Abstract
Despite the fact that there is an extensive scientific literature regarding the importance of exposure to psychosocial stressors, the assessment of such stressors is often neglected in clinical work. The present review summarizes the scientific literature on critical life changes and work-related stressors. Particular emphasis has been on somatic outcomes and physiological processes that have been shown to be affected by exposure to stressors. Although the relationships are highly complex, it could be concluded that exposure to stressors may often determine the onset of many illnesses. Standardized well-functioning assessment instruments that could be used in clinical practice exist and should be used.
The interplay between the individual characteristics of a patient and the situation surrounding him/her determines the effects of stressors on the patient's health. This has led to the misinterpretation that the nature and intensity of stressors could be disregarded because the only thing that really counts is the individual reaction. It is enough to point out for instance that deaths related to the loss of a loved person would not occur if loved persons would never die. Death of a close relative does not affect all individuals in the same way. That, however, does not mean that the event per se is unimportant. Critical life events may have a decisive role in determining the onset of a disease [see for instance 1].
An external condition that demands excessive adaptation from an individual is ‘filtered’ through the individual coping program. This is constituted by genetic factors and previous experiences [see 2]. The interaction between stressor and individual coping program determines the behavioral, psychological and physiological response to the stressor. One of the possible responses is a general arousal reaction. If this arousal reaction lasts for a long time, the response will gradually change. In Hans Selye's terminology, this is the general adaptation syndrome that has three stages, arousal, defense and exhaustion. According to McEwen [3], long-lasting adaptation to high arousal leads to allostasis (for a more detailed discussion, see below).
Throughout life, there is a constant flow of experiences. This means that the individual program is never ‘ready’, it is going through changes throughout life.
The aim of this chapter is to provide a theoretical framework for the assessment of life events and chronic stressors and to review the literature supporting or refuting the influence of stressors on illness development. A second aim is to propose assessment tools that could be used in clinical practice. Since the psychosomatic approach relates psychosocial processes to physiology and somatic illness, the examples that I choose will have ‘objective’ outcomes, for instance myocardial infarction (which has a defined onset in time and therefore is more accessible for studies of time relationships between stressor and illness) or physiological repeated outcome measures whenever this is possible.
Stressors
Stressors are always divided into critical events with a defined onset and duration on one hand and chronic stressors, or chronic life difficulties [1] with undetermined onset and duration, on the other hand. Chronic life difficulties are typically conditions that last for a lifetime or for very long periods such as life-long poverty and unresolved marital discord. The border may not be easy to define in all cases however. For instance, a natural disaster such as a tsunami, an earthquake or a nuclear power station emergency are examples of disasters that may result in chronic difficulties such as physical handicap, posttraumatic stress disorder or in a series of secondary events (loss of residence, loss of relatives, loss of job opportunities, etc). An effort to disentangle chronic life difficulties from critical life events has been made by Brown et al. [1].
Critical Life Events
There are two ways of recording critical life events.
First of all, subjects who go through a defined life event such as job loss, bereavement or reorganization at work could be studied separately. In such studies, the ideal is to follow different phases before, during and after the event. There are numerous such studies that have been published. For instance, longitudinal studies of subjects going through job loss [4, 5], bereavement [6] and hurricanes [7] have been published that relate the different phases of these critical life changes to the development of mental symptoms and physiological reactions. This is of relevance for clinicians who frequently encounter patients that go through such crisis situations but also for public health officers in charge of prevention programs.
Secondly, critical life change in general can be studied as they occur spontaneously in the flow of people's life. This is particularly relevant for practitioners who could use clinical interview techniques for all of their patients. Two quite different assessment theories have been established, one of them based upon self-administered questionnaires and the other one on systematic interviews.
