Friday, March 12, 1999
12:30 pm - 2:00 pm
Interleukin 6 and Cortisol in Acute Musculoskeletal Disorders - Results from a Case-Referent Study in Sweden
Töres Theorell, Professor, Ph.D., M.D. (1,2), Hans-Martin Hasselhorn M.D. (2), Eva Vingård, Ph.D., M.D. (3), Birger Andersson, Ph.D., M.D. (4) and the MUSIC-Norrtälje Study Group; (1) National Institute for Psychosocial Factors and Health, (2) Division of Psychosocial Factors and Health, Dept. of Public Health Sciences, Karolinska Institutet, (3) National Institute for Working Life, (4) Nova Medical Calab. All: Stockholm, Sweden
Introduction: According to modern stress theory, the hypothalamo-pituitary-adrenocortical system (HPA axis) is an important part of the body’s arousal system in physically straining (such as in infections) as well as in psychologically arousing situations. When demands for arousal have been excessively high for a long time, the HPA may show disturbance in its regulatory ability. Interleukins (IL) constitute an important part of the communication between the HPA axis and the immune system. The aim of the present study was to explore part of the circadian serum cortisol variation as well as Interleukin 6 (IL-6) in relation to acute low back and neck/shoulder pain and potential psychosocial stressors.
Method: In a defined geographical area, men and women who consulted any caregiver (traditional as well as alternative) for an acute spine pain condition preceded by six months without consultations constituted the case group. Age and gender matched men and women from the population register constituted the referent group. During an examination day all subjects went through ergonomic and psychosocial examinations (questionnaires, interviews and clinical examinations). Blood samples were drawn in the morning (8 am) as well as midday (noon) in a subsample of 243 individuals (86 cases and 157 referents) and analyzed for IL-6 (midday sample) and cortisol (morning and midday sample).
Results: Female cases had higher IL-6 levels (p<.05) and those seeking care for low back pain had a smaller cortisol variation between morning and midday values referents. In men, low self-reported decision latitude at work (possibility to issue control) was significantly associated with higher IL-6 concentration and low social support at work with low morning cortisol values. In women, high job satisfaction was associated with low IL-6 concentrations and lower midday cortisol levels. Statistically strong correlations were found between IL-6 and the degree of cortisol decrease between morning and midday (p<.01).
Conclusions: Reduced cortisol variation among female low back pain cases may indicate a disturbance in the HPA axis. A high IL-6 level may arise in the course of a long-lasting aroused psychosocial situation with concomitant impaired cortisol regulation. It could be speculated that the hormonal disturbances found reflect an impaired psychosocial state. Serum IL-6 may be a useful serum marker for psychosocial adverse conditions in epidemiological studies.
Job Stress, Psychophysiological Mechanisms and Musculoskeletal Disorders: Preliminary Results
Lawrence M. Schleifer*, Ed.D., CEP, University of Maryland, Ronald Ley, Ph.D., University at Albany, Thomas W. Spalding, Ph.D., University of Maryland
Psychosocial job-stress factors such as mental work load, work pressure, and lack of job control are presumed to contribute to musculoskeletal disorders through chronic low-level muscle tension, especially in computer work. However, a definitive cause-effect relationship has not been established and the psychophysiological mechanisms by which this process occurs have not been adequately explained.
Schleifer and Ley (1996) have postulated that stress-induced overbreathing may be a possible psychophysiological mechanism by which job-stress factors in computer work contribute to musculoskeletal disorders. If stress-induced breathing is sufficient to exceed the metabolic requirements for oxygen, there is an excessive loss of CO2. This physiological imbalance results in heightened neuronal activity (i.e., depolarization of neurons caused by the migration of CO2 from cells for the purpose of maintaining balance of pH in extracelluar fluids diminished by excessive loss of CO2 through hyperventilation), heightened muscle potentials, increased muscle tension/spasms, paresthesia, and suppression of parasympathetic activity with consequent sympathetic dominance of CNS. This condition results in amplified responses to catecholamines. These effects of chronic hyperventilation may be contributing factors in work-related musculoskeletal disorders.
The purpose of the present study is to demonstrate that increases in EMG activity of the trapezius and forearm extensor muscles correspond with reductions in end-tidal CO2 as a result of performing a mentally demanding computer-based task. Thirty healthy, young adults free of musculoskeletal problems are to be assigned on a random basis to either a stress-induced hyperventilatory conditioning session or a non-conditioning control session. In the conditioning session, the unconditioned stimulus (UCS) consists of a mentally demanding task (counting backwards by sevens rapidly). The unconditioned response (UCR) is the resultant strain induced by the UCS, as measured in terms of physiological changes consisting of a decrease in end-tidal CO2, and an increase in EMG activity of the trapezius and extensor muscles, and corresponding increases in skin conductance and heart rate. The conditioned stimulus (CS) consists of an auditory signal (i.e., a neutral tone).
