SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Pregnancy is a critical time window for evaluating weight gain on subsequent risk for obesity among women of childbearing age. The purpose of this investigation was to determine whether symptoms of depression, anxiety, stress, self-esteem, and fetal health locus of control (FHLC) beliefs were significant risk factors for adequacy of gestational weight gain (GWG) when maternal sociodemographic characteristics and health behaviors were considered. A total of 1,605 women were prospectively followed from ≤20 weeks' gestation through delivery. Participants completed standard self-report questionnaires. Adequacy of GWG was expressed as the ratio of observed/expected weight gain based on Institute of Medicine recommendations. Multivariate ANOVA models were conducted and generalized linear models were performed to calculate risk ratios. Higher depressive symptoms reported throughout pregnancy were significantly associated with higher adequacy ratios. Stronger beliefs in chance in determining fetal health predicted inadequate relative to adequate GWG and was positively associated with larger GWG ratios overall. Several relationships were attenuated when adjusted for covariates. The relationship between psychosocial status and adequacy of GWG is significantly impacted by maternal sociodemographic factors and health practices engaged in during pregnancy. Women who tend to believe that external factors primarily determine fetal health appear to be more vulnerable to nonadherence to clinical GWG guidelines. These results have important implications for targeting prevention and intervention efforts for improving maternal and fetal outcomes secondary to GWG patterns.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Recent analysis of surveillance data from the National Health and Nutrition Examination Survey indicated rates of obesity (defined as a BMI of ≥30 kg/m2) among US women may have reached a plateau at roughly 33% (1). While this trend is encouraging, women, especially those of ethnic minority status, still bear the disproportionate burden of obesity and are especially vulnerable to weight gain during the childbearing years (2).Researchers have given increasing attention to clarifying the roles of excessive weight gain during pregnancy and persistent postpartum weight retention as contributors to women's increased risk of obesity and the long-term health consequences (2,3).

In 1990, the Institute of Medicine (IOM) provided gestational weight gain (GWG) recommendations that were individualized to pregravid body mass index categories (4). These recommendations evolved from decades of observations made regarding birth outcomes; higher maternal weight gains have led to better overall infant outcomes (5). However, in spite of the amended guidelines and clear benefits to mother and child, only 30–40% of US women gain in accordance with advised clinical recommendations (5,6). Between 35 and 60% of pregnant women reported receiving no advice from their providers about appropriate pregnancy weight gain (7,8).

In the face of persistent nonadherence to expert consensus together with the wide variability in the amount of weight gained during pregnancy, more researchers have acknowledged that the determinants of the adequacy of GWG are multi-factorial in nature (7,8,9). More comprehensive analyses are needed that include biological characteristics and modifiable psychosocial and contextual indicators as predictors of weight gain during pregnancy. Whereas, the bulk of research in this area has emphasized the former (e.g., pregravid BMI, parity, height, race, age, age at menarche), the value of identifying pertinent psychological, sociocultural, and behavioral risk factors for inadequate and excessive GWG has gained momentum in the scientific literature in recent years (10,11).

The biopsychosocial model (12) provides a flexible framework from which to explore the complex, multidimensional determinants and causal mechanisms that support inadequate or excessive GWG. Psychosocial factors may affect weight gain in the prenatal months through multiple conduits. Such pathways may include through the direct impact on physiological weight control mechanisms (e.g., reduced metabolic efficiency, neuroendocrine mediation of abdominal fat accrual; 13,14,15). Additionally, psychosocial factors may indirectly exert influence on weight regulation in the gestational period through lifestyle practices (e.g., nutritional habits, level of physical activity engagement, frequency of tobacco use; 10). While lifestyle factors are forming a solid research base within the literature (9,16,17,18), the former remains an important candidate for scientific inquiry.

An accumulating body of work has evaluated the relationship between psychosocial status variables and adequacy of prenatal weight gain with mixed results. Some investigations have found significant associations between self-reported stress and low maternal weight gain during pregnancy (19) whereas others have not (9). Similarly, comparisons involving the level of reported depressive symptoms have yielded robust (20,21) as well as equivocal (11) associations with the adequacy of GWG in the existing literature. This conflicting evidence may be explained by regional variance in rates of GWG, differences in the sociodemographic diversity (e.g., ethnicity, age, pregravid BMI) of the samples in conjunction with variation in study design and execution (e.g., sample size, assessment methodology, weight gain in the continuous vs. categorical form, inclusion of tobacco use, dietary and physical activity variables as covariates, analytic procedures implemented).

