SEARCH

SEARCH BY CITATION

Keywords:

  • postpartum depression;
  • body image;
  • perfectionism

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Objective

To explore body dissatisfaction and maladaptive perfectionism as risk factors for postpartum depression symptomatology.

Design

Prospective, cohort design.

Setting

Urban and suburban hospital-based obstetrician and midwifery offices.

Participants

Forty-six (46) adult ethnically diverse pregnant women with gestational age greater than or equal to 28 weeks.

Method

Body dissatisfaction and maladaptive perfectionism were assessed in the offices of health care providers during the third trimester of pregnancy. Postpartum depression symptomatology was assessed at least 2 months postpartum using the Edinburgh Postnatal Depression Scale (EPDS) via postal mail or e-mail.

Results

Body dissatisfaction predicted postpartum depression symptoms, even after controlling for previously established risk factors. No main effect emerged between maladaptive perfectionism and postpartum depression symptoms.

Conclusions

Body dissatisfaction in the third trimester of pregnancy serves as a risk factor for postpartum depression. Therefore, assessment of body image during the third trimester of pregnancy may help health care providers identify women at risk of postpartum depression. Body dissatisfaction also may be an important target for postpartum depression prevention and treatment programs.

During the transition from pregnancy to motherhood, a woman experiences a myriad of changes, all of which can affect her emotional well-being. During this transition, approximately 13% of all new mothers will develop postpartum depression; however, it is estimated that 80% of cases go untreated and/or undetected (Flynn, 2005). Postpartum depression is of unique concern to the health care community because of its association with numerous adverse effects on mothers, spouses, and children. Effects include reduced maternal caregiving activities (e.g., breastfeeding, sleep routines, well-child visits, vaccinations, and safety practices), maladaptive parenting behavior, and behavioral problems and cognitive and linguistic delays in children (Field, 2010; Grace, Evindar, & Stewart, 2003; Reck et al., 2004). Furthermore, maternal depression in the postpartum period is correlated with paternal depression (meta-analytic estimate is .308), which also has deleterious effects on parenting behaviors (Paulson & Bazemore, 2010; Rosenthal, Learned, Liu, & Weitzman, 2013; Wilson & Durbin, 2010).

Research aimed at understanding postpartum depression is needed to guide treatment and prevention programs. Several risk factors for postpartum depression have been established in the literature. The results of a meta-analysis in which the author examined these risk factors indicated that the following factors had moderate effect sizes in predicting postpartum depression (the mean effect size indicator ranges for the risk factors are shown parenthetically): prenatal depression (.44–.46), low self-esteem (.45–.47), child care stress (.45–.46), prenatal anxiety (.41–.45), life stress (.38–.40), poor social support (.36–.41), strained marital relationships (.38–.39), history of depression (.38-.39), difficult infant temperament (.33-.34), and maternity blues (.25–.31) (Beck, 2001). Three risk factors emerged with small effect sizes: marital status (.21-.35), low socioeconomic status (.19–.22), and unplanned or unwanted pregnancy (.14–.17) (Beck, 2001).

Postpartum depression has deleterious effects on the health of the entire family and needs to be better understood to provide efficacious treatment and prevention.

In another synthesis of the research, Robertson, Grace, Wallington, and Stewart (2004) reported that the following antenatal risks factors had strong to moderate effective sizes in the prediction of postpartum depression: depression and anxiety during pregnancy, stressful life events, social support, and previous history of depression. Neuroticism and martial relationships were found to have moderate effect sizes, whereas socioeconomic status and obstetric factors had small effect sizes. A more recent systematic review of risk factors for depressive symptoms during pregnancy also may shed light on risk factors for postpartum depression because depression during pregnancy is one of the greatest risk factors for postpartum depression. Medium to large effect sizes were found for anxiety, negative life events, and lack of social support from partner (Lancaster et al., 2010). Additional risk factors not included in meta-analyses have begun to emerge and include history of intimate partner violence (Urquia, O'Campo, Heaman, Jansen, & Thiessen, 2011) and infants born prematurely (Vigod, Villegas, Dennis, & Ross, 2010).

