To determine whether "job strain" (defined as high psychological demands and low decision latitude on the job) is associated with increased workplace diastolic blood pressure and the left ventricular mass index, we conducted a case­control study at seven urban work sites of 215 employed men aged 30 to 60 years without evidence of coronary heart disease. After comprehensive blood pressure screening of male employees (N=2556) at the work site, 87 cases of hypertension and a random sample of 128 controls were studied. In a multiple logistic regression model, job strain was significantly related to hypertension, with an estimated odds ratio of 3.1, after adjusting for age, race, body­mass index, type A behavior, alcohol intake, smoking, work site, 24­hour urine sodium excretion, education, and physical demand level of the job. Controlling for the above variables in subjects aged 30 to 40 years with job strain, we found that the echocardiographically determined left ventricular mass index was, on average, 10.8 g/m2 greater than in subjects without job strain. We conclude that job strain may be a risk factor for both hypertension and structural changes of the heart in working men. (]AMA. 1990;263:1929­1935)
A role for psychosocial factors, such as environmental stress, in the etiology of essential hypertension has been suspected for many years but remains unproved. One reason for this may be that the assessment of the two principal variables involved-stress and blood pressure (BP)-has been problematic. Traditional methods of BP measurement, particularly those done in a clinic, have a low reliability because of error in measurement and high biological variability between measurements. An alternative is to measure BP at the work site, either casually or with the new technology of ambulatory BP monitors (ABPMs). The ABPMs have addressed several problems associated with casual BP measurements. They can obtain readings that are more reliable than casual BP measurement, due to the absence of observer error and the increased number of readings. In addition, ABPM estimates of BP may be a more valid reflection of an individual's true BP, because ABPMs can sample BP during subjects' usual activities. Using ABPMs, we recently showed that BP tends to be highest during working hours and that BP measured on a working day was more highly correlated with echocardiographically determined left ventricular mass than BPs measured either in the clinic or on a non-working day. Most of the subjects in these studies had sedentary jobs, suggesting that the higher BPs at work were due to job­associated psychosocial factors rather than to increased physical activity. One model of stress at work has been shown to be predictive of an increased risk of coronary heart disease as well as psychological symptoms, such as exhaustion and depression. "Job strain," in this model, is a variable according to which jobs characterized by both high levels of psychological work demands (working fast and hard) and low levels of control over the work process (low job autonomy and little use of skill discretion) are stressful (Fig 1). In these studies, however, the effect of job strain on BP and heart mass was not examined. The present study was designed to test the hypothesis that job strain is a risk factor for hypertension and for increased left ventricular mass. Repeated casual measurements of BP taken at the workplace were used to diagnose hypertension and determine our sample. Ambulatory BP monitoring was then used in a further analysis to verify the validity of the diagnoses of hypertension based on casual BP measurements.
SUBJECTS AND METHODS
This is a case­control study of working men conducted at seven New York, NY, work sites employing at least 150 men. These sites were a newspaper typography department, a federal health agency, a stock­brokerage firm, a liquor marketer, a private hospital, a sanitation collection and repair facility, and a department store warehouse. To maximize the generalizability of the findings, work sites and departments within work sites were selected that included a wide range of occupations (both blue-collar and white­collar). However, a number of types of occupations were not sampled (high­strain jobs such as manual laborers, assembly­line workers, and minimum­wage workers and low strain jobs such as scientists and artists) because of problems of logistics or management resistance at a number of work sites. At least 80% of the employees in a department had to participate in the screening for the employees from that department to be eligible for the study. All employed men underwent an initial BP screening procedure conducted by specially trained staff; this included three sitting readings of BP (using the American Heart Association protocol) and a brief evaluation of medical and demographic variables. Altogether, 2556 male employees were screened at the seven work sites. From this screened sample, subjects were eligible for the study if they were between 30 and 60 years old, were employed more than 30 hours per week, were educated in the United States and able to read English, had a body­mass index (BMI) of 30 kg/m2 or less, had no second job of 15 or more hours per week, and had been in their current job for at least 3 years. Cases with a history of high BP had to have entered their current job at least 3 years prior to diagnosis. Subjects were excluded if they had a history of coronary, cerebrovascular, or peripheral vascular disease; electrocardiographic evidence of myocardial infarction, ischemia, or atrial fibrillation; funduscopic changes; evidence of any secondary cause of hypertension; diastolic BP (DBP) greater than 105 mm Hg, or systolic BP greater than 160 mm Hg at screening. In addition, subjects who reported taking any drug that might affect their BP (higher or lower) were also excluded from the study, unless they were taking antihypertensive medication for hypertension. Subjects with diagnosed hypertension were eligible only if they could have their medication stopped for at least 3 weeks before wearing the ABPM, with a DBP below 105 mm Hg. Altogether, 1291 men were eligible for the study after the first screening.
