Preterm delivery is the major cause of perinatal morbidity and mortality in the United States, accounting for 85% of adverse perinatal outcomes.1 Preterm birth rates in the United States have increased from 9.4% in 1991 to 11.6% in 2000,2 despite the Healthy People 2010 goal of achieving a preterm birth rate of less than 7.6%.3 In addition, the rate of preterm birth among African Americans is nearly double that of both non-Hispanic whites and Hispanics. In 2000, African-American women accounted for 14.9% of all births,4 yet experienced 37.7% of the preterm births before 28 weeks of gestation.2 Given that preterm delivery accounts for the large majority of neonates with low birth weight (less than 2,500 g), it is reasonable to assume that similar racial or ethnic disparities exist for the frequency of low birth weight (less than 2,500 g) deliveries. In 2001, non-Hispanic black women had twice the rate of low birth weight deliveries (11.2%) as non-Hispanic white women (5.0%).2 Yet, the basis for the ethnic disparities remains poorly understood.5
One potential etiology of preterm birth and low birth weight delivery is psychosocial stress. There are multiple physiologic pathways linking stress and preterm birth that have been proposed and make the association biologically plausible. Stress is known to activate the hypothalamic-pituitary-adrenal axis stress response, resulting in increased corticotropin-releasing hormone and increased levels of estrogen. It has been proposed that these increases in hormone levels may be linked to the onset of preterm contractions and labor.6 Additionally, chronic stress has been shown to negatively impact the immune system. Given the known connection between infection, inflammatory pathways, and preterm birth, stress-related immunosuppression may play an additional role in the pathology of preterm birth.7–9 Moreover, the finding that African-American women have greater chronic stress than women of other ethnic groups suggests that stress also may contribute to the ethnic disparities in the adverse outcomes of pregnancy.10
Although biologically plausible, the epidemiologic evidence linking stress, ethnic status, and preterm delivery has been somewhat inconsistent.11–22 In the present study, this relationship was further explored using data derived from a population of low-income women who were participating in the Illinois Family Study, a longitudinal investigation of the effect of transitioning from welfare to work on maternal and child health. The fact that all participants in this study faced the socioeconomic pressures of poverty helped ensure that this study population was relatively uniform with respect to other socioeconomic factors predictive of preterm birth and thus assisted in optimizing the assessment of the independent relationship between psychosocial factors and low birth weight delivery.
PARTICIPANTS AND METHODS
The Illinois Family Study is a longitudinal, cohort study that was designed to assess the effects of welfare reform on families in Illinois who were receiving Temporary Assistance for Needy Families in 1998.23 The sample population consisted of 1,899 Temporary Assistance for Needy Families grantees in Illinois, randomly selected from administrative records. Nine Illinois counties were selected for the study: Cook (Chicago and suburbs), St. Clair (East St. Louis and suburbs), Peoria, Fulton, Knox, Marshall, Woodford, Tazewell, and Stark. Combined, these nine counties represent nearly 80% of the Illinois caseload of Temporary Assistance for Needy Families. The sample was drawn by using a stratified random sampling design. The stratification was based on two geographic areas: Cook County and the remainder of the state. Approximately 950 Temporary Assistance for Needy Families grantees were randomly selected per stratum. Illinois Family Study interviewers conducted yearly in-person surveys with participants between 1999 and 2004. The surveys measured changes in socioeconomic status, employment, assistance from government programs, life events, and self-reported stress. Approval for this study was obtained through the institutional review board at Northwestern University.
Participants who gave birth to singletons during the first 4 years of the Illinois Family Study were identified. For each participant, only the first reported birth in the study period was included. Participants were asked to report the birth weight for each child. Each birth weight was then validated through a review of the child's medical record. For this study, low birth weight was defined as less than 2,500 g, the criterion used by the World Health Organization. Low birth weight has been highly correlated with preterm delivery and perinatal morbidity.2,24
The relationship between both demographic and psychosocial variables and low birth weight delivery was investigated. In an effort to identify the degree of chronic stress most temporally proximate to the birth, we analyzed the survey data that was collected closest to the reported delivery date. Thus, all data were from surveys conducted within 6 months of each identified delivery. Psychosocial factors that occurred in temporal proximity to the delivery were grouped into four categories, based on a previously reported conceptual model of chronic stress11: 1) external stressors, 2) buffers of stress, 3) enhancers of stress, and 4) perceived stress. The questions and validated scales from the Illinois Family Study questionnaire used to reflect each category are shown in Table 1.
