can lead to job dissatisfaction. Ganster also emphasizes the enduring nature of job insecurity and its relationship to stress. Even many of those employees who enjoy ongoing, long-term paid employment will feel the strain of the economic distress detailed by Conger and Rueter. The labor saving capacity of information software, such as expert systems helps facilitate a decline in wages and shift of

income to the more wealthy (Allen, 1996).

Financial concerns will be a constant source of

stress for most workers, even for many of the Haves. How can these stressors be mitigated? Do any workplace trends work in favor of jobholders? Yes, employee empowerment. For companies to keep pace with rapidly changing customer preferences and innovative competitors, they will have to unleash the full productivity of their workforces. To do so, they will need to empower individual employees by giving them the decision-making authority to accomplish corporate objectives. In addition, the decision to gain productivity advantages through the use of information technologies makes the remaining human contributors relatively more precious. Therefore, companies will need to place more emphasis on human engineering because the failure of one component -- one person-- can be critical for corporate survival. With fewer permanent workers, companies will play an expanded role in developing their workforces. It will be in their interest to keep employees’ talents abreast of an ever-changing work environment. Individuals in the highly educated workforce of the 1990s and early 21st Century know that they have the skills to accomplish complex, creative tasks. They will expect-- they will demand-- a voice in work decisions. Information technologies such as groupware are facilitating the collection of input from diverse individuals, thereby giving more and more people a voice in what work they do or do not do. The speed of marketplace changes will also contribute to the expansion of employee empowerment. No longer will employers have the luxury of allowing information to flow up and down a hierarchical chain of command. The people who are in closest contact with customers and the relevant information will be expected to make more and more decisions. If they do not, opportunities will be lost and competitors with less cumbersome management processes will gain the advantage. Purely for business survival reasons, employee empowerment could become a reality in the workplace of the early 21st century. This shift would yield both marketplace and health benefits. As Dan Ganster outlines, workers who possess both “decision authority” and “skill discretion” suffer less from harmful stress levels. People who care about their work enjoy higher levels of psychological satisfaction as well as greater productivity. Employees, who perceive that their employers trust them and are concerned with their needs, demonstrate a greater commitment to the shared goals and vision of the organization (Rosen, 1995).

Implications for Individuals The implications of the three areas highlighted by these chapters-- job loss, job seeking, and job holding—can be expected to differ for the Haves, the Used-to-Haves, and the Can’t Haves. Unemployment, for example, is a constant for the Can’t Haves. They are excluded from the information economy, and acquiring the symbolic skills needed to become viable knowledge workers will be very difficult. Moving from Can’t Have to Have can be viewed as trying to climb a ladder from which all of the bottom rungs have been removed. This group can be expected to suffer fully from the economic and associated health consequences of unemployment. Because they are excluded from paid work, they will often seek other means of creating an identity. The stressors for this group will stem from the environment of poverty and will be increasingly dissociated from any links to well-paid employment. The people most vulnerable to these trends will be the new workforce entrants, minorities, the lesser educated, and the poor. This socioeconomic divergence between the Haves and the Can’t Haves may have major consequences. As the gap grows between rich and poor, the rich may increasingly want to live a life apart from the poor. The Haves may come to feel threatened by the expanding cohort of the Can’t Haves (Frank and Cook, 1995). Cut-off from access to employment, the Can’t Haves may actively reject the system that has rejected them. Already we see increases in indicators of such anti-system activities as poverty-driven crime, gangs, and drug culture (Didsbury, 1996). These choices merely switch one set of stressors (unemployability) for another (the threat of violence). Stemming this trend will be difficult, but perhaps such programs as JOBS can enable Can’t Haves to acquire personal mastery skills. The value of acquiring these skills may be less in preparing for re-employment -- a dearth of jobs may be an endemic feature of the early 21st Century workplace -- than in limiting the domain of life in which income through paid employment matters (Robertson, 1990). The JOBS program and its progeny may be able to maintain a cultural link between the Haves and Can’t Haves. The mental health benefits of such programs as JOBS-- their ability to increase self-confidence and personal mastery-- could prepare the Can’t Haves to contribute in a creative rather than a destructive way to changing