Using data obtained from women recruited in the Pregnancy, Infection, and Nutrition longitudinal cohort study, the present analysis addresses some of the limitations of earlier work in this area. Informed by the biopsychosocial systems approach, the current study tested whether traditional (e.g., perceived stress, depression, anxiety, self-esteem) and more innovative (e.g., fetal health locus of control (FHLC)) psychosocial scales predict adequacy of GWG when adjusted for standard maternal biodemographic and health behavior covariates.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

The Pregnancy, Infection, and Nutrition study conducted in central North Carolina enrolled 2006 English-speaking women age ≥16 between January, 2001 and June, 2005. Eligible women were identified through an initial review of medical records obtained from prenatal clinics comprising the UNC Hospital system. Study personnel recruited women at ≤20 weeks' gestation at their second prenatal visit. Interested women were invited to consent to participate in two in-person research clinic visits and two telephone interviews during their pregnancies. Women were retained for the current analysis if they met the following criteria: singleton pregnancy, live birth, and had GWG data available. Only one pregnancy per woman was included. This resulted in an analysis file of 1,605 women.

Research clinic visits took place before 20 weeks' gestation and between 24 and 29 weeks' gestation. At these visits, women were given self-administered questionnaires that included several psychosocial scales. Completed questionnaires were returned by mail. Telephone interviews were conducted between 17–22 weeks' gestation and 27–30 weeks' gestation, and included collection of additional psychosocial scales and information about lifestyle practices. Study protocols were approved by the Institutional Review Boards of UNC Chapel Hill's School of Medicine.

Demographic characteristics

Maternal sociodemographic information was obtained from the medical chart or during the first telephone interview, including participant's date of birth, race, ethnicity, marital status, level of education attained, and reproductive history (e.g., parity). Household income and the number of adults and children in the household were used to construct the percent of poverty index (22). All demographic variables were analyzed in the categorical format.

Pregravid BMI (kg/m2) was calculated based on height routinely measured at the first research clinic visit and on self-reported pregravid weight obtained at recruitment. Although this estimate relies on retrospective accounts, research has indicated that self-reported weight is a reliable proxy for measured weight (23). Women were classified using IOM pregravid BMI cut point ranges: underweight (<19.8), normal weight (19.8–26.0), overweight (26.1–29.0), and obese (>29.0) (4).

GWG

Weight gain at each prenatal visit was abstracted from the medical chart after delivery. The absolute amount of weight (kg) gained throughout pregnancy was determined by subtracting the weight recorded at the last prenatal visit prior to delivery from the self-reported pregravid weight. The primary outcome of adequacy of GWG was expressed as the ratio of observed over expected weight gain using the IOM recommended weight gains1 to calculate the expected for a given gestational age as previously described (24,25). This ratio was devised in recognition of the individual differences in gestational age at delivery and how weight at delivery is rarely documented (24). To illustrate this, suppose a woman's adequacy ratio was 1.5. This number is interpreted as her having gained 50% in excess of recommended weight gain based on her pregravid BMI.

In addition to the continuous form, the adequacy of GWG variable was also evaluated in the categorical form according to IOM guidelines (4). The upper and lower limits of each respective pregravid BMI weight recommendation cut points were divided by the expected GWG at 40 weeks and then multiplied by 100 to yield inadequate, adequate and excessive weight gain percentage ranges as previously used (24,25).

Perceived stress

The Perceived Stress Scale (PSS; 26,27) assesses individuals' stress appraisals and ability to successfully cope with the demands of stressors over the previous month. The 14-item PSS was administered during the 17–22 weeks' gestation telephone interview, and the 10-item version was administered between 27 and 30 weeks' gestation. Respondents were instructed to rate each item on a five-point scale ranging from 0 (never) to 4 (almost always). After reverse scoring positively worded items, a total score is calculated by summing across all items. Higher scores reflect greater distress and lower confidence in personal coping resources.

Depression

The Center for Epidemiological Studies-Depression (CES-D; 28) scale is a 20-item measure of symptoms of depression manifest in the general population. It has been widely validated as a screening instrument for psychological distress in diverse community samples (29). This instrument requires respondents to endorse each item with respect to the last week using a Likert scale with total scores ranging from 0 to 60. Higher scores indicate greater reported depressive symptoms. The CES-D scale was self-administered at ≤20 weeks' gestation and between 24 and 29 weeks' gestation.

State and trait anxiety

The Spielberger State-Trait Anxiety Inventory (STAIS/T; 30) is a well-established measure of both transitory and dispositional anxiety. Each instrument includes a set of 20 items with Likert response categories. Higher scores on the state measure reflect greater current anxious affect, whereas higher trait scores indicate a more stable tendency to experience elevated levels of anxiety. Both the state and trait forms were self-administered before 20 weeks' gestation. The state form was re-administered between 24 and 29 weeks' gestation.