Baseline psychophysiological measures are established while presenting an auditory tone during a 90-sec period. During the conditioning session, the participants sit in front of a video-display terminal (VDT) and enter the remainders from backwards- sevens subtractions (1000 - 7 = 993, 993 - 7 = 986, etc.) using a numeric keypad. The CS (auditory signal) is presented with the UCS (backwards sevens) during four 90-sec. conditioning trials. In the non-stress control session, no mental calculations are performed. Participants read from a printed sheet successive backwards-sevens subtractions. The remainders are entered using a numeric keypad while the auditory tone is presented during four 90-sec. trials.
Following the four trials, the auditory tone is presented to both groups in the absence of the UCS (i.e., no backwards sevens). It is hypothesized that a decrease in end-tidal CO2 and a corresponding increase in EMG activity will be greater for the conditioning group than the control group. Such results would be consistent with the view that hyperventilation provides a psychophysiological pathway that mediates the relationship between psychosocial job-stress factors and work-related musculoskeletal problems.
CORRESPONDING AUTHOR: Lawrence M. Schleifer, Ed.D., CEP, Department of Health Education, Suite 2387, University of Maryland, College Park, Maryland 20742-2611.
Repetitive Work Stress, Muscle Tension and Musculoskeletal Disorders
Ulf Lundberg, Department of Psychology, Stockholm University, Sweden
It is well documented that physically monotonous or repetitive work is associated with an increase in musculoskeletal disorders; however, recent studies also report an association between psychosocial factors in the workplace and, for example, shoulder/neck and lower back pain problems. Time pressure, lack of influence over one's work, and constant involvement in repetitive tasks of short duration often characterize jobs associated with a high risk for muscle pain syndromes.
In order to explain the high prevalence of musculoskeletal disorders in psychologically stressful, but physically less demanding jobs, new psychobiological models have been proposed. Possible mechanisms involve ongoing psychological stress that induces long-lasting activation of small, low-threshold motor units that may lead to degenerative processes, damage, and pain. These models are consistent with a new stress model called the "Allostatic Load Model." Allostasis refers to the ability to achieve stability through change. Activation of various biological systems in response to stress is necessary for successful coping and survival; however, over- or underactivity of the allostatic systems may add to the wear and tear on the organism. Thus, it is relevant to explore to what extent mental aspects of a work task contribute to sustained muscle activity, in sedentary work with low physical demands as well as in more physically demanding jobs.
Laboratory experiments show that mental stress significantly increases activity in the trapezius muscle, as measured by electromyographic (EMG) activity. Field studies show that repetitive jobs leading to a high number of musculoskeletal disorders are associated with highly elevated physiological stress levels. Additional information is obtained from analysis of short periods of very low muscular electrical activity (EMG gaps). Female workers with a high frequency of EMG gaps seem to have less risk of developing myalgia problems than do workers with fewer gaps. Muscle pain associated with psychosocial factors in the workplace could be explained by a lack of pauses in muscle activity, and women are likely to be at a greater risk than men are because of the combined load from paid and unpaid work.
In the modern work environment, with its emphasis on time pressure, effectiveness, competitiveness, lean production, and downsizing, it is possible that lack of relaxation is an even more important health risk than is the absolute level of contraction or the frequency of muscular activation because of the way it prevents rest and recovery of the allostatic systems.
CORRESPONDING AUTHOR: U. Lundberg, Ph.D., Department of Psychology, Stockholm University, S-106 91 Stockholm, Sweden
Predicting Improvement in Hand Symptoms & Function From Job Characteristics
Julia Faucett, RN, PhD*, Paul Blanc, MD, MSPH, & Eeward Yelin, PhD – University of California, San Francisco
This study investigates changes in symptoms and hand function associated with carpal tunnel syndrome (CTS) over an eighteen month period. The study is drawn from a larger investigation of the outcomes of CTS that used a community sample of patients with occupational CTS and non-occupational CTS. In the larger study, data were collected from 113 subjects over the telephone by trained interviewers using structured and pilot tested questionnaires at baseline and again 18 months later. Ninety-four subjects provided complete data on the variables used in the current study. A modified version of the NIOSH criteria for a surveillance definition of CTS was used (CTS-SENSOR); however, the majority of subjects provided evidence from nerve conduction studies to confirm their diagnoses. CTS symptom and function scales, Karasek’s Job Content Questionnaire, and Faucett’s Work Interpersonal Relationships Inventory were included in the questionnaires. Additional information was gathered about patients’ demographics, medical histories, and jobs, including repetitive and forceful work tasks. We also applied a scoring system derived from the National Health Interview Survey to evaluate the effects of repetitive and forceful activities on outcomes (Blanc et al., 1996). Subjects whose hand symptoms or hand function had improved over the 18 months of the study were compared to those who failed to improve in terms of job characteristics at baseline (not necessarily the job held when the symptoms of CTS first appeared) using Student’s t-tests. As compared to those who failed to improve (n=31), patients whose symptoms improved (n=63) demonstrated significantly (p< .05) less exposure to repetitive work activities and greater baseline pain and symptom severity. Patients whose hand function improved (n=53) demonstrated significantly worse function, more frequent muscle weakness, and higher family incomes at baseline as compared to those who did not improve (n=41); and less exposure to repetitive work activities was marginally predictive. Self reports of health status at follow-up were associated with less exposure to repetitive activities (r=.26), greater job satisfaction (r=.25), greater coworker support (r=.21), less coworker conflict (r=.21), and less depression (r=.24) at baseline; and greater improvement in symptoms over time (r=.24). Improvement status was not significantly related to having had surgery or a worker compensation claim, or gender. In this longitudinal study, physical and interpersonal job characteristics, in addition to personal factors, were shown to be significantly associated with later improvements in symptom severity and hand function, and overall health.
