This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Theorell, T. Articles by Weingarten, A.-M. Search for Related Content PubMed PubMed Citation Articles by Theorell, T. Articles by Weingarten, A.-M. Related Collections Stress and Coping

Employee Effects of an Educational Program for Managers at an Insurance Company

Division for Psychosocial Factors and Health at the Department of Public Health Sciences, Karolinska Institute, and the National Institute for Psychosocial Factors and Health (T.T., R.E.) and Hälsonätet—Psykosomatiskt Centrum (A.-M.W.), Stockholm; and Department of Public Health and Caring Sciences Section of Social Medicine (B.A.), Uppsala, Sweden.

Address reprint requests to: Töres Theorell, MD, PhD, Professor, National Institute for Psychosocial Factors and Health, Box 230, s-171 77 Stockholm, Sweden. Email: tores.theorell{at}ipm.ki.se

	ABSTRACT

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
BACKGROUND: Possible health effects for employees of efforts to improve the psychosocial competence of managers have not been studied scientifically in the past.

OBJECTIVE: To explore how efforts to improve management will change the work environment and health of the employees.

METHODS: Managers of the experimental department in a large insurance corporation underwent 2-hour biweekly training sessions for 1 year—altogether, 60 hours. A control group of employees in other departments in the corporation not affected by the modification was followed with the same assessments. Morning blood samples for the assessment of serum cortisol were collected both at baseline and after 1 year in 155 participants in the experimental group and in 147 subjects in the control group. Liver enzymes and lipids were also assessed. In the questionnaire part of the 1-year follow-up study, there were 119 participants in the experimental group and 132 in the control group.

RESULTS: When repeated-measures ANOVA was used, a significant interaction effect was found for the level of serum cortisol; serum cortisol levels were decreased in the intervention group and were unchanged in the control group (ANOVA two-way interaction, p = .02; after exclusion of the managers, p = .005). A significant interaction effect was also observed for decision authority, with increased decision authority in the intervention group and, conversely, a decreased level in the control group (p = .001; after exclusion of managers, p = .02).

CONCLUSIONS: The study indicates that a moderately intensive psychosocial manager program lasting for 1 year can be beneficial for the employees with regard to both lowered serum cortisol and improved authority over decisions.

Key Words: intervention • educational program • managers • cortisol • decision latitude

Abbreviations: ANOVA = analysis of variance;; GGT = gamma glutamyltransferase.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
The Swedish work situation, in general, resembles that in other countries in the west, with the important difference that since the end of the 1980s, there has been a more rapid acceleration in the pace of adjustment to market conditions. Since the early 1990s, there has been a higher level of unemployment and a substantially increased demand on effectiveness and the amount of hours worked. Other countries in the west have had a more continuous process during the 1980s up to the present.

Recently, according to statistics (1), the feeling of being stressed and having nonlocalized trouble/pain was reported to have increased strikingly from the early 1990s to 1997.

In a challenging world, the perception of having control over one’s own work situation is particularly important. A recent study of middle-aged men (age 45–54 years) showed that a pronounced decrease in decision latitude was associated with an elevated risk of developing a myocardial infarction during follow-up within 5 years. This increased risk could not be explained by general standard risk factors such as smoking habits (2).

Both Swedish and other international investigations have indicated that little opportunity to exert control over one’s own work situation is a risk factor for myocardial infarction (3–5), gastrointestinal disorder (6), sick leave (7, 8), and psychological symptoms such as tiredness and depression (9).

If low decision latitude is a risk factor for morbidity and mortality, then an obvious question would be: how is it possible to improve authority over decision-making for the employees?