Standardized Methods for Recording Typical Critical Life Events That People May Experience
In the most frequently used methodology for recording critical life events, subjects are confronted with lists of critical life events and asked to report whether they have experienced any of them during a defined period of time retrospectively. Such questionnaires were introduced by Holmes and Rahe [8]. The lists have been refined and put into theoretical context. For each change, a score has been assigned based upon population studies. This score indicates mean level of adaptation that the average subject in the population would require for coping with the change. For the time interval studied, these scores are added. For instance, if a person has lost his/her job and divorced during the past 6 months, the population adaptation score for job loss and the corresponding score for divorce are added for this time interval and a total life change unit sum is achieved [for further introduction and theoretical discussion see Rahe 9]. Even short time intervals were used for the study of weekly life events in relation to physiological change. In a study of male patients who were back to work after a myocardial infarction, Theorell et al. [10] showed that a change in weekly life change sum from one week to the next (calculated on the basis of repeated weekly standardized interviews) explained statistically 10% of the concomitant change in urinary output of adrenaline during work hours in this kind of clinical sample. It should be pointed out however that there was a pronounced interindividual variance in strength of association; for one third of the subjects no relationship was found. Prominent events in this study were conflicts at work and at home, changes in work hours or conditions and personal outstanding success.
More recent research has shown that although the life change unit sums are helpful in predictions of ill health, they are highly correlated with number of critical life events, and the theoretical rationale for additivity has therefore been questioned. New lists of critical life changes have been renewed and validated [11, 12]. In more recent studies of objective health outcomes such as myocardial infarction, shorter lists of high impact life events are usually used and summing of weights has been abandoned. Weighting is still used but rather in the form of self-rated impact ratings from the participants themselves than from the population.
An example of more recent life change lists utilized in a study of myocardial infarction patients is from Möller et al. [13] (see table 1 in online suppl. appendix). In this study (the so called SHEEP, Stockholm HEart Epidemiology Programme), all cases of myocardial infarction were contacted within a defined geographical area during a defined time period. At the same time, matched population controls were contacted. There were 1,000 men and 500 women in each group. Weight sums of life changes during the 12 months preceding myocardial infarction (and the corresponding period) did not differentiate the groups. However, a crude rating was performed by each subject (for instance conflict ‘affected me strongly/fairly/not much’), and this rating turned out to be important. ‘Impaired economic situation’, ‘conflict at work’ and ‘increased responsibility at work’ - if they were reported to ‘affect fairly or very strongly’ - differentiated clearly both male and female myocardial infarction cases from controls. A relevant critique against this kind of retrospective case control study is that retrospective ‘search for meaning’ could have influenced the findings. The findings are however consistent also with a prospective study in which healthy men were followed during a one-year period with regard to incidence of myocardial infarction [14]. The life event ‘increased responsibility at work during the past 12 months’ predicted increased risk of myocardial infarction but a weighted life event sum score of life changes in general was not associated with increased myocardial infarction risk. These findings are also consistent with another large case-control study of myocardial infarction patients [15] in which a standardized interview was used (see below).
Conclusion
This review has shown that stressors are very important in disease etiology. There is a large scientific literature supporting this notion. Both psychiatric and somatic disorders are affected by long-lasting exposure to adverse stressors. Plausible underlying mechanisms have been studied. Standardized assessments for stressors have also been published and could be widely used clinically. The assessment of work stressors has often been neglected in clinical work. Given the etiological importance that work stressors have according to the literature, this neglect has not been scientifically justified. Methodological work has resulted in several easily administered and extensively tested questionnaires for work environment and life events. In the online supplementary appendix, the reader can find easily administered instruments that are free for use: table 1, a short questionnaire regarding serious life events; table 2, an interview guide regarding life events, and table 3, a list of criteria for AL. In addition, the reader may contact web pages which present other instruments which could be used clinically or for research. See for instance questionnaires about life events, coping, social support: www.drrahe.com - a questionnaire regarding working conditions in general, www.mentalhealthpromotion.net/resources/english_copsoq_2_ed_2003-pdf.pdf - a questionnaire specifically designed to measure effort reward imbalance at work, and www.workhealth.org/UCLA%20OHP%20class%202004/ERI%202004.pdf - a questionnaire designed to assess working conditions in general, particularly psychological demands, decision latitude and social support at work.