Data analysis was performed with SPSS 12.0 for Windows (SPSS Inc, Chicago, IL). Univariable analysis was used to explore the relationship between the dependent variable of low birth weight delivery and the independent demographic and psychosocial variables. The relationship between the exposure and outcome was determined by using χ2 and Fisher exact analysis, as appropriate, and reported as odds ratios (ORs) with 95% confidence intervals (CIs). The Breslow-Day test35 (SPSS 12.0 for Windows) was used to evaluate homogeneity of odds ratios after stratification. Variables that were found to be significant in univariable analysis were further analyzed by using multivariable logistic regression to assess their independent contribution to the low birth weight outcome.
Of the 1,363 women interviewed, 294 women (21.6%) became pregnant and delivered singletons during the study period. Of the 294 infants born, 39 (13.3%) were of low birth weight. English was the primary language for 99% of the sample. Non-Hispanic black women accounted for 76.8% of the sample, while 4.5% of women were Hispanic, 16.1% were non-Hispanic white, and 2.7% were classified as “other.” The mean household income of the 165 participants who reported this information was $13,416. There was no significant difference in incomes between women who delivered low birth weight neonates ($15,397±10,274, n=22) and those who did not ($13,111±10,838, n=143), P=.9. The women ranged in age from 19 to 47 years of age, with approximately 34% younger than 22 years, 33% aged 22–26 years, and 32% more than 26 years. Women with a low birth weight delivery had a significantly greater mean age (28.0±5.9 years) than those who did not have a low birth weight delivery (25.1±4.6 years), P<.01. The relationships of other demographic factors with the delivery of a low birth weight neonate are shown in Table 2. No other demographic variable was significantly associated with the low birth weight outcome.
Conversely, as illustrated in Table 3, multiple indicators of chronic psychosocial stress were associated with the low birth weight outcome. Women who reported food insecurity (OR 3.2, 95% CI 1.4–7.2), living in a home with at least one child with a chronic illness (OR 3.4, 95% CI 1.5–7.9), living in a crowded home (OR 2.7, 95% CI 1.3–5.6), being unemployed (OR 3.1, 95% CI 1.2–7.9), or having poor coping skills (OR 3.8, 95% CI 1.7–8.7) were significantly more likely to have a low birth weight neonate. The existence of poor social support (OR 2.0, 95% CI 0.97–4.1), a diagnosis of depression (OR 2.2, 95% CI 0.98–5.1), and perceived economic hardship (OR 2.1, 95% CI 0.98–4.3) also were reported more frequently by those women who delivered a low birth weight neonate, although these associations did not reach statistical significance.
To optimize the identification of chronic stressors most temporally proximate to the delivery, we used data from the survey most proximal to the identified delivery. Therefore, some survey data were obtained before delivery (n=169), and other data were obtained after delivery (n=125) and may have introduced some recall bias. In an effort to assess whether recall bias affected the responses, the associations of the chronic stress variables with low birth weight delivery were stratified by the time of the interview in relation to the delivery. Using the Breslow-Day test for homogeneity,35 we found that, regardless of whether women were interviewed before or after the identified delivery, the association of the chronic stress variables with low birth weight was not significantly different (P>.05 for all comparisons).
Multivariable logistic regression was used to evaluate the independent association of each psychosocial variable with the low birth weight outcome after adjusting for maternal age. Thus, each regression included a single psychosocial variable and maternal age as the independent variable. All psychosocial variables were not included as dependent variables in a single regression given their extensive collinearity. In these regressions, each psychosocial variable, with the exception of home crowdedness, continued to be independently associated with delivery of a low birth weight neonate even after adjusting for maternal age (Table 4).
In this analysis, several measures of self-reported chronic stress in low-income women were significantly associated with delivery of a low birth weight neonate. Specifically, among low-income women, both the presence of chronic stressors (indicated by having difficulty obtaining food, caring for a child with a chronic illness, living in a crowded home, and being unemployed) and having fewer coping skills to withstand these stressors were associated with having a pregnancy with the adverse outcome of a low birth weight neonate.