The feeling of being in control stands for something fundamental. Several attempts have been made to increase the authority over decision making at work. Evaluations of such attempts have been made. These studies investigated whether there is any relationship between change in the experience of control at work and improved health of the employees. For example, in one study (10), hospital outpatient facilities were randomly allocated to an experimental and a control population. To increase decision latitude for the employees, the frequency of staff meetings was substantially increased to 2 per month. To improve social support, the staff were trained in participatory group problem techniques. Follow-up data in the two populations after 6 months showed that there was a significant drop in role ambiguity and role conflict in the experimental population, but not in the control population. As a result, the staff in the experimental population experienced significantly reduced emotional strain, job dissatisfaction, absenteeism, and intentions to leave the job. In another study, which represents a less conclusive but useful kind of evaluation (11, 12), mail deliverers were offered the possibility to move to a new postal office in the same region—an adaptation to a marked increase in the size of the population in the area. The employees in both stations were subjected to a follow-up study. When the new station started, a number of psychosocial work environment changes were instituted that aimed at increased decision latitude and improved social climate for the employees—for instance, more responsibility for a particular area for the working group that aimed at increased cohesiveness. The results indicated that the development of physical ergonomic conditions did not differ between the groups during the year of follow-up but that social climate improved more in the new station. Possibly as a consequence, the prevalence of symptoms from the neck/shoulder region decreased in the new station but not in the old one. In a third study (13), the occupational health care team offered a 2-day course in psychosocial stress for the employees in offices and also explored the psychosocial work environment, as well as individual conditions, by means of standardized questionnaires. On the basis of the findings, a number of changes that involved the employees were started, all of which aimed at improved decision latitude and social support. The findings after 8 months indicated significantly improved autonomy and intellectual discretion, as well as almost significantly improved social climate, in the experimental group and no changes in the control group. No significant changes in personal habits (such as smoking and diet) were found. Despite this, a significantly improved serum lipid pattern was found, indicative of reduced coronary heart disease risk, in the experimental group but not in the control group. There are several other experiences in the field (14) that indicate that psychosocial improvements of this kind take time, usually several months, and that collective feedback and support belong to the necessary conditions for successful organizational changes.

Usually, simple organizational solutions constitute the framework of the changes. One example would be an attempt to change the role of a foreman to become more of a coordinator rather than a supervisor or (as in the first example described above) the introduction of regular, structured meetings for organized information exchange.

There have been attempts to increase cohesiveness in a work team by allowing the group to take responsibility over a large number of diverse tasks in the working process—and, accordingly, less specialization—as in the second example. A common result with such an approach is an improvement in perceived authority over decisions as well as improved social support.

Thus, whereas several studies have evaluated the health effects of psychosocial interventions that involved the whole staff and of interventions that focused on foremen, no studies have evaluated the employee health effects of interventions specifically focused on higher level managers.

	Situation at the Insurance Company and the Actual Educational Program for Managers

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
The organization studied is the insurance company for salaried employees in Sweden. All the members of the union of salaried employees in Sweden pay part of their union fees for a supplemental pension that will be paid to them when they retire. In addition, the company has been marketing other kinds of insurance in competition with other insurance companies. There was an anxiety-provoking social situation at the company during the spring 1998. The whole company was affected by a thorough discussion about the basic conditions for supplementary pension for the individual. The company had occupied a monopoly with regard to this kind of insurance for white-collar workers. The question was whether competitors would be allowed to enter the market.

It was a period of waiting and anxiety—politicians and the trade unions were about to change the basic conditions. The employees were worried about the ongoing discussions, because the result of the discussions could lead to a major reorganization. The company considered that it would be possible to prevent a further deterioration of the work climate in relation to the expected reorganization by an attempt to improve the managers’ competence. A consultant was contacted. A program was planned for the 42 managers at one of the departments (intervention site) to have a mandatory meeting once every second week for 2 hours during the fall of 1998 and the spring of 1999. At the start and end of the program, the managers spent a whole day discussing psychosocial work environment factors.

During the first 30 minutes of a single session of the intervention program, there was a structured lecture. The remaining time was set up for discussion. The topics of lectures and discourses were organized in four successive periods:

1. Individual: medical and psychological knowledge regarding the individual’s function
   
2. Group: social psychological knowledge regarding the group
   
3. Social-psychological processes: this part was designed to initiate practical applications and
   
4. Reorganization: knowledge regarding psychosocial redesign and ways of starting processes for psychosocial improvement.
   

The intervention sessions took place once every second week during a whole working year, with mandatory participation for all managers in the organization. There was good attendance among participants, with sick leave and traveling as the only reasons for absence. Only two managers refused participation after some sessions. Both of these managers stopped working in the department shortly afterward. The discussions took place in groups of seven to eight persons, with a consultant as a discussion leader in each group. The program was continuously monitored and adapted to the opinions about it among the participants. One result of the continuous assessment of the program was that a regrouping of participants in the discussion groups took place before phase 4 to energize the discussions.

