|
The role of social support in prevention of preterm birth: A case/ control study of pregnant women in Southwest and West Philadelphia A Senior Thesis in Anthropology Neena Desai April 9, 1999
Introduction Preterm birth is the most common cause of infant mortality and morbidity. Preterm birth is traditionally defined as a livebirth prior to 37 completed weeks gestation. The incidence of preterm birth in Western countries is between 6 and 10 percent and has not decreased despite a decline in the incidence of low birth weight. Infants with low birth weights (preterm and small for gestational age) account for less than 10% of all live births in the United States, but the death rate of this population comprises a significant portion of the country's overall infant mortality rate (Goldenberg, 1998). The national infant mortality rate has experienced a significant decline in the last half-century with rates of death per 1000 births dropping from 47.0 in 1940 to 7.6 in 1995. The rate of African-American infant mortality currently is double that of whites, as 15.1 infants deaths were experienced by African American mothers, compared to 6.3 for their White counterparts (U.S DHHS, National Center for Health Statistics, 1996). Traditional explanatory models have not been able to explain the persistence of low birth weight despite early entry into prenatal care. In a review of the vital statistics of 1994, Guyers et al. noted that prenatal care utilization rates have increased for African-Americans despite the persistence of a racial differential in the infant mortality rate. Through observation of the limitations of previous biologic research in explaining the rates of preterm birth, Rowley et al. (1993) have argued for a more sociobiologic model for studying the health and pregnancy experiences of women: We propose that future research focus on the environmental exposures and the social context that create a higher risk of preterm delivery for African-American women. This evolving model requires an interdisciplinary approach that expands the contribution of the social sciences to the biophysical model. This approach emphasizes the need (1) to describe more fully the sociocultural, psychological and behavioral influences of maternal health during pregnancy, and (2) to improve the definition and measurement of the psychosocial constructs and the physical environmental stressors that may be identified with pregnancy outcome in African-American women, and to identify statistical approaches that permit estimates of the contribution of both individual behavior and social forces to the risk of preterm delivery (p.3). Many studies have looked at the black-white differential in infant mortality and have focused on the medical risk factors such as cigarette smoking, alcohol, drug use and time of entry into prenatal care that contribute to this disparity. However, a growing body of literature focuses on the structural risk factors such as cultural and economic issues that affect the distribution of these medical risk factors within a population (Williams, 1990). These structural factors have been overlooked as being relevant determinants of poor birth outcomes. Researchers have proposed many hypotheses to explain the high rates of preterm birth in Western countries. The role of psychosocial factors such as stressful life events has been of interest in this research. A wealth of studies has explored a relationship between preterm birth and stress, though no paper has specifically addressed the correlation between stress and infant mortality. This paper uses preterm birth as a proximate variable for infant mortality because preterm birth is the leading cause of infant mortality. This paper adopts the sociobiologic approach in determining the causation of preterm birth. Although the definitions of stress vary, stress has been conceptualized as the inability of an organism to adapt or respond to external stimuli or maintain homeostasis. Decades ago, Selye focused his research efforts on the patterned physiological effects of laboratory animals in reaction to noxious stressors (Selye, 1956). The body of research devoted to stress and health grew significantly after the development of a checklist of life events from which stress was measured. Holmes and Rahe, who recognized a relationship between life stress and the onset of illness, are credited with the formation of this system (Rahe, Meyer, Smith, Kjaer, and Holmes, 1964). The pair later developed a social readjustment rating scale which scored events such as death of spouse, divorce, marital separation and jail term on a scale of 0-500 based on the degree to which they required readjustment (Holmes and Rahe, 1967). The basis of this research lies in the fact that environmental changes pose unfavorable consequences to health. A significant body of research has focused its efforts towards understanding the pathway by which stress influences health outcomes or how it "gets under the skin." Although this relationship is not clearly understood, it is known that increased stress suppresses immune function. The experiencing of stressful events associated with increased anxiety or depression such as bereavement, divorce, surgery and academic examinations has been related to changes in cell subsets of T-cell lymphocytes, B-cell lymphocytes, NK cells and monocytes (Gennaro and Fehder, 1996). Exposure to stressors has been identified with other health affecting behaviors such as smoking, hypertension and drug use as maintaining a significant association with low birth weight (Orr, et.al 1996). Researchers have also addressed the linkage between socio-economic status and health. The conclusion the individuals of lower socio-economic status have poorer health outcomes is universal. According to Anderson and Armistead (1995), "the SES gradient extends to a wide array of health problems, including heart disease, cancer, stroke, diabetes, hypertension, infant mortality, arthritis, back ailments, mental illness, kidney disease and many others" (p.213). The most direct evidence of the relationship between stress and pregnancy outcomes has been documented in animal studies. One study compared the fetal outcomes of pregnant Wistar rats receiving random uneven shocks with control rates that did not receive signaled shocks. Rats receiving shocks gave birth to infants of lower birthweight than their control counterparts as well as displayed a higher fetal mortality rate (Pollard, 1984). Other stressors in animals such as temperature changes during pregnancy have elicited similar results of lowered birthweight of offspring (Istvan, 1986). The animal studies have provided the most consistent source of evidence regarding stress and low birthweight. Social isolation has been associated with an increased risk of numerous health problems. This body of research has generally defined social isolation as the absence of contact or sustained interaction with individuals and institutions the represent mainstream society (Cobb, 1976). The observed association between social isolation and poor health outcomes has compelled researchers to the formation of a hypothesis that social support functions as a link between poverty and poorer health, especially when a disease is stress related. This premise raises the idea that social support has the potential to serve as a buffer between stress and disease by decreasing vulnerability to stress and increasing host resistance to disease (Cassell, 1976). The basis of this hypothesis rests in the concept that an individual's susceptibility to disease is intensely affected by both the social and physical environments. He or she is protected from the negative effects of stressful exposures when he is included in a network of individuals that provide both emotional and instrumental support. Social support is difficult to define and is subject to a great deal of cultural variation. In his address to the American Psychosomatic Society in 1976, Dr. Sidney Cobb defined social support as "information leading the subject to believe that he is cared for and loved, esteemed, and that he belongs to a network of communication and mutual obligation." Cobb addresses the fact that the idea that supportive interactions among people are important is not a new one. However, the idea that adequate social support may protect people from a variety of pathological states is important (Cobb, 1976). In a review article of 6 prospective large sample field studies, House et al. found that mortality rates are higher among those in the sample who are lowest in social integration (House, Landis, Umberson, 1988). Stress has been defined as "a particular relationship between the person and his or her environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being" (McAnarney and Stevens-Simon, p.789, 1990). The premise of this association is that social support can serve as an intervening variable that buffers against the deleterious effects of stress by creating a environment of shared comfort and emotion. The purpose of this paper is to examine the potential buffering effect of social support on stress among poor pregnant women. All relevant literature that has investigated the association between stress, preterm birth and social support will be reviewed. Case/control qualitative data that focuses on the psychosocial health status of these women during their pregnancy will be utilized in order to assess this association. This paper will describe the stressful domains that African-American, inner-city, pregnant women are exposed to and determine if their available social support networks can buffer against the harmful effects of these stressors and reduce the incidence of preterm birth. Literature Review This section will review the relevant literature that investigated the impact of psychosocial stress and depression on poor birth outcomes through both prospective studies and random intervention trials. Baker et al.,(1997) investigated the impact of social support on depression and health during pregnancy. 