The link between maternal stress and low birth weight or preterm delivery has been inconsistently demonstrated in previous studies.11–13,15–17 Some of the variation in prior results may be related to methodological inconsistencies in the measurement of stress. For example, many studies that have not demonstrated any association between maternal stress and low birth weight have examined either acutely stressful life events or a limited number of indicators of chronic stress during pregnancy.13,22,36–39 These studies have not, however, typically focused on a broad profile of chronically stressful experiences nor the many conditions that buffer or enhance these experiences.40,41 When investigators have examined chronic psychosocial stress in a wider context and included in the analysis such factors as coping skills, daily stressors, neighborhood safety, and perceived racial discrimination, a relationship between stress and adverse pregnancy outcome has been discerned.11,12,42 Additionally, some investigators have studied women from diverse socioeconomic backgrounds and, as a result, have had difficulty isolating the role of chronic stress associated with poverty from other confounding socioeconomic factors.43 In this study, the relationship between chronic stress and adverse pregnancy outcome was analyzed in a population limited to women who were or had been receiving welfare cash assistance. By examining only these women, the potentially confounding effects of differing income levels and social status were greatly limited. In addition, multiple aspects of chronic stress, including chronically stressful life events, enhancers and buffers of stress, and differences in perception of stress, were included in the analysis. Using this study design, we found results similar to those of Dole et al11 and identified a relationship between self-reported chronic stress and low birth weight delivery.
In further support of the effect of chronic stress, we found increasing maternal age to be associated with low birth weight delivery. This finding contrasts with published data showing that younger maternal age is associated with higher rates of low birth weight delivery in the general population and points to the cumulative effects of chronic stress in a subpopulation of low-income women.10,44,45 Indeed, in studies of birth certificate data in Chicago, Michigan, and New York, investigators have also noted evidence of an increased risk of a low birth weight delivery as maternal age increases among socioeconomically disadvantaged women.46 The “weathering” hypothesis, as proposed by Geronimus,10 suggests that the health status of disadvantaged women “may begin to deteriorate in early adulthood as a physical consequence of cumulative social disadvantage and prolonged active coping with stressful circumstances.” In this population of low-income women, we believe that maternal age serves as a proxy for additional chronic psychosocial stressors that we are not able to easily quantify and is evidence of the weathering effect.
There are several limitations to this study. Although this study did identify significant associations between low birth weight and several measures of stress, we cannot exclude the possibility that other significant associations also could have been discerned with a larger sample size. In addition, while the use of data from the Illinois Family Study allowed for the inclusion of women who had been or were receiving welfare cash assistance in Illinois over a 5-year period, not all factors potentially reflective of chronic stress were collected. For example, a measure of stress related to perceived racial discrimination was not available. In addition, this study assessed only one pregnancy outcome, namely, low birth weight delivery, and did not evaluate other adverse pregnancy events such as miscarriage. We chose low birth weight because it is associated with the majority of perinatal morbidity and also limits the possibility of ascertainment bias in the outcome variable. Although other factors that could be related to adverse pregnancy outcome, such as literacy rate, smoking, drug and alcohol abuse, and domestic violence, were included in the survey, the low rates of positive response did not permit an assessment of associations with the outcome of interest. Lastly, as with any self-reported measure, it is possible that response bias could affect the ascertainment of the association of interest. Given our desire to assess evidence of a chronically stressful environment closest to delivery of the index neonate, we chose to use the survey data from the interview that was most temporally proximate to delivery. Theoretically, women with low birth weight neonates, if interviewed after delivery, may be more likely to have and express greater chronic stress. Yet, this was not the case. When stratified according to the timing of the interview, the demonstrated relationship between chronic stress and a low birth weight delivery was unchanged.
This analysis of chronic maternal stress and its association with adverse birth outcome in a low-income population of women has demonstrated that increased chronic external stressors and a lack of coping skills to buffer these stressors are associated with the delivery of low birth weight neonates. The association of increasing maternal age and low birth weight delivery is further evidence of the cumulative adverse effects of chronic psychosocial stress in this population. These data suggest that additional research to develop methods to measure chronic stress in the perinatal setting may be necessary to better understand the cause of low birth weight deliveries and the reason for the social disparity that exists with respect to this outcome. A better understanding of this relationship may allow the development of more effective risk assessment measures or even interventions that could mitigate the damaging effects of the stress response in these women.