BP Criteria for Selection of Cases and Controls
Based on the average of the last two (of three) casual BP measurements taken during the work&SHY;site screening, Subjects who met the above eligibility criteria and had none of the exclusion criteria were divided into two groups: (1) those who had a DBP greater than 85 mm Hg or who were taking antihypertensive medication for hypertension (n = 155) vs. (2) those who had a DBP of 85 mm Hg or less (n= 1136) and were not taking antihypertensive medication. At each work site, all subjects in the first group and a random sample of the second group were invited to a recruitment session. This stratified design was chosen to maximize our ability to test the average within&SHY;site association between job strain and hypertension. At the time of recruitment, 4 to 6 weeks after the initial screening, casual BP measurements were again made at the work site, using the same American Heart Association protocol. Subjects from the first group willing to participate in the study whose recruitment DBP again exceeded 85 mm Hg or who were taking antihypertensive medication were defined as cases, while those willing to participate from the second group whose DBP was again 85 mm Hg or less were defined as controls. Cases and controls were recruited in a ratio of two cases to three controls. Subjects whose BP crossed over at the recruitment visit (initial screening DBP >85 mm Hg and recruitment session DBP </=85 mm Hg, or initial screening DBP </=85 mm Hg and recruitment session DBP >85 mm Hg) were not invited to participate (Fig 2). Based on the initial work&SHY;site screening, 155 men who had a DBP greater than 85 mm Hg or who were taking antihypertensive medication met all the eligibility and exclusin criteria; all of these men were then invited to a recruitment session. Of this group, 87 individuals who had a DBP still above 85 mm Hg or who were taking antihypertensive medication at the recruitment session agreed to participate in the study; all of them subsequently completed the protocol. From the larger population of individuals (n=1136) with a DBP of 85 mm Hg or less at screening, 233 subjects who met the eligibility and exclusion criteria were randomly selected (stratified by work site) and invited to the recruitment session. Of this group, 136 individuals had a recruitment DBP of 85 mm Hg or less and agreed to participate; 128 completed the protocol. The combined group of cases (n=87) and controls (n=128), totaling 215 subjects, constitutes the case control sample. All subjects gave informed consent. Ambulatory BP monitoring was included as part of the evaluation of all 215 subjects, as a check on the validity of our standardized casual BP measuremnts to identify elevated BP and also to determine whether the observed association between job strain and case-contol status was related to the severity of hypertension. A secondary analysis based on ambulatory BP results was performed, in which three subgroups of cases were defined according to their work-time ambulatory DBP: group 1, above 85mm Hg (n=67); group 2, above 90 mm Hg (n=45); and group 3, above 95 mm Hg (n=23). Controls had an ambulatory DBP of 85 mm Hg or less (n=100). Subjects whose worktime ambulatory SBP was inconsistent with their original classification as a case or control (eg, for group 1, controls with an ambulatory DBP >85 mm Hg or cases with an amblatory DBP.