Of particular interest in this study is emerging literature that suggests that maladaptive perfectionism and body image also may serve as risk factors for postpartum depression. Body image represents a combination of cognitive, behavioral, and perceptual conceptions of one's own body and often plays a large role in a woman's feelings of self-worth and self-esteem. Women experience body dissatisfaction when there is a perceived discrepancy between their current bodies and their ideal bodies (Rallis, Skouteris, Wertheim, & Paxton, 2007; Skouteris, Carr, Wertheim, Paxton, & Duncombe, 2005). Researchers have identified a positive correlation (Range [r] = .24–.39) between body dissatisfaction and depressive symptoms during pregnancy and have shown that body dissatisfaction is relatively stable throughout the perinatal period (Duncombe, Wertheim, Skouteris, Paxton, & Kelly, 2008; Skouteris et al.). Furthermore, there is evidence to suggest a relationship between body dissatisfaction and depressive symptoms measured during the postpartum period (r = .22–.38) (Birkeland, Thompson, & Phares, 2005; Rallis et al.). Changes in body dissatisfaction between pregnancy and the postpartum period also have been observed, with body dissatisfaction increasing postpartum (Rallis et al.). Prospective studies also have found that body dissatisfaction measured directly after childbirth or during the third trimester of pregnancy is correlated with symptoms of depression postpartum (r = .19–.58); however, these studies are methodologically limited by small sample sizes and the use of the Beck Depression Inventory–II (BDI-II), which is not an ideal measure of postpartum depression because of its focus on somatic symptoms that are a normal part of the postpartum period (Anderson, Fleming, & Steiner, 1994; Walker, Timmerman, Kim, & Sterling, 2002).

In addition to body dissatisfaction, the literature suggests that maladaptive perfectionism also may play a role in the development of postpartum depression symptoms. Perfectionism is a multidimensional stable personality construct that encompasses adaptive and maladaptive traits (Slaney, Rice, Mobley, Trippi, & Ashby, 2001). Adaptive perfectionism is characterized by a tendency to set high personal standards, prefer order and organization, and be unwilling to procrastinate (Rice & Ashby, 2007; Rice, Ashby, & Slaney, 1998; Rice & Mirzadeh, 2000). Maladaptive perfectionism is marked by these characteristics and also includes excessive concerns about making mistakes, self-doubts, distress about perceived discrepancies between actual performance and self-set standards, and a tendency to procrastinate (Rice & Ashby; Rice et al.; Rice & Mirzadeh). Maladaptive perfectionism has been linked to various forms of psychological distress, including anxiety, depression, and eating disorders whereas no such correlation has been found between adaptive perfectionism and psychopathology (Enns & Cox, 2005; Rice & Ashby; Rice et al.; Rice & Mirzadeh; Scott, 2007; Wu & Wei, 2008).

A strong link between maladaptive perfectionism and depression has been demonstrated in the literature. Specifically in samples of healthy college students, maladaptive perfectionism is positively correlated with symptoms of depression (r = .38-.68) (Ashby, Rice, & Martin, 2006; Chang, 2002; Huprich, Porcerelli, Keaschuk, Binienda, & Engle, 2008; Rice et al., 1998). This link has been also been replicated in clinical populations that show that maladaptive perfectionism is correlated with depression symptoms as measured by the BDI-II (r = .31–.51) (Clara, Cox, & Enns, 2007; Enns & Cox, 1999). Additionally, Rice and Mirzadeh (2000) found that individuals classified as maladaptive perfectionists scored approximately one half a standard deviation over the cutoff score on the Center for Epidemiological Studies–Depression Scale, which indicates clinically significant depression. Maladaptive perfectionists also report significantly more symptoms of depression than individuals classified as adaptive perfectionists (Rice & Ashby, 2007). Researchers have demonstrated that individuals with maladaptive perfectionistic traits may be more likely to develop depression because they need more reassurance from others, have less ability to provide their own self-reinforcement, and are more inclined to perceive stress or negative life events as their own personal failures (Dunkley, Sanislow, Grilo, & McGlashan, 2006; Enns & Cox, 2005; Flett, Hewitt, Blankstein, & Mosher, 1995; Wu & Wei, 2008).