9,208 pregnant women filled out a questionnaire at two stages of their pregnancy; at eight weeks before birth and eight weeks after birth. Baker's study differs from other reviewed in this paper in that it measures the buffering potential of social support by self-reported rates of urinary infection, backaches, and depression, not by birth outcomes such as low birth weight and gestational age at delivery. The survey instrument contained 8 questions that dealt with emotional and instrumental support levels from family, friends and neighbors. Material deprivation was measured using household indicators of relative deprivation of tenure and use of car (Baker, et al., p.1327, 1997). The study's findings supported a positive correlation between health status and deprivation of adequate transportation and housing conditions as measured by rates of depression and urinary infection. Levels of depression were significantly different among women who reported high social support that women who reported to having lower social support levels. Baker concludes that " if social support does act to reduce vulnerability to disease, then in this study it had a much more potent effect on mental as oppose to physical health" (p.1333, 1997). Baker raises the possibility that social support and depression are related variables that contribute to overall emotional well being and may not be distinguished from one another. Many studies have measured the effects of stress on pregnancy outcomes. In a prospective study of 340 primiparous white, military wives, Nuckolls et al. (1972) measured the psychosocial assets of women at the time of entry into prenatal care- at or around 24 weeks gestation. Nuckolls' team defined psychosocial assets as the "ability to adapt to pregnancy" and used a survey instrument designed to measure this construct. The investigators mailed the Holmes and Rahe Schedule of Recent Life Events to study participants at about 32 weeks gestation. Pregnancy outcomes were grouped as being either normal or complicated, the criteria for complicated was comprised of birthweight under 2500 grams and preterm delivery at less than 37 weeks gestation. Results from the study showed that women with high psychosocial assets as well as high scores on the SRE were 1/3 less likely to have obstetric and neonatal complications compared to women who had high scores on the SRE and low psychosocial assets. Because the psychosocial assets and SRE were measured at 32 weeks gestation, before the onset of delivery, a causal relationship between low assets and increased rate of pregnancy complications can be engendered (Cobb, 1976). Limitations of the study are found in the demographic homogeneity of the study population (military wives) whose lifestyle may be very different from women who live in a more natural support setting (Norbeck and Tilden, 1983, Culhane, 1999). Additional problems include the exclusion of births prior to 32 weeks gestation, which rejected women who may be subjected to high levels of stress and who experience an extremely preterm birth. Another study which found a positive correlation between social support and pregnancy outcomes was done by Norbeck and Tilden, 1983. This study was based on a theoretical model of stress proposed by Sarason, Johnson, and Siegel (1978). This model theorizes that a life situation is followed by a call to action, if the individual feels that the situation is beyond his personal capabilities and available resources, a stressful reaction occurs. Sarason et al. postulates that social support is a critical variable during the appraisal stage and may prevent a situational overload of too many calls to action, which may interfere with optimum individual performance. Norbeck and Tilden classify pregnancy as a situation that requires an appraisal process, thus social support has the potential to act as a resource during the appraisal period. The study sample was composed of 117 women who received care from a large urban medical center at 12-20 weeks gestation. Participants were excluded if they had any preexisting medical risk factors such as preterm delivery, fetal death, drug abuse or other conditions affecting medical history. The investigators employed instruments designed to measure depression, anxiety, self-esteem, tangible support, emotional support, and recent life experiences. Outcome measures included delivery prior to 37 weeks gestation, birthweight of less than 2500 grams and neonatal death within 3 days of delivery. Information was collected by self-reported questionnaires at the onset of prenatal care and six weeks before delivery as well as a review of the medical chart after the delivery. Results showed that high life stress over the past year resulted in an increase in gestational complications and that high emotional dis-equilibrium predicted infant complications. Norbeck and Tilden related the social-support and ego functioning variables used in their study to the "psychosocial assets" measured by Nuckolls et al. (1972) and concluded that a stress- buffering effect of these "psychosocial assets" existed. The investigators concluded that the findings provide significant, but modest relationships between psychosocial variables and pregnancy complications. The model did not include race or class into its determination of stress levels, as well as the exclusion of drug abusers from the study sample may have served to hide the role that substance abuse may play as a behavioral means of coping with stress. In a randomly selected, prospective study of stress, low birthweight and preterm birth, Newton and Hunt used a life events inventory measurement at three stages of pregnancy. Subjects were interviewed at the point of entry into prenatal care at about 30 weeks gestation, again at 37-40 weeks and for a third time after delivery. Life events were dichotomized into major life events, objective major life events and self-rated major life events in order to minimize subjectivity. For example, an objective life event would be Œhusband lost his job', whereas a self-rated life event would be Œincrease in number of family arguments.' The study had a relatively small sample size of 224, with very low results of preterm birth (12) and low birth weight (19). Significant results include that women who experience 2 or more objective life events while pregnant had births with considerably lower mean birth weights as compared to women who did not experience objective life events while pregnant. The study excluded any foreign (outside of the UK) born participant due to language barriers, which limited the heterogeneity of the sample. White middle class, married women composed 95% of the sample. Newton and Hunt address the fact the objective major life events have a more significant association with low birthweight than self-rated events, suggesting that the actual event may be of more importance than the subjective response to it, and may be a more reliable measure of stress. Pritchard and Mfphm (1994) addressed the need for attention to be given to the strains of daily life, or on-going chronic stressors that are often overlooked as being affective of psychosocial health. They investigated the impact of "on-going role strain" on a sample of 393 women who were pregnant with their second child. The investigators defined role-specific strains as "perceived problems within a social role, frustrated expectations, demands and difficulties." (p.89) Questionnaires were sent to women at 20 and 30 weeks gestation and were based on the Kandel scale. The instrument was comprised of agree/disagree statements identifying potential role strain such as time overload, non-reciprocity, strain, and perception of household role as unrewarding. Data was also collected on the husband's social class using the Registrar generals social class scheme as well as self-reported smoking behavior. Low birthweight was defined as less than 2500 grams and preterm birth as less than 37 weeks gestation. Results of the study indicated that women with high role strain scores had (1) husbands or partners in manual employment, (2) single parents, and (3) rented accommodations. Incidence rates of both preterm and low birthweight births were significantly higher in women with high levels of role strain. However, the relationships between these variables were exaggerated by the high incidence of smokers who reported high levels of role strain. High strain smokers represented only 8.5 percent of the sample and account for over 35.3 percent of preterm births and 46.2 percent of the low birthweight. Pritchard and Mfphm call for special attention to be given to this high-risk group. In a prospective random investigation, Orr et al. evaluated the association between psychosocial stressors and low birth weight in a sample of urban women. A sample of 2000 women was enrolled into the study at their first prenatal visit. The investigators designed their own instrument that measured maternal exposure to stressors called the