| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry (K.A.M., J.F.O.), and Department of Psychology (J.D.F.), University of Pittsburgh, Pittsburgh, PA; Department of Psychology (K.R.), University of Helsinki, Helsinki, Finland; Department of Epidemiology (S.A.E.), University of Michigan School of Public Health, Ann Arbor, MI; Department of Psychology (C.A.M.), Worcester State College, Worcester, MA; and The Open University/Birmingham Heartlands Hospital (C.E.L.), Birmingham, United Kingdom.
Address reprint requests to: Karen A. Matthews, PhD, Department of Psychiatry, University of Pittsburgh, 3811 O’Hara Street, Pittsburgh, PA 15213. Email: matthewska{at}msx.upmc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: Participants were 50 men and 50 women matched for occupational prestige who were healthy and middle-aged and who completed measures of mood and conflict simultaneously with measures of ambulatory blood pressure and heart rate recorded every 30 minutes during waking hours of two workdays and one nonworkday; at the end of each day, overall ratings were made. Work strain was assessed by the Work Section of the Self-Evaluation and Social Support Interview Schedule. Multiple level random regression coefficients analyses were conducted.
RESULTS: Men and women with low-prestige occupations experienced more interpersonal conflict, b = -0.03,
p = .04, and higher ambulatory heart rate, b = -4.83, p = .004, throughout the three days of the study. Relative to those with low work strain, those reporting high work strain experienced negative emotion, b = -0.41, p < .0001, and boredom, b = -0.17, p < .0004. End of the day ratings of negative mood were more influenced by work strain among men than among women. No effects of occupational prestige or work strain were obtained for ambulatory blood pressure readings after adjustment for physical activity, posture, and location.
CONCLUSIONS: Individuals in low-prestige occupations experience greater exposure to interpersonal conflict and arousal as indexed by heart rate, which might increase risk for stress-related illnesses often associated with social class. Individuals who report work strain experience negative mood and boredom, both at work and at home. The absence of work effects on ambulatory blood pressure may be due to the participants being healthy.
Key Words: social class • work strain • ambulatory bloodpressure • heart rate
Abbreviations: CHD = coronary heart disease; BP = blood pressure; HR = heart rate; LEDS = Life Events Difficulties Schedule; SES = socioeconomic status.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Investigations of mechanisms underlying the SES-health association focus on the effects of severe poverty (eg, poor hygiene and inadequate medical care) as well as on life-style factors (eg, cigarette smoking and alcohol consumption). Adler et al. (6) pointed to the importance of understanding the psychological mediators by which SES affects health. One potential psychological mediator of SES and health is differential exposure to and experience of stress. McLeod and Kessler (7) found small but consistent associations between SES and reports of negative life events. Kessler (8) reported that distress was more highly associated with life events among lower-SES individuals than among upper-SES individuals, suggesting affective responses to stress may vary because of SES variations in vulnerability or in resilience.
Occupational stress may be particularly relevant to understanding the effects of SES on health. Recently, the Whitehall Study showed that men and women in the lower grades of the British civil service reported having less control and less demanding work than did those in higher grades (9). Job control was measured by items such as "I have a good deal of say in decisions about work; My job requires me to take the initiative"; job demand was measured by items such as "The job often involves working very fast. Different groups at work demand things which are difficult to combine." Low levels of control were related to high risk of newly reported CHD during follow-up; job demands were unrelated to new CHD. Other prospective studies have found an association between the combination of high demands and low control and risk of CHD (10).
Several investigations have evaluated the influence of perceived demand and control at work on ambulatory BP. Momentary assessments of perceived task control were correlated with ambulatory diastolic BP in healthy men and women, although perceived task demands were unrelated (11). Men in jobs characterized as high in demand and low in control have the highest mean ambulatory BP compared with other groups (12–14). These effects are not routinely obtained in women (12, 14–15). Women who were high in both family responsibilities and job strain had elevated blood pressure, but only if they had a university degree (16). A follow-up study reported that hypertensive and normotensive men employed in jobs that were perceived as consistently high in demand and low in control across time had the highest BP compared with men employed in jobs that changed in their characteristics or were high in control or low in demand (17). The last study categorized approximately 20% of the sample as having jobs that were high in demand and low in control.