Self-esteem

The 10-item Self-esteem Scale (31) was used to evaluate participants' general estimations of self-worth and degree of positive or negative orientation toward the self. This measure has been extensively used in research that traverses multiple health-related disciplines. Respondents indicate their level of item agreement on a 6-point scale2. Higher scores reflect more positive self-appraisals. This instrument was self-administered between 17 and 22 weeks' gestation.

FHLC

The 18-item FHLC (32) scale assesses the degree to which a woman believes the health of her fetus is principally under her own behavioral control (Internality Scale), is primarily influenced by authoritative figures (e.g., health professionals, a higher power; Powerful Others Scale) or is chiefly due to fate or chance (Chances Scale). Scores are calculated for each component with higher scores reflecting stronger beliefs consistent with that particular locus of control for determining fetal health. This instrument was self-administered before 20 weeks' gestation.

Dietary intake

Intake over the previous three months was assessed using a modified version of the Block-98 Food Frequency Questionnaire (33) which was self-administered between 24 and 29 weeks' gestation. The Block is a widely used 110-item measure of food intake that is well validated in clinical and community samples. In this study, only the total average daily calories (kcal) consumed were included in the multivariable analyses.

Physical activity

The amount of physical activity engaged in during the gestational period was assessed during both telephone contacts (i.e., between 17 and 22 weeks' gestation and 27 and 30 weeks' gestation) using a measure that was specifically designed for this study (34). This instrument assesses the frequency and duration of a variety of reported physical activities participated in over the last 7 days at either a moderate or vigorous intensity level. Domains incorporated the following settings and/or roles: at work, for recreation, for transportation, during care giving and as a part of indoor and outdoor household tasks. The average activity level for each domain is expressed in terms of standard metabolic equivalent hours per week (MET-HR). For this analysis, only recreational, indoor and outdoor physical activity levels at both time points were included.

Cigarette smoking

Tobacco use over the first six months of pregnancy was assessed during the telephone interview at 27–30 weeks' gestation. The number of cigarettes smoked daily during that time period was categorized as follows: none, 1–9 cigarettes, 10–19 cigarettes and 20+ cigarettes.

Statistical analysis

Standard point estimates (i.e., means and standard deviations) and frequency counts were used to describe the distributions of continuous and categorical variables, respectively, within the sample. An alpha level of 0.05 was established as the criteria for statistical significance for each series of analyses. All analytic procedures were performed using the SAS Enterprise Guide version 3.0 statistical software package (35).

Separate multivariable linear regression models were used to evaluate possible confounding by maternal sociodemographic and lifestyle factors. Confounding was defined as occurring when the adjusted beta coefficient for the original psychosocial indicator was modified by ≥10%. Identified confounders for each psychosocial variable were included in subsequent regression or ANOVA models. Measures of perceived stress, self-esteem, and fetal health locus of control were transformed into tertiles (low, moderate, and high). Cut points for state anxiety were based on tertiles at the first questionnaire administration. CES-D cut points were at 0–16, 17–24, and 25+, which were slightly higher than in a general population because scores may be inflated because of physical manifestations of pregnancy (36,37). The low category was used as the reference in all cases except self-esteem and the Fetal Health Locus of Control-Internality Scale in which the high category served as the reference group.

Separate one-way ANOVA models were used to evaluate the relationship between the level of psychosocial status variable and the mean ratio of adequacy of GWG in the continuous form. Bartlett's test for the homogeneity of variances was performed to ensure the basic assumptions for using this standard parametric procedure were tenable. In cases where these assumptions were not met, a Welch's variance-weighted ANOVA F-test was used. The Bonferroni correction was used to adjust for multiple comparisons in the post-hoc analyses. The likelihood ratio χ2-test from the multivariable regression models was subsequently used to assess the strength of the associations when adjusted for relevant covariates.

To examine the variables in their categorical forms, we performed a multivariable analysis using a generalized linear model to estimate the adjusted risk of inadequate or excessive gain associated with each psychosocial variable. Inadequate and excessive gains were modeled separately, with adequate gain as the referent variable in each model. All covariates were in the categorical form for this set of analyses. Thus, dietary and physical activity variables were quantified according to tertiles with the latter including a fourth “no activity” category. Crude and adjusted risk ratios and corresponding 95% confidence intervals were calculated.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Table 1 presents select sociodemographic, behavioral, anthropometric, dietary, physical activity and psychosocial characteristics of the sample. Approximately 20% of the women were African American, a majority reported being married (74%), having attained at least a college education (59%), and living well above the poverty level (75%). Half of the sample was age ≥30 and 48% were nulliparous. Eleven percent of women reported smoking at least 1 cigarette per day during the first two trimesters. Regarding gestational weight changes, participants gained an average of 15.1 kg and 64.2% gained in excess of clinical guidelines. Only 21% of women studied gained within the recommended ranges. Women's activity levels decreased uniformly from early to late pregnancy.