PSEI- Prenatal Social Environmental Inventory. The instrument assessed both major life events that require long-term adjustment and chronic stressful life conditions by assigning a one-point value to for each affirmative response out of 41 total stressor items. Clinical information was collected by medical record abstraction after delivery and provided data on health behaviors such as substance abuse, and weight gain. Results showed a significant association between exposure to stressors and low birthweight among African-American women. African-American women did not experience a greater prevalence of high levels of exposure to stressors than their Caucasian counterparts, rather they appeared to have an increased susceptibility to the deleterious effects of psychosocial stressors (Orr et al., 1996). Significant methodological issues have arisen in the study of stress and pregnancy complications. The main limitation of the prospective studies reviewed that measured stress by the Holmes and Rahe Schedule of Recent Life Events or some other similar checklist is the lack of inclusion of a measurement of chronic or ongoing stressors. Lower income women may suffer from repeated exposure to stressful life events in a manner that is not adequately measured by a life events checklist. Furthermore, the tremendous range in definitions of adverse birth outcomes is problematic because it makes it hard to determine the specific relationship between stress and low birthweight. The next body of literature evaluated the efficacy of social support via intervention strategies. These studies developed a program from which a support intervention was delivered and subsequently tested the ability of this program to affect birthweight and or gestational age at delivery. The conclusion of the majority of these studies is that support interventions are overwhelmingly ineffective in increasing birthweight. However, instead of concluding that support may not play a critical role in prevention of low birthweight the limitations of randomized intervention trials must be recognized. The hypothesis that support can act as a buffer between stress and health outcomes may not be effectively measured through the use of this methodology. This section will review the findings of four randomized control trials designed to measure the effects of social support on pregnancy outcomes. Oakley et al.,(1990) investigated the impact of intervention within a group of socially disadvantaged women with as history of low birth weight births. The use of the term "socially disadvantaged" stems from the fact that 77 percent of the study participants were working class women and the incidence of stressful life events such as inadequate housing, unemployed partners and cigarette smoking was high. Selected women received standard prenatal care and were also paired with a midwife who provided support. The midwife was responsible for providing home visits, information and advice when asked, collected medical information and provided referrals when needed. Findings reported that babies of intervention mothers had a mean birth weight 38 grams higher than that of control group babies. The gestational age at birth was identical among intervention and control group women. Oakley concludes "the provision of social care for pregnant women has the capacity to affect a range of pregnancy outcomes," though this conclusion is not well supported by the findings presented in the paper (p.160). Although women in the study reported that they felt "more in control" during their pregnancy, experienced less worrying before and after birth, as well as an overall satisfaction with the medical care they received, the direct effects of the support intervention on outcomes such as birthweight and preterm delivery are unclear. Oakley et al., (1990) states that "whatever the results of other social support in other pregnancy trials, it is unlikely that social support will ever override the cumulative effects and problems of social disadvantage" (p.161). Other studies of support interventions have come to similar conclusions. Heins conducted a study that investigated the impact of an organization called the Resource Mothers Program that attempts to improve perinatal outcome through social support. The group is comprised of non-professional community women who have five main roles: teacher, role model, reinforcer, friend, and facilitator. Each woman had about 35 cases and made monthly home visits during pregnancy, daily visits during the hospital stay, and regular home visits during the infants 1st year of life. 89% of the study population was African-American, and 93% of the population was single. The caseworkers served as reinforcements of the importance and roles of the mother in child development, as well as provided many instrumental means of support such as transportation to prenatal appointments. The results of the study show that the home visiting group exhibited no reduction in the incidence of low birth weight births or preterm deliveries. Heins concluded that the improvement in perinatal outcomes might have been due to the comprehensive prenatal care received by the cases. The importance of the Resource Mothers input is unclear, but may have played a key role in reinforcing the information learned through prenatal care. A third study reviewed in this paper agreed with the conclusions of Oakley et al., (1990) and Heins et al (1990). Bryce et al., (1991) studied the impact of intervention on women with a history of low birth weight or preterm births. The intervention consisted of home visits and telephone calls to participating women by midwives who were trained in the provision of social support. The intervention began after consent was given and continued at 4-6 week intervals throughout the duration of the pregnancy. Results of the study showed that the rates of preterm birth among women who received additional care (126/981) did not differ significantly from women who did not receive additional care (147/986). Limitations of the study included the fact that the midwives were not allowed to give information that interfered with the care received by the prenatal care provider and that restriction may have confused participating women (Bryce, et al., 1991). Bryce's conclusions echoed that of his peers in that social support interventions do not prove to be successful in the prevention of preterm birth. Villar et al., (1992) studied the efficacy of social support interventions among Latin American women using birth weight and gestational age at delivery as the primary measurements of outcome. This study was original in the sense that it utilized a member of the women's social network such as the mother, husband or friend to participate with the mother in all intervention activities. This support person encouraged the mother to discuss any concerns she may have about the pregnancy or her prenatal care. Support was provided at each home visit, as well as at a special support office that was established at each participating hospital. Participating women could visit the support office without appointment to receive information about their care. Subjects were interviewed at three stages of the antenatal and postpartum experience: during the 36th week of gestation, immediately after the delivery and 40 days post-partum at home (Villar, et al., 1992). The rate of preterm birth was 11.1 percent in the intervention group and 12.5 percent in the control group, while the rate of low birth weight was 8.7 percent in the intervention group and 9.4 percent in the control group (Villar, et al., 1992). Villar's results echo that of the other studies reviewed in this paper in that the intervention did not effect outcomes in a significant manner. Villar et al., echoes Oakley in concluding that "it is evident that the intervention was not sufficient to overcome a lifetime of disadvantage and poor health (p.1269, 1992). However, results did show that women in the intervening group had increased knowledge of pregnancy complications and had positive attitudes about their delivery. It is possible that these women will have better outcomes in their next pregnancy and the knowledge gained from this intervention will present itself in the long-term. The differential findings in this body of research provide mixed conclusions on the impact of stress, social support on pregnancy outcomes. Although the results of the prospective studies do show a positive correlation between stress, social support and pregnancy outcomes, studies designed to improve outcomes through support interventions have been futile in providing an adequate means of measuring the effects of support. Recall Cobb's definition of social support, as "information leading the subject to believe that he is cared for and loved, esteemed, and that he belongs to a network of communication and mutual obligation." Although this construct does not mention that support must come from family alone, inherent in this definition is the concept that the individual is located within this network thus the support is also derived from within the network, not from a detracted third party provider. The lack of success achieved via intervention may mean that home visits are not an adequate means of provision of support. While the "effects of a lifetime of social disadvantage" are tremendous, researchers should rule out the health benefits of support interventions. The socially disadvantaged population at hand may be wary of outside intervention and resistant