CORRESPONDING AUTHOR: Julia Faucett, RN, PhD – Box 0608, University of California, San Francisco, CA – 94143-0608.
Job Stress and Health in Japan: Baseline Findings From Japan Work Stress And Health Cohort Study Group
Chair: Norito Kawakami, M.D., Gifu University School of Medicine; Presenters: Norito Kawakami, M.D., Gifu University School of Medicine, Takashi Haratani, Ph.D., National Institute of Industrial Health, Japan Ministry of Labour, Fumio Kobayashi, M.D., Aichi Medical University, Takeshi Hayashi, M.D., Hitachi Health Care Center, and Shogo Miyazaki, M.D., Center for Occupational Health at Keihin, NKK Corporation
An increased attention has been paid on job stress among Japanese workers, because of their extremely long working hours, "karoshi" (death due to overwork) and a different cultural background underlying job stress from the Western countries. However, findings from previous small-sized studies in Japan were conflicting and inconclusive. A large-scale prospective cohort study was designed to know the effects of job stress on health among Japanese employees by a researcher group called the Japan Work Stress and Health Cohort Study Group (http://www.gifu-u.ac.jp/~norito/cohort.html). Recently, the baseline survey was completed for 25,143 employees in Japan. Based on cross-sectional analyses of the large-scale baseline data from Japan, this symposium intends to introduce evidence on distribution of job stress and its adverse effects on physical and mental health in Japan.
The symposium includes five speakers from members of the Japan Work Stress and Health Cohort Study Group. Dr. Kawakami, as the first speaker, will introduce the study design and subjects of the Japan Work Stress and Health Cohort Study, as well as distributions of job stressors by sex and occupation, comparing them with those in the U.S. Dr. Haratani will present experience with two major measures of job stressors used in the study, including their reliability/validity. Then Prof. Kobayashi will present findings on cross-sectional associations of the job demands-control model and the demand-control-support model with self-reported depression (CES-D) and sick absence in Japanese workers. Dr. Hayashi will present evidence for job stress effects on coronary heart disease risk factors, which are a topic most associated with "karoshi". Dr. Mizayaki will present a new finding on the association of job stress and chronic atrophic gastritis, which has been paid as a high risk condition for gastric carcinoma.
This symposium will be the first opportunity to show a systematic overview of the distribution and association of job stress with health in Japanese workers, covering a variety of topics, e.g., job stress assessment, adverse health effects of job stress and cultural difference, thus and it will contribute to goals of the APA/NIOSH job stress conference.
CORRESSPONDING AUTHOR: Norito Kawakami, M.D., Department of Public Health, Gifu University School of Medicine, 40 Tsukasa-machi, Gifu 500-8705, Japan
Study Design, Job Stress Patterns and Implications for Policy Development
Norito Kawakami, M.D., Gifu University School of Medicine, and The Japan Work Stress and Health Cohort Study Group*
The Japan Work Stress and Health Cohort Study is a research project supported by the Japanese government. This paper gives an overview of the research project, as well as distribution of job stress and possible implication for policy development.
Study Design: The study is a 5-year prospective cohort study to assess the effects of job stressors on health, including mortality, major illnesses and sick absence in Japan. Study subjects at baseline were 21,000 men and 4,000 women employees working at nine companies in Japan. The baseline questionnaire included two set of job stressor scales, i.e., the Job Content Questionnaire (JCQ, Karasek 1985) and the Generic Job Stress Questionnaire complied and expensively used by the U.S. NIOSH (Hurrel & McLaney, 1988), as well as other scales/questions on psychological distress (e.g., depression), health habits and personality. Physical check-up data at baseline were also collected.
Distributions of Job Stressors from the Baseline Data: The proportions of those who worked overtime greater than 50 hours per month were 18% for men and less than 1% for women. Among men, managers had the highest job overload (or job demands) and highest job control (both in JCQ and the NIOSH questionnaire); machine operators had lower job overload (or job demands) but lower job control. The patterns were quite similar with those observed in the U.S. Quality of Employment Survey in 1970’s (Karasek & Theorell, 1990). It seems that Japanese men worked longer than U.S. men but perceived similar levels of job overload. Women had lower levels of job overload (or job demands), job control, skill utilization and supervisor support than men. Women seem engaged in passive jobs than men in Japan, while women service workers experienced relatively higher job overload compared with their levels of job control.
