Low birth weight has received major attention in recent years because of the continuing severity of its consequences. Several factors established to affect birth weight, such as smoking and gravidity, do not account for all of the variance and contribute little to the explanation of ethnic differences. 1 We designed this study to identify the psychosocial effects of work on pregnancy outcome among black and white southern women. While employment in general appears to have no adverse consequences 2 and possibly even a protective advantage 3,4 for pregnancy, job strain during pregnancy may be deleterious.
Recent studies have applied the Karasek 5 job strain model to birth outcomes. 6–11 It posits that high psychosocial demands (task requirements) will compromise health if not countered by high control (the ability to negotiate demands). Typical women’s jobs such as nurse’s aide and cook are found in the high strain category. 5 Nevertheless, studies of job strain on birth outcomes based on Karasek have not found a direct effect on birth weight or preterm delivery. 6–11
African-Americans have higher rates of low birth weight than European-Americans. 12 The enduring gap has not been satisfactorily explained by conventional factors such as socioeconomic status, health behaviors, and stress levels, despite the increased attention recently given to the problem. This study addresses and partly explains the observed black-white differences in birth weight.
Subjects and Methods
We collected prospective data in Tuscaloosa from March 1993 through May 1996, at four public and private clinics serving persons of various socioeconomic levels. All eligible women were interviewed within the first 14 weeks of pregnancy and again at or after the 28th week. Eligibility criteria eliminated high risk patients by selecting only patients: between the ages 20 and 34, receiving early prenatal care, and free of chronic hypertension, chronic diabetes mellitus, cardiopathy, actively-treated epilepsy or thyroid disease. Response rate was 66%.
In-depth private interviews of 30 to 60 minutes were conducted at the clinics by trained interviewers, including the authors. We translated questions to below the 8th grade reading level to preclude comprehension problems. The interview schedule addressed demographics, social support, relationships, stressful life events, pregnancy history, illegal substance use, income, work schedule, job history, and job conditions.
Of the 557 patients who completed the first interview, 500 delivered a viable singleton infant. Three patients were lost to follow-up, two were missing data on job conditions, and four were omitted because of genetic abnormalities in offspring. Women of other ethnicities were excluded from analyses (N = 11), resulting in a total sample of 480 cases.
Job Status and Job Strain
Following Karasek, we tested the hypothesis that women experiencing job strain would deliver lower birth weight babies. All women, employed or not, were selected into the study. Employment was defined as working for 8+ hours a week for more than 2 weeks during the first trimester.
We based the job strain questions on the original Karasek 13 questionnaire. Two represented psychosocial demand (At your job, are you always on the move? Does the work you do on the job cause you to worry a lot?) and four represented the control dimension (Can you make a 10-minute personal phone call whenever you wish? Can you receive a personal visitor for 10 minutes? Can you leave work and return later during work hours? Must you ask for permission first before you leave?). Women with 1 or more Job Demands were scored as high, those with 2 or fewer Job Controls as low, and those with high demands and low controls as having Job Strain.
We measuredphysical job demands (present/absent) by a six-point scale: work that was (a) physically difficult or that (b) caused sore muscles, a job environment involving (c) offensive odors, (d) a high noise level, or (e) chemicals, and (f) standing for ≥6 hours/day. The scale was then dichotomized with two or more physical demands coded as high (0 = low, 1 = high). While physical demands had no direct effect on birth weight before the entry of job strain into the model, it was included in the regression model for its possible protective effect. 5
We assessed sociodemographic and behavioral factors such as age, ethnicity (0 = black, 1 = white), income (ranked by insurance type: public assistance; working poor on Medicaid; privately insured) and smoking (0 = no, 1 = yes) at interview. Our review of clinic and hospital charts provided the standard covariates: sex of baby (0 = male, 1 = female); mother’s height (cm); pregravid body mass index (kg/m2); gravidity; spontaneous and induced abortions. We measured the outcome, birth weight, in gm.
We used SPSS for Windows to calculate bivariate associations and ordinary least squares regression. Preliminary regression analysis was run on birth weight with mother’s height, sex of baby, smoking, gravidity, body mass index, ethnicity and physical demands as covariates and the no strain and job strain categories entered separately as categorical variables (unemployed = referent category). The small difference between the unemployed and employed-no-strain categories allowed us to combine them as “no strain” for further analyses. We used regression to assess job strain on birth weight with the same covariates.