systemic income inequality, thereby restoring agency to those who are currently excluded from the economy. The Used-to-Haves, those individuals displaced by technology or a suddenly obsolete profession, however, will feel the stress of loss of identity and status that both Ganster and Price, Friedland and Vinokur describe. They can also be expected to feel economic stress as they are pushed into sporadic, low paying assignments. Their lack of control over the nature and timing of these assignments will make them prime candidates for the “untoughening” effect that Ganster discusses. Additionally, as they become marginalized and only intermittent participants in the information economy, their skills will begin to lose relevance in an everchanging marketplace. Over time, the Used-to-Haves risk slipping into the Can’t Have category. Stressors for the Haves will center in work. Recognizing the void that awaits them outside of paid employment, they will feel enormous pressure to attain higher standards of performance. With the shift to employee empowerment comes the expectation to produce results. Workers can expect a continuation of the trend toward longer hours and rising performance demands (Shor, 1992).

More than 45% of workers now contend that they are overwhelmed at work and

under more stress than they were only 3 years ago (Gillian, 1995). “Stress has become one of the most serious health issues of the 20th Century” (Rifkin, 1995). This concern can be expected to continue into the 21st Century as well. The desire of many companies to reduce health care costs could have the indirect effect of mitigating workplace stress. Clearly, companies whose management is insensitive to the needs of their employees will suffer greater health-related losses than those organizations whose management is more focused on the value of relationships within the work context (Courter, 1995). Health Maintenance e Organizations may also seek to reduce expenses by cutting demand for services, thereby creating incentives for companies to create less stressful work operations. Another possibility is that national health care policy will extend further into the workplace. For instance, the Occupational and Safety Health Administration’s regulations might begin to cover nonphysical aspects of work conditions, guided by validated findings of mental health researchers.

Part of the health maintenance process for individuals may well include personal biomonitors, small, computerized noninvasive devices that can be worn unobtrusively and that monitor bodily processes and biochemistry. The miniaturization of the circuit chip and other computing devices will turn such everyday items as watches, necklaces, earrings and eyeglasses into bio-monitoring devices that measure pulse, blood pressure, blood glucose levels, and O2 content. Prototypes of eyeglasses that are also computer screens have been developed. These hold particular promise for bio-monitors because the eye is the only place on the body where one can actually see the blood flow-- it is a window into as well as out of the body. The use of bio-monitors can make individuals aware of their peripheral catecholamines or cortisol levels, both important health measures as Dan Ganster points out. Related advances in the pharmaceutical industry and the Human Genome Project may contribute to quicker recognition of mental health problems. For example, oncologists have clearly linked changes in the p53 gene to a range of cancers. Inexpensive self-testing kits are already on the market, such as the p53 GeneKit by Visible Genetics, Inc., which allows users to quickly detect possible mutations in the DNA code of the p53 gene. In the future, biochemistry and genetics may identify a tool chest of early biomarkers for stress and mental disorders. Once made aware of their physiological response to a stressor, people will be able to take immediate action. They may recognize stressful situations as opportunities for personal health promotion and maintenance, rather than merely accepting such situations as unalterable. Biomonitors could help restore personal responsibility and agency to stress-laden circumstances, thereby decreasing the chronic acceptance of unhealthy levels of stress and increasing the use of effective coping responses.

Next Steps: Issues and Opportunities Several promising preventive measures emerged in these chapters. Reducing social stressors and subsequent mental health problems can be usefully approached from either a fundamental cause or a mediator model. The following measures are needed: 1.

More specific information is needed on high risk groups, businesses, and

occupations. Preventive interventions can then be cost-effectively targeted at those high-risk individuals, and their workloads and stress levels can be adjusted accordingly.

2.