to a third party- who is now interested in their life and well-being, if a lifetime of social disadvantage proved otherwise. An evaluation of the methods and design of support interventions is necessary in order for to determine the efficacy of these trials. Methods In 1995, the infant mortality rate among African-Americans in Philadelphia was 17.1 deaths per 1000 births, over twice the rate of their White counterparts (Philadelphia Department of Public Health, 1996). The black-white differential in infant mortality is not well understood despite a significant body of research devoted to this topic. The Philadelphia Infant Mortality Review (PIMR) was initiated to explore the non-medical risk factors that contribute the high rates of infant death among African-Americans. The project employed a case-control study design. Cases were comprised of women who experienced an infant death in a seven zip code area of West and Southwest Philadelphia during a three-year period of 1994, 1995, and 1996. West and Southwest Philadelphia are two regions of particular importance due to the fact that the infant mortality rate of this area is slightly higher than that of the overall Philadelphia rate, at 17.6 deaths per 1000 births. Case women were compared to women who gave birth at the same hospital during the same time period, live in the same zip code area, but gave birth to a healthy infant. The two groups were compared on individual and composite variables from four non-medical domains: social support, instrumental support, housing inadequacy and instability and exposure to violence (Culhane, 1999). Case-control methodology has both advantages and disadvantages. The methodology is of a retrospective nature, in which cases are selected because they carry the particular disease under study. The cases are then compared to a group that has similar demographic variables but does not possess the disease under investigation. As subjects of study, both cases and controls maintain the autonomy to decline participation in the study. This is a major disadvantage, because participants may feel that the interview is invasive of privacy. Another drawback is the retrospective nature of the interview in which subjects are not identified until after the event under investigation has taken place. The time period between the event and the interview can be of significant length. This can affect a subject's memory of sensitive topics such as stress, exposure to violence and social support as well as the order of events. There is also an inherent "recall bias" in that information recalled after the fact has the potential to be altered and enhanced by the subjects as they determine which events may have been more relevant in their specific disease. In other words an incident that may not have been thought to be very important at the time may be amended by the knowledge of the outcome of the event in question. Case women were identified through the help of the 15 hospitals in Philadelphia who provided routine fetal death reports, the Pennsylvania SIDS center, as well as the State of Pennsylvania Office of Vital Records. Controls were matched by date of birth, hospital of birth and residence in the same zip code area as the case mother. Controls were selected and matched after the case mother is interviewed. The PIMR employed four community women who contacted the prospective subjects to arrange for an interview. The trained community women interviewed the participating subjects in their own homes. The interviews were taped and transcribed, and were both qualitatively and quantitatively coded where appropriate. The interview instrument was a comprehensive survey of both forced choice and open-ended questions. It consists of six sections: (a) infant death, (b) prenatal psychosocial profile, (c) prenatal care utilization attitudes, (d) labor and delivery experience and outcomes, (e) the post-partum period and demographic variables. The first section, "Infant Death" asks the interviewee to address the events prior to the death of the infant- both the emotional and physical states of life at the time. Questions such as "Since the baby's death, how have you been coping?" provide the subject with the opportunity to give her opinion about the death as well as describe her means of coping with the loss. This section also focuses on the woman's feelings about the quality of care that she received at the time of the infant's death. If the interviewee is a control subject, this section of the interview is ignored. The second section of the interview, "Prenatal Psychosocial Model" asks about the subjects overall life experiences during her pregnancy. The first question is an open-ended inquiry that asks, "During your pregnancy, how would you say that things in your life were going?" This is followed by a long series of forced choice questions in which the subject states of a variety of specific aspects of her life such as money, transportation, housing, and safety were a major or minor concern during the pregnancy. The subject's description of the conditions that were stressful during the pregnancy is very important. This section allows the interviewer to assess how a woman conceptualizes stressful conditions. It asks the woman to describe her social support during the pregnancy and asks a series of forced choice and open-ended questions on both her instrumental and emotional support networks. Questions such as, Did you know someone that would loan you $100 if you needed it? And Could you rely on family and friends for the things you needed, like money?, indicate the availability of resources within the subjects network. Other questions such as Did you know someone that would watch your children when needed?, Did you know someone that would help you with your daily chores if you were sick?, and Did you know someone that would take you to the doctor, if needed? indicate the presence or absence of people who can engage in "doing work." This section examines her exposure to verbal, sexual, physical and emotional abuse. This part of the interview also explores the adequacy of the subjects housing situation during the pregnancy using a series of forced choice questions that probe specific conditions such as: did she own or rent her housing, and her access to basic necessities such as hot water, electricity and a stove. This segment also asks about her pregnancy wantedness and birth control utilization, such as "Thinking back to just before you became pregnant, how did you feel about becoming pregnant?" and "Did you ever consider not continuing your pregnancy?," to determine her attitude towards being pregnant. Included in this section is a query into the nature of the mother's relationship with the father of the baby, and the emotional support he provided during the pregnancy. The prenatal history of the mother and the steps of her entry into the prenatal care system are the focus of the third section of the interview. If she did not seek care until the second trimester of her pregnancy, questions are asked about the reasons for her delay. The section asks about her experience with alcohol, drugs, and tobacco as well as other lifestyle behaviors such as eating, sleeping, and exercise both during and before the pregnancy. The fourth section of the interview focuses on the sequence of events leading to the delivery of the baby, as well as the birthing process. The mother's attitudes about her treatment of care as well as her support levels while in the hospital are ascertained. This subdivision attempts to understand the level of integration between her prenatal providers and the hospital staff in an attempt to determine the continuity of care these women receive as they move within various healthcare facilities. The fifth segment, entitled "Post Partum Module" asks about the discharge from the hospital and the information on taking care of her baby that she received from the hospital. Questions also try to determine the impact of the new baby on the mother's lifestyle in terms of quality of life indices such as housing, childcare, and financial resources. This section also covers the mother's previous job history as well as educational background. Interviews were taped and transcribed when appropriate. After transcription, interviews were entered onto a software program designed for qualitative analysis called QSR NUDIST (Non- Numerical, Unstructured, Data, Indexing, Systematizing and Theorizing). All transcripts were read and all forced choice questions were coded on a blank maternal interview. The numerical codes for the responses to forced choice questions were recorded and entered in SPSS, a database program. Results In the three-year birth cohort of 1994, 1995, and 1996, 13,556 total live births occurred in West and Southwest Philadelphia. Out of this total figure, 11,380 babies were born to African-American mothers. 212 infant deaths occurred during this time period; 190 of those deaths were of babies born to African-American mothers. 