Race-related stress and smoking among pregnant African-American women

Authors


Anita Fernander, University of Kentucky, Department of Behavioral Science, 103 Medical Behavioral Science Bldg, Lexington, Kentucky 40511, United States. E-mail: afern2@uky.edu

Abstract

Objectives. To examine the association between the psychosocial construct of race-related stress and smoking among pregnant African-American women. Design. Inferential statistical analyses were performed. Setting. Participants were recruited primarily at a medical clinic as well as through word-of-mouth consistent with the snowball sampling technique. Population. Seventy pregnant self-identified African-American women (32 smokers and 38 non-smokers) 18 years or older participated in the study. Methods. Participants completed self-report measures of the Index of Race-Related Stress and an investigator-developed demographic and smoking questionnaire. Main outcome measures. Smoking status of each participant was established through self-report. Results. Significant associations were found between the smoking status of pregnant African-American women and the frequency and perceptions of overall race-related stress (p < 0.03 and 0.02, respectively), the frequency and perceptions of both individual and cultural race-related stress (p < 0.01, 0.03, 0.05, and 0.03, respectively). No associations were found between institutional race-related stress and smoking status. Conclusions. The findings suggest that integrating race-related stress relieving and coping activities into smoking cessation intervention programs for pregnant African-American women may reduce smoking and subsequent smoking-related reproductive health disparities in the population.

Introduction

The prevalence of smoking during pregnancy has significantly decreased over the past two decades, with 10.7% of pregnant women smoking (down almost 42% since 1990 when 18.4% reported smoking) (1). Increased awareness of the dangers of smoking and implementation of tobacco control policies have likely contributed to declines in smoking during pregnancy, but smoking prevalence remains unacceptably high in the American population. For example, 14% of white American and 8.5% of African-American pregnant women still smoke, which is well above the Healthy People 2010 goal to decrease the percentage of women who smoke during pregnancy to 1.2% (1, 2).

While the prevalence of smoking among pregnant African-American women is less than that of their white American counterparts, African-American women experience poorer smoking-related reproductive health in comparison (3–7). Therefore, reducing or eliminating smoking during pregnancy remains an important goal for both maternal and child health among African-Americans. Smoking during pregnancy has numerous adverse effects on prenatal, perinatal, and postnatal health, including ectopic pregnancy, increased risk of abruption, placenta previa, premature rupture of membranes, spontaneous abortion, preterm delivery, stillbirth, fetal growth restriction, birth defects, asthma, and sudden infant death syndrome (8–11).

Women who quit smoking before or early in pregnancy significantly reduce the risk for several adverse health outcomes to herself and her fetus. However, some research has demonstrated that African-American women are less likely to quit smoking during pregnancy compared to pregnant women of other racially classified social group status and it is unclear why this is the case (12). Modifiable and non-modifiable factors associated with smoking, such as not having a partner or friends who smoke, being married, high levels of education, absence of physical and sexual abuse, lack of weight concerns, emotional health, religiosity/spirituality, reduction, and/or elimination of alcoholic and caffeinated beverages, and low levels of stress are all associated with smoking abstinence as well as likelihood of leading to successful smoking cessation among both white and African-American pregnant women (12–21). Because abstaining from smoking is the single most important preventive health behavior a woman can engage in to significantly reduce her chances of harming herself and her unborn fetus, identifying factors that contribute to why African-American pregnant women smoke during pregnancy is critically needed, particularly given that African-American women are less responsive to smoking cessation interventions and have higher rates of smoking-related morbidity and mortality than their other racially classified social group counterparts (22).