Little is known about the ordinary, everyday experiences of individuals who are in high vs. low occupational prestige positions or who are in positions of high work stress. Especially relevant to the SES-stress relationship is understanding ongoing negative mood states and exposure to interpersonal conflicts, which is highly correlated with distress. With the experiential sampling method, individuals can be prompted to record throughout the workday and at home their stress levels, mood, and activities. In the present investigation, we used this method to evaluate the effects of stressful job characteristics on men’s and women’s daily experiences across two work days and one nonwork day in men and women matched for occupational prestige. BP and heart rate were recorded at each experiential assessment.
We tested the hypotheses that individuals who had low-status jobs would find their work stressful, which, in turn, would lead them to experience ongoing negative mood and conflictual interactions compared with their high-status counterparts. Those in low-status jobs also might find their jobs unstimulating, so they would be more bored and fatigued than those in high-status jobs. The nonwork day was included in the protocol to examine whether the proposed effects were obtained only at work or spilled over to nonwork locations. If the obtained effects were at work only or if they were stronger at work than nonwork, it would argue for the effects being due to work characteristics. If the effects were equally strong regardless of location, it would argue for the effects being due to other stable individual differences correlated with occupational prestige or work stress. Work stress was measured by a standardized semistructured interview, which can be objectively coded for overall work strain. We evaluated the hypotheses according to gender because some studies show that different indicators of occupational stress are important for women and men and according to whether the measurements were taken at work or home (18).
We explored the possibility that BP and heart rate would be elevated at times of negative emotion or interpersonal conflict among participants who had low-prestige occupations or reported high work stress, although all participants had to be nonobese, nonsmoking, and normotensive to participate in our study. The selection of healthy individuals made it less likely that obtained results were due to healthy individuals being better able to compete for higher prestige jobs, ie, the (un)healthy work effect. Based on a subset of the participants, we previously reported that, among employees in low-prestige occupations, those who attributed enduring malintent to the actions of others exhibited a larger difference between average peak and low BP levels at work than those who did not attribute enduring malintent (19). In the present report, we used multilevel random regression modeling (20), which yields regression estimates for each individual (within-subject) and for groups of individuals and can take into account different error structure assumptions (including serial autocorrelation effects between adjacent residuals when multiple observations are made of the same variable over time).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
6) (21) and the other half worked in technical or clerical positions (Hollingshead occupation scores 
5), with men and women matched for their level of occupational prestige. Examples of jobs classified as 6 include computer programmer, nurse, and lawyer; examples of jobs classified as <6 include mechanic, data entry clerk, secretary, and dental hygienist. Relatively few participants had unskilled jobs, ie, occupational prestige scores of 1 or 2, or an occupational prestige of 9, which usually requires an advanced graduate degree. Eighty-eight percent of the subjects were white and 10% were African American. All subjects worked more than 20 hr/wk and were employed during daytime hours. A telephone interview screened potential subjects to ensure that they were nonsmoking, free of chronic disease including hypertension, <20% overweight (weight: women, 64.3 ± 9.9 kg; men, 79.2 ± 11.0 kg), and not taking medications with known cardiovascular or metabolic effects (psychotropic medications, hormone replacement, and contraceptives). Additional eligibility criteria included resting (laboratory) BP assessments in the normal range (systolic BP 140 mm Hg and diastolic BP 91 mm Hg).