Improvements are needed in the procedures for information exchange and

coordination of functions and activities between medical and non-medical services. Full advantage should be taken of the information-gathering capacities of HMO’s and the linking capacities of tomorrow’s information technologies. 3.

Companies need ongoing information and training for managers, personnel

departments, and employees aimed at heightening awareness and recognition of the signs of excessive mental pressure and stress. Until reliable biomarkers can be identified, greater knowledge and new attitudes toward stress and mental health will be the best early warning system. 4.

Mental health research and practice teams should be multidisciplinary.

Investigators need to accommodate holistic health perspectives. Just as tomorrow’s corporations are viewing themselves as elements of a business ecosystem, the definition of health is broadening. Multidisciplinary health care approaches will become the norm. Mental health research teams will include biochemists, human genome researchers, economists, and perhaps futurists. 5.

The growing influence of holistic perspectives is part of a larger trend that

recognizes the expanded interconnectivity and complexity of modern society. No longer will the industrial paradigm of cause and effect be sufficient. Instead, discussion of interventions will center on interrelationships and processes, rather than inputs and outputs. As Bertrand Russell (1918) said, “Every advance in a science takes us further away from the crude uniformities which are first observed into a greater differentiation of antecedent and consequent and into a continually wider circle of antecedents recognized as relevant” (p. 188). 6.

A desire to unveil intricate relationships will characterize mental health research

and will link it with emerging studies in the fields of chaos and complexity. Insights from the study of complexity have implications for psychology and mental health research. For example, complexity theory indicates that seemingly minor events are as important over time as major life events. This observation is particularly relevant to the issues raised by the paper writers about high-risk and low-risk populations. Attention needs to be given to the antecedent conditions that make an individual resilient or susceptible to stress. Changes or conditions that rarely register on the researcher’s scope might result in significant differences in resilient behavior later (Duke, 1995).

Whether a person is a Have, a Used-to-Have, or a Can’t Have, stress will be an ever-present threat in my projected scenario. The pathogenic factors in the work situation are part of systemic organizational change. In a global information economy, the transition from jobholder to unemployment can be instantaneous and come without warning. Prevention of mental health distress under these circumstances will require a concerted effort to bring about social and cultural change (e.g., changes in attitudes toward work and identity and changes in attitudes toward mental health problems induced by stress). I am hypothesizing that such interventions as the JOBS program can be an important part of that broader change, and that education and training may lead to strengthened stress immunity because successful forms of behavior will tend to be successful on both individual and organizational levels. Research sponsored by the National Institute of Mental Health could be a key component in those successes.

References Allen, G. E. (1996, August/September). Science misapplied, Technology Review, pp. 23-31. Anderson, W. T.(ed.) (1995). The truth about the truth: De-confusing and re-constructing the postmodern world. New York: G. P. Putnam & Sons. Bowman, K. H. (1996, February/March). Reaffirmation of self reliance? A new ethic of selfsufficiency, The Public Perspective, p. 8. Boyett, J., & Boyett, J. (1995). Beyond Workplace 2000.. New York: Dutton. Coates, J. F. (1995, Spring). Work and pay in the twenty-first century: An impending crisis,. Employment Relations Today, p. 19. Collins, J., & Porras, J. (1994). Built to last. New York: Harper Business. Courter, C. L., & Poe, R. (1995, November/December). Fast forward, Across the Board. Didsbury, H. (Ed.). (1996). Future vision. Bethesda, MD: World Future Society. Duke, M. P. (1995, August). Genetic, Social, and General Psychology Monographs, 120, 267285. Frank, R., & Cook, P. (1995). The winner take all society. New York: The Free Press. Gillian, F. (1995, August). For your information, Personnel Journal, 74, 22. Harmon, W. W. (1996, July/August). Reassessing the economic assumption, The Futurist, p. 14. Levine, L. W. (1985, Summer). American Culture and the Great Depression, The Yale Review, p. 203. Nardone, T. (1995, Summer). Part-time employment: Reasons, demographics, trends, Journal of Labor Research, 16, 287. Rifkin, J. (1995). The end of work. New York: G. P. Putnam & Sons. Rinehard, V. (1995, May 14). Odd jobs, The Washington Post, p. H9.