The conventional explanation for why individuals smoke is that the act of smoking relieves stress. Research has revealed that stress is a more prominent feature in the daily lives of African-American women than among white American women (23). Semmes points out that culture is vital in the way a person responds to and copes with psycho-social stress (24). Although stress may occur in different forms, a unique and relatively common stressor that African-American women face is that of race-related psycho-sociocultural stress. Race-related stress has been described as ‘the race-related transactions between individuals or groups and their environment that emerge from the dynamics of racism, and that are perceived to tax or exceed existing individual and collective resources or threaten well-being’ (25, p. 44). Woods-Giscombe and Lobel recently confirmed that race-related stress is a component of the general life stress that African-American women experience (26). Studies conducted among pregnant African-American women have shown that increases in racism are associated with increased stress (23, 27). Furthermore, a major study conducted among pregnant African-American women found that exposure to race-related stress was independently associated with very low birthweight infants (28). Conceptual models have been delineated to elucidate the contribution that stress has on pregnancy. For example, Hogue, Hoffman and Hatch have proposed the classic host (i.e., chronic social stressors) and agent (i.e., immediate emotional, behavioral, or physical response to stressors) association of epidemiological causality (29). It is likely that race-related stress may also contribute to adverse pregnancy outcomes through its impact on unhealthy behaviors such as smoking.

Little is understood about the association between stress and smoking among pregnant African-American women. While researchers have examined the influence of race-related stress on smoking among non-pregnant African-American women, as well as associations between general life stress and smoking among pregnant African-American women, no studies appear to exist that have examined the association between race-related stress and smoking among pregnant African-American women (15, 30–33). With a better understanding of this psycho-sociocultural construct, health practitioners can improve the design and effectiveness of psychosocial smoking cessation programs for pregnant African-American women. Therefore, the purpose of this study was to examine the association between the psychosocial construct of race-related stress and smoking among pregnant African-American women.

Material and methods

Procedures

The study was granted approval by the medical institutional review board at the University of Kentucky. Seventy pregnant African-American women (32 smokers and 38 non-smokers), 18 years of age or older, were recruited primarily at a medical clinic through word of mouth and other venues consistent with the snowball sampling technique. The study utilized the following self-report indices: the Index of Race-Related Stress-Brief Version (IRRS-B) and an investigator-developed demographic and smoking status questionnaire (34). Smoking status of each participant was established through self-report in response to the question ‘Do you smoke?’ with a yes/no response option on the questionnaire.

The IRRS-B is a 22-item scale used to measure lifetime perceptions of and the frequency of race-related stress experienced by African-Americans (34). In addition to an overall score the IRRS-B assesses three subscales of race-related stress: institutional stress, cultural, and individual race-related stress. Institutional race-related stress refers to policies and practices of a given institution that are intentionally implemented under the ideology that one racially classified socially group is superior to another. Individual racism derives from an individual's belief that their racially classified social group is superior to others. Finally, cultural race-related results from the cultural practices of one racially classified social group being lauded as superior than another. The frequency of race-related experiences endorsed is calculated by summing all of the race-related events endorsed by each participant for overall race-related stress, as well as for each subscale. The combined scores of institutional, individual, and cultural race-related events are tallied for overall frequency of race-related events. Mean scores are tallied for each of the experiences of institutional, individual, and cultural race-related perceptions, respectively. Second, perceptions of race-related events are determined by assessing the impact of each event endorsed. Participants responded on a five-point Likert-type scale, ranging from 0 to 4 (0 = this has never happened to me, 1 = event happened but did not bother me, 2 = event happened and I was slightly upset, 3 = event happened and I was upset, and 4 = event happened and I was extremely upset). Mean scores are tallied for the overall score as well as each of the subscales. Previous studies utilizing the IRRS-B reported Cronbach alpha reliability coefficients for the scale ranging between 0.77 and 0.92 among samples of African-American women (30, 31, 35).

Statistical analyses

Descriptive analysis of the total sample and the subsample of smokers involved calculating frequencies as percentages of categorical variables and means with standard deviations of continuous variables. There was less than 3% missing data, so calculations for scaled measures required data for more than three-quarters of the items; otherwise a missing value was assigned. Inferential analyses involved independent sample t-tests for continuous variables and chi-squared analyses for nominal data comparing smokers and non-smokers. All analyses were conducted using SAS® and a conventional alpha level of p < 0.05.