Procedure
After the initial recruitment interview to ensure their eligibility for the study, all participants attended an assessment session in the laboratory. At this session, the protocol was explained, and informed consent was obtained; resting BP levels, height, weight, demographic information, and physical activity were measured. Participants then completed a battery of psychological questionnaires and interviews. Half of the participants completed the semistructured interview, Work Section of the Self-Evaluation and Social Support Schedule (22) at this session, whereas half completed it when they returned for the second session. After training in how to complete the diary, participants then completed a diary entry at approximately 30-minute intervals on two working days from 8:00 AM to 10:00 PM and on one nonworking day from 10:00 AM to 6:00 PM. They wore an ambulatory BP monitor, and each time the cuff deflated, the participants noted in the diary whom they were with, mood states, activities, location, etc. At the end of each of the 3 days the participants completed an "end-of-day" assessment in the diary, including overall measures on psychological and behavioral states while wearing the monitor on that particular day. After wearing the monitor for 3 days, they returned to the laboratory for a second assessment session. At this session, they completed another set of psychological questionnaires and participated in interviews that were not administered at the first session. After the second assessment session, any questions the participant had about the study were answered. The subjects were then paid $100 for their participation in the study.
Measures
Diary of Mood and Events.
The ambulatory diary was designed to measure behavioral and psychological states of potential relevance to describing the work and nonwork days of people varying in occupational prestige or to BP and was adapted from the diary developed by Krantz et al. (23). Participants indicated their location (home, work, or driving a car), posture (reclining, sitting, or standing), physical activity level (low, mild, moderate, or strenuous), substance use (alcohol, caffeine, tobacco, and drugs), interpersonal interaction (yes-no), and the type of interpersonal interaction (positive, neutral, negative, or an argument) at the time of the BP reading. The participants responded to seventeen words describing their current mood states (eg, stressed, angry, happy, and tired) on a scale of 1 (Not at all) to 4 (A Lot). For the present study, the average of the 75 diary entries for each of the 17 mood states were subjected to factor analysis using principal factors method with varimax rotation. Three factors with eigenvalues >1 emerged: the first factor consisted of high positive loadings for stressed, irritable, angry, resentful, nervous, worried, impatient, and sad and was labeled as negative mood; the second factor consisted of high positive loadings for pleased, happy, content, energetic, in control, and interested/involved and was labeled as positive mood; whereas the third factor consisted of high positive loadings for bored, apathetic, and tired, a mood state labeled as bored mood. The items were unit weighted and summed, individually at each of the 75 data collections, according to the positive loadings on the three factors to designate periods of negative, positive, and bored mood during the ambulatory monitoring.
We have previously reported the within-subject correlations between mood and the negative interpersonal interaction (24). The r-to-z–transformed correlations showed that the greater the negative mood the less the positive mood (r = -.36, p < .001) and the greater the negativity of the interpersonal interactions (r = 0.56, p < .001). Positive mood correlated negatively with bored mood (r = -.32, p < .01) and with negative interpersonal interactions (r = -.30, p = .01). Correlations between bored mood and negative mood (r = 0.03) and type of interpersonal interaction (r = 0.04) were nonsignificant.
The end-of-day assessment in the diary included information regarding mood: the subjects responded to the same 17 mood states that had been rated throughout the day. However, they were instructed to respond to these mood items in terms of how strongly they felt each mood during that day. They were also asked to record the most stressful or difficult (and the best) event that had happened to them while wearing the monitor that day and to rate the severity of that event on a scale from 1 (very mild) to 4 (very severe).
Work Interview.
Questions from the Work Section of the Self-Evaluation and Social Support Schedule (22) were administered by trained interviewers to assess attitudinal and behavioral aspects of the work environment. During the interview, employees were encouraged to give detailed descriptions of incidents that happened to them at work, and these behaviors were compared with anchoring examples from a detailed code book of ratings. Speech tone and affect were considered when making the attitudinal ratings. An overall index of work strain was assessed, which is a composite of four scales: identity destructive character of work, security decreasing character of work, negative interaction with colleagues, and work overload. Identity destructive character of work refers to whether the participant thinks she/he is working below her competence or receives critical comments from others regarding her or his performance. Security decreasing character of work refers to the job’s potential to decrease the participant’s financial, physical, emotional, and/or structural security. For example, a job that was temporary, had poor benefits, or was likely to be eliminated would be considered to be security decreasing. Negative quality of interaction measures the extent of negative interactions with colleagues that is not part of the work itself. Work overload measures the participant’s belief that she/he neglects other areas of her life because of work or that she/he feels pressure from work responsibilities.