1.Arias E, MacDorman M, Strobino D, Guyer B. Annual summary of vital statistics–2002. Pediatrics 2003;112:1215–30
2.Martin JA, Hamilton BE, Ventura SJ, Menacker F, Park MM, Sutton PD. Births: final data for 2001. Natl Vital Stat Rep 2002;51:1–102.
3.U.S. Department of Health and Human Services. Healthy People 2010. 2nd ed. With Understanding and Improving Health and Objectives for Improving Health. Washington, DC: U.S. Government Printing Office; 2000.
4.Martin JA, Hamilton BE, Sulton PD, Ventura SJ, Menacker F, Munson ML. Birth: final data for 2002. Natl Vital Stat Rep 2003;52:1–114.
5.Mercer B, Goldenberg RL, Moawad A, Meis PJ, Iams JD, Das AF, et al. The Preterm Prediction Study: effect of gestational age and cause of preterm birth on subsequent obstetric outcome. Am J Obstet Gynecol 1999;181:1216–21.
6.Lockwood CJ. Stress-associated preterm delivery: the role of corticotrophin-releasing hormone. Am J Obstet Gynecol 1999;180:S264–6.
7.Rich-Edwards JW, Grizzard TA. Psychosocial stress and neuroendocrine mechanisms in preterm delivery. Am J Obstet Gynecol 2005;192:S30–5.
8.Wadhwa PD, Culhane JF, Rauh V, Barve SS. Stress and preterm birth: neuroendocrine, immune/inflammatory, and vascular mechanisms. Matern Child Health J 2001;5:119–25.
9.Wadhwa PD, Culhane JF, Rauh V, Barve SS, Hogan V, Sandman CA, et al. Stress, infection and preterm birth: a biobehavioural perspective. Paediatr Perinat Epidemiol 2001;15 suppl:17–29.
10.Geronimus AT. The weathering hypothesis and the health of African-American women and infants: evidence and speculations. Ethn Dis 1992;2:207–21.
11.Dole N, Savitz A, Hertz-Picciotto I, Siega-Riz AM, McMahon MJ, Buekens P. Maternal stress and preterm birth. Am J Epidemiol 2003;157:14–24.
12.Hogue C, Bremner J. Stress Model for research into preterm delivery among black women. Am J Obstet Gynecol 2005;192 suppl:S47–55.
13.Hedegaard M, Henriksen T, Secher N, Hatch MC, Sabroe S. Do stressful life events affect duration of gestation and risk of preterm delivery? Epidemiology 1996;7:339–45.
14.Lobel M, Dunkel-Schetter C, Scrimshaw S. Prenatal maternal stress and prematurity: a prospective study of socio-economically disadvantaged women. Health Psychol 1992;11:32–40.
15.Nordentoft M, Lous H, Hansen D, Nim J, Pryds O, Rubin P, et al. Intrauterine Growth Retardation and premature delivery: the influence of maternal smoking and psychosocial factors. Am J Public Health 1996;86:347–54.
16.Copper R, Goldenberg RL, Das A, Elder N, Swain M, Norman G, et al. The Preterm Prediction Study: maternal stress is associated with spontaneous birth at less than thirty-five weeks' gestation. Am J Obstet Gynecol 1996;175:1286–92.
17.Rini C, Dunkel-Schetter C, Wadhwa PD, Sandman CA. Psychological adaptation and birth outcomes: the role of personal resources, stress, and sociocultural context in pregnancy. Health Psychol 1999;18:333–45.
18.Collins JW Jr, David RJ, Symons R, Handler A, Wall S, Andes S. African-American mothers' perception of their residential environment, stressful life events, and very low birthweight. Epidemiology 1998;9:286–9.
19.Bryce R, Stanley F, Garner J. Randomized Controlled trial of antenatal social support to prevent preterm birth. Br J Obstet Gynaecol 1991;98:1001–8.
20.Oakley A, Rajan L, Grant A. Social support and pregnancy outcomes. Br J Obstet Gynaecol 1990;97:155–62.
21.Orr S, Miller C. Maternal Depressive Symptoms and the risk of poor pregnancy outcomes: review of the literature and preliminary findings. Epidemiol Rev 1995;17:165–71.