Given the stress often experienced during the transition into motherhood, it is not surprising that perfectionism also is correlated with postpartum depression. Maladaptive perfectionism, specifically high concern over mistakes, measured during the postpartum period has been associated with a fourfold increase in risk of developing a major depressive disorder (Gelabert et al., 2012). Self-criticism, a specific aspect of maladaptive perfectionism, also is predictive of postpartum depression symptoms even after controlling for prenatal depression (Priel & Besser, 1999; Vliegen, Luyten, Meurs, & Cluckers, 2006). Furthermore, Mazzeo et al. (2006) found that women who had higher levels of maladaptive perfectionism retrospectively reported higher levels of postpartum depression symptoms. It may be that new mothers with perfectionistic tendencies experience stress and concern over not meeting their own high standards and, therefore, are more vulnerable to postpartum depression. For instance, they may have high expectations and standards for what it means to be the “perfect” mother or how a “good” infant should behave and view any deviation from these standards as their own personal failure. Women with perfectionistic tendencies may have greater difficulty adapting to and accepting changes in their bodies that result from pregnancy. These women may then experience greater distress when their high standards for perfection are not met, which produces a reduction in self-esteem and contributes to postpartum depression.

The purpose of this study was to explore relationships between body dissatisfaction and postpartum depression symptoms. We expanded upon previous research on these constructs by using a prospective design and validated measures of postpartum depression symptoms. We also controlled for previously established risk factors to determine if body dissatisfaction and maladaptive perfectionism uniquely contribute to the development of postpartum depression symptoms. After controlling for established risk factors for postpartum depression symptoms, the following two hypotheses were examined: body dissatisfaction in the third trimester of pregnancy would predict postpartum depression symptoms 2 months postpartum, and maladaptive perfectionism in the third trimester of pregnancy would predict postpartum depression symptoms 2 months postpartum.

Method

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Participants

Data were collected from 140 women during their third trimesters of pregnancy as part of a larger project in which researchers examined women's experiences during pregnancy and the postpartum period. Participants were recruited from two hospital-based obstetrics and gynecology offices in the greater Philadelphia area, one serving a primarily insured, middle- to upper-middle-class population and the second serving a primarily urban, low-income minority population. Additional participants were recruited from a suburban midwifery office serving primarily insured, middle- to upper-middle-class women. Eligibility criteria included the following: age 18 or older; ability to read, write, and understand English; and gestational age greater than or equal to 28 weeks. Data from 10 participants were not included in analyses because of incomplete measures. Because we focused on predictors of postpartum depression symptoms, only women who completed additional follow-up measures (2 months postpartum) were included in final analyses. The final sample comprised 46 women. Demographic descriptive statistics were calculated on the final sample of 46 women. These women ranged in age from 18 to 42 years (mean [M] = 27.17, standard deviation [SD] = 6.59). Most participants were middle class and married; 47.8% were identified as an ethnic minority (see Table 1 for complete demographic information).

Table 1. Demographic Information for Postpartum Sample
 Percentage 
 of Sample 
Demographic Variable(= 46)n
Note
  1. BMI = body mass index; SES = socioeconomic status.

  2. Number of children includes current pregnancy.

Ethnicity
African American26.112
Asian4.32
Caucasian non-Hispanic52.224
Hispanic8.74
Other8.74
SES
High15.27
Middle76.135
Low8.74
Marital Status
Single26.112
Married/Cohabitating65.230
Partnered8.74
Number of Children
One47.822
Two32.615
Three or more19.59
BMI Before Conception
Normal (18–24)50.023
Overweight (>25)50.023
History of an eating disorder2.21
History of major depressive disorder4.32

Measures

Demographics Questionnaire

Demographic information was collected during the third trimester of pregnancy using a nonvalidated demographics questionnaire created by the researchers. Participants were asked to report the following information: age, ethnicity, number of children, height, and weight. Participants also were asked whether they previously had been diagnosed with major depressive disorder or an eating disorder (no distinction was made between diagnoses of anorexia nervosa, bulimia nervosa, or eating disorder, not otherwise specified). Participants were not asked about a history of any other psychiatric diagnoses.