167 cases and 93 controls were interviewed for this project. QSR-NUDIST was used to generate reports on specific stressful events that women described during the interview. An example of a report: Actual physical violence among case women would entail commanding the program to run node 17-1-2, a numerical value that corresponds to physical violence. These reports were created for over 15 different variables, and the quotes were analyzed thoroughly. The passages used in this section are a small sampling of the type of events described by women in the sample. These descriptions were chosen because they represent a wide variety of the stressors that women in this group were exposed to during their pregnancies, and also for the fact that they create a mural of deeply expressive remarks about the experience of pregnancy in an urban setting. In order to determine the levels of stress among the sample population, a variety of stressful domains were qualitatively assessed. These domains included; actual physical violence, threatened physical violence, housing inadequacy and instability in terms of both the individuals housing conditions and the perceived quality of the condition of the neighborhood. These domains will be addressed using a comparison of the textual responses of both case and control women in the sample population. Violence Both case and control women report high levels of domestic violence within their familial networks. When asked about the nature of arguments in her household, one case woman responded: Kayla's father was very abusive, he fought me a lot, and I wondered would it cause me to have a miscarriage or any birth defects? He gashed me in my eye, which caused me to get seven stitches, he chipped my tooth, and forced me to have a lot of unwilling sex with him. He beat me for breakfast, lunch, dinner and snack. Whenever he felt like it and for reasons that only he knew. Another case subject describes the fighting that took place between she and her mother as a result of her attempts to prevent her mother from engaging in drug abuse. When asked about her relationship with her mother, she answers: Well me and my mom was fighting, literally fist fighting when I was pregnant. She was doing drugs again and I felt that being her only child and daughter, I'm not going to sit back and let her harm herself. Later in the interview she is asked if anyone ever hit or physically harmed her in any way, and she responds, "my mom hit me. But it was like a mutual fight so...it wasn't like to harm me or nothing." Control women report similar violent exposures during their pregnancies. One woman describes the violence in her household as "Well he [The father of the baby] hit me with the radio in my stomach, and I didn't get a restraining order because he was married to someone else." Later she says, "it only happened when I hit him, he hits me back. But being pregnant he shouldn't hit me anyway you know." Women also report personal violence and abuse from their siblings and other family members. When asked if there was ever any physical fighting or abuse in the household one control woman reports: Respondent: I was fighting with my little sister. She kicked me in the stomach, and my mother just automatically jumped on her side. Interviewer: You all had a disagreement or something, you and your sister? Respondent: Yea. Cause she ate something of mine and I popped her and she kicked me. Interviewer: You think she deliberately kicked you in the stomach because she knew you were carrying a baby? Respondent: Oh yea. The scenarios described above lend a hint into the violent incidents that seem to pervade the lives of women in this population. The episodes are not limited to the partner or the father of the baby but may originate from mothers, siblings and other relatives. Physical violence may be regarded as a culturally accepted behavior within urban families. Perhaps the fact that pregnant women initiate some of these episodes, limits them from blaming their partners when they are hit back. In sum, these occurrences may not take place everyday, but are present in many of the testimonies of the sample population. It appears that the pervasiveness of violence in the lives of urban women has created a culture that is desensitized to violence. This behavior may serve to decrease the awareness of the harmful effects of abusive episodes. The quote of the woman who's mother hit her is evidence of this- "it was a mutual fight, so it wasn't like to harm me or nothing." There appears to be a lack of conceptualization of harmful behaviors and actions, but she does understand that drug abuse is a harmful behavior and she expresses a desire to prevent her mother from engaging in such an act. The survey instrument also inquires about the threat of violence that may occur during pregnancy. Both case and control women describe threats of physical violence that occurred during their pregnancies. One control woman reports Interviewer: During your pregnancy, did anyone ever threaten to hurt you in any way? Respondent: Yea, the baby's father did. It was just stupid stuff. He threatened to put that statue over there- to put that over top of me and knock me out. One day we was arguing, I don't even know what we was arguing about, And he threatened to hit me upside my head. That's how stressed I was. I was just sitting right there and he was standing over top of me, had that statue, while I was pregnant. Another control describes a threat of violence that transpired between herself and a friend of her mothers over childcare. Interviewer: At any time during your pregnancy did any one threaten to hurt you in any way? Respondent: Uh a babysitter. She was watching my kids and my husband didn't have any money to pay her so she threatened to come to my house with a gun. One control woman describes an incident that took place between herself and her cousins over her housing situation: The arguments didn't get bad until 4 weeks before I had her. I got locked out of my apartment and I eventually was forced out. When I went up there to get my stuff and I was with my mom you know I couldn't do anything because I was eight months pregnant. I had to seriously choose between getting into a fight with one of my cousins or letting my mom try to handle it. But, I didn't like to see them putting their hands on my mom so that stressed me out and it actually ended up- I ended up having contractions that night which caused her to come two weeks early because I was that stressed out. Case women also describe threats of violence they received during their pregnancy. One woman described the threat of violence posed by the father of the baby: He didn't want me there he didn't want the baby. He acted like everything was okay in front of his mom, like you know buy the little baby clothes and show her, but then when he go back to the room he was like, I don't want this baby anyway, you're gonna have to find somewhere to go and stuff like that. It was like a mental abuse thing, I knew something was gonna happen physically you know. I was scared. Another control describes the threat of violence posed by her neighbors: "My neighbors threatened to kill me. All of my neighbors had problems with me and my family. Only threats though, because they were too afraid of me." Threats of physical endangerment while pregnant are present throughout the sample population in a variety of forms. The most striking aspect of these quotes is that these threats are posed from individuals close to the subject. The attempts to downplay the importance of these threats resurfaces in the quote from the case subject who describes father of the baby's threat to drop a statue on her head as "just stupid stuff." The control subject who was threatened with a gun over a financial dispute with her babysitter may indicate the level of poverty in this population. This threat was posed by the babysitter, a friend of her mother's who was known and trusted with her children. Another control subject believes that the threat of violence between her cousins and her mother brought on enough stress to cause her to give birth two weeks early. These passages question the nature and strength of the familial ties in this population, which may take a backseat when issues of housing and finances are addressed. Housing Detailed descriptions of housing inadequacy and instability are prevalent throughout the sample population, who depicts residential and environmental issues as a major source of stress during pregnancy. The conditions of the surrounding environment had both positive and negative effects on women in this sample. Women cite a number of aspects of their neighborhoods that made them feel safe during their pregnancy. When asked, "Were there things about your neighborhood that made you feel especially good while you were pregnant? One control woman responds, "The neighbors, we all sorta helped each other out when we needed to and we had an alarm system that really helped us with the safety issue. I felt comfortable with the people, even the people at the corner stores knew me and I felt they were helpful." Another control woman cites her neighbors as being helpful during the pregnancy, "There's a lot of unity on the block. I felt safe. I could leave my door open all night. It's a peaceful neighborhood, no noise." A third control woman cites the help of one particular neighbor, "Betty, she looks out. Sometimes these little hookers walk up and down the streets, and gangbangers might try to stand on the corner. But she (Betty) looks out and keeps this whole block clean of all of that." However, not all control women report such positive feelings about the safety of their neighborhoods and their surrounding communities. One woman describes a fear of going out in her neighborhood while she was pregnant: Every time you opened your windows there was a group of people standing on the corner. And you know you're pregnant and you're carrying a pocketbook, how fast can you go? Or if I got in my car, I was worried about