Implications for Policy Development: Although scientific evidence derived from the project is essential for development of a policy for job stress control, there is a gap between scientists’ view and policy makers’ need. A possible strategy linking research findings to policy development is to combine analytical and integrating roles within a research project.
* The Japan Work Stress and Health Cohort Study Group includes T. Haratani, Ph.D., F. Kobayashi, M.D., M. Ishizaki, M.D., T. Hayashi, M.D., O. Fujita, M.D., Y. Aizawa, M.D., S. Miyazaki, M.D., H. Hiro, M.D. S. Hashimoto, Ph.D., S. Araki, M.D., as well as the first author.
CORRESSPONDING AUTHOR: Norito Kawakami, M.D., Department of Public Health, Gifu University School of Medicine, 40 Tsukasa-machi, Gifu 500-8705, Japan
Questionnaires: JCQ and NIOSH Job Stress Questionnaire
Takashi Haratani, Ph.D., National Institute of Industrial Health, Akiko Miki, M.S., University of Tokyo, and The Japan Work Stress and Health Cohort Study Group*
Purpose:The Japan Work Stress Cohort Study Group conducted the baseline survey using two major job stress questionnaires: the Job Content Questionnaire (JCQ) and the NIOSH Job Stress Questionnaire (JSQ). A cross-sectional analysis was performed to evaluate psychometric properties of JCQ and JSQ using the data from the baseline survey of five worksites in 1996.
Methods: Each subject was asked to complete a self-administered questionnaire package including JCQ and JSQ. A total of 10,914 workers (8,601 males, 2,783 females and 70 of unknown gender) responded to the questionnaire in 1996. We analyzed 9 scales of JCQ and 20 scales of JSQ.
Results: Cronbach's alpha coefficient was used to measure the internal reliability of each scale. The alphas of JCQ ranged from .45 for job insecurity to .88 for supervisor support with a mean of .73 (SD = .14) for the 9 scales of JCQ, and ranged from .44 for nonwork activities to .95 for job control with a mean of .80 (SD = .10) for the 20 scales of JSQ. Principal component analysis of each scale was conducted. Each first component explained the item variance from 28% for psychological demands to 89 % for isometric loads with a mean of 57% (SD = 21) for JCQ, and from 26% for nonwork activities to 78% for responsibility for people with a mean of 54% (SD = 15) for JSQ. Six first components (67%) of the 9 scales of JCQ explained 40% or more of the item variance, and 16 (80%) of the 20 scales of JSQ did. Thirty-two (73%) of 44 items of the 9 scales of JCQ showed .60 or more first component loadings, and for JSQ, 115 (81%) of 142 items. The construct validity of the two questionnaires was assessed by performing principal component factor analyses with varimax rotation. Most of items showed high factor loadings to the same factor corresponding to the scale. Factor structure supported the construct validity of the scales of JCQ and JSQ.
Conclusion: The results of this study indicated that both JCQ and JSQ had acceptable reliability and validity and they are considered to be reliable and useful in epidemiologic studies for Japanese workers.
*The Japan Work Stress and Health Cohort Study Group includes N. Kawakami, M.D., F. Kobayashi, M.D., M Ishizaki, M.D., T. Hayashi, M.D., O. Fujita, M.D., Y. Aizawa, M.D., S. Miyazaki, M.D., H. Hiro, M.D., S. Hashimoto, Ph.D., S. Araki, M.D., as well as T. Haratani, Ph.D.
CORRESPONDING AUTHOR: Takashi Haratani, Ph.D., Division of Work Stress Control, National Institute of Industrial Health, 6-21-1 Nagao, Tama-ku, Kawasaki 214-8585, Japan.
Job Stress, Depressive Symptom and Sick Absence in Japan
Fumio Kobayashi, M.D., Aichi Medical University, and The Japan Work Stress and Health Cohort Study Group*
A cross-sectional relations of job stress to depressive symptom, and to sick absence for the past one year were examined among Japanese working populations.
Study Design: A questionnaire survey was conducted at 9 enterprises which were located in Kanto, Tokai and Hokuriku area of Japan. The response rate were from 50 to 95%, and 22164 employees returned the questionnaire. The subjects younger than the age of 17 or older than the age of 61 were omitted. Data from 13601 men and 2766 women who filled out all questions were analyzed (mean ages, 40.8 and 36.5 years, respectively). Job demand, job control, supervisor and coworker support were assessed using the Japanese version of Job Content Questionnaire (JCQ). Mean scale scores were calculated for each sex, and the subjects were divided into two groups for each scale using a median split. Depressive symptom was defined as the score above 16 in the CES-D scale. The criteria for sick absence was set to be 7 days for the past one year (self report based). Age-adjusted rates of depressive symptom and of sick absence were compared among the groups which were formed based on the social support scale, and the Karasek's theoretical model, i.e., low-strain, active, passive, and high-strain. The relations were also assessed using a multiple-logistic analysis controlling for age, neuroticism (EPQ), and social desirability scale.