Table 1.  Characteristics of women in the Pregnancy, Infection, and Nutrition Study, 2001–2005 (N = 1,605)
inline image

Adequacy of GWG by level of psychosocial status

Table 2 presents the crude and adjusted means for adequacy of GWG in the continuous form by psychosocial predictors in the categorical form. The Bonferroni-corrected results from unadjusted one-way ANOVA models showed that women scoring high in depressive symptoms at both <20 weeks, (P = 0.05) and between 24 and 29 weeks, (P = 0.03) gestation gained on average a significantly greater percentage of weight in excess of recommendations when compared to women reporting lower levels of depression at both assessments. However, even women endorsing few depressive symptoms on average gained in excess of clinical guidelines. It is worth noting that women classified as having gained an excessive amount of weight during the gestational period reported greater elevations of depressive symptoms than women who gained weight in adherence with clinical guidelines (P < 0.01; data not shown).

Table 2.  Crude and adjusted means for adequacy of gestational weight gain categorized by psychosocial status
inline image

Greater beliefs in chance as a controlling factor in the health of the fetus were marginally associated with higher mean adequacy of weight gain achieved (P = 0.06). Similar trends emerged for self-esteem (P = 0.08) and trait anxiety (P = 0.07) in the uncorrected models (see Table 2). As shown in Table 2, results from likelihood ratio χ2-tests yielded from multivariable linear regression models indicated that when adjusted for important covariates these relationships were attenuated. Overall, pregravid BMI appeared to be the confounder with the greatest impact on results. This is illustrated here for the association between CES-D (≤20 weeks) and adequacy of GWG: (i) without confounders: B = 0.007, P <0.001, (ii) with pregravid BMI: B = 0.001, P = 0.002, (iii) with parity: B = 0.008, P < 0.001, (iv) with marital status: B = 0.005, P < 0.05, (v) with race: B = 0.006, P < 0.01, (vi) with education: B =0.004, P < 0.05, (vii) with smoking: B = 0.006, P < 0.01, (viii) with total calories: B = 0.006, P < 0.01, (ix) with indoor MET-Hrs (27–30 weeks): B = 0.006, P < 0.01, (x) with outdoor MET-Hrs (27–30 weeks): B = 0.006, P < 0.01, and (xi) with recreation MET-Hrs (27–30 weeks): B = 0.006, P < 0.05. Finally, women who gained less weight than recommended on average reported marginally greater beliefs in the influence of powerful others on the health of their fetus in comparison to women who gained weight within the recommended range (P = 0.09; data not shown).

Risk of inadequate or excessive GWG by level of psychosocial status

Table 3 presents the crude and adjusted risk ratios for dichotomized adequacy of GWG by psychosocial predictors in the categorical form. The first series of comparisons involved inadequate vs. adequate GWG outcomes. As illustrated in Table 3, the relationship between the level of psychosocial status variables and inadequate relative to adequate weight gain during pregnancy tended to be modest at best when examining both crude and adjusted risk ratios. Women who reported moderate (RR: 1.75; 95% CI: 1.3, 2.4) and high (RR: 1.57; 95% CI: 1.2, 2.1) relative to low belief that chance influenced the health of her fetus tended to gain less weight than clinically advised. However, the risk ratios were significantly attenuated when adjusted for covariates (see Table 3). Finally, it would appear that the relative risk of excessive weight gain as compared to expected weight gain levels in this participant sample is similar across varying degrees of psychosocial exposure during the gestational period (see Table 3).

Table 3.  Crude and adjusted risk ratios for categorized psychosocial variables and adequacy of gestational weight gain for women in the Pregnancy, Infection, and Nutrition Study, 2001–2005
inline image

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Pregnancy serves as a unique life experience that poses increased risk for negatively affecting both psychological and physical wellness among women. Psychologically, women may face an amplification of stress and affective disturbance in attempting to emotionally and physiologically adjust to the transition to motherhood. For instance, an estimated 9–20% of women report experiencing clinically significant levels of depressive symptoms during pregnancy (38,39) which in turn increases their vulnerability for subsequent episodes of depression in the postpartum period (39). From a physiological standpoint, childbearing also has major public health implications for enhancing the probability of short- and long-term adverse maternal and fetal outcomes with risk of overweight and obesity as primary concerns (1). Guided by the integrative biopsychosocial model, the present analysis is one of a growing number of studies that aim to describe the relationships between modifiable psychosocial risk factors assessed early and late in gestation to the adequacy of total weight gain achieved.