A major aspect of the program was that the 2-week period between the sessions was intended to be a period of reflection and discussion of the program between the manager and the employees.

	Cortisol

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
Cortisol is a steroid hormone that is produced by the adrenal cortex. An increase in the concentration of cortisol level mostly has a relationship with energy-mobilizing situations (15–19). Intensive long-lasting, challenging situations are, on the other hand, associated with a lowered concentration of cortisol, and the normal diurnal rhythm, which is characterized by a high value during morning hours, is extinct. The increase in the cortisol level in arousal situations is also attenuated. In psychiatric depression (20), serum cortisol concentrations may be elevated throughout the day and night. This particular study population was a relatively healthy group with high work intensity. With a lowered tension in the work situation, a lowered mean serum cortisol concentration could be expected.

	Serum Lipids

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
Serum lipids are part of the available fuel for energy in aroused situations. The concentration of them is affected both by diet and, to some extent, by arousal. The total serum cholesterol level has been shown to be affected by long-lasting adverse psychosocial conditions (21, 22). In this study, a decreasing total serum cholesterol ratio was predicted, but it was also anticipated that a lowered level could arise in both groups, given that all subjects were continuously informed about their concentration. They had contact with their physicians, and, hence, diet and physical exercise could be affected.

	GGT

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
GGT is a serum liver enzyme that reacts relatively slowly to excessive alcohol consumption. It was anticipated that the program could affect alcohol habits, and, hence, lowered serum GGT would be observed after a year.

	Aims

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
The aim has been to explore whether the management improvement program affects the work environment and health of the whole organization, particularly the situation for the employees.

	METHODS

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
Our plan was to recruit as many employees as possible from the intervention and the comparison departments and to do the measurements in June 1998 (baseline measurement) and in June 1999 (measurement after the end of the program).

Subjects
Table 1 shows the numbers of employees who were asked to participate in the two groups, divided into managers and other employees, respectively. At baseline, 483 employees were asked to participate in the study: 223 were in the intervention department, and 260 were in the control divisions. Of these, 344 employees (215 women and 129 men), 71% participated in the baseline blood-sampling study: 78% in the intervention group (176 persons) and 65% in the control group (168 persons).

At the 1-year follow up of cortisol concentration, the participation rate was 63%: 69% in the intervention group (N = 153, 51 men and 102 women), and 56% in the control group (N = 146, 68 men and 78 women).

The intervention group consisted of employees in the department for individual supplementary insurance.

The control population consisted of three departments in the same company—the financial, marketing, and administrative departments. The proportion of participants from these departments in the cortisol assessments after 1 year was 18%, 47%, and 35%, respectively, with a similar gender distribution and dropout frequency in the three subgroups.

The participation rate in the questionnaire part of the study was smaller than that in the assessment of blood parameters. For the key variable, authority over decisions, the numbers of subjects who participated both in the baseline and the 1-year follow-up in the intervention and control groups were 119 (53%; 38 men and 81 women) and 132 (51%, 63 men and 69 women), respectively.

The number of managers who participated in the blood sampling both at baseline and after 1 year was 23 (12 men and 11 women) in the intervention group and 19 (12 men and 7 women) in the control group. The corresponding numbers for the 1-year follow-up in questionnaires of authority over decisions were 20 (11 men and 9 women) in the intervention group and 15 (10 men and 5 women) in the control group. The ratio of managers to employees who participated in the blood sampling was similar in the intervention and control groups, respectively—15% in the former and 13% in the latter. The 1-year follow-up participation rate in the group of managers was lower than that among other employees—55% (blood tests) and 48% (questionnaire) in the intervention group and 45% for the blood tests and 36% for the questionnaires in the control group.

The participants filled out a questionnaire and left a blood sample in the morning between 8:00 AM and 10:00 AM. The blood sample (10–20 ml) was drawn from the antecubital vein in the sitting position by experienced licensed nurses who were qualified in anesthesia (which makes them particularly competent to perform blood sampling). The blood samples were centrifuged and the serum transported to the hospital laboratory for analysis.

All laboratory analyses were performed at the Central Laboratory for Clinical Chemistry at the Karolinska Hospital according to established standard methods. This laboratory has a high standard and is continuously monitoring assay errors and variability.