Postpartum Depression Predictors Inventory–Revised

Established risk factors for postpartum depression were assessed during the third trimester of pregnancy using the Postpartum Depression Predictors Inventory–Revised (PDPI-R) (Beck, 2002). The PDPI-R is a self-report measure that assesses the 10 prenatal risk factors that have been established as predictors of postpartum depression (prenatal depression, self-esteem, prenatal anxiety, life stress, social support, marital relationships, a history of depression, marital status, socioeconomic status, and unplanned/unwanted pregnancy) (Beck). Each section is scored by adding points for endorsed answers with higher scores, which indicates a strong propensity for developing postpartum depression. The PDPI-R demonstrates strong reliability (Cronbach's alpha = .83), a validity sensitivity of 76%, and specificity of 54% (Beck, Records, & Rice, 2006; Records, Rice, & Beck, 2007). The PDPI-R also demonstrated strong reliability in the present study (Cronbach's alpha = .84).

Body Attitudes Questionnaire

Body dissatisfaction was assessed during the third trimester of pregnancy using the Body Attitudes Questionnaire (BAQ; Ben-Tovim & Walker, 1991). The BAQ is a 44-item self-report measure designed to assess women's attitudes toward their bodies that has been validated with clinical and nonclinical populations. The BAQ demonstrates good reliability (Cronbach's alpha = .87) and has shown strong correlations with other measures of body dissatisfaction. Factor analyses have revealed six reliable subscales of the BAQ. These subscales include feeling fat, body disparagement, strength and fitness, salience of weight and shape, attractiveness, and fat on lower body. Participants are asked to respond to statements on a 5-point Likert-type scale, which range from 1 (strongly disagree) to 5 (strongly agree). Items can be summed to determine total and subscale scores with higher scores indicating higher levels of body dissatisfaction (Ben-Tovim & Walker). The BAQ has been used in samples of women during pregnancy and the postpartum period (Duncombe et al., 2008; Rallis et al., 2007; Skouteris et al., 2005). The BAQ demonstrated strong reliability in this study (Cronbach's alpha = .86).

Almost Perfect Scale–Revised and Frost's Multidimensional Perfectionism Scale

Maladaptive perfectionism was assessed during the third trimester of pregnancy using the Discrepancy subscale of the Almost Perfect Scale–-Revised (APS-R; Slaney et al., 2001) and the Concern Over Mistakes subscale and Doubts About Actions subscale of Frost's Multidimensional Perfectionism Scale (FMPS; Frost, Marten, Lahart, & Rosenblate, 1990). Combining these subscales of the APS-R and FMPS to measure maladaptive perfectionism has been supported using structural equation modeling and used in recent research (Blankstein & Dunkley, 2002; Wu & Wei, 2008). The Discrepancy subscale of the APS-R comprises 12 items designed to measure the tendency to set extremely high standards and to be displeased when these standards are not met. The Discrepancy subscale has good reliability (Cronbach's alpha = .91), has been found to be a strong measure of maladaptive perfectionism, and is related to poor adjustment, achievement, and self-esteem (Slaney et al.). Participants respond to statements on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). The Discrepancy subscale also demonstrated good reliability in this study (Cronbach's alpha = .96). The Concern Over Mistakes and the Doubts About Actions subscales of the FMPS have been shown to identify maladaptive perfectionists and to be correlated with depression (Dunkley, Blankstein, Halsall, Williams, & Winkworth, 2000; Slaney et al.). Participants respond to statements on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The Concern Over Mistakes subscale has good reliability (Cronbach's alpha = .88) and comprises nine items that assess tendencies to believe mistakes are personal failures and to overestimate the negative consequences of mistakes. This subscale also demonstrated strong reliability in this study (Cronbach's alpha = .90). The Doubts About Actions subscale also has strong reliability (Cronbach's alpha = .77) and comprises four items designed to measure doubts about one's own ability to complete tasks (Frost et al.). The Doubts About Actions subscale performed with good reliability in this study as well (Cronbach's alpha = .84). Participants’ responses on the Concern Over Mistakes subscale and Doubts About Actions subscale of FMPS were converted to a 7-point metric by multiplying each response by 7/5. Total scores on the Discrepancy subscale of APS-R, Concern Over Mistakes subscale of FMPS, and Doubts About Actions subscale of FMPS were then summed for a total score for maladaptive perfectionism, with higher scores indicating higher levels of maladaptive perfectionism. The combination of these three subscales demonstrated strong reliability in this study (Cronbach's alpha = .92).

