Matern Child Health J (2012) 16:929935

DOI 10.1007/s10995-011-0816-7

Depression and Pregnancy Stressors Affect the Association

Between Abuse and Postpartum Depression
D. Yvette LaCoursiere Kathryn P. Hirst
Elizabeth Barrett-Connor

Published online: 17 May 2011

Springer Science+Business Media, LLC 2011

Abstract To determine how psychosocial factors affect

the association between a history of abuse and postpartum
depression (PPD). Women at four urban hospitals in Utah
were enrolled B48 h of delivering a live-born infant. At
enrollment, pregravid history of physical or sexual abuse
was obtained via self-report. Psychosocial covariates such
as pregnancy stressors and depression were also collected.
Pregnancy stressors were categorized using stressor
questions from the Pregnancy Risk Assessment Monitoring
System. The primary outcome measure, a pre-specified
Edinburgh Postnatal Depression Scale score of C12 was
obtained 68 weeks postpartum. Among the 1,038 women
studied, psychosocial risk factors were common: abuse
history 11.7%, pregnancy stressorsfinancial 49.1%, emotional 35.0%, partner-associated 19.8%, and traumatic
10.3% and depression history 16.7%. While abuse was
associated with a ?PPD screen in a preliminary model
[aOR 2.05 (1.28, 3.26)], adding psychosocial covariates
reduced the unadjusted association of abuse and PPD [aOR
1.12 (0.66, 1.91)]. After adjustment, PPD was associated
with depression history [aOR 2.85 (1.90, 4.28)], prepregnancy BMI [aOR 1.04 (1.01, 1.07)] multiple stressors [3
D. Yvette LaCoursiere (&)
Division of General Obstetrics and Gynecology, Department of
Reproductive Medicine, University of California, 200 West
Arbor Dr., #8433, San Diego, CA 92103, USA
e-mail: ylacoursiere@ucsd.edu
K. P. Hirst
Department of Psychiatry and Department of Family and
Preventive Medicine, University of California, San Diego,
CA, USA
E. Barrett-Connor
Division of Epidemiology, Department of Family and Preventive
Medicine, University of California, San Diego, CA, USA

categories aOR 4.35 (2.00, 9.46)]; 4 categories [aOR 6.36

(2.07, 19.49)] and sum of stressors * history of abuse [aOR
1.50 (0.92, 2.46)]. Interestingly only women with a moderate number of stressors were sensitive to an abuse history.
Abuse and pregnancy stressors are common and interact to
influence the likelihood of screening positive for PPD.
Keywords Abuse
Pregnancy stressors
Depression and
postpartum depression

Introduction
Emerging data suggests that a history of abuse is a risk
factor for postpartum depression (PPD). Women with an
abuse history tend to experience other psychosocial
stressors, therefore these potential covariates should be
included in abuse-PPD analyses. However, to date these
more complex analyses are few.
A recent systematic review identified only eight studies
using standardized measures to assess the impact of an
abuse history on PPD [1]. Seven of these studies found an
association between an abuse history and either PPD or
elevated scores on screening tests for depression. The
remaining study did not find an overall association, but a
subgroup analysis revealed that sexual abuse was associated with depression during pregnancy [2]. We identified
an additional study of 248 Middle Eastern women that
showed an increase in self-reported depressive symptoms
during pregnancy in women reporting an assault victim
history [3]. The majority of these studies reported only a
univariate association and did not explore the influence of
other psychosocial factors that are likely to affect this
association such as a preexisting diagnosis of depression or
the presence of stressors during pregnancy.

123

930

Depression has been associated with abuse in non-pregnant women and is a strong risk factor for PPD [4, 5].
However, only two of the studies included in the above
mentioned review included pregravid depression as potential
covariate in the abuse-PPD association [6, 7]. Cohen et al. [6]
studied 200 women and confirmed the univariate association
between emotional abuse during pregnancy and PPD, however the association was no longer demonstrable in a multivariate model which included a history of depression
before pregnancy. Records and Rice found no association
between depression and history of abuse, but the study was
limited by small sample size (n = 28) and attrition [7].
Another set of potentially important factors affecting the
abuse-PPD association are pregnancy stressors. Stressors
have been implicated as a risk factor for both depression
and PPD [4, 810]. However, we could not identify any
studies assessing how pregnancy stressors impact the
abuse-PPD association.
Not only are abuse and stressors linked to depression in
women outside of pregnancy, but recent univariate analyses describe a link between abuse and PPD. These studies
have only minimally evaluated the impact of depression
history and have not assessed pregnancy stressors as
covariates. Given these findings and acknowledging that
both depression history and pregnancy stressors are quite
common, an assessment of how these covariates affect the
abuse-PPD link seems warranted [1113]. Consequently,
this study aims to quantify the association between abuse
history and PPD and assess how depression history and
pregnancy stressors affect this association.