Results: In men, the high-strain group, and the low supervisors' support group showed significantly high rates (p<0.01) in depressive symptom. The rate of sick absence longer than 7 days was also significantly high in the high-strain group. In women, the low supervisors' support group had a significantly high rate both in depressive symptoms and in sick absence. The group with low score in social support from coworkers showed significantly high rates in depressive symptom both in men and women, and in sick absence in men. A multiple logistic analysis for depressive symptom in men proved low job control (odds ratio, OR 1.48), low support from supervisors (OR 1.27), and low support from coworkers (OR 1.55) to be significant. In women, low support from supervisors (OR 1.51), and low support from coworkers (OR 1.35) were significant. Concerning to sick absence, high job demand (OR 0.80), interaction between high job demand and low job control (OR 1.33), and low support from coworkers (OR 1.18) showed significant relationship in men. In women, low support from supervisors (OR 1.70) was significant.
* Including N Kawakami, T Haratani, M Ishizaki, T Hayashi, O Fujita, Y Aizawa, S Miyazaki, H Hiro, S Hashimoto and S Araki, as well as the first author.
CORRESPONDING AUTHOR: Fumio Kobayashi, M.D., Department of Health and Psychosocial Medicine, Aichi Medical University, 21 Karimata Yazako, Nogakute-cho, Aichi, 480-1195, Japan
Job Stress and Coronary Heart Disease Risk Factors in Japan
Takeshi Hayashi, M.D. Hitachi Health Care Center, and The Japan Work Stress and Health Cohort Study Group*
Occupational factors have been considered to relate coronary heart disease(CHD) as work-related disease. To investigate the relationship between CHD risk factors and job stress and occupational factors, we make the following three studies.
1. A study of Relationship between occupational factors and CHD risk factors.
We investigated the relationship between occupational factors and CHD risk factors which include blood pressure, total cholesterol, triglyceride, body mass index. This study subjects were 7,226 Japanese male electric company employees (40 to 60 years old). The main results were as follows: (1) Work style was strongly related to life style. (2) Overtime work and walking time from home to the workplace did not relate to CHD risk factors. However, frequency of business trips, working posture and job category were related to most of CHD risk factors. (3) Smoking and frequency of drinking were strongly related to CHD risk factors in both blue-color and white-color workers.
2. A study about Relationship between job stress and CHD risk factor.
Job stress is estimated by a self-administered questionnaire of the Job Content Questionnaire (JCQ) and the NIOSH Generic Job Stress Questionnaire. The target subjects were 6,539 Japanese male workers aged 35 or older working in electric manufacturing company. Between high and low strain group, there was a significant difference in body fat ratio and uric acid. Using the NIOSH Generic Job Stress Questionnaire, job control, physical environment, responsibility for people and quantitative workload showed a significant correlation with CHD risk factors.
3. A study of CHD patients using the JCQ
To investigate the risk factors of recurrent CHD by job stress, we studied job stress of CHD patients. The target subjects were 92 Japanese male workers who were operated on coronary angiography. Cases(31) of 75% or more stenosis were positive group, another cases(61) were negative group. There were not significant difference between both groups at Ratio of high strain.
*The Japan Work Stress and Health Cohort Study Group includes N. Kawakami, M.D., T. Haratani, Ph.D., F. Kobayashi, M.D., M. Ishizaki, M.D., O. Fujita, M.D., Y. Aizawa, M.D., S. Miyazaki, M.D., H. Hiro, M.D., S. Hashimoto, Ph.D., S. Araki, M.D., and the first author.
CORRESPONDING AUTHOR: Takeshi Hayashi, M.D., Hitachi Health Care Center, 4-3-6 Ohse-cho, Hitachi, Ibakraki 317-0076, Japan
The Association Between Psychosocial Job Conditions And Chronic Atrophic Gastritis
Shogo Miyazaki, M.D., Seich Horie, M.D., Takeshi Masumoto, M.D., Center for Occupational Health at Keihin, NKK Corp., and The Japan Work Stress and Health Cohort Study Group*
Purpose: Chronic Atrophic Gastritis (CAG) has been recognized as a precursor of gastric carcinoma and serum pepsinogen levels provide a precise measure of the extent of CAG. It is also known that there is evidence linking between psychological job conditions and gastric ulcers. In this study, we investigated the association between psychological job strain and CAG.
Materials and Method: A blood sample was taken from male employees aged from 39 to 60 years old (n=496) of a steel company in Japan, and serum levels of pepsinogens were measured using pepsinogenⅠandⅡRiabead Kits (Dinabot Co.Ltd.). At the same time, a mailed questionnaire was sent to these subjects. CAG was determined when pepsinogenⅠlevel was less than 50ng/dl and pepsinogen I/II ratio was less than 3.0. Job overload, skill use and job control were assessed by the NIOSH Generic Job Stress Questionnaire. Multiple logistic regression of CAG on age, occupation, job overload, skill use, job control, alcohol consumption, number of cigarettes smoked per day and body mass index (BMI) was conducted.