Edinburgh Postnatal Depression Scale

Postpartum depression symptoms were assessed using the Edinburgh Postnatal Depression Scale (EPDS; Cox, Holden, & Sagovsky, 1987). The EPDS also was used to assess depressive symptoms during the third trimester of pregnancy. The EPDS is a screening instrument commonly used to detect postnatal depression in women after childbirth. The instrument purposely excludes certain symptoms that have been associated with new motherhood (e.g., fatigue). This self-report measure has 10 questions that specifically evaluate depressive symptoms experienced in the context of the past 7 days. Each item is rated on a 4-point Likert-type scale ranging from 0 to 3. Higher scores indicate endorsement of more depression symptoms with significance indicated at a score of 12 or more. Validity sensitivity is identified as 86%, with specificity at 78%, a positive predictive value of 73% and a split-half reliability of 0.88 with a standardized alpha coefficient of 0.87 (Cox et al.). Reliability of the EPDS was also strong in this study during pregnancy (Cronbach's alpha = .83) and the postpartum period (Cronbach's alpha = .86). Additionally, researchers have suggested that women who score between 9 and 11 may be a risk of postpartum depression. Scores at this level provide strong evidence for postpartum depression, whereas scores 12 and greater provide convincing diagnostic evidence (Gibson, McKenzie-McHarg, Shakespeare, Price, & Gray, 2009).

Procedure

This study was approved by university and hospital Institutional Review Boards. Women in their third trimesters of pregnancy (28 weeks or beyond) were recruited from waiting rooms of obstetrics and gynecology offices and midwifery offices and completed the following assessment measures: demographics questionnaire, PDPI-R, EPDS, BAQ, the Discrepancy subscale of the APS-R, and the Concern Over Mistakes subscale and the Doubts About Actions subscale of the FMPS. As part of the larger research project, five additional measures were administered to participants during pregnancy. These measures were given to test hypotheses associated with the larger research project but are not relevant to this study and, therefore, were not included in analyses.

Women who completed baseline assessments were contacted either via e-mail or postal mail 2 months after their expected due dates and asked to complete the EPDS. As part of the larger research project, data from three additional measures were also collected 2 months postpartum. Participants were given the option of completing follow-up measures using an online confidential database designed for social science research or completing paper-and-pencil measures and returning completed packets to the researchers via postal mail. In an attempt to reduce attrition, participants were given the option to receive reminder calls and/or reminder e-mails. All women who provided contact information received two reminders calls or e-mails at one week and one month after follow-up measures were sent out. Participants were mailed a $20 gift card after completion of the follow-up measures.

Body dissatisfaction during the third trimester of pregnancy is uniquely related to the development of postpartum depression.

Data Analysis

Preliminary analyses of descriptive statistics were run to determine the prevalence of postpartum depression symptoms and established risk factors, body dissatisfaction, and maladaptive perfectionism. Additionally, an ANOVA was run to determine if women who completed follow-up measures differed on prenatal assessments from women who failed to follow up. A multiple regression was conducted to determine if age or ethnicity accounted for any variance in postpartum depression symptoms. Other demographic variables of interest (e.g., marital status and socioeconomic status) were controlled for in the primary analysis using the PDPI-R.

The primary hypotheses were tested using a hierarchical multiple regression to determine whether body dissatisfaction and maladaptive perfectionism during pregnancy predicted postpartum depression symptoms. Hierarchical multiple regression was chosen to control for established risk factors for postpartum depression. Zero-order correlations and collinearity statistics were run to determine the presence of multicollinearity. Finally, a power analysis was run post hoc to determine power in this study (Faul, Erdfelder, Buchner, & Lang, 2009). Results of power analysis demonstrated adequate power (power = .83).