Methods
This is a secondary analysis of data from a published study
describing the influence of body mass index on PPD [14].
After Institutional Review Board approval was obtained,
the primary study enrolled 1,054 women delivering liveborn singleton infants from 2005 to 2007. These women
were invited to participate 2448 h postpartum from four
urban hospitals in Utah. Only those women who completed
the abuse screening were included in this current analysis
(n = 1,038).
Initial enrollment data were collected via a self-administered questionnaire. Upon completion of the questionnaire the study staff confirmed accuracy of the data by
interviewing the participant and reviewing the inpatient
medical record.

Matern Child Health J (2012) 16:929935

(physical and sexual), (2) history of depression and (3)

pregnancy stressors. Women were classified as having a
history of abuse if they answered yes to Have you ever
been physically or sexually abused? in the portion of the
questionnaire that inquired about events prior to this
pregnancy. Depression history was obtained by written
report of depression (outside of pregnancy or the postpartum period) or depression diagnosed in pregnancy or
postpartum. Pregnancy stressors were identified using the
13 pregnancy stressor questions from the Centers for Disease Control and Preventions Pregnancy Risk Assessment
Monitoring System (PRAMS). These questions, completed
in written form by the study participant, ask about stressful
life events during pregnancy. PRAMS stressors were then
categorized into four maternal stressor constructs: financial, partner-associated, emotional and traumatic stressors.
These categories have been previously identified by a
principal component analysis and described in the literature
[15].
Additional independent variables including demographic information, anthropometric data (as the primary
study demonstrated an association between increased BMI,
PPD and stressors), medical history (including other psychiatric and obstetric history), and habits such as smoking,
alcohol and recreational drug use were obtained at intake.
The demographic data collected included maternal age,
education, race/ethnicity, marital status, and insurance
status. Maternal age (years), gravidity and education (years
completed) were self-reported and reported as a continuous
variable. Race/ethnicity was self-reported and only White
non-Hispanics and Hispanics were included in the regression models because there were very few Blacks, Asian/
Pacific Islanders, or other ethnic groups in the study,
reflecting the general population in Utah. Marital status at
enrollment was categorized into one of four categories as
single, married, separated, or divorced. Insurance status
was defined as one of three groups: insured if women
reported private insurance, underinsured if women reported
publically funded prenatal or medical insurance and uninsured if women did not report health insurance during the
prenatal period. Prepregnancy self-reported body mass
index was calculated and pregnancy weight gain was
recorded. Lastly, cigarette smoking, alcohol use and drug
use were defined dichotomously (yes or no) if the women
described any use during pregnancy. These variables were
included to assess the co-occurrence of remediable factors
associated with abuse.
Dependent Variable

Independent Variables
At the intake session we collected the key independent
variables including (1) history of prepregnancy abuse

123

The primary outcome measure was depressed mood at

68 weeks postpartum as determined by a score of C12 on
the Edinburgh Postnatal Depression Scale (EPDS). The

Matern Child Health J (2012) 16:929935

EPDS was administered after delivery via a mailed, selfadministered questionnaire. The EPDS is a well-validated
10 question instrument which identifies women at risk for
PPD [16]. A score of C12 was selected as it has been
shown to have good sensitivity, specificity and positive
predictive value for PPD [17].
Statistical Analysis
Proportions were compared via Chi-square tests and continuous variables were compared using independent t tests.
All tests of hypotheses were 2-tailed, with a type 1 error
fixed at 5%. The percent of women screening positive for
PPD were calculated and stratified by the presence of abuse
history, depression history and stressor frequency. To
better describe the influence of multiple variables the data
were analyzed in a cohort design using five multiple
logistic regression models with age, gravidity, race/ethnicity, education, prepregnancy BMI, and history of abuse
in all models and with 4 models varying the inclusion of
recent pregnancy stressors and depression history as
explanatory variables. EPDS C12 was the dichotomous

931

outcome variable for logistic modeling. Maternal age,

gravidity, education and prepregnancy BMI were included
as continuous variables. Ordinal scaled data were coded
using dummy variables. The interaction of both depression
history and sum of stressors by history of abuse was tested
by the introduction of cross products into the regression
models, P \ 0.15 was used for testing significance of
interaction. The analyses were performed with Statistical
Package for the Social Sciences (SPSS), version 16.