More than 60% of the women in our sample gained weight in excess of recommendations based on pregravid BMI; this rate falls somewhat above those cited in previous investigations (7,8). We suspect this disparity may in part be attributable to the time period of when this study was conducted as well as differences in whether initial BMI and weight gained over the course of pregnancy were measured prospectively by research or medical staff or were obtained retrospectively based solely on patient reports. Regardless, this finding supports continued advocacy for greater awareness and education among health-care providers and women and their families in the communities at large to promote better adherence to the IOM GWG guidelines and consequently, improve clinical outcomes. One important dimension of this health promotion initiative is by fostering greater empowerment for both patients and health care providers through identifying and managing modifiable psychosocial risk factors for excessive weight gain during pregnancies.

Toward this end, we found that select psychosocial status variables were significantly related to the adequacy of GWG attained in unadjusted models. Experiencing elevated depression symptoms both at earlier and later stages of pregnancy were positively related to gaining more weight than is clinically recommended based on prepregnancy BMI. Similarly, women classified as reporting high levels of depressive symptoms at both time points were shown to have significantly higher adequacy ratios relative to those women reporting minimal depressive symptoms at either administration. Both self-esteem and trait anxiety also played modest roles in predicting excessive GWG in this sample in the uncorrected models. However, the effects of psychosocial status on exceeding GWG clinical guidelines were overshadowed by maternal sociodemographic and health behaviors engaged in during pregnancy. This may suggest that they mediate the effects of psychosocial factors though this was not directly tested in our analysis.

In comparison with our initial findings regarding self-reported depressed mood, an analysis conducted by Walker et al. of Medicaid-eligible women receiving care at a local community hospital in the Southwest failed to find a strong relationship between depressive symptoms and GWG (11). This more socioeconomically homogenous population reported notably higher levels of depressive symptoms on the CES-D than were documented in the current sample. Therefore, it is likely that the distribution of CES-D scores had a restricted range which precluded observing a more meaningful effect in this earlier study.

Additionally, the present study utilized a more systematic approach to capturing the adequacy of GWG achieved that was derived from the ratio of observed to expected weight gain based on pregravid BMI and gestational age at the last weight measurement. This is in contrast to the Walker study (as well as others) in which only the absolute amount of GWG attained at term was used as a primary outcome or dependent variable (11). Finally, the participants in our sample were administered the CES-D during the second and early third trimester while the Walker sample completed this instrument in the immediate postpartum period. Other experts have noted that the responses to stress and affective instruments may heavily depend upon where in the trajectory of pregnancy a woman is when these dimensions are assessed (40).

We generally did not find a robust association between the appraisals of stress and sufficiency of coping resources and adequacy of GWG in crude or adjusted models. However, when evaluating the risk ratio differences observed between women who gained inadequate or excessive weight (relative to women who gained adequate weight), the former tended to have a stronger, albeit modest, link to perceived stress than the latter. This distinction is comparable to a pattern also cited previously in another investigation (7). In that study, women who reported high stress during pregnancy tended to gain weight below clinical guidelines in contrast to those who did not report significant prenatal stress. This relationship failed to emerge among those who gained weight in excess of clinical advice. This finding is also consistent with earlier work that typically reported higher stress in relation to insufficient GWG (19).

The underlying biopsychosocial mechanisms giving rise to these differential relationships between indicators of psychological status and adequacy of GWG remain to be elucidated. Further, given the subjective nature of stress in particular, authors have urged scientists in this area to give careful consideration to measuring stress in this population using instruments that are sensitive to diverse cultural meanings and life events (10) and that include items relevant to the experience of pregnant women specifically (41). Focusing efforts on enhancing both the identification of distinct biophysical mechanisms and the validity of measures may improve the consistency and interpretation of data obtained across participant samples and may serve as benchmarks for evaluating the efficacy of targeted interventions.

One of the novel aspects of this analysis was the inclusion of an instrument that uniquely assessed fetal health locus of control beliefs as determinants of the adequacy of GWG achieved. The degree to which a woman perceives that she, powerful authority figures (actual or symbolic), or chance factors are responsible for the health of her fetus has been correlated with several health risk- and resilience-promoting behaviors in the existing literature. On the beneficial side, women reporting a greater internal fetal health locus of control demonstrated better adherence with limiting caffeine consumption (32), not smoking during pregnancy (32), attending childbirth classes (42) and intending to breastfeed (43).

Adversely, women who typically endorsed beliefs that fetal health is due to chance tended to also have higher rates of smoking (44) and heavy alcohol use (45) during pregnancy. Low personal responsibility for the health of the fetus was also particularly high in one study among pregnant smokers who were classified as precontemplative along the stages of change toward consideration of smoking cessation (46). Similarly, lower perceptions of personal control in affecting the health of the unborn baby were associated with higher scores on a composite self-care inventory reflecting poorer dietary, exercise, sleep, and hygiene practices, in conjunction with indexing greater problems with substance abuse and reckless behavior during pregnancy (13).