Results

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Analysis of differences on prenatal assessments between women who completed follow-up measures and those who failed to follow-up demonstrated no significant differences for measures of body dissatisfaction, F(1, 124) = .01, = .948), maladaptive perfectionism, F(1, 124) = .06, = .802), or established risk factors for postpartum depression, F(1, 124) = 1.06, = .305). However, women who completed follow-up measures reported significantly less symptoms of depression, as measured by the EPDS during pregnancy (M = 7.09, SD = 4.36), than women who did not complete follow-up measures (M = 9.45, SD = 5.39), F(1, 124) = 6.41, = .013. No differences in postpartum depression symptoms were observed based on age or ethnicity, F(2,43) = .20, = .817.

As can be seen in Table 2, EPDS scores measured during the postpartum period ranged from 0 to 18 with a mean of 5.87 (SD = 4.46). Several women reported a clinically significant level of symptoms associated with postpartum depression, with 17.4% indicating a level of symptoms that put them at risk of developing postpartum depression (EPDS = 9–11). Furthermore, 10.9% of women reported a level of symptoms that is likely consistent with a diagnosis of a depressive disorder (EPDS ≥ 12). Means, standard deviations, and ranges for predictor variables are provided in Table 3.

Table 2. Means, Standard Deviations, and Ranges for Independent and Dependent Variables
VariableMSDRange
Note
  1. N = 46; APS-R = Almost Perfect Scale–Revised (Discrepancy subscale); BAQ = Body Attitudes Questionnaire; EDPS = Edinburg Postnatal Depression Scale; FMPS = Multidimensional Perfectionism Scale (Concern Over Mistakes and Doubts About Actions subscales); PDPI-R = Postpartum Depression Predictors Inventory–-Revised; SD = standard deviation.

Established risk factors for postpartum depression (PDPI-R)6.064.651–19
Body dissatisfaction (BAQ)110.3714.3786–153
Maladaptive perfectionism (APS-R, FMPS)65.3624.4630.20–123.6
Postpartum depression symptomatology (EDPS)5.874.460–18
Table 3. Zero-Order Correlations among Predictor Variables
Predictor123
Note
  1. N = 46; APS-R = Almost Perfect Scale–Revised (Discrepancy subscale); BAQ = Body Attitudes Questionnaire; FMPS = Multidimensional Perfectionism Scale (Concern Over Mistakes and Doubts About Actions subscales); PDPI-R = Postpartum Depression Predictors Inventory–Revised.

Established risk factors for1.00−.087.536
postpartum depression (PDPI-R)   
Body dissatisfaction (BAQ)1.00.389
Maladaptive perfectionism1.00
(APS-R, FMPS)   

A significant model emerged in the primary multiple regression analysis, F(4, 32) = 5.652, = .002. Results indicated that previously established risk factors for postpartum depression explained 11.2% of the variance in postpartum depression symptoms (R= .11; F(1, 44) = 5.55, = .023). An additional 17.6% of variance was accounted for by body dissatisfaction and maladaptive perfectionism, which suggests that together these variables were significant predictors of postpartum depression symptoms after controlling for established risk factors (R2 change = .18, F(2, 42) = 5.16, = .010). As can be seen in Table 4, body dissatisfaction emerged as a significant predictor of postpartum depression symptoms whereas maladaptive perfectionism did not contribute any unique variance to the model. There was little concern for multicollinearity: PDPI-R (tolerance = .66, variance inflation factor [VIF] = 1.61), Maladaptive perfectionism (tolerance = .53, VIF = 1.89), BAQ (tolerance = 0.73, VIF = 1.38).

Table 4. Summary of Hierarchical Multiple Regression for Examined Predictors of Postpartum Depression Symptoms
PredictorΔR2β
Note
  1. N = 46; APS-R = Almost Perfect Scale–Revised (Discrepancy subscale); BAQ = Body Attitudes Questionnaire; FMPS = Multidimensional Perfectionism Scale (Concern Over Mistakes and Doubts About Actions subscales); PDPI-R = Postpartum Depression Predictors Inventory–Revised.