somebody hitting me, although its just a phobia I guess. They (drug pushers) gang up on the corner and everybody's gotta say something smart when you walk by. You're scared to say anything you know. And if someone says something, you gotta pick up your pace. And being pregnant, how fast could I walk? Another woman describes an incident that occurred during her pregnancy: One time I was pregnant and I had got stuck up during my pregnancy. After that I didn't feel like going out, I was like five months then. They was selling drugs on the corner and they saw me and my girlfriend and walked up to me, and put the gun to my stomach. Told me if I wouldn't give the money up they'd kill me, so I gave the money up. Case women give similar descriptions of their neighborhood, with a mix of positive and negative attributes. When asked, "What were the things about your neighborhood that made you feel safe during your pregnancy?," One woman responds: "Yeah, this block is like all family. There is this house, then my aunt house and I have two more aunts on this block also. Everybody was around. I mean when you have your family that's close to you, there's a sense of security." Another case woman reports, " The cops always around here. During the school hours, the cops would always walk around everybody's houses. Make sure that everybody was in school you know, so I felt pretty safe." However, many case women describe the problem of crime assaults and burglaries in their neighborhood. One extreme case is described here: We had people around here getting stuck up, stuck up. Earrings and stuff snatched. You be scared to go outside, around the area. You just had like people coming in jumping out of cars with masks and stuff on while I was pregnant I had seen these guys they jumped out the car and I had to drop down and they were robbing this man on the corner. And I ducked down and they jumped back in the car. They didn't get nothing. They beat him for $10.00. When asked, "were you concerned for your safety in the neighborhood where you lived, did you ever feel unsafe going out?" When I was pregnant, the whole time I was pregnant I was writing and calling the councilwomen for that area telling her about the guns and the shooting. I was asking her, could they help me with some kind of low income housing or better neighborhood, because I always felt as though one of my children would be caught up in a cross fire and I was always concerned- thinking one of my sons will get shot.
Although neighborhood conditions are a major source of environmental stress, the internal housing situations of the sample population were numerous. Problems with landlords, lack of utilities, overcrowding and unsanitary conditions were often cited as sources of stress during pregnancy in both case and control interviews. One case woman reports of her situation: During the time that I was pregnant, the heater broke and the Gas Company shut my gas off because they said the heater was a violation and if it was turned back on it could blow up. So I had to be in the cold so the Salvation Army put us up in a hotel for two nights. It was cold and we didn't have any hot water. Another woman describes the conditions of her home during her pregnancy: It was definitely unsafe to live in, electrical problems, the roof about to fall in and me being on strict bed rest, trying to do everything that we could possibly do to make the pregnancy peaceful and easy on me, I couldn't rest with all that on my mind. A control woman described one extreme report of poor housing conditions: Interviewer: Were you ever afraid you might be forced to move because of problems with the landlord or people you were living with? Respondent: Yes, the landlord. Gosh that house was disgusting. You know, he didn't want to take care of the house. He didn't want to do the plumbing, didn't want to fix things. It was flea infested, roach infested. So my biggest fear was, here I am pregnant, I just moved in and I am going to have to move again? Where am I going to go after this?
Part of the interview asks women to classify their housing problems as either major or minor problems. When asked about the nature of her housing situation, one control woman responded, "it was a minor problem. Well, the electric, I mean the gas was turned off and it was cold, you know at nights it was really cold." The quotes displayed here illustrate the depth of housing inadequacies that this sample population confronts on a regular basis. A lack of basic necessities such as heat, hot water, a stove or refrigerator are common occurrences in the lives of this population. These quotes give a hint to the mechanism by which the sample population perceives stress. To a middle-class American citizen, even a night without heat may be described as a major problem. However, these women have dealt with far greater stressful exposures in their lives and do not conceptualize even a year without heat as a major issue. Although poor housing conditions are prevalent among the study participants, many of the women do reside in well-maintained housing. A limitation of the survey instrument is that it does not focus on the positive aspects of housing, instead it asks women to describe deteriorated conditions and lack of utilities. To compensate for this shortcoming, all interviewers were required to describe the conditions of the place where the interview occurred, which was usually the woman's home, as well as the conditions of the surrounding neighborhood. The interviewer comments are useful in detecting discrepancies in the information reported during the interview and observations about the woman's lifestyle. Interviewer comments are used to describe the housing conditions of both case and control women who did reside in good conditions. The residences of two case women are described: "The physical conditions of her home were excellent. Her house was very nice, very well kept very modern furniture and wall hangings and it was clean and shiny and everything was just very nice." Another interview commentary reads: "the general impressions of her home were very, very nice. She lived in a great big house with a lot of nice furniture and all kinds of antiques." Control women also resided in residencies of high quality during their pregnancies: "The physical condition of the mother's house is very good. Beautiful walls, nice floors, painted walls, nice clean kitchen. Everything seemed to be new," and "The physical condition of the mother's home as I said was unusually neat and clean. Good furniture, good baby stuff all around the place. A newly rehabilitated house, beautiful kitchen and floors and everything. Beautiful cabinets and all the things a house would need," are only samples of the interviewer's housing comments, which provide evidence of sufficient resources. The events and scenarios described in the quotes above display the exposure to stressful domains of housing inadequacy and instability, neighborhood conditions and violence experienced by inner-city pregnant women. Both case and control women in the sample confronted many similar stressful experiences during their pregnancy. If both groups in the sample population were exposed to the same types of stressful exposures, why did some women achieve a healthy pregnancy? Is it possible than an intervening variable was present in the lives of control women that served to buffer the harmful effects of such negative exposures? An analysis of the support levels of both cases and control women will provide insight into this discrepancy. There was a tremendous range of variation in support networks described by the sample. Social support was measured through both instrumental and emotional descriptions, as well as statement on the attitudes of the family of the mother towards her pregnancy. Two specific relationships are identified as important sources of support. These figures are the mother of the adult daughter and the partner of the adult daughter. Both the active participation of both of these figures in the mother's life during her pregnancy, as well as the consequences of negative attitudes and poor support networks are evaluated. Both case and control women describe the detrimental effects of a deficient support network on their emotional and mental states during their pregnancy. These effects are manifested in the great deal of stress and depression that is reported throughout the interview transcripts. The absence of support from the father of the baby is addressed throughout the sample. Issues contributing to this lack of involvement during the pregnancy include denial of fatherhood, lack of pregnancy wantedness and instability of the mother's relationship with the father of the baby. The issues are described below, as both case and control women grapple with the desertion of the father of the baby from their lives during the pregnancy. One control woman reports: Interviewer: In general how were things going in your life? Respondent: I was depressed, really depressed. My son's father, everything he said to me was negative. And it just seemed like my family on my father's side they was closer to me where I lived at, they didn't come check on me at all, only my aunt, she was the only person that was there for me. I was all alone and stuff. I was so depressed I even had suicidal thoughts. I used to dry a lot and stay in the house. I don't know how I got though it, it was just so hard. Not physically but mentally. That's why I say to anybody that gets pregnant, just try to be married, cause you have somebody there for you. Because if you don't, you going through a whole lot. Another control subject comments on the abandonment of the father of the baby during the pregnancy: [The father of the baby] was scared that I was pregnant. He couldn't like, deal with it, like couldn't believe it. Like, what am I gonna do, I can't take in no baby. That's what he kept saying. Like I can't take care of no baby now...I mean after I got pregnant, he tried to avoid me. He lives up Kensington and he used to come down here like every day and stuff and see me. But then, after I found out, he found out I was pregnant he started like getting scared. Like he just couldn't take like being a man, talking to my mom and dad and so he just kinda be like that. So I mean I just started messing with my old boyfriend again.