In addition, responses also were coded for commitment to work (ie, importance of work in the participant’s life in providing a focus or in defining who she/he is, and plans for working over the next 10 years) and for competence (ie, viewing her/his work as positive).
Each of the scales was rated on a scale of 1 (marked) to 4 (little/none). The code for work strain is equal to the most extreme code on the four work strain scales (excluding commitment and competence). For example, if the participant is given a rating of 1 (marked) on security decreasing character of work, her work strain code is also coded 1. The work interviews were tape recorded and rated at a later time by trained personnel. All interviewers and raters were trained by a representative (C.E.L.) from Brown’s LEDS unit in London, who also rated the first 20% of the interviews. Interrater reliability between the LEDS unit representative and the chief rater on the work strain scale was 0.72. The remaining 80 interviews were coded by two trained raters, and any discrepancies were resolved by the LEDS unit representative.
Cardiovascular Measures.
BP and HR were assessed outside the laboratory environment with the Colin ABPM-630 ambulatory monitor. This monitor has the major advantage of using a silent CO2 cartridge in cuff inflations, rather than an inflation pump that can produce varying degrees of noise. The measurement may be less intrusive to subjects when wearing the monitor in the company of others and therefore is less likely to cause subjects to deviate from their normal behavior patterns (25). The BP cuff (child, adult, or large adult size) was placed on the participant’s nondominant upper arm and positioned so that the microphone was over the inner aspect of the arm. The Colin monitor allows for both auscultatory (detected via microphone) and oscillometric determinations of BP. Oscillometric values were used in favor of auscultatory ones because this measurement technique is less prone to contamination if the microphone is not precisely mounted on the artery, and thus serves as a more reliable estimate of ambulatory BP. In addition, the oscillometric method produced less missing data than the auscultatory method. Heart rate was determined based on the number of pulses detected during cuff deflation.
Data Reduction
Cardiovascular Measures.
Approximately 16% of the readings (1,223 of 7,500) were lost due to problems with the monitor or because the participant failed to wear the monitor during the predetermined times. Additional readings were excluded based on criteria developed in our laboratory for assessment of BP in normotensive, healthy populations. Twenty-one readings were excluded because either diastolic BP was >140 mm Hg or <40 mm Hg, or pulse pressure (ie, the arithmetic difference between systolic BP and diastolic BP) was <15 mm Hg. Forty-two systolic BP readings <80 mm Hg and three systolic BP readings >200 mm Hg were excluded because the questionable BP was greater or less than two SDs of the participant’s average BP and inspection of the physical activity and/or posture, as indicated in the ambulatory diary, could not explain an unusually high or low systolic BP. It should be noted that if any of these exclusion criteria were met, all data for a particular point in time were excluded. That is, if a diastolic BP reading was >140 mm Hg, the corresponding systolic BP and HR data were also excluded from the analyses. Thus an average subject produced 62 ± 6.81 valid BP and HR readings; the valid data overall consisted of 82.9% (6,214 of 7,500) of total observations.
Diary of Mood and Events.
An average subject missed fewer than 3.08% ± 3.16% of the mood observations and fewer than 2.60% ± 2.82%, 1.79% ± 2.37%, 2.53% ± 3.28%, 2.40% ± 3.22%, 3.40% ± 3.50%, and 1.99% ± 2.2% of the observations on location, posture, physical activity, substance use, interpersonal interaction, and the type of interpersonal interaction, respectively. Altogether, 85.3% (6,397 of 7,500) of ambulatory diary observations were available for the study.