22.Lu MC, Chen B. Racial and ethnic disparities in preterm birth: the role of stressful life events. Am J Obstet Gynecol 2004;191:691–9.
23.Personal Responsibility and Work Opportunity Reconciliation Act of 1996. 42 U.S.C. 1305 (1996).
24.Iams J, Mercer B; National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. What have we learned about antenatal prediction of neonatal morbidity and mortality? Semin Perinatol 2003;27:247–52.
27.Bickel G, Nord M, Price C, Hamilton W, Cook J. Measuring food security in the united states: guide to measuring food security, revised. Washington, DC: United States Department of Agriculture; 2000. Available at: http://www.fns.usda.gov/fsec/FILES/FSGuide.pdf
. Retrieved October 24, 2006.
28.Johnson JH, McCutcheon SM. Assessing life stress in older children and adolescents: preliminary findings with the Life Events Checklist. In: Sarason IG, Spielberger CD, editors. Stress and anxiety. Vol 7. Washington, DC: Hemisphere Publishing; 1980. p. 111–25.
29.Winston P, Angel RJ, Burton LM, Chase-Lansdale P, Lindsay C, Andrew J, et al. Welfare, children, and families: a three city study, overview and design. Baltimore (MD): Johns Hopkins University; 1999.
30.Snyder CR, Sympson S, Ybasco F, Borders T, Babyak M, Higgins R. Development and validation of the State Hope Scale. J Pers Soc Psychol 1996;70:312–35.
31.Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. J Appl Psychol Meas 1977;1:385–401.
33.Iowa State University. Strengthening families program: for parents and youth 10-14. Available at: www.extension.iastate.edu/sfp
. Retrieved October 25, 2006.
34.Ware JE, Snow KK, Kosinski M. SF-36 Health survey: manual and interpretation guide. Boston (MA): The Health Institute, New England Medical Center; 1993.
35.Breslow NE, Day NE. Statistical methods in cancer research: the analysis of case-control studies. IARC Scientific Publication No. 32. Vol 1. 3rd ed. Lyon, France: World Health Organization, International Agency for Research on Cancer; 1989.
36.Goldenberg RL, Cliver SP, Mulvihill FX, Hickey CA, Hoffman HJ, Klerman LV, et al. Medical, psychosocial, and behavioral risk factors do not explain the increased risk for low birth weight among black women. Am J Obstet Gynecol 1996;175:1317–24.
37.Villar J, Farnot U, Barros F, Victora C, Langer A, Belizan JM; The Latin American Network for Perinatal and Reproductive Research. A randomized trial of psychosocial support during high-risk pregnancies. N Engl J Med 1992;327:1266–71.
38.Hoffman S, Hatch M. Stress, social support and pregnancy outcome: a reassessment based on recent research. Paediatr Perinat Epidemiol 1996;10:380–405.
39.Whitehead N, Hill H, Brogan D, Blackmore-Prince C. Exploration of threshold analysis in the relation between stressful life events and preterm delivery. Am J Epidemiol 2002;155:117–24.
40.Lobel M. Conceptualizations, measurement, and effects of prenatal maternal stress on birth outcomes. J Behav Med 1994;17:225–72.
41.Paarlberg K, Vingerhoets A, Passchier J, Dekker GA, Van Geijn HP. Psychosocial factors and pregnancy outcome: a review with emphasis on methodological issues. J Psychosom Res 1995;39:563–95.
42.Collins JW Jr,David RJ, Symons R, Handler A, Wall SN, Dwyer L. Low-income African-American mothers' perception of exposure to racial discrimination and infant birth weight. Epidemiology 2000;11:337–9.
43.Dole N, Savitz DA, Siega-Riz AM, Hertz-Picciotto I, McMahon MJ, Buekens P. Psychosocial factors and preterm birth among African American and White women in central North Carolina. Am J Public Health 2004;94:1358–65.
44.Geronimus AT. The effects of race, residence, and prenatal care on the relationship of maternal age to neonatal mortality. Am J Public Health 1986;76:1416–21.
45.Geronimus AT, Korenman S. Maternal youth or family background? On the health disadvantages of infants of teenage mothers. Am J Epidemiol 1993;137:213–25.
46.Geronimus AT. Black/white differences in the relationship of maternal age to birthweight: a population-based test of the weathering hypothesis. Soc Sci Med 1996;42:589–97.