Results
Of the 1,054 women enrolled in the primary study, 1,038
responded to the questions on abuse and completed the
EPDS 68 weeks after delivery. A comparison of the
demographic and psychosocial history of women with and
without an abuse history is presented in Table 1. The 121
women (11.7%) with a history of abuse were more likely to
be unmarried, underinsured, and have a history of depression (outside of pregnancy and PP) and other psychiatric
disorders. In fact women with a history of abuse were twice

Table 1 Distribution of
maternal demographics and
psychosocial history by reported
history of personal abuse

Total

No history
of abuse
N = 917

History
of abuse
N = 121

P value

Age

27.7 4.9

27.7 4.9

27.4 5.2

0.520

Gravidity

2.5 1.6

2.5 1.6

2.7 1.7

0.138

Education

14.7 2.7

14.8 2.8

14.3 2.5

0.064

Prepregnancy BMI

24.4 5.5

24.2 5.3

25.4 6.7

0.017

Race/ethnicity
White, non-Hispanic

907 (87.4)

805 (87.8)

102 (84.3)

White, Hispanic

73 (7.0)

61 (6.7)

12 (9.9)

Black

5 (0.5)

4 (0.4)

1 (0.8)

Asian, Pacific Islander

22 (2.1)

21 (2.3)

1 (0.8)

Other

31 (3.0)

26 (2.9)

5 (4.2)

0.570

Marital status
Married

929 (91.3)

834 (92.7)

95 (80.5)

Single

83 (8.2)

62 (6.9)

21 (17.8)

Divorced/separated

\0.001

6 (0.6)

4 (0.4)

2 (1.7)

Insurance status
Insured

722 (69.6)

653 (71.2)

69 (57.0)

Underinsured

208 (20.0)

170 (18.5)

38 (31.4)

Uninsured

108 (10.4)

94 (10.3)

14 (11.6)

0.002

History of depression

173 (16.7)

133 (14.5)

40 (33.1)

\0.001

History of PPD (multiparas) n = 616

87 (14.1)

65 (11.9)

22 (31.4)

\0.001

History of other psychiatric disorders

11 (1.1)

7 (0.8)

4 (3.3)

Family history of psychiatric disease

157 (15.1)

131 (14.3)

26 (21.5)

0.032

Tobacco use

23 (2.2)

16 (1.7)

7 (5.8)

0.012

0.011

Alcohol use

135 (13.0)

113 (12.3)

22 (18.2)

0.072

Recreational drug use

7 (0.7)

3 (0.3)

4 (3.3)

0.002

123

932

as likely to report a history of depression (P \ 0.001) and

over 2.5 times more likely to report a history of PPD
(P \ 0.001). Additionally their BMI was higher on average
than women who did not report an abuse history. They
were also more likely to have smoked or used recreational
drugs during pregnancy compared to women without an
abuse history. However these events were quite infrequent
accounting for 2 and \1% of the sample, respectively. As
shown in Table 2, the frequency of stressful life events and
all categories of PRAMS stressors were more common
among women with a history of abuse.
Overall, 17.3% of all women screened positive for
PPD. Several factors increased the risk of screening
positive for PPD such as abuse before pregnancy, a physical fight during pregnancy, and initially race/ethnicity.
Nearly twice as many women with a pregravid abuse
history screened positive for PPD compared to those
without (28.1% vs. 15.4%, P = 0.001). Women reporting a
physical fight during pregnancy had a fourfold increased
odds of screening positive for PPD [OR = 4.09 (1.23,
13.54)]. Hispanic women screened positive for PPD
more commonly than white, non-Hispanic women. (28.4%
vs. 16.2%, P = 0.007), but this association was no
longer significant after controlling for other risk factors
(Table 3).