Statistical Analyses
In addition to simple descriptive statistics, data were analyzed using multilevel random coefficients regression analysis (PROC MIXED) (26). The random regression analysis yields regression estimates for each individual (on a within-subject basis) and for the group. The analysis requires no equality of slopes and intercepts of all variables across the sample and allows different error structure assumptions, including the autocorrelation effects between the adjacent residuals. Because one of the dependent variables was dichotomous, which is not ideal for PROC MIXED, we also applied the GLIMMIX macro (26). The results were virtually identical.
Occupational prestige was associated with reporting a marital/partner status, 
2 = 5.0, p = .03, with 50% of individuals employed in low-prestige occupations being married or with a partner compared with 72% of individuals with high-prestige occupations. Thus, we tested simultaneously whether the diary ratings varied according to gender (0: women, 1: men), occupational prestige (1: low, 2: high), number of children living at home (0: none, 1: one or more), marital/partner status (0: no, 1: yes), and location (0: home, 1: work). The random regression analyses were conducted with the repeated mood ratings as dependent measures with negative, positive, and bored mood and type of interpersonal interaction (scored as 0 = positive or neutral; 1 = negative interaction or an argument) analyzed in separate models. Gender, occupational prestige, number of children living at home, and marital/partner status served as between-person grouping factors, and the home vs. work location served as a time-varying, within-person predictor for these models. The within-person predictor data (eg, location) were person-centered, ie, transformed to deviations from the person means to reduce the covariance between slope and intercept terms across subjects.
Then we added to the above model continuous ratings of Work Strain as a between-person predictor. Interaction terms between Work Strain and gender, occupational prestige, home vs. work location, and negative interpersonal interaction were each tested in separate models of mood consisting of the two main effects of the lower-order components. We did not test the influence of Competence and Commitment ratings because of lack of variance and because of the theoretical focus of this paper.
We used a maximum likelihood estimation method, set up a Variance Components model for the parameter covariance matrix, and modeled the residual covariance matrix according to the first-order autoregressive errors, AR (1). In each model, the covariance parameter estimate for the AR (1) effect was significant (p < .001 for all estimates). We also modeled the residual covariance matrix according to 30-minute units and specified the spatial power error structure. The latter did not produce a significantly better fit to the BP and HR data, although it was a better fit to the mood data. We chose to use the AR (1) modeling, though, because of the fixed time intervals between data collection.
The analyses of BP and HR were similar to the method described above. Thus, our analysis determined, on a within-subject basis, the extent to which momentary changes in BP and HR were attributable to momentary changes in physical activity, posture, and location as these factors significantly affected BP and HR. Physical activity, posture, and location were person-centered and treated as random effects in the models. The analyses first tested the effects of occupational prestige and gender on BP and HR, followed by analyses that also included Work Strain. Next, we tested whether the mood states and interpersonal interaction would moderate any association between occupational prestige, Work Strain, and cardiovascular activity. In addition to within-person fluctuations, we were interested in whether experiencing more negative or bored mood, or less positive mood across the 3 days (between-person effect: average of 75 data points) would act as a moderator in these analyses (see Ref. 20 for "cleanly" differentiating the within-person and between-person effects). When Work Strain yielded significant effects on mood, we explored which subscale of the Work Strain measure contributed to the overall pattern of results. The "end of the day" ratings of mood and stress were analyzed by multiple linear regression analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Effects of Occupational Prestige and Work Strain on Ongoing Mood States and Conflict
Table 1 shows the results of testing the simultaneous effects of occupational prestige, gender, number of children living at home, living with someone, and home vs. work location. Individuals without a spouse or partner were more bored than those with a spouse/partner. In general, participants were in a more negative mood at work than at home, although they were less bored and experienced fewer conflicts at work than at home.
|
Consistent with our hypotheses, individuals who had low-prestige jobs reported more conflictual interactions throughout the 3 days, although they did not experience more negative mood than their high-prestige counterparts. They also tended to be more bored (b = -0.17, t(6,986) = 1.81, p < .07) when the model did not include marital/partner status.