Results: We received 456 responses to the questionnaire (response rate, 92%). After excluding one subject with a past history of gastric cancer, 14 subjects who currently received treatment for gastric ulcer, and 35 subjects who had any missing response in the questionnaire, data from 406 responders (average age, 47.7 with standard deviation, 4.9) were analyzed. Forty-four (10%) of the responders were classified as having CAG. According to the multiple logistic regression analysis, those with lower scores of skill use had a significantly higher risk of CAG than those who with higher scores of skill use (odds ratio, 2.36; 95% CI, 1.19-4.81). Age, decreased BMI were also significantly associated with higher risk of CAG. Other psychosocial job variables did not reach at significance level (p>0.05).
Conclusion: Our finding suggested that skill underutilization, which is one of the key dimensions of the job strain model, is associated with the development of CAG. The findings may be explained by decreased regenerative ability of gastric mucosa or greater vulnerability to Hericobactor pylori infection through decreased immune functions associated with psychological strain.
*The Japan Work Stress and Health Cohort Study Group includes N. Kawakami, M.D., T. Haratani, Ph.D., F. Kobayashi, M.D., M. Ishizaki, M.D., T. Hayashi, M.D., O. Fujita, M.D., Y. Aizawa, M.D., H. Hiro, M.D., S. Hashimoto, Ph.D., S. Araki, M.D., as well as S. Miyazaki, M.D.
CORRESPONDING AUTHOR: Shogo Miyazaki, M.D., Center for Occupational Health at Keihin, NKK Corporation, 1-1 Minami-watarida, Kawasaki 210-0855, Japan
Preventing Stress, Improving Work, Health and Productivity: Eleven European Case Studies: Symposium
Chairs: Michiel A.J. Kompier, Ph.D, University of Nijmegen & Cary L. Cooper, Ph.D., University of Manchester; Presenters: Cary L. Cooper, Ph.D., Raija Kalimo, Ph.D., Finnish Institute of Occupational Health, Tores Theorell, Ph.D., Karolinska Institute, Susan Cartwright, Ph.D., University of Manchester, Michiel Kompier, Ph.D.
The current practice of stress prevention and intervention seems disappointing. Although a great deal of activity is going on in this field, these stress prevention programs are predominantly reactive and aimed at individuals. Only teaching employees to cope with stress may treat the effects while ignoring the cause. Therefore, it is a less desirable and perhaps less ethical strategy than improving the (psycho-social) work environment. Another peculiarity in the practice of stress prevention concerns the lack of a systematic risk assessment as well as of serious research into the effects of all these activities.
The demonstration of 'examples of good preventive practice', that may provide an alternative to this 'post hoc' individual oriented tradition, is considered as a conditio sine qua non for developing effective stress-prevention procedures, and for the involvement of both social partners in this field (i.e. employers and employees). These examples may be crucial when trying to answer the question: 'How well - and why - do stress prevention programs work?'
Hence the major purpose of this symposium is to analyze and compare such case studies from various European countries. Therefore eleven cases in stress prevention were analyzed and compared (Kompier & Cooper, 1999). These eleven countries are: Sweden, Finland, the Netherlands, Belgium, Denmark, United Kingdom, Germany, Ireland, Portugal, Greece, and Italy.
After an introduction of this field by Cary Cooper (first presentation), three of these European cases are presented in this symposium: from Sweden (T�res Theorell), Finland (Raija Kalimo), and the United Kingdom (Susan Cartwright). Each of these presentations briefly characterizes the current state with respect to stress prevention in the country under study. Second, each speaker introduces and analyzes a national case study with a view to motives, risk assessment, interventions, implementation and outcomes (effects, costs and benefits, follow-up, lessons). The final presentation (Michiel Kompier) presents a more systematic analysis and comparison of these eleven countries and eleven cases. After these presentations, this project and its' studies are open for discussion.
CORRESPONDING AUTHOR: Michiel A.J. Kompier, Ph.D, Department of Work and Organizational Psychology, University of Nijmegen, P.O. Box 9104, 6500 HE Nijmegen, The Netherlands
A Strategy for Organizational Stress Interventions
Cary L. Cooper, Ph.D., University of Manchester
A great deal of research has been undertaken into identifying sources of workplace stress (e.g., Cooper & Payne, 1988; Schabracq et al., 1997), but little systematic work has been carried out on organizational stress interventions. This paper outlines the costs of organizational stress, what research has shown are the main organizational stressors and a three-prong strategy for stress management interventions. In terms of the latter it will highlight primary, secondary and tertiary approaches to organizational stress interventions, as well as emphasizing the potential effectiveness of each type of strategy. Primary preventions has to do with taking action to modify or eliminate sources of stress inherent in the work environment (by identifying the stressors through stress audits or risk assessments) and so reduce their negative impact on the individual. Secondary prevention is concerned with managing stress by increasing awareness and improving the stress management skills of the individual (usually by training approaches). And tertiary prevention is concerned with the treatment, rehabilitation and recovery process (e.g. Employee Assistance Programs) of those individuals who have suffered or are suffering from job related stress.