Matern Child Health J (2012) 16:929935

As the frequency of stressors increased the prevalence of

screening positive for PPD also increased. Approximately
10% (30 of 293) of women without any stressors screened
positive for PPD, this increased to 15.4% (62 of 402) of
women with 1 stressor category, 17.2% (39 of 188) with 2
stressor categories, 41.3% (31 of 75) with 3 different categories of stressors and 61.9% (13 of 21) of women
reporting stressors in all four categories (Chi-square test for
trend P \ 0.001). Figure 1 depicts these percents stratified
by the presence or absence of abuse before pregnancy.
Differences between the percentages of women screening
positive for PPD by abuse history existed only when 2 or 3
stressor (moderate) categories were reported. In women
with 01 stressors (few) and all 4 stressors (marked) during
pregnancy, history of abuse was not associated with
screening positive for PPD.
Models were constructed to examine the effect of various exposure variables on PPD while controlling for
associated covariates. Insurance status did not significantly
alter the models and was not included. Table 3 contains
five models of screening positive for PPD. Abuse history is
associated with PPD, but pregnancy stressors and history of
depression significantly reduce this association. The interaction of abuse history with depression history was
not statistically significant (P = 0.638). However, the

Table 2 Frequency of stressful life events during pregnancy in 1,038 women by history of abuse prior to pregnancy
Total (%)

No history
of abuse (%)

History
of abuse (%)

P value

Financial stressors

501 (49.1)

432 (47.9)

69 (58.0)

0.040

Partner associated stressors

202 (19.8)

164 (18.2)

38 (31.9)

\0.001

Emotional stressors

357 (35.0)

303 (33.6)

54 (45.8)

0.009

Traumatic stressors

105 (10.3)

74 (8.2)

31 (26.1)

\0.001

402 (39.3)

350 (38.8)

52 (43.7)

0.300

78 (7.6)

67 (7.4)

11 (9.2)

0.477

137 (13.4)

109 (12.0)

28 (23.5)

0.001

44 (4.3)

35 (3.9)

9 (7.6)

0.063

Financial stressors
You moved to a new address
Your husband or partner lost his job
You had a lot of bills you couldnt pay
You lost your job even though you wanted to go on working
Partner associated stressors
You got separated or divorced from your husband or partner

32 (3.1)

24 (2.7)

8 (6.7)

0.017

170 (16.7)

139 (15.4)

31 (26.1)

0.003

32 (3.1)

21 (2.3)

11 (9.2)

305 (29.8)
153 (15.0)

256 (28.3)
127 (14.0)

49 (41.5)
26 (21.8)

You were homeless

16 (1.6)

9 (1.0)

7 (5.9)

\0.001

You were in a physical fight

11 (1.1)

6 (0.7)

5 (4.2)

\0.001

You or your husband or partner went to jail

24 (2.3)

17 (1.9)

7 (5.9)

0.007

Someone close to you had a bad problem with drinking or drugs

84 (8.2)

59 (6.5)

25 (21.0)

You argued with your husband or partner more than usual

Your husband or partner said he didnt want you to be pregnant

\0.001

Emotional stressors
A close family member was very sick and had to go into the hospital
Someone very close to you died

0.003
0.025

Traumatic stressors

Denominator may vary secondary to missing data

123

\0.001

Matern Child Health J (2012) 16:929935

933

Table 3 Multiple logistic regression models of screening positive for postpartum depression by history of abuse with and without the inclusion
of recent pregnancy stressors and history of depression
Model 1
history of
abuse only

Model 2
history of
abuse and pregnancy
stressors

Model 3
history of
abuse and history
of depression

Model 4
history of abuse,
pregnancy stressors
and history
of depression

Model 5 full model

with interaction
of abuse
and pregnancy
stressors

Age

0.95 (0.91, 0.99)

0.97 (0.92, 1.02)

0.95 (0.91, 1.00)

0.97 (0.92, 1.02)

0.97 (0.92, 1.02)

Gravidity

1.19 (1.05, 1.35)

1.16 (1.02, 1.32)

1.15 (1.01, 1.30)

1.13 (0.99, 1.28)

1.13 (0.99, 1.29)

White, Non-Hispanic

Reference

Reference

Reference

Reference

Reference

White, Hispanic

1.39 (0.74, 2.61)

1.30 (0.66, 2.54)

1.45 (0.76, 2.75)

1.39 (0.70, 2.78)

1.41 (0.70, 2.82)

Education
Prepregnancy BMI

0.91 (0.85, 0.98)

1.04 (1.01, 1.08)

0.92 (0.85, 1.00)

1.04 (1.01, 1.08)

0.93 (0.86, 1.01)

1.04 (1.01, 1.07)

0.94 (0.87, 1.03)

1.04 (1.01, 1.07)

0.94 (0.87, 1.03)

1.04 (1.01, 1.07)

History of abuse

2.05 (1.28, 3.26)

1.45 (0.87, 2.43)

1.51 (0.92, 2.48)

1.12 (0.66, 1.91)