Bivariate analyses revealed differences by ethnicity and employment status for several of the demographic and standard covariates, as well as outcomes (Tables 2, 3). The mean birth weight was 3229 gm, with a mean of 3340 for whites compared with 3125 for blacks. Low birth weight was more than twice as prevalent for blacks (10.1%) as for whites (4.7%). Women with job strain had babies with lower birth weights than either the no strain or non-employed women. Work in a high strain job was associated with an adjusted difference in birth weight of 190 gm (95% CI = 48–333; Table 4). The effect of job strain on adjusted birth weight was greater for blacks than whites (273 vs. 88 gm).
In this sample of healthy adult women, we found a clear effect of job strain on birth weight. The predictive strength of job strain is equivalent to that of behavioral variables such as smoking, and biological effects like baby’s sex. Exposure to job strain can be viewed as similar to other toxins or deleterious factors—those exposed run a higher risk than those not exposed.
Other factors that might be expected to influence the relation, such as hours worked per week or length of time working, did not. While a gap in birth weight by ethnicity was still evident under the no strain condition (220 gm), it was only about half that found in the presence of job strain (405 gm). This finding is consistent with other findings of a greater risk from job strain for black women. 11 Those who felt discriminated against at work were nearly three times as likely to be under job strain (17.2%) as those who were not (6.8%). Racism 14 and weathering 15 —the physical toll from hardships brought on by social inequalities—are two potential avenues to pursue in the search for an explanation for the persistent black-white gap. Table 1
1. Kramer MS. Intrauterine growth and gestational duration determinants. Pediatrics 1987; 80: 502–511.
2. Chamberlain GV. Work in pregnancy. Am J Industr Med 1993; 23: 559–575.
3. Saurel-Cubizolles MJ, Kaminski M. Work in pregnancy: its evolving relationship with perinatal outcome (a review). Soc Sci Med 1986; 22: 431–442.
4. Poerksen A, Petitti DB. Employment and low birth weight in Black women. Soc Sci Med 1991; 33: 1281–1286.
5. Karasek RA, Theorell T. Healthy Work. New York: Basic Books; 1990.
6. Homer CJ, James SA, Siegel E. Work-related psychosocial stress and risk of preterm, low birthweight delivery. Am J Public Health 1990; 80: 173–176.
7. Brandt LPA, Nielsen CV. Job stress and adverse outcome of pregnancy: a causal link or recall bias? Am J Epidemiol 1992; 135: 302–311.
8. Henriksen TB, Hedegaard M, Secher NJ. The relation between psychosocial job strain, and preterm delivery and low birthweight for gestational age. Int J Epidemiol 1994; 23: 764–774.
9. Fenster L, Schaefer C, Mathur A, Hiatt RA, Pieper C, Hubbard AE, Von Behren J, Swan SH. Psychologic stress in the workplace and spontaneous abortion. Am J Epidemiol 1995; 142: 1176–1183.
10. Cerón-Mireles P, Harlow SD, Sánchez-Carrillo CI. The risk of prematurity and small-for-gestational-age birth in Mexico City: the effects of working conditions and antenatal leave. Am J Public Health 1996; 86: 825–831.
11. Brett KM, Strogatz DS, Savitz DA. Employment, job strain, and preterm delivery among women in North Carolina. Am J Public Health 1996; 87: 199–204.
12. Blackmore CA, Ferre CD, Rowley DL, Hogue CJ, Gaiter J, Atrash H. Is race a risk factor or a risk marker for preterm delivery? Ethn Dis 1993; 3: 372–377.
13. Karasek PA, Baker D, Marxer F, Ahlbom A, Theroett T. Job decision latitude, job demands, and cariovascular disease: a prospective study of Swedish men. Am J Public Health 1981; 71: 694–705.
14. David RJ, Collins J.W. Bad outcomes in black babies: race or racism? Ethnicity Dis 1991; 1: 236–244.
15. Geronimus A. Black/White differences in the relationship of maternal age to birthweight: a population-based test of the weathering hypothesis. Soc Sci Med 1996; 42: 589–597.