This framework will provide the basis on which work directed and employee-directed interventions in the further four presentations can be contextualized. It will also highlight the need for systematic research on evaluating primary, secondary and tertiary interventions, so that practice can be refined and developed to reduce stress in the workplace.
corresponding author: Cary L. Cooper, Ph.D., Manchester School of Management, UMIST, P.O. Box 88, Manchester M60 1QD, UK.
Psychosocial Environment at Work - The Swedish Arena: Mail Processing
T�res Theorell, Ph.D., Karolinska Institute, and Kurt Wahlstedt, M.D., University Hospital, Uppsala, & The Swedish foundation for occupational health and safety among state employees
For a long period, Sweden - a European country not exposed to two world wars and with material affluence that created a good climate for social research and reforms - used to be the Mecca of work environment. Four decades of uninterrupted social democratic rule contributed to this climate, which resulted in the creation of a special agency for the protection of workers, the National Board of Occupational Safety and Health, a large institute for Research on Occupational Health and Safety, and a national research council, the Work Environment Fund, for research grants. The latter has contributed financially to a large number of research projects. Due to a recent societal financial crisis, however, health promotion work in relation to work organizations has had a weaker position during later years.
In comparison to many other European countries, Swedish research had been more oriented towards public perspectives, emphasizing epidemiological research. There is a tradition of regularly recurring national surveys of living and working conditions in representative population samples (national bureau of statistics). New links were created between basic psychological and physiological research, and work organization research. Stress at work has been an important issue since the 1970's. The Work Environment Act states a.o. that working conditions shall be adapted to people's differing aptitudes, and that technology, work organization and job content shall be designed in such a way that the employee is not subjected to physical or mental strains which can lead to illness.
The interventions described deal with mail processing in two Swedish urban regions. The job redesign processes were prompted by current demands for effectiveness and speed in public service, but also by concerns caused by a high sick leave prevalence among the employees. Mail processing has been a public service given by the state in Sweden. No competition has existed so far, but recently small-scale efforts to create competing mail service organizations have started and due to national financial problems, discussions regarding the effectiveness in the governmental mail system took place. These discussions have dealt both with the postal sorting systems - which are now concentrated at postal sorting terminals - and with the delivery, i.e. the work of the mailmen. Accordingly, a number of job redesigns have been tested in both these areas.
The case description illustrates that it is possible to improve the work organization of public service and that such changes may benefit employee health.
CORRESPONDING AUTHOR: T�res Theorell, Ph.D., Department of public health, Karolinska Institute/National Institute for psychosocial factors and health, P.O. Box 60205 S-104 01, Stockholm
Finland: Organizational Well-Being, Ten Years of Research and Development in a Forest Industry Corporation
Raija Kalimo, Ph.D., and Salla Toppinen Ph.D, Finnish Institute of Occupational Health
In Finland stress prevention is considered an important policy issue, and increasingly as a means to improve work, health and productivity. The prevention of occupational stress is the legal duty of the employer, and fits in the overall care for health and safety at work (Occupational Health Care Act, Occupational Safety Act). These laws focus on protecting employees from noxious effects of work, on prevention of work-related diseases and on the maintenance and improvement of the ability to work. Coverage of occupational health care is 92 per cent of salaried employees. According to these regulations, working conditions, including work organization and psychosocial factors have to be monitored systematically. Several monitoring systems have been developed and are in use both at national and organizational levels.
In 1984, a large forest industry enterprise (app. 19,000 employees) took the initiative to collaborate with the Finnish Institute of Occupational Health in order to analyze working conditions and, after extensive negotiations and a pilot study, to start a complex action-research program. Therefore two committees with members of the company and the research institute were formed to take care of the undertaking.
On the basis of a first basis questionnaire study (1986), risk factors and risk groups were identified and preventive measures were discussed in the company. Two basic approaches were chosen: 1.) reorganization of work and training of management through the mobilization of actions as a long-term process in the organisation (e.g. work redesign, developing leadership, personnel development, encouraging shop-floor participation) and 2.) implementation of specific developments projects (e.g., for office personnel and fore-men). Follow-up took place by a.) a sample survey 2.5 years later, b.) a management report after 4 years, c.) comprehensive survey 10 years after the base line study. Again, recommendations on improving work and well-being were given to company management. Results (over time developments) from these evaluation studies are generally positive. Still, time pressure increased steadily, as generally in Finland. Currently sickness absence data and developments are being studied. One major and positive outcome of the project is that it has initiated a development process, which has been adopted as a continuous company policy. The extent to which the project accounts for these positive outcomes and other related methodological aspects will be discussed.