Results

Table I provides a complete description of the sample characteristics. In summary, the total sample consisted of 70 pregnant African-American women, of whom 46% were smokers and 54% non-smokers. Participants were almost equal in responses to where they resided, and ranged from 18 to 37 years in age, with the average age of 25 years. The majority were single, divorced or widowed. Of the women 60% had only a high school education or less, while 74% earned less than $25,000/year. Only 34% reported regular employment. Sample demographics were examined to determine whether age, marital status, income, education, employment, and residence were associated with the smoking status. No associations in the variables of interest between smokers and non-smokers were revealed.

Table 1. Demographics of the pregnant African-American women by smoking status.
VariablesTotal (n = 70)Smoker (n = 32)Non-smoker (n = 38)Significance level
  1. Note: For categorical variables the chi-squared test was used. For continuous variables independent t-tests were used. Percentages may not add up to 100 due to missing data in some variables.

Residence
 Rural32 (46%)17 (53%)15 (39%)0.33
 Urban34 (49%)14 (43%)20 (53%) 
 Age25 (4.92)26 (4.72)24.66 (5.06)0.26
Marital status
 Married/significant other12 (17%)4 (12.5%)8 (21.05%)0.34
 Single/divorced/widowed58 (83%)28 (87.5%)30 (78.95%) 
Income
 Less than $24,99952 (74%)29 (91%)33 (87%)0.62
 More than $25,0006 (9%)3 (9%)3 (8%) 
Educational level
 High-school graduate or less42 (60%)18 (56%)23 (60%)0.87
 Some college24 (34%)12 (38%)12 (32%) 
 College degree3 (4%)0 (0%)3 (8%) 
Employment
 Employed24 (34%)10 (31%)14 (37%)0.62
 Seeking work, student, home-maker, retired, or disabled46 (66%)22 (69%)24 (63%) 

Table 2 provides an overview on the cultural, institutional, and individual and overall race-related stress and smoking status. Significant associations were found with the overall frequency of race-related stressful events as well as overall perceptions of race-related stressful events [t(67) = 2.27, p < 0.03 and t(67) = 2.32, p < 0.02, respectively], i.e., pregnant African-American women smokers reported a greater number of race-related events as well as more negative perceptions of race-related events than non-smoking pregnant African-American women.

Table 2. Race-related stress and smoking status among pregnant African-American women.
VariableSmokers M (SD)Non-smokers M (SD)p-Value
  1. *Significant. Note: F, frequency of race-related events; IRRS, Index of Race-Related Stress; P, perceptions of race-related events.

IRRS-cultural-F7.90 (2.31)6.60 (3.16)0.05*
IRRS-cultural-P2.40 (1.01)1.82 (1.11)0.03*
IRRS-institutional-F3.29 (1.88)2.66 (1.94)0.18
IRRS-institutional-P1.60 (0.91)1.18 (1.02)0.08
IRRS-individual-F4.61 (1.31)3.50 (1.81)<0.01*
IRRS-individual-P2.11 (1.09)1.54 (1.06)0.03*
IRRS-total-F15.81 (4.79)12.74 (6.17)0.03*
IRRS-total-P2.10 (0.91)1.60 (0.97)0.02*

It was further hypothesized that the smoking women would report a greater number of and more negative perceptions of institutional, individual, and cultural race-related stress than non-smoking pregnant women. The African-American women smokers reported more experiences of both individual and cultural race-related stress [t(67) = 2.86, p < 0.01 and t(67) = 1.95, p < 0.05, respectively] as well as more negative perceptions of individual and cultural race-related stress [t(67) = 2.20, p < 0.03 and t(67) = 2.25, p < 0.03, respectively]. However, no significant associations were found between the smoking status and the frequency of institutional race-related stress experienced [t(67) = 1.37, p < 0.18] and reported perceptions of institutional race-related stress [t(67) = 1.75, p < 0.08].