Our results showed marginally significant relationships between dimensions of fetal health locus of control and adequacy of GWG in the unadjusted ANOVA models. Women who were categorized as having reported high beliefs in chance factors gained a considerably higher than anticipated proportion of weight during pregnancy than women who were less likely to attribute fetal health to chance. Finally, the association between chance fetal health locus of control beliefs and GWG was further modified by the level of adequacy as indicated by the observed risk ratios. Women who gained weight below clinical expectations based on pregravid BMI reported stronger beliefs in chance factors relative to women who gained within the recommended ranges. An analogous relationship was not discerned from comparing risk for excessive (vs. adequate) GWG based on the degree of reported beliefs in chance occurrences in impacting fetal health.

Importantly, these relationships were attenuated when adjusted for maternal sociodemographic (e.g., pregravid BMI, marital status, race, education) and lifestyle factors. Thus, it will be essential for future researchers to consider the cultural context of these health-specific loci of control beliefs. We believe adopting a greater culturally sensitive stance toward clarifying health beliefs during pregnancy will enhance attempts to promote positive maternal and fetal outcomes secondary to GWG among women of diverse ages, ethnicities and socioeconomic backgrounds.

Building upon prior work, subsequent research would further benefit from more formally testing specific modifiable weight-regulating behaviors (e.g., energy intake, the frequency of participating in moderate-vigorous physical activity) as mediators in the association between fetal health locus of control beliefs and the adequacy of weight gained during pregnancy. In addition to the biopsychosocial model, future work may also help extend the contributions of this line of research by utilizing existing theoretical models to inform psychological and behavioral risk assessment for problematic GWG and to identify predictors essential to successful prevention and therapeutic change efforts (e.g., the Health Belief Model (47), the Transtheoretical Model (48), the Theory of Planned Behavior (49)).

Finally, not only does excessive weight gain in the prenatal period serve as a potential harbinger for detriments to physical well-being, it may conversely lead to more frequent and/or severe experiences of psychological and emotional distress throughout pregnancy, in the early postpartum and beyond. Our preliminary evidence revealed that women who gained a greater proportion of weight during pregnancy than clinically advised reported significantly more symptoms of depression than women who gained within the appropriate range. Likewise, DiPietro and colleagues found that when stratified by adequacy of GWG, women in their sample who gained an excessive amount of weight had more negative body image attitudes during pregnancy (16). More specifically, these overgainers experienced greater embarrassment when weighed, generally felt less attractive, were embarrassed about their weight and were more concerned about becoming fat than their adequate and undergaining counterparts (16). Undoubtedly, the bidirectional nature of this relationship is a rich area for further scientific scrutiny.

The current study has several notable strengths including a large, prospective cohort design, and inclusion of energy intake and physical activity as covariates. It is also important to consider limitations. Although the participant sample was more diverse than some others studied in previous research, the women were mostly white, well educated, married, financially stable, and received early prenatal care at University-based clinics. These characteristics may limit the generalizability of our findings. Although we obtained measures of psychological functioning at multiple times throughout pregnancy, we did not have access to information about psychological status prior to pregnancy. Conceivably, a history of clinical depression, anxiety, body image disturbance or significant life stress in the face of poor coping resources could potentiate risk for experiencing negative mood and affective changes in pregnancy, which could subsequently impact the adequacy of GWG achieved.

In summary, in the unadjusted models, symptoms of depression, anxiety, self-esteem and dimensions of fetal health locus of control all predicted significantly deviating from established GWG clinical guidelines. To a large extent, maternal sociodemographic and health behavior characteristics reduced these relationships thus suggesting their critical role in determining higher than expected weight gain ratios in the gestational period. Although additional work is needed, our findings contribute to a body of research that has important clinical implications for the early identification and treatment of women at risk for both undergaining and overgaining during pregnancy.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

This study received support from the National Institute of Child Health and Human Development, National Institutes of Health (HD37584, HD39373), the National Institute of Diabetes and Digestive and Kidney Diseases (DK61981), and the National Cancer Institute (CA109804). Dr Webb's data analysis and manuscript preparation efforts were supported by a National Institute of Mental Health T32 Post-doctoral Training Award (T32-MH65742-05) administered through the Duke University Medical Center's PREMIER Program. Biological specimen collection was supported by the General Clinical Research Center of the Division of Research Resources and funded by the National Institutes of Health (RR00046) at the University of North Carolina at Chapel Hill.

Footnotes
  • 1

    The IOM GWG recommendations (4) based on pregravid BMI are as follows: underweight (BMI < 19.8 kg/m2) = 28–40 lbs; normal weight (BMI = 19.8–25.9 kg/m2) = 25–35 lbs.; overweight (BMI = 26–29 kg/ m2) = 15–25 lbs.; obese (BMI > 29 kg/m2) = at least 15 lbs. There is no established upper limit for women in this last category.