For serum cortisol, a fluoroimmunoassay technique was used (Auto Delfia, Wallac Oy, Turku, Finland). The total (intra- and interassay) coefficient of variation for the lower ranges was 9.0%. For the higher ranges, it was 7.8%. The normal range was 200 to 700 nmol/liter at 8 AM.

The questionnaire covered the following areas: Work environment with respect to the possibility of exercising control over what should be done and how it should be done (authority of decision making); the possibility of developing competence (skill discretion); psychological demands (14); and other basic aspects of the work environment—social climate, psychic energy, clarity of goals, and work pace. The three first dimensions have been assessed by means of the Swedish short version of the demand-control questionnaire (23, 24). Four-point Likert scales were used for all questions that ranged from "never" to "almost always." For authority over decisions, two questions with four graded-response categories (1–4) were used—the first one dealing with influence over what to do at work and the second one with how to it. These were summed to an index ranging from a minimum of 2 (very little) to a maximum of 8 (very much). For intellectual discretion, four questions were used regarding the possibility of learning new things, creativity, and repetitiveness. The total score ranged from 4 (very little) to 16 (very much). Psychological demands finally were summarized on the basis of five questions regarding working hard, working fast, and having conflicting demands. The index thus ranges from 5 (very small demands) to 20 (very high demands). These questions have been used extensively in epidemiological studies. For psychometric properties, see Reference (24.

Work climate and work pace were three indices in a questionnaire designed for the assessment of organizational and staff well being. Each index consists of multipoint questions with standard Likert check-off scales. Response alternatives were graded on a scale of four, from "disagree strongly" to "agree strongly" or "no, never" to "yes, most of the time." For psychometric properties, see References (25, 26.

The subjects were queried about alcohol consumption. The total consumption of alcohol in centiliters of pure alcohol consumed per week was calculated on the basis of responses to questions about consumption of beer, wine, and hard liquor. The regional ethical research committee at the Karolinska Institutet approved the study.

	RESULTS

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
Table 2 shows the results of the analyses of questions regarding authority over decisions, intellectual discretion, and psychological demands. Two-way analyses of variance have been performed for men and women together, but Table 2 shows means and standard deviations for men and women separately. The results from the total groups (managers and employees together), as well as from managers and employees separately, are presented. For authority over decisions, the findings show a statistically significant interaction for the total groups, with similar means in the intervention and control groups at baseline and then decreased level in the control group and slightly increased level after 1 year in the intervention group. The findings also are significant for managers and employees separately. The interaction is, in general, stronger for men than for women, although parallel tendencies are seen for the genders.

The findings for intellectual discretion are similar to those for authority over decisions, although the interaction is not significant. There are two main effects for employees: among employees in the intervention as well as in the control group, there is an increased intellectual discretion after 1 year. Furthermore, the mean intellectual discretion is higher for the control group than for the intervention group.

With regard to psychological demands, there are no significant interactions. There is one main effect: in the total groups (but not in managers or employees separately), the demands were reportedly higher after 1 year in both groups.

In summary, the work environment became more "active" (increasing demands and intellectual discretion) in both groups, but both managers and employees reported different developments in the intervention and control groups with regard to authority over decisions.

Table 3 shows the results for social climate and work pace. There were no reported significant changes in social climate. In general, however, the social climate in the control group was reported to be better than in the intervention group.

Work pace reportedly decreased significantly in both groups. Among employees, the tendencies were similar for the intervention and control groups. Among managers, there was reportedly a more pronounced decrease in work pace in the intervention group, but this failed to reach statistical significance in the interaction analysis.

Table 4 shows the analyses of serum cortisol, GGT, and total serum cholesterol. For serum cortisol, there was a highly significant interaction effect among employees, with decreased cortisol levels in the intervention group and increased levels in the control group. In the manager group, no significant changes were observed. The patterns were parallel in women and men. For serum GGT, there was also a significant, albeit weaker, interaction effect in the total group, with decreased levels in the intervention group and increased levels in the control group. It was parallel in managers and employees as well as among employed men and women.

Total serum cholesterol decreased significantly both in the intervention and control groups. This pattern was consistent across genders and employees/managers, although the main effect failed to reach statistical significance for the managers.