  2. *< .05

  3. < .01

Step 1.11* 
Established risk factors for .34*
postpartum depression (PDPI-R)  
Step 2.18** 
Body dissatisfaction (BAQ) .49**
Maladaptive perfectionism −.28
(APS-R, FMPS)  
Total R2.29** 

Discussion

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Our findings provide additional information regarding risk factors for postpartum depression symptoms in a diverse sample of women. Using an EPDS cutoff score of 12, approximately 10.9% of women in this sample reported a level of symptoms consistent with a diagnosis of postpartum depression, which is slightly below the established population prevalence rate of 13% (Flynn, 2005). The lower prevalence of likely postpartum depression observed in this sample may be attributed to the finding that women who completed follow-up measures were significantly less depressed during pregnancy than women who did not complete follow-up measures. Prenatal depression is one of the strongest correlates of postpartum depression, and therefore it is likely that rates of postpartum depression in this sample were affected by attrition (Beck, 2001). Additionally, researchers have suggested that optimal cutoff scores vary depending on ethnicity, and EPDS cutoff scores should be lowered at least 2 to 3 points to accurately identify depression in low-income, urban, Black mothers (Chaudron et al., 2010; Gibson et al., 2009). Therefore, the prevalence of postpartum depression in this sample may actually align with population rates given that a substantial portion of the sample was Black and living in an urban environment.

This study was strengthened by the ethnic diversity of the sample. Notably, postpartum depression symptoms did not differ based on ethnicity or age. As expected, the previously established risk factors for postpartum depression measured by the PDPI-R significantly predicted postpartum depression symptoms in this sample. This lends support to the use of the PDPI-R to identify women at risk of postpartum depression symptoms in ethnically diverse samples and also allows for the examination of additional risk factors that may be important to consider when assessing a woman's propensity to develop postpartum depression.

Consistent with hypotheses, after controlling for established risk factors, body dissatisfaction during pregnancy emerged as a significant predictor of postpartum depression symptoms. This finding supports previous research and expands upon these studies by using a prospective design and validated measures of postpartum depression symptoms (Anderson et al., 1994; Birkeland et al., 2005; Rallis et al., 2007; Walker et al., 2002). Researchers have shown that body dissatisfaction may increase in the postpartum period (Rallis et al.). Women who reported higher body dissatisfaction during pregnancy may experience body dissatisfaction postpartum to an even greater degree. This potential increase in body dissatisfaction postpartum may serve as a contributing factor in the postpartum depression symptoms observed in women who reported high levels of body dissatisfaction during pregnancy.

Health care providers should assess women's satisfaction with their bodies during pregnancy to identify women at risk of postpartum depression and provide necessary treatment.

Contrary to hypotheses and previous research, maladaptive perfectionism during pregnancy did not predict symptoms of postpartum depression symptoms (Gelabert et al., 2012; Mazzeo et al., 2006; Priel & Besser, 1999; Vliegen et al., 2006). This difference may reflect the high attrition rate in this study and the fact that women who completed follow-up measures were significantly less depressed during pregnancy than women who did not complete follow-up measures. Our results may differ from the previous research because of differences in the time of measurement of maladaptive perfectionism and postpartum depression symptoms. Prior studies that demonstrate a relationship between these constructs have either used retrospective reports or measured both constructs at the same time point. Of note, a significant and positive correlation between maladaptive perfectionism and prenatal depressive symptoms measured by the EPDS was found in this sample (r = .72, < .001).

Although maladaptive perfectionism is thought of as a stable construct, the degree that the traits are expressed may be dependent on other variables, such as stress, depression, or achievement-related tasks. In particular stress, problem solving, self-esteem, shame, and ineffective coping have been found to interact with perfectionism in the prediction of depression and may account for the lack of relationship between maladaptive perfectionism and postpartum depression symptoms in this sample (Ashby et al., 2006; Chang, 2002; Rice et al., 1998; Wei, Heppner, Russell, & Young, 2006; Yang & Jiang, 2008). Furthermore, maladaptive perfectionism is a multidimensional construct and dimensions not assessed in this study also may explain lack of the significant findings. Specifically, given the relational nature of postpartum depression and the importance of social support, it may be that dimensions of perfectionism that are socially prescribed (e.g., beliefs that the approval of others is dependent on being perfect) are more salient in the development of postpartum depression symptoms (Stoeber, Kempe, & Keogh, 2008).