When Work Strain was added to the model, results showed that individuals who reported high work strain reported more negative mood and boredom, but experienced no more conflictual interactions, relative to their low-strain counterparts. Subanalyses showed that these effects were obtained because of both the Work Overload (b = -0.48, p < .001) and Negative Interactions at Work (b = -0.55, p < .001) subscales for negative mood, and because of the Work Overload subscale for boredom (b = -0.19, p < .001). (Recall that these scales are scored such that 1 is high.) Work Strain had a stronger effect on participants’ negative mood (p < .001) and boredom (p < .0001) at work than at home (due primarily to the Work Overload subscale, p values < .03). However, separate analyses of mood measured at home also showed that participants who reported high work strain also experienced more negative mood (p < .01) and boredom (p < .03) at home as well (b = -0.19, t(7,025) = -2.8, p < .006; b = -0.11, t = -4.9, p < .0001 for Work Strain x Location in the analyses of negative mood and boredom, respectively).
The significant interaction between Occupational Prestige and Work Strain for negative mood (b = -0.44, t(6,927) = 2.2, p < .03) showed that the negative mood of participants in high-prestige occupations was more affected by work strain (p < .01) than was the negative mood of those in low-prestige occupations (p < .15) (Figure 1). Subscale analysis showed that this interaction was primarily due to the Work Overload subscale (b = -0.48, p = .06). Neither Occupational Prestige nor Work Strain influenced positive mood ratings.
|
Individuals who reported severe stress at the end of each day experienced more ongoing negative mood and boredom, both at work and at home, independent of Work Strain and Occupational Prestige (p values < .001). Stated differently, individuals who reported negative mood and boredom at work and at home throughout the day reported at the conclusion of the day that they had experienced severe stress.
Ambulatory Blood Pressure and Heart Rate
The random coefficients regression analyses showed no significant effects involving either Occupational Prestige or Work Strain on BP (all p values > .15). We previously reported that within-subject elevations in BP were correlated with elevations in negative and positive mood and low boredom (24). There were no significant interactions of work variables with the mood states or interpersonal interaction in the BP analyses (p values > .11).
In the analyses of HR, employees in low-prestige occupations had elevated heart rate (b = -4.83, t(5,512) = -2.88, p < .004), particularly if they reported experiencing on average more negative mood (p < .001), across the three days (aggregate of 75 data points), relative to employees in high-prestige occupations (p values < .55) (b = -4.1, t(6,114) = -2.60, p < .009 for Occupational Prestige by Negative Mood interaction) (Figure 2). The HR analysis controlled simultaneously for both the within-person and between-person effects of activity level and posture by entering the deviations of the person-specific mean and the means themselves into the regression equations. However, it was possible that the Occupational Prestige effects were due to differences in fitness or body mass index. Analyses of physical activity in the last week reported on the Paffenbarger scale (27) or body mass index revealed no significant results involving Occupational Prestige (p values > .11). Work Strain had no effects on HR (p < .32).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Contrary to expectation, occupational prestige and work strain were not associated. As noted earlier, few participants were either very high or very low in occupational prestige, and all participants were healthy. Taken together, the present sample was not representative of the normal distribution of occupational prestige, and this fact may account for the absence of association. On the other hand, a post hoc examination of the previous literature does suggest that occupational prestige and work strain may not necessarily covary. In the Whitehall study, high occupational prestige was associated with high work demands, which is similar to perceptions of work overload (9), and the combination of work demands and low control was not related to one index of SES, educational attainment (17). Moreover, in the present sample, most reported feeling competent and committed to their work, perhaps because they were stably employed and middle-aged. Thus, the present data indicate that low occupational prestige does not necessarily translate into elevated job stress or strain, and this part of the model was not confirmed.