CORRESPONDING AUTHOR: Raija Kalimo, Ph.D., Finnish Institute of Occupational Health, Topelinksenkatu 41 aA, FIN-00250 Helsinki, Finland
United Kingdom: Evaluation of a Stress Management Program in the Public Sector
Susan Cartwright, Ph.D., University of Manchester
Stress prevention activities in the UK tend to be confined to large organizations (>500 employees). Although there are many examples of secondary and tertiary level interventions, rather less emphasis has been placed at the primary level. The effectiveness of stress interventions generally is rarely evaluated in any systematic way.
This paper reports on the stress intervention strategies introduced in a large government department with 25,000 employees. In common with other public sector organizations, it had undergone a period of major restructuring and cultural change. In this particular department, the sickness/absence rates were starting to rise above the public sector average. A study was conducted to evaluate the effectiveness of stress management programs designed to improve the coping skills of individual employees and thereby reduce reported symptoms of ill health and strain. Whilst it is acknowledged that action which focuses on addressing and changing identified organizational stressors is necessary to minimize sources of stress in the workplace (Cooper & Cartwright, 1994), such initiatives generally require longer term implementation. Skills training targeted at the individual can be introduced within a short time span and can therefore have a more immediate effect on employee strain. The study compares the effectiveness of training based on different theoretical perspectives i.e. education and awareness; exercise and cognitive restructuring. It incorporates measures to evaluate program effectiveness on individual outcomes (e.g. physical and mental health) and on organizational variables (e.g., self reported absenteeism and commitment). 270 Individuals participated in the study. Measures were taken pre-intervention (baseline), three months post-intervention and six months post-intervention. Type of intervention acted as the independent variable, with three types of stress management training plus a wait-list control and full control conditions.
At the three month time point, exercise group participants reported improvements in physical health symptoms, anxiety, depression and general mental health. The education/awareness group also improved in depression, physical and mental health states. No improvements were reported amongst those who participated in the cognitive restructuring module. At the six month time point, the exercise participants demonstrated the most benefits from training. Whilst improvements in somatic anxiety and physical health were sustained, improvements in anxiety, depression and mental health were not maintained. However, the exercise group did record improved sickness/absence rates.
The results of this study and its wider implications for the design and delivery of stress management interventions are discussed. Finally, the paper will also report on the success of subsequent interventions which the department has introduced.
CORRESPONDING AUTHOR: Susan Cartwright, Ph.D., Manchester School of Management, UMIST, P.O. Box 88, Manchester M60 1QD, UK.
Stress Prevention: European Countries and European Cases Compared
Michiel A.J. Kompier, Ph.D, University of Nijmegen
Through a network approach eleven European cases studies were collected. For each country, national specialists 1. characterized the current status with respect to occupational stress and its' prevention, and, 2. systematically analyzed and evaluated a case study. Country profiles demonstrated three European clusters. The first (Sweden, Finland, Netherlands) pays a lot of attention to (preventing) work stress and considers it a relevant health and safety issue. The second (Belgium, Denmark, U.K., Germany, Ireland) considers stress a policy issue, but lacks the national monitoring system and prevention examples of the first. In the third cluster (Italy, Greece, Portugal) occupational stress is not 'a real issue'.
To evaluate the eleven cases, each case received a research design rating (Murphy, 1996): one *-rating (descriptive evidence), one **-rating (no intervention), six ***-ratings (no control group but evaluation), two ****-ratings (control group without randomization), and one *****-rating (randomised control group). Main project motives were high absence figures, prevention of work related health problems, and jointly improving working conditions and productivity. A wide range of instruments was used for assessing risk factors and groups, most projects combining several instruments. Work directed measures (e.g. work redesign, social support) and person directed measures (mostly training or other 'HRM' measures) were taken. Eight cases decided to combine both work and person directed measures. Most organisations chose to integrate the interventions in the regular company and management structure. In various cases explicitly principles of worker participation were chosen.
An evaluation of these cases first reveals that stress prevention is no 'one time event' nor merely a technical process. Obstructing factors were a.o. time constraints, difficulties in keeping employees involved, and differences between practical and scientific aims. No evaluation data were available in two cases (* and **). Four of the other cases offer more objective data on sickness absence, in three of them this was significantly reduced. Subjective outcomes were available in seven cases, in general with positive outcomes, e.g. less constraints in the work situation, and decreased health complaints. There are clear indications that several cases were successful from a financial perspective. Five critical success factors may be derived from these cases: 1. a stepwise and systematic approach, 2. an adequate diagnosis or risk analysis, 3. a combination of work-directed and worker directed measures, 4. a participative approach, assuring the involvement and commitment of employees and middle-management, and 5. top management support. In sum, these cases strongly suggest that stress prevention may be beneficial to both the employee and the organisation.
CORRESPONDING AUTHOR: Michiel A.J. Kompier, Ph.D, Department of Work and Organizational Psychology, University of Nijmegen, P.O. Box 9104, 6500 HE Nijmegen, The Netherlands