There are several limitations that should be considered when interpreting these findings. The sample size is relatively small, and although power analysis revealed adequate power, a larger sample size would have strengthened the study and allowed for additional analyses (e.g., subcomponents of body image and any interaction effects). In an attempt to reduce attrition, the length of follow-up measures was minimized, monetary incentive was only provided to women who completed follow-up measures, and participants were contacted twice by phone or e-mail to encourage completion of follow-up measures. Despite these efforts, almost two thirds of the original sample failed to complete follow-up measures, and women who completed follow-up measures were significantly less depressed during pregnancy than women who did not complete follow-up measures. Therefore, it is likely that women who did not complete follow-up measures were more depressed postpartum and would have been particularly valuable to include in final analyses. This also may explain discrepancies between the findings of this study and those in the previous literature. Body dissatisfaction and maladaptive perfectionism were only measured during pregnancy, which limited our ability to explore changes in body image or perfectionism from pregnancy to postpartum and their relationship with depressive symptoms at both time points. Finally, although we used a prospective design, correlational analyses were conducted and thus causation cannot be assumed.

In summary, in an ethnically diverse sample of women, body dissatisfaction during the third trimester of pregnancy appears to be uniquely related to the development of postpartum depression symptoms after controlling for established risk factors. Therefore, it may be beneficial for health care providers to assess women's satisfaction with their bodies during pregnancy to identify women at risk of postpartum depression, particularly because less than one half of obstetricians currently indicate that they routinely assess body image (Leddy, Jones, Morgan, & Schulkin, 2009). Health care providers may want to encourage women that present with disturbances in body image to engage in prevention programs designed to reduce the risk of postpartum depression. Additionally, it may be helpful to target body dissatisfaction when working with women who struggle with depression during the antenatal and postnatal periods.

These findings also can help guide future research. In future studies with larger sample sizes, researchers could explore the relative contribution of subcomponents of body dissatisfaction to the development of postpartum depression symptoms to tailor risk assessments and interventions. Additionally, it would be beneficial to measure body dissatisfaction and maladaptive perfectionism during pregnancy and postpartum to further explore the relationship between these constructs. Additional studies on the role of maladaptive perfectionism in the prediction of postpartum depression symptoms are needed. der to gain a better understanding of the relationship between maladaptive perfectionism and postpartum depression symptoms, researchers should use larger sample sizes and explore additional variables that may affect the development of postpartum depression symptoms, such as stress, self-esteem, coping, and shame. Researchers also should consider variables that affect a woman's likelihood to complete assessment measures during the postpartum period. Women with significant levels of depression or environmental stress may have more difficulty independently completing follow-up measures. Researchers may want to assess women during health care appointments or establish protocols for home visits to increase follow-up participation.

The results of this study provide valuable information for health care providers that work with women during the prenatal and postpartum periods. This study significantly expands upon previous research by using a prospective design, diverse sample, and validated measures of the constructs of interest. It is essential for the health care community to continue to explore the effect of body image and maladaptive perfectionism on the development of postpartum depression symptoms to improve identification of populations at risk. This study also adds to the existing literature on possible targets for prevention and treatment of postpartum depression and thus allows for increased quality of care for mothers and families affected by this debilitating illness.

Acknowledgment

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Supported by La Salle University Research Grant.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies

Biographies

  1. Top of page
  2. ABSTRACT
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgment
  7. REFERENCES
  8. Biographies
  • Alison C. Sweeney, PsyD, is a staff psychologist in the Women's Health Center, Michael E. DeBakey VA Medical Center, Houston, TX and assistant professor in the Menninger Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine, Houston, TX.

  • Randy Fingerhut, PhD, is Director of Clinical Training and an associate professor in the Department of Psychology, La Salle University, Philadelphia, PA.