No change in total alcohol consumption was observed for any group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 
The societal debate regarding competition in the field of supplementary pension for white-collar workers affected the whole company. The central management of the company had declared that no employee would be laid off as a consequence of these discussions, given that the financial basis of the company was very stable. Thus, there was no fear among the employees of becoming redundant. However, any change in the conditions would have affected the whole company, and there was anticipation that substantial reorganizations could occur if conditions changed. When the intervention and control groups were assigned, it was felt that the debate would affect the employees in both groups in similar ways. The debate culminated during the spring of 1999. Employees were continuously informed about the discussions. However, in May 1999, the final decision was made by central unions and parliamentary politicians that no changes would be instituted.

The program for the managers in the intervention group was mandatory. The managers did participate throughout the period, with exceptions for illness absence and traveling. Two of them, however, declared that they refused participation. Both of these managers stopped working in the organization shortly after this decision.

In the modern work world, it is impossible to find companies doing nothing in the field of work organization. In this case, there were competing projects that are of relevance to the interpretations of the results of the present study.

In the experimental group, a project with the nickname "Our company on its way" was going on during the spring 1999. The program, which involved more than half of the participants in the intervention group, was designed first of all to improve contacts between the customers and the employees. The other aim was to improve possibilities to recruit new employees by creating attractive conditions (for instance, home-based computers) and improved information. There were no explicit psychosocial goals of this program, although, of course, such effects could arise as a consequence of it. According to the personnel management, there were positive feelings among the personnel when "Our company on its way" was launched but disappointment when it turned out that few substantial improvements were obtained.

In the marketing division of the control group, a program for "concretizing management policies" was instituted during the period September 1998 to June 1999 (almost the same period as the manager program in the intervention group). This program for managers was not as extensive as the program in the intervention group. Furthermore, it was much more individually oriented. Its main goal was to strengthen the individual manager’s ability to institute management policies. Assessments of this ability were made before and after the program.

In the financial division of the control group, finally, working hours were reorganized during the study period. Initially, several employees were severely affected by the international orientation in this part of the organization. A well-organized schedule was instituted during the period, and this was perceived as a major improvement because the number of working hours could be kept to more reasonable levels.

Thus, there were three competing programs that may have affected our interpretations, two of them (in the control group) possibly decreasing and one of them ("Our company on its way") possibly increasing the difference between the intervention and control group. The manager program in the intervention group, however, was unique because of its explicit psychosocial goals, its emphasis on social interaction, and its long-lasting building up of knowledge and group discussion. This makes it likely that the observed differences between the intervention and control group in the development of psychosocial and biochemical parameters have been greatly influenced by this program. Because it is impossible to disentangle the effects of different processes going on, however, it should be emphasized, for instance, that the manager program in the intervention group and the program "Our company on its way" could have mutually strengthened one another.

The mean scores of psychological demands, intellectual discretion, and authority over decisions were all high compared with the population means in Sweden (27). This means that the group could be characterized as a group with "active work" (high demands and high decision latitude). The mean work pace score was also high (28, 29). As in most studied working populations, women reported lower authority over decision than men, which probably reflects higher positions in the hierarchy for men in the organization. The biochemical and hormonal data indicate that we are dealing with an essentially healthy population. The serum cortisol levels as well as the cholesterol and GGT levels were clinically normal. Because we were not dealing with a group with a high prevalence of chronic stress-related disorders, although the group was experiencing high demands, it is possible on a group level to test the hypothesis that an improved psychosocial climate with lowered tension should be associated with lowered serum cortisol.

Two work climate factors were reportedly worse in the intervention group compared with the control group at baseline (women, "social climate" and men, "psychic energy"). Apart from these differences, the two groups were very similar at baseline.

One problem was a relatively large dropout frequency. Comparisons were made between subjects who participated only in the baseline assessment and those who participated both in the baseline assessment and the 1-year follow-up with regard to serum cortisol and authority over decisions, with adjustment for age and gender. No significant differences were observed. In the analyses of serum cortisol, the participation rate was higher in the intervention group (69%) than in the control group (56%). In the analyses presented above, 63% in the intervention group and 52% in the control group showed decreased cortisol concentration. Making the (unlikely) assumption that all the 13% excess nonparticipants in the control group would have had decreased levels (contrary to the hypothesis), the corresponding percentage of participants with cortisol decrease would have been 63% (unchanged) in the intervention group and 61% in the control group—accordingly, a much smaller difference but still in the expected direction. On the other hand, if we make the contrasting (also unlikely) assumption that all the excess nonparticipants in the control group would have had an increased level, the corresponding numbers would have been 63% and 42%. Thus, it is likely that there is a true difference between the groups with regard to cortisol change, but the magnitude of this difference is uncertain. With regard to decision latitude, on the other hand, the difference in participation rate is much smaller, 53% in the intervention group and 51% in the control group. In this case, the possible effect of group differences between participants and nonparticipants in the two groups is very small. The participation rate in the group of managers was lower than that of the other employees. Accordingly, the findings regarding these groups should be regarded with additional caution.