Discussion

While the study does not suggest that race-related stress causes pregnant African-American women to smoke, the findings suggest that for some of the women higher frequencies and perceptions of race-related stress and smoking are synchronous. The study does, however, add to other research recognizing the presence of stress in this segment of the African-American population. The current study's findings, of an association between race-related stress and the smoking status of pregnant African-American women, differ somewhat from a previous study examining the association between race-related stress and smoking status among non-pregnant African-American women that found no direct association between race-related stress and smoking. Instead, the study found that race-related stress predicted general life stress, which, in turn, predicted smoking status (30). In that study the average number of race-related events among both smokers and non-smokers was similar to the smokers in the current study, i.e., a mean of 16. However, average reaction scores reported by Fernander and Schumacher were lower than the average reaction scores of smokers in the current study (average of 1 compared to 2, respectively). Although no statistical comparisons can be made between the two studies, it is possible that being pregnant may enhance the physiological, behavioral, and emotional response to race-related stress in such a way to lead them to adopt negative health behaviors, such as smoking. Future studies, including both pregnant and non-pregnant women in a single cohort may provide additional insight into such a hypothesis.

Pregnant African-American smokers in the current study experienced significantly higher frequencies and perceptions of cultural race-related stress than their non-smoking counterparts. Some of the cultural race-related events included perceptions of how African-American culture is projected through the media. This could be an indication that pregnant African-American smokers perceived cultural racism as more threatening to their identity as an African-American. Previous studies have linked racial identity with race-related stress among African-Americans. For example, one study found that the more important being African-American was to self-concept, the more racial discrimination the women reported to be experiencing (36). The media is a powerful tool and can magnify stereotypes and negative images of African-Americans. Studies suggest that some individuals experiencing high levels of race-related stress may view events as threats instead of challenges and are more likely to respond to race-related stress with negative health behaviors (37, 38).

Pregnant African-American women smokers also reported significantly higher frequencies and perceptions of individual race-related stress than non-smokers. Items on the subscale assess racism experienced on an interpersonal level. It is clear that the more personal the experience is, the more of an impact it will have on an individual's psyche and therefore their psychological and physical health. In this case, it is possible that the interpersonal race-related stress experienced by the African-American women was such that it may have caused them to take on the negative health behavior of smoking, supporting Hogue's causal hypothesis (29).

Analyses to determine differences between the frequency and perception of institutional race-related stress between pregnant African-American smokers and pregnant African-American non-smokers yielded no significant differences. These results do not indicate the lack of perception or frequency of institutional race-related stress; they only suggest that there are no significant differences in the race-related institutional experiences of pregnant African-American women who smoke compared to their non-smoking counterparts. Institutional racism primarily occurs in environments such as schools, hospitals, social agencies, etc., and is often subtle and immediately unrecognizable, thus making it more difficult to detect the influence of this type of race-related stress on health and health behavior.

Because of concerns about the safety and efficacy of pharmacological therapies to aid pregnant women in smoking cessation, psychosocial interventions (counseling, cognitive, and behavioral therapy) are the preferred methods for intervening among pregnant smokers (39, 40). Studies have shown that psychosocial interventions have a positive effect on rates of smoking cessation and studies document that pregnant women who have received psychosocial smoking cessation counseling are more likely to quit smoking (41–43). Stress in and of itself can have deleterious effects on the reproductive health in pregnant women (44). In fact, several studies have found associations between stress and pregnancy outcome and programs that teach coping with stress skills may have a direct impact on reproductive health as well as indirectly by preventing the uptake or continuation of negative health behaviors such as smoking (45–50). Researchers have posited that psychosocial intervention counseling programs targeted toward specific populations need to be psycho-socioculturally tailored. Tailoring intervention programs for pregnant African-American women to cope with the specific challenges of race-related stress should decrease the negative emotional and physical health responses to such stressors and may move progress closer to the United States Department of Health and Human Services Healthy People 2010 tobacco control goals.

This study focused on the association of race-related stress and smoking status among pregnant African-American women. Although this study offers a promising addition to the research on race-related stress and smoking among these women, some caution must be utilized in the inference of its findings to similar populations. First, the study did not establish causality, and secondly, responses were based upon self-report. However, the findings suggest that integrating race-related stress relieving and coping activities into smoking cessation intervention programs for African-American women may reduce smoking and subsequent smoking-related reproductive health disparities among pregnant African-American women.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Ancillary