Robertson, J. (1990). Future wealth: A new economics for the 21st century. New York: The Bootstrap Press. Rosen, R. (1991). The healthy company. Los Angeles: Jeremy P. Tarcher, Inc. Russell, B. (1918). Mysticism and logic. London: Longman, Green, & Co. Schor, J. (1992). The overworked American.. New York: Basic Books. Swoboda, R. (1996, August 25). Turning paychecks into a performance art, The Washington Post, p. H4. World trends and forecasts. (1996, May/June). The Futurist, p. 44.

Commentary on Economic Contexts: Implications for Future Priorities in Disease Prevention Research Virginia S. Cain, Ph. D. Office of Behavioral and Social Sciences Research, National Institutes of Health

Policy and Social Change These chapters on socioeconomic disparities in mental health and mental disorders elucidate the repeated findings in the literature of persistent differences among social classes in a wide variety of health areas, with lower socioeconomic status (SES) most consistently being associated with poorer health outcomes. Even when a disease is less prevalent among lower SES individuals, the rates of mortality associated with the disease are higher among those in lower socioeconomic statuses. Some papers suggest that the health differences associated with SES cannot be substantially ameliorated without major social change. Clearly, if a social change model is a prerequisite for improved health outcomes, then the interventions must affect factors at the fundamental cause level- i.e. the social and economic resources available to individuals and to the communities in which they live. Decisions are made daily on policies regarding such resources. Changes in the tax code and in policies or laws regarding welfare, family leave, child care, employment and the environment can have profound impacts on health; seldom, however, is the link established and highlighted between such policies and health outcomes. It is critical for both policy makers and researchers to understand that broad social change can affect a number of diseases and conditions. Such breadth of impact cannot be obtained with interventions limited to a particular disease or a particular

mechanism. Several papers take a step toward proposing such broad-based approaches. The case for major social change as a preventive health strategy could be substantially strengthened by further research on the economic impact of disease and the economic benefit of disease prevention.

Cost Effectiveness Research In addition to assessing the multiple health benefits of a broad social intervention, it is critical to examine the cost-effectiveness of different intervention strategies compared with each other and compared with no intervention. Intervention costs have become increasingly important as budgets have tightened in the federal, state and local agencies primarily responsible for supporting services, the costs of interventions have become increasingly important. In the area of physical health, primary prevention strategies have generated dramatic cost savings. For example, interventions aimed at preventing osteoporosis ultimately prevent hip fractures and, frequently, subsequent nursing home care; the savings are substantial. As mental health administrators confront fiscal pressures, the advantages of well-grounded primary prevention and treatment should be strongly apparent-- for behavioral and social as well as pharmacological interventions. The papers in this volume touch on issues involving both direct and indirect health care costs. However, this body of prevention research and the policies that derive from it could benefit by further research on the cost effectiveness of mental health interventions that prevent adverse health outcomes related to events such as violence, family dissolution, and alcoholrelated automobile accidents. A broadly defined range of consequences of mental illness will underscore the impact of prevention activities. The evaluation of interventions traditionally shows an improvement in quality of life, but it is no longer sufficient to assert that an intervention reduces stress or facilitates coping. Interventions are considered for broad implementation when they produce the desired outcome, but they have even greater value if the benefit to society, particularly in reduced costs, can be demonstrated. Clearly, not all interventions will result in cost savings. In many cases, the cost is well justified by improvements in health and quality of life. However, a strong research base that associates major cost savings with prevention would be extremely persuasive for policy makers. Despite a recent shift toward less government involvement in a wide range of areas, support for improving health and supporting health research is still strong. The challenge is to show how social

factors and health are linked and how broad social changes affect health. By demonstrating that improved health of the population can reduce costs for the society, researchers can increase the likelihood that policies aimed at preventing mental illness will be adopted.

Suggest Documents