It is interesting to note that the 1-year change in decision latitude in the managers in the two study groups was very similar to the 1-year change among employees. On the other hand, there was no significant difference in serum cortisol development between the two manager groups.

One problem in many studies of this kind is that employees may leave the company or that there may be flows of participants from the experimental group to the control group, etc. This was no significant problem in the studied company during the study period. The external turnover (outflow and inflow of staff to/from other companies) both in 1998 and 1999 in the intervention department was 4%. In the control departments, the external turnover was 7% in 1998 and 6% in 1999. The corresponding numbers for internal flow (staff going to or coming from other departments in the company) was 7% in 1998 and 6% in 1999 in the intervention department and 3% in 1998 and 1% in 1999 in the control departments. Fewer than five persons moved between the intervention and control departments during the study period. Our goal was to follow everyone who participated in the baseline examination regardless of where they worked after 1 year. Without exception, they were grouped into their own initial group.

It is important to observe that the main difference between the groups with regard to the development of authority over decisions was that there was a very clear decrease in authority over decisions in the control group, whereas some increase—albeit of a smaller degree—took place in the intervention group. This may indicate that the employees in the control group were experiencing an anxiety-provoking situation during the year of study. This was particularly the case among the female employees in this group. This situation may have been relevant for the whole organization, and the manager program may accordingly have prevented this anxiety and feeling of decreasing possibility to influence the situation among the employees in the intervention group.

The educational program for the managers aimed to improve the managers’ psychological insight. We did not ask the participants specifically about their opinion about the managers, because those types of questions could have affected the development of the whole process. Obviously the managerial initiatives taken and the processes started (triggered by the program) in the intervention department could have been affected by the educational program. In a group interview performed by the main author, the personnel department spontaneously reported that such processes had started during the latter part of the study period—changes that were considered to be triggered by the program.

It could be argued that one serum sample for the assessment of cortisol is a weak basis for conclusions regarding effects on cortisol levels, because serum cortisol shows a clear circadian variation. According to several studies, however, the spontaneous variation in serum cortisol concentration from 8:00 AM to 10:00 AM, when the blood samples were collected, is of a relatively small magnitude (30). The differences are much larger when very early morning samples are compared with evening samples. Furthermore, the mean hour as well as the distribution of hours of blood sampling were the same for the two occasions. Thus, changes in mean serum cortisol could not be attributed to this variation. There is no reason in this study to suspect systematic bias in the sampling procedures that could be related to the findings on the group level, and the relatively large number of subjects compensates for the error variance added due to circadian variation.

The methodological variation in cortisol assay also adds to error variance but could not possibly introduce systematic bias, because intervention and control subject blood samples were randomly mixed throughout the study period.

In this study, the biochemical changes were not significantly related to the psychosocial ones. The correlations between the 1-year change in each one of the psychosocial parameters with corresponding change in cortisol, cholesterol, and GGT were computed. No significant correlation was found. Accordingly, no inference could be made regarding psychosocial factors that cause physiological changes, although of course the changes in cortisol are likely to have been mediated by psychophysiologic mechanisms. At this stage, the two groups of changes should be regarded as different parts of a general improvement.

Two of our biological hypotheses and only one of the psychosocial hypotheses were supported by findings of significant interactions. A somewhat unexpected conclusion is, accordingly, that the support for a physiological effect was stronger than the support for a psychosocial effect. Because the physiological effect must have been mediated by a psychosocial one, it may be that we did not assess the psychosocial effects in a proper way. Perhaps other psychosocial variables should have been included.