Another case subjects describes the effects of the father of the baby's actions on her mental state: He treated me downright wrong. He could see her or whatever. He just did me so wrong. Since I was like 4 months pregnant, I just been so depressed. I cried basically every night. I asked him to come see me. He cheated on me. He did everything a man was not supposed to do he did to me. The beneficial effects of the involvement and engagement of the father of the baby during the pregnancy is described by both case and control women. One case woman reports on the help she received from the father of the baby when she was put on bed rest to avoid an early delivery: He been there 100 percent, he done just about everything. Cooking, cleaning, washing clothes. Working. He was working two jobs and he would come home and make sure me and the kids eat. Make sure their clothes was clean. He would be late to work just to make sure the kids was ready for school. He love me and he was there and he wanted me to make sure that I was all right. Control women also describe the help they received from the father of the baby while pregnant. One woman states, "even though we are separated, my husband would take me back and forth to the prenatal clinic, give me food and food stamps." When asked, "Were you being treated badly by the baby's father? " one woman reports: "He stuck be my all the time. Rubbed my back, talked to the baby. Make sure I ate. Made sure I made it to all my appointments. Anything I needed, he done got it for me. He really took care of me." Instrumental and emotional support is provided by the mother of the adult daughter (baby's grandmother), who's role is of critical importance. Throughout the interview, women describe the positive feelings and help given by their mother during the pregnancy. Their mother's acceptance of the pregnancy is a critical determinant of their health during this time. When asked, "At the time you became pregnant, did you have close friends or relatives who made you feel good about the pregnancy?" one control woman responds: "My mom. She is my best friend. I can go to her for anything, she just always made me feel good. She said she would baby-sit and everything. Like they loved her already." A case subject gives an account of the importance of the mother during her pregnancy: Whenever I needed something, or whenever I needed my mother to watch my daughter or if I needed someone to talk to, my mother would say, well let's talk. We'd stay on the phone, or I'd come up there. Just to get away for a couple of days, that like- took away some of the stress for me. One control woman poignantly expresses the effect of the absence of a mother during pregnancy as she states, "I wanted to die when I was pregnant with Lillian. I felt alone, and I felt that nobody cared. My mother died, and I felt like my best friend was gone and that I didn't have anybody." Feelings of isolation and abandonment also are created by the attitudes of other family members towards the pregnancy. Women throughout the sample discuss the effects of their mother's feelings towards their daughter's pregnancies. One sixteen- year old control describes the treatment she received from her parents, especially her mother: It was rough. No one was supporting me. They disowned me. Everything I wanted I had to do on my own. I had to go out and get a job, and pay rent to them [her parents]. I mean it was a big problem. I'd be in the kitchen and she'd [mother] act like I wasn't there. She'd walk around me. No communication at all. A case woman describes her mother's negative attitude towards her pregnancy, "I got angry with her a lot, she kept telling me, you don't want those babies. You need more babies like you need a hole in the head. I tried to stay away from her, because I didn't feel like hearing her." Supportive networks are cited as sources of encouragement of healthy behaviors and lifestyles. As one case participant reports: Interviewer: What kind of exercise did you do? Respondent: Walking, boy did my mom have me walking. I never walked so much in my life. Interviewer: Where were you like walking to? Respondent: We just walked from here up City Line Avenue. Then my boyfriend, he lives down there by 52nd and Market. So its better him letting me catch the bus. He said, no you need the exercise. You gonna walk. Support is also of critical importance when a mother is engaging in behavior that may harm the fetus. Both case and control women cite the importance of family and friends in encouraging them to stop smoking or drug abuse. As one control woman reports, "I had one good friend. He was constantly telling me like, get off the drugs, think of your baby's sake. Think of your other children. You know, that was like a once in a lifetime shot for having somebody out there that cares." A case woman describes the intervention of her mother in helping her move out of the crack house she was living in, "She didn't want her grandson living in no crack house. She was very concerned for his well-being and she let us stay with her." Another control woman describes how she stopped smoking during her pregnancy: Interviewer: To quit or cut down smoking can be hard to do. Did you get any help to change your smoking habits? Respondent: Yup Interviewer: Yes? What kind of help did you get? Respondent: My boyfriend just kept telling me to stop. The passages and quotes display that pregnant women in this sample were exposed a myriad of stressors within their environments. Their response to these stressors as she seeks the comfort of protective social relationships and attempts to reduce stress through exercise or other healthful mechanisms or through smoking, drinking, or other unhealthful mechanisms. The availability of supportive relationships plays a critical role in the mental and emotional health of the mother during her pregnancy, though is not a powerful enough buffer to influence the birth outcomes when measured by infant mortality. Social support can not be measured quantitatively by outcome of pregnancy. It is a qualitative construct that elicits emotional and psychological responses to a pregnant mother. However, it can act as a deterrent from a variety of behaviors that can negatively affect pregnancy outcome such as smoking, drug and alcohol abuse. Discussion This study addresses the significant exposures to stressors that were experienced by inner-city women during their pregnancy. It is important to acknowledge that the stressors assessed in this paper; violence, housing instability and financial deprivation are of a chronic nature. Chronic exposure to stressors is a difficult topic to surmount. When a woman is constantly dealing with stressful situations in virtually all aspects of her life, that stress may manifest itself in an observed increase of harmful behaviors such as cigarette smoking, drug or alcohol abuse. The buffering role of social support has been observed in a various studies of the literature reviewed earlier. However, this study was of a purely qualitative nature. It was designed to assess the exposure to stressors of inner-city women and the to evaluate the available emotional and instrumental support resources they had to aid in coping with those stressors. However, it was illustrated here that women who did not experience an infant death were exposed to stressors of a similar nature during their pregnancies as women who did experience an infant death. The protective role of social support in acting as a buffer from the negative exposures was not proven here. Case and control women had similar stressful exposures, as well as similar levels of support. While the protective aspects of social support are recognized, these aspects were not strong enough to override the depth of stressful exposures. In this study, the presence of supportive networks did not protect against a negative health outcome such as infant mortality. Although many women in the study did possess seemingly adequate networks of support during their pregnancy, the isolation and depression described by a good portion of the sample must be addressed. Potentially health protective social relationships may occur less often in high poverty environments. African-Americans residing in high poverty areas have a higher percentage of individuals reporting being unmarried, having no current partner, and having no best friend compared with those living in non poverty areas (Anderson and Armistead, 1995). As evidenced by the gripping descriptions of the lives of urban woman, the conditions of the neighborhood can serve as a critical variable. The descriptions provide insight into the role that the environment can play in either