  • 2

    In the standard administration of this measure respondents traditionally use a 4-point scale. A nonstandard protocol was used here for the purposes of maintaining consistency of response scales across the variety of self-report measures administered to participants in the larger parent study.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES
  • 1
    Ogden CL, Carroll MD, Curtin LR et al. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006; 295: 15491555.
  • 2
    Siega-Riz AM, Evenson KR, Dole N. Pregnancy-related weight gain—A link to obesity? Nutr Rev 2004; 62 (Suppl):S105S111.
  • 3
    Gore SA, Brown DM, Smith West D. The role of postpartum weight retention in obesity among women: a review of the evidence. Ann Behav Med 2003; 26: 149159.
  • 4
    Institute of Medicine. Nutrition during pregnancy, weight gain and nutrient supplements. Report of the Subcommittee on Nutritional Status and Weight Gain During Pregnancy, Committee on Nutritional Status During Pregnancy and Lactation, Food and Nutrition Board. (1990).
  • 5
    Abrams B, Altman SL, Pickett KE. Pregnancy weight gain: still controversial. Am J Clin Nutr 2000; 71 (Suppl):1233S1241S.
  • 6
    Caulfield L, Witter F, Stolzfus R. Determinants of gestational weight gain outside the recommended ranges among black and white women. Obstet Gynecol 1996; 87: 760766.
  • 7
    Brawarsky P, Stotland NE, Jackson RA et al. Pre-pregnancy and pregnancy-related factors an the risk of excessive or inadequate gestational weight gain. Int J Gynecol Obstet 2005; 91: 125131.
  • 8
    Olson CM, Strawderman MS. Modifiable behavioral factors in a biopsychosocial model predict inadequate and excessive gestational weight gain. J Am Diet Assoc 2003; 103: 4854.
  • 9
    Hickey CA, Cliver SP, Goldenberg RL, McNeal SF, Hoffman HJ. Relationship of psychosocial status to low prenatal weight gain among nonobese black and white women delivering at term. Obstet Gynecol 1995; 86: 177183.
  • 10
    Hickey CA. Sociocultural and behavioral influences on weight gain during pregnancy. Am J Clin Nutr 2000; 71 (Suppl):1364S1370S.
  • 11
    Walker LO, Minseong K. Psychosocial thriving during late pregnancy: relationship to ethnicity, gestational weight gain, and birth weight. J Obstet Gynecol Neonatal Nurs 2002; 31: 263274.
  • 12
    Engel GL. The need for a new medical model: a challenge of biomedicine. Science 1977; 196: 129136.
  • 13
    Bjorntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2001; 2: 7386.
  • 14
    Kyrou I, Chrousos GP, Tisgos C. Stress, visceral obesity, and metabolic complications. Ann NY Acad Sci 2006; 1083: 77110.
  • 15
    Lamounier-Zepter V, Ehrhart-Bornstein M, Bornstein SR. Metabolic syndrome and the endocrine stress system. Horm Metab Res 2006; 38: 437441.
  • 16
    DiPietro JA, Millett S, Costigan KA, Gurewitsch E, Caulfield LE. Psychosocial influences on weight gain attitudes and behaviors during pregnancy. J Am Diet Assoc 2003; 103: 13141319.
  • 17
    Walker LO, Cooney AT, Riggs MW. Psychosocial and demographic factors related to health behaviors in the 1st trimester. J Obstet Gynecol Neonatal Nurs 1999; 28: 606614.
  • 18
    Hurley KM, Caulfield LE, Sacco LM, Costigan KA, DiPietro JA. Psychosocial influences in dietary patterns during pregnancy. J Am Diet Assoc 2005; 105: 963966.
  • 19
    Orr ST, James SA, Miller CA, Barakat B. Psychosocial stressors and low birthweight in an urban population. Am J Prev Med 1996; 12: 459466.
  • 20
    Stevens-Simon C, McAnarney ER. Determinants of weight gain in pregnant adolescents. J Am Diet Assoc 1992; 92: 13481351.
  • 21
    Siega-Riz AM, Hobel CJ. Predictors of poor maternal weight gain from baseline anthropometric, psychosocial, and demographic information in a Hispanic population. J Am Diet Assoc 1997; 97: 12641268.
  • 22
    Proctor BD, Dalaker J. Poverty in the United States, 2001. US Bureau of the Census, Current Population Reports, Series P60-219 p5. US Government Printing Office: Washington, DC, 2002.
  • 23
    Engstrom JL, Paterson SA, Doherty A, Trabulsi M, Speer KL. Accuracy of self reported height and weight in women: an integrative review of the literature. J Midwifery Womens Health 2003; 48: 338345.
  • 24