The results, in general, support the view that it is possible to improve the work environment and to decrease the arousal level of employees by means of a manager program (31). An important aspect of this program was that it had a moderate intensity (2 hours once every second week) but long duration (a whole year). This creates a good opportunity for gradual processes to start. Because participation was mandatory, the managers had to spend time discussing important organizational problems. The nature of the changes that arose may have to do with many aspects of human communication among managers, between managers and employees, and perhaps also among employees.

Received for publication June 21, 2000.

Revision received January 3, 2001.

	REFERENCES

TOP ABSTRACT INTRODUCTION Situation at the Insurance... Cortisol Serum Lipids GGT Aims METHODS RESULTS DISCUSSION REFERENCES

 

1. SCB-ASS. Arbetsorsakade besvär. Stockholm: National Board of Safety and Health at Work and Statistics Sweden; 1999.
2. Theorell T, Tsutsumi A, Hallquist J, Reuterwall C, Hogstedt C, Fredlund P, Emlund N, Johnson JV,and the Sheep Study Group. Decision latitude, job strain, and myocardial infarction: a study of working men in Stockholm. Am J Public Health 1998; 88: (3) 382–8.[Abstract/Free Full Text]
3. Theorell T, Karasek RA. Current issues relating to psychosocial job strain and cardiovascular disease research. J Occup Health Psychol 1996; 1: 9–26.[Medline]
4. Bosma H, Marmot M, Hemingway H, Nicholson A, Brunner E, Stansfeld S. Low job control and the risk of coronary heart disease in the Whitehall II study. BMJ 1997; 314: 558–65.[Abstract/Free Full Text]
5. Kristensen TS. Job stress and cardiovascular disease: a theoretic critical review. J Occup Health Psychol 1996; 1: 246–60.[Medline]
6. Westerberg L, Theorell T. Working conditions and family situation in relation to functional gastrointestinal disorders. Scand J Prim Health Care 1997; 15: 76–81.[Medline]
7. Kristensen TS. Sickness absence and work strain among Danish slaughterhouse workers: an analysis of absence from work regarded as coping behaviour. Soc Sci Med 1991; 32: 15–27.
8. North F, Syme L, Feeney A, Head J, Shipley MJ, Marmot MG. Explaining socioeconomic differences in sickness absence: the Whitehall II study. BMJ 1993; 306: 361–6.
9. Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q 1979; 24: 285–307.
10. Jackson S. Participation in decision making as a strategy for reducing job related strain. J Appl Psychol 1983; 68: 3–19.
11. Wahlstedt KGI, Nygård CH, Kemmlert K, Torgen M, Gerner Bbjörksten M. Påverkan av en organisationsförändring på arbetsmiljöfaktorer och upplevd hälsa inom brevbäring. Stockholm: Arbete och hlsa 1996; 15: 1–29.
12. Wahlstedt KGI, Edling C. Organizational changes at a postal sorting terminal—their effects upon work satisfaction, psychosomatic complaints and sick leave. Work Stress 1997; 3: 279–91.
13. Orth-Gomr K, Eriksson I, Moser V, Theorell T, Fredlund P. Lipid lowering through work stress reduction. Int J Behav Med 1994; 1: 204–14.
14. Karasek RA, Theorell T. Healthy work. Stress, productivity and the reconstruction of working life. New York: Basic Books, Inc.; 1990.
15. Lundberg U, Frankenhauser M. Pituitary-adrenal and sympathetic adrenal correlates of distress and effort. J Psychosom Res 1980; 24: 125–30.[Medline]
16. Demitrack MA, Gold PW, Dale JK, Krahn DD, Kling MA, Straus SE. Plasma and cerebrospinal fluid monoamine metabolism in patients with chronic fatigue syndrome: preliminary findings. Biol Psychiatry 1992; 15: 1065–77.
17. Demitrack MA, Dale JK, Strause SE, Laue L, Listwak SJ, Kruesi MJ, Chrousos GP, Gold PW. Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients with chronic fatigue syndrome. J Clin Endocrinol Metab 1991; 73: 1224–34.[Abstract/Free Full Text]
18. Mason JW. Psychological influence on the pituitary adrenal cortical system. Horm Res 1959; 15: 345–89.
19. Chrousos GP, Gold PW. The concept of stress and stress system disorders. Overview of physical and behavioral homeostasis. JAMA 267: 1244–52.
20. Rubin AL, Price LH, Charney DS, Heninger GR. Noradrenergic function and the cortisol response to dexamethasone in depression. Psychiatry Res 1985; 15: 5–15.[Medline]
21. Dimsdale JE, Herd JA. Variability of plasma lipids in response to emotional arousal. Psychosom Med 1982; 44: 413–30.[Abstract/Free Full Text]
22. Mattiasson I, Lindgårde F, Nilsson J, Theorell T. Threat of unemployment, sleep disturbance and elevated serum cholesterol. BMJ 1990; 301: 461–6.
23. Theorell T, Perski A, Akerstedt T, Sigala F, Ahlberg-Hultn G, Svensson J, Eneroth P. Changes in job strain in relation to changes in physiological state. Scand J Work Environ Health 1988; 14: 189–96.[Medline]
24. Theorell T. The demand-control-support model for studying health in relation to the work environment: an interactive model. In: Orth-Gomr K, Schneiderman N, editors. Behavioral medicine approaches to cardiovascular disease prevention. Mahwah NJ: Erlbaum; 1996.
25. Arnetz BB. Staff perception of the impact of health care transformation on quality of care. Int J Qual Health Care 1999; 2: 345–55.
26. Petterson I-L, Arnetz BB. Measuring psychosocial work quality and health: development of health care measures of measurements. J Occup Health Psychol 1997; 2: 229–41.[Medline]
27. Theorell T, Michelsen , Nordemar R, Stockholm MUSIC 1 Study Group. Validitetsprövning av psykosociala indexbildningar. In: Hagberg M, Hogstedt C, editors. Stockholmsundersökningen 1. Stockholm: MUSIC Books; 1993.
28. Arnetz BB. Physicians’ view of their work environment and organization. Psychother Psychosom 1997; 66: 155–62.[Medline]
29. Arnetz BB. Techno-stress: prospective psychophysiological study of the impact of a controlled stress-education programe in advanced telecommunication system design work. J Occup Environ Med 1996; 38: 53–65.[Medline]
30. De La Torre B, Sjöberg B, Hedman M, Bartfai G, Diczfalusy E. A study of the short-time variation and interrelationship of plasma hormone levels reflecting pituitary, adrenocortical and testicular function in fertile men. Int J Androl 1981; 4: 532–45.[Medline]
31. Kompier M, Cooper C. Preventing stress, improving productivity. European case studies in the workplace. London, New York: Routledge; 1999.