On the other hand, consistent with expectations, men and women in low-prestige positions experienced more conflictual interactions than did those in high-prestige positions, both at work and at home. They also tended to experience boredom more frequently, although they did not report more negative emotion. Employees in low-prestige jobs did exhibit elevated heart rate across the three days, especially if they also reported high negative emotion. The physiological significance of this pattern of results is not established. One may speculate that heightened heart rate reflects a more generalized style of vigilance to the environment, which would be required if on guard for potential conflict, as evidenced in our study to be associated with occupational prestige. The fact that the associations between job prestige and conflictual interactions and heart rate did not vary by work vs. home location could suggest that stable characteristics of the individual account for the results as opposed to specific characteristics of the work environment.
The findings regarding men and women who varied in work strain were quite striking. Participants with high work strain reported considerably more negative emotion and boredom than did their low-strain counterparts throughout the three days. Although the association between work strain and emotional states was stronger for mood at work than at home, it still was the case that men and women who reported strain at work also had more negative mood and boredom at home than did those who reported less strain at work. This pattern suggests a spillover effect of work strain on home life. It is also consistent with the notion that stable characteristics of the individual account for both work strain and emotional states at home and at work.
The subscales contributing to the pattern of results for overall work strain scale were perceptions of work overload and, to a lesser extent, perceptions of negative interactions with colleagues at work and not the identity decreasing or security destructive character of work. Work overload, ie, high pressures from work responsibility or neglecting other areas of life because of work, would seem to lead to daily experiences of negative mood and boredom, regardless of occupational prestige.
Reports of mood measured at the end of each day were generally consistent with the data collected during the day. Those under high work strain reported at the end of each day that the most stressful event of the day was more severe and that they had experienced more negative mood and boredom compared with their less-strained counterparts. Although episodic negative emotion was correlated with episodic elevations in BP, there was not an association between BP or heart rate and work strain, ie, independent of or in interaction with negative mood.
The study design allowed for testing whether gender influenced the pattern of results, independent of or in interaction with the work characteristics. There were almost no significant effects involving gender, suggesting that, at least in this sample of employees, sex differences were not critical. Recall that we carefully matched men and women for occupational prestige, which is usually unmatched in the general population. The only significant effect involving sex was that men’s ratings of negative mood at the end of the day were more influenced by work strain ratings than were women’s ratings, suggesting that retrospective reports of men’s mood are more associated with work variables than are retrospective reports of women. Stated differently, a woman’s sense of well being reported retrospectively may be more influenced by nonwork variables than a man’s sense of well being.
We explored the possibility that occupational prestige and work strain might be associated with ambulatory BP measures. In retrospect, this possibility was unlikely because we had carefully selected a study group that was healthy, middle-aged, nonsmoking, stably employed, and felt positively about their work. Previous studies of job strain and ambulatory BP included hypertensive men or healthy young adults. Although our study group was not representative of the general population of those in high- vs. low-prestige occupations, it did have the advantage of any obtained differences not being due to poor health, unemployment, or lack of job involvement.
In summary, middle-aged men and women holding low-prestige jobs did experience more interpersonal conflict, to some extent boredom, and elevated heart rate throughout 3 days of the study. Those reporting high work strain, although highly committed to their work, experienced negative emotion and boredom, which spilled over to nonwork settings. These experiences, if repeated daily and often, may have a high emotional cost to their sense of well being and drain resources from meeting the inevitable challenges of daily living.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication January 8, 1999.
Revision received October 21, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Steptoe, S. Kunz-Ebrecht, N. Owen, P. J. Feldman, G. Willemsen, C. Kirschbaum, and M. Marmot Socioeconomic Status and Stress-Related Biological Responses Over the Working Day Psychosom Med, May 1, 2003; 65(3): 461 - 470. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Steptoe and M. Marmot The role of psychobiological pathways in socio-economic inequalities in cardiovascular disease risk Eur. Heart J., January 1, 2002; 23(1): 13 - 25. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