    Bodnar LM, Siega-Riz AM, Arab L, Chantala K, McDonald T. Predictors of pregnancy and postpartum haemoglobin concentrations in low-income women. Public Health Nutr 2004; 7: 701711.
  • 25
    Saldana TM, Siega-Riz AM, Adair LS, Suchindran C. The relationship between pregnancy weight gain and glucose tolerance status among black and white women in central North Carolina. Am J Obstet Gynecol 2006; 195: 16291635.
  • 26
    Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983; 24: 385396.
  • 27
    Cohen S, Williamson GM. Perceived stress in a probability sample of the United States. In: Spacapan S, Oskamp S (eds). The Social Psychology of Health. Sage: Newbury Park, CA, 1988, pp 3167.
  • 28
    Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385401.
  • 29
    Roberts RE, Vernon SW. The Center for Epidemiologic Studies Depression Scale: its use in a community sample. Am J Psychiatry 1983; 140: 4146.
  • 30
    Spielberger CD. Manual for the State-Trait Anxiety Inventory. Consulting Psychologists Press: Palo Alto, CA, 1983.
  • 31
    Rosenberg M. Society and the Adolescent Self-Image. Princeton University Press: Princeton, NJ, 1965. For further information, see: <www.bsos.umd.edusocyRosenberg.html>.
  • 32
    Labs SM, Wurtele SK. Fetal Health Locus of Control Scale: development and validation. J Consult Clin Psychol 1986; 54: 814819.
  • 33
    Borodulin K, Evenson KR, Wen F, Herring AH, Benson AM. Physical activity patterns during pregnancy Med Sci Sport Exerc 2008; 40: 19011908.
  • 34
    Block G, Wakimoto P, Block T, A revision of the block dietary questionnaire and database, based on NHANES III data [online]. <http:www.nutritionquest.com>.
  • 35
    SAS Institute. SAS Enterprise Guide Version 3.0. SAS Institute: Cary, NC, 2004.
  • 36
    Orr ST, Miller CA. Maternal depressive symptoms and the risk of poor pregnancy outcome: review of the literature and preliminary findings. Epidemiologic Reviews 1995; 17: 165171.
  • 37
    Hoffman S, Hatch MC. Depressive symptomatology during pregnancy: evidence for an association with decreased fetal growth in pregnancies of lower social class women. Health Psychol 2000; 19: 535543.
  • 38
    Orr ST, Blazer DG, James SA. Racial disparities in elevated prenatal depressive symptoms among black and white women in Eastern North Carolina. Ann Epidemiol 2006; 16: 463468.
  • 39
    Rich-Edwards JW, Kleinman K, Abrams A et al. Sociodemographic predictors of antenatal and postpartum depressive symptoms among women in a medical group practice. J Epidemiol Community Health 2006; 60: 221227.
  • 40
    Da Costa D, Larouche J, Dritsa M, Brender W. Variations in stress levels over the course of pregnancy: factors associated with elevated hassles, state anxiety and pregnancy-specific stress. J Psychosom Res 1999; 47: 609621.
  • 41
    Kendall A, Olson CM, Frongillo EA. Evaluation of psychosocial measures for understanding weight-related behaviors in pregnant women. Ann Behav Med 2001; 23: 5058.
  • 42
    Serafine M, Broom BL. Predicting low-risk pregnant women's attendance at a preterm birth prevention class. J Obstet Gynecol Neonatal Nurs 1998; 27: 279287.
  • 43
    Haslam C, Lawrence W, Haefeli K. Intention to breastfeed and other important health related behaviour and beliefs during pregnancy. Fam Pract 2003; 20: 528530.
  • 44
    Stewart DE, Streiner DL. Cigarette smoking during pregnancy. Can J Psychiatry 1995; 40: 603607.
  • 45
    Stewart DE, Streiner D. Alcohol drinking in pregnancy. Gen Hosp Psychiatry 1994; 16: 406412.
  • 46
    Haslam C, Lawrence W. Health-related behavior and beliefs of pregnant smokers. Health Psychol 2004; 23: 486491.
  • 47
    Strecher VJ, Rosenstock IM. The health belief model. In: Baum A, Newman S, Weinman J, West R, McManus C (eds). Cambridge Handbook of Psychology, Health and Medicine. Cambridge University Press: Cambridge, 1997, pp 113117.
  • 48
    Prochaska JL, DiClemente CC, Norcross JC. In search of how people change: applications to addictive behaviors. Am Psychol 1992; 47: 11021114.
  • 49
    Azjen I. From intentions to actions: a theory of planned behavior. In: Kuhl J, Beckman J (eds). Action-Control: From Cognition to Behavior. Springer: Heidelberg, 1985, pp 1139.