This article has been cited by other articles:

				A Nyberg, L Alfredsson, T Theorell, H Westerlund, J Vahtera, and M Kivimaki Managerial leadership and ischaemic heart disease among employees: the Swedish WOLF study Occup. Environ. Med., January 1, 2009; 66(1): 51 - 55. [Abstract] [Full Text] [PDF]

				L. Dellve, K. Skagert, and R. Vilhelmsson Leadership in workplace health promotion projects: 1- and 2-year effects on long-term work attendance Eur J Public Health, October 1, 2007; 17(5): 471 - 476. [Abstract] [Full Text] [PDF]

				J. R. B. Halbesleben, H. K. Osburn, and M. D. Mumford Action Research as a Burnout Intervention: Reducing Burnout in the Federal Fire Service Journal of Applied Behavioral Science, June 1, 2006; 42(2): 244 - 266. [Abstract] [PDF]

				J. Head, M. Kivimaki, P. Martikainen, J. Vahtera, J. E Ferrie, and M. G Marmot Influence of change in psychosocial work characteristics on sickness absence: the Whitehall II study J Epidemiol Community Health, January 1, 2006; 60(1): 55 - 61. [Abstract] [Full Text] [PDF]

				M. Kivimaki, J. E Ferrie, J. Head, M. J Shipley, J. Vahtera, and M. G Marmot Organisational justice and change in justice as predictors of employee health: the Whitehall II study J Epidemiol Community Health, November 1, 2004; 58(11): 931 - 937. [Abstract] [Full Text] [PDF]

				T. Pickering Job Stress, Control, and Chronic Disease: Moving to the Next Level of Evidence Psychosom Med, September 1, 2001; 63(5): 734 - 736. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Theorell, T. Articles by Weingarten, A.-M. Search for Related Content PubMed PubMed Citation Articles by Theorell, T. Articles by Weingarten, A.-M. Related Collections Stress and Coping