strengthening a woman's resistance to various conditions that may threaten the health of the pregnancy, or increase her exposure to negative conditions by exposure to psychosocial insults. The issues of environmental violence and concern for the well-being of themselves and their families as evidenced in the testimonies above show the degree to which external factors can affect one's psychosocial health and well-being. These concerns were found in both case and control groups, illustrating that the exposure to environmental stressors was equal in this domain. It is of critical importance that neighborhood violence and overall environmental conditions are regarded as a source of chronic stress. Women in our sample population illustrate feelings of frustration and helplessness as they attempt to improve their situations. The failure of "the system" to maintain the resources which can adequately provide help to these women when the reach out for it must be addressed. This study highlighted the profound importance of psychosocial stressors in the lives of inner-city African American women, who confront financial deprivation, neighborhood and interpersonal violence, poverty, joblessness and housing inadequacies on a daily level. The chronic and ubiquitous exposure to stress has severe consequences for health and preterm birth. This population has been shown to have an increased susceptibility to the deleterious effects of psychosocial stressors (Orr et al., 1996). If the levels of stressful exposures are high, and the risk for harmful effects of these exposures is also high, the outcomes of such a lifestyle are overwhelmingly negative. Researchers must continue to focus on the psychosocial health of this population in their struggle to create new explanatory models of preterm birth among African-American women. The majority of studies that investigate the role of stress on birth outcomes use measures of life events to measure maternal stress. Life events are regarded as occurrences that require major change or adjustment- both positive and negative situations. These events are thought to be potentially stressful situations or environmental conditions. However, using the life event measurement proves to be difficult when investigating the stressors of an urban population. There seemed to be a significant discrepancy in the self-reporting of resource deprivation within study participants, indicating that the perception of stress is warped within this group. It is possible that if chronic stress is endemic to this population then the instruments from which stress is measured are inadequate. When significant life events occur on a regular basis in the form of job instability, housing inadequacy, death of a loved one, and burglary, the perception and definition of a stressful event is skewed. The development of an instrument from which chronic stress can be measured is necessary in order increase comprehension of the relationship between stress and preterm birth in an African-American population. Conclusion Traditional explanatory models for the racial disparity in infant mortality have been limited to medical behaviors such as access and quality of prenatal care services, time of entry into prenatal care, smoking and substance abuse. While these models are important, they alone can not account for the persistence of this black-white differential. In recent years, researchers have called for a paradigm shift that moves away from a focus on medical behaviors and into a more socially constructed arena. This redirection has widened the scope of research to include psychosocial, environmental and cultural issues as critical variables that influence birth outcomes. The link between stress and poor health has been applied to pregnancy and birth outcomes. This study addressed the presence of chronic and ubiquitous stress in the lives of pregnant African-American women living in West and Southwest Philadelphia. Exposure to stressors such as housing inadequacy and instability, as well as domestic and neighborhood violence was prevalent throughout the sample population. Women who did not experience an infant death were subjected to very similar conditions as women who did experience an infant death. The mitigating role of social support as a buffer between stress and poor health outcomes was addressed. However, it was found that women who had many protective social relationships during their pregnancies still experienced an infant death. The importance of both emotional and instrumental support was described by both case and control women as affecting both their mental and behavioral states. Thus, the role of social support should not be minimized. The significance of these relationships may not be measured quantitatively, in the form of infant birthweight or overall birth outcome. The effects of supportive networks in increasing pregnancy wantedness, encouragement of healthy behaviors, and providing an outlet from which to release feelings are of critical importance in the mental and emotional health of the mother. Psychosocial risk factors for preterm birth and infant mortality must continue to be addressed. The descriptions given by these women depict a depressing portrayal of the daily hassles and frustrations of life in an urban, poverty-ridden environment. In is no wonder that alcohol, smoking and illicit substance abuse are used in the struggle to cope with such exposures. Strategies aimed to alleviate the mental and emotional health status of these women must be developed to further the understanding of as well as reduce the racial differential in infant mortality that exists in the United States. Bibliography Anderson N, Armstead C. Toward understanding the association of socioeconomic status and health: a new challenge for the biopsychosocial approach. Psychosomatic Medicine. 1995; 57:213-225. Baker D, Taylor H, and the Alspac Survey Team. The relationship between condition specific morbidity, social support and material deprivation in pregnancy and early mother hood. Social Science Medicine. 1997; 45 (9) 1325-1336. Bryce R, Stanley F, Garner J. Randomized control trial of antenatal social support to prevent preterm birth. British Journal of Obstetrics and Gynecology. 1991; 98:1001-1008. Cassel JC. The contribution of the social environment to host resistance. American Journal of Epidemiology. 1976; 104:107-23. Cobb S. Presidential address, 1976 Social support as a moderator of life stress. Psychosomatic Medicine.1976; 38:300-314. Culhane, JF. Psychosocial risk factors associated with death among infants and extremely premature infants: a case/ control study. 1999 Dissertation for a Doctorate of Philosophy, Temple University. Gennaro S, Fehder W. Stress, immune function and relationship to pregnancy outcome. Nursing Clinics of North America. 1996; 31 (2): 293-303. Goldenberg R, Rouse D. Prevention of preterm birth. New England Journal of Medicine. 1998 339:313-320. Guyer B, Strobino DM, Ventura SJ, Singh GK. Annual summary of vital statistics, 1995. Pediatrics. 1995; 96(6) 1029-1039. Heins H, Nance N, Ferguson J. Social support in improving perinatal outcome: the resource mothers program. Obstetrics and Gynecology. 1987; 70 (2): 263-266. Holmes TH, Rahe RH. The social readjustment rating scale. Journal of Psychosomatic Research. 1967; 11: 213-218. House J, Landis K, Umberson D. Social relationships and health. Science. 1988 24 (July) 540-544. Istvan J. Stress, anxiety and birth outcomes: a critical review of the evidence. Psychology Bulletin. 1986; 100:331-348. McAnarney ER, Stevens-Simon C. Maternal psychological stress/depression and low birth weight: is there a relationship? AJDC. 1990; 144(July): 789-792. National Center for Health Statistics, 1996 data. The Office of Maternal and Child Health. Newton R, Hunt L. Psychosocial stress in pregnancy and its relation to low birth weigh. British Medical Journal. 1984; 288 (April): 1191-1194. Norbeck J, Tilden V. Life stress social support and emotional disequilibrium in complications of pregnancy: a prospective, multivariate study. Journal of Health and Social Behavior. 1983; 24 (March): 30-46. Oakley A, Rajan L, Grant A. Social support and pregnancy outcome. British Journal of Obstetrics and Gynecology. 1990; (97) 155-162. Orr ST, James SA, et al. Psychosocial stressors and low birthweight in an urban population. American Journal of Preventive Medicine. 1996; 12(6): 459-466. Pollard I. Effects of stress administered during pregnancy on reproductive capacity and subsequent development of the offspring of rats: prolonged effects on the litters of a second pregnancy. Journal of Endocrinology. 1984; 100:301-306. Pritchard CW, Mfphm PYK. Preterm birth, low birthweight and the stressfulness of the household role for pregnant women. Social Science Medicine. 1994; 38 (1) 89-96. Rahe RH, Meyer M, Smith M, Kjaer G, Holmes TH. Social stress and illness onset. Journal of Psychosomatic Resiliency. 1964; 8 (35): 61-69. Rowley DL, Hogue CJR, Blackmore CA, Ferre CD. Preterm delivery among African-American women: a research strategy. American Journal of Preventive Medicine. 1993; (6) supplement 1-4. Selye, Hans. 1956. The Stress of Life. New York: McGraw-Hill. Villar J, Farnot U, Barros F, Victoria C, Langer A, Belizan J. A randomized trial of psychosocial support during high-risk pregnancies. New England Journal of Medicine. 1992; 327 (18) 1266-1271. Williams DR. Socioeconomic differentials in health: a review and redirection. Social Psychology Quarterly. 1990; 57 (2) 81-99.
|