0.55 (0.19, 1.56)

Race/ethnicity

PRAMS stressors (# categories positive)

0

Reference

Reference

Reference

1.76 (1.07, 2.91)

1.73 (1.04, 2.88)

1.67 (1.00, 2.77)

1.54 (0.87, 2.76)

1.43 (0.79, 2.57)

1.31 (0.72, 2.37)

5.56 (2.81, 10.99)

5.71 (2.84, 11.50)

4.35 (2.00, 9.46)

11.24 (4.04, 31.25)

8.63 (3.01, 24.76)

6.36 (2.07, 19.49)

2.84 (1.93, 4.19)

2.78 (1.86, 4.16)

2.85 (1.90, 4.28)

0.41

0.15

0.15

Sum of stressors * history of abuse

1.50 (0.92, 2.46)

History of depression
Constant

0.58

0.18

70
60
50

es
str
4

str

str

es

so

rs

so

so
es

es
str
1

rs

rs

so

so
es
str
No

History of abuse
No history of abuse

rs

40
30
20
10
0

No history of abuse

History of abuse

Fig. 1 Percent of women with PPD by history of abuse (absent and

present) and number of PRAMS stressor categories reported during
pregnancy

interaction of sum of stressors and history of abuse as

tested by the introduction of a cross product in model #4
was statistically significant (P = 0.103) and the significance of the main effects of abuse and the sum of stressors
remain statistically unchanged.

Discussion
In this study assessing the association of abuse and PPD,
we found that both were quite common and related. Nearly

12% of women reported a history of physical or sexual

abuse prior to pregnancy and 17% screened positive for
PPD. Supporting recent publications, our univariate analyses revealed a twofold increased odds of PPD in women
with an abuse history compared to those without a history
of abuse. The two covariates of interest, depression history
and pregnancy stressors, were also remarkably common
and affected the univariate association. Depression before
pregnancy and pregnancy stressors in a multivariate model
significantly decreased this association by up to 90%. In the
adjusted model, depression history, prepregnancy BMI,
multiple stressors, and the interaction between stressors
and abuse predicted PPD.
Our analyses suggest that women with an abuse history
who now report an environment with few stressful life
events are not at elevated risk of PPD. Furthermore,
women who now experience an excess of stressful events
are at maximum risk of PPD (nearly 60%) regardless of
abuse history. However it is the women who see themselves as being in the moderate risk categories (2 or 3
stressor categories) that are most sensitive to a history of
abuse.
These findings of a decrease in the abuse-PPD association with the introduction of other psychosocial risk factors
are consistent with those of Cohen et al. [6] who described
no association between abuse and PPD when depression
history was included. Interpreting the attenuation of the

123

934

abuse-PPD association is challenging. Admittedly, given

that women with a history of abuse are more likely to
report stressors during pregnancy including prenatal abuse,
adjusting for abuse may be over-adjusting. Schisterman
and colleagues described overadjustment bias as controlling for an intermediate variable on the pathway between
exposure and disease. This could result in a diminution in
the size of the association [18]. It is important to consider
this when interpreting these data, however causal pathway(s) between abuse, psychosocial stressors, depression,
and screening positive for PPD have not been elucidated
and as such exploratory models are necessary to begin to
understand these pathways.
There are several strengths of the study including its
large sample size and the use of validated instruments to
measure pregnancy stressors and PPD. In contrast to
earlier studies, we captured the interaction between prior
episodes of abuse and depression as well as current
pregnancy stressors. However some limitations should be
noted. We did not determine the type, relationship of
perpetrator, or specific timing of abuse. Likewise, for
women reporting both abuse and depression before
pregnancy we do not know which came first. It is possible
that some women had undiagnosed depression during
pregnancy which was misidentified as PPD. Also, the
characteristics of the study population reflect their distribution within the state of Utah, as such the results may
not be generalizable to other populations. For example,
the overall low rates of smoking and recreational drug use
in this study population does reflect the population in
Utah, and allows for less confounded estimates of variables which precede or follow depression, but may also
decrease the generalizability of the findings.
From a practical perspective, the PRAMS stressor
questions are readily available and easy to administer.
Given this and the findings that (1) exposure to fewer
stressful life events among women with a history of abuse
may return their PPD risk to baseline, (2) women in the
moderate stressor risk categories are most sensitive to
abuse history and (3) all women with excess stressors are at
maximum risk of PPD, perhaps this tool could be used in
late pregnancy or immediately postpartum to identify
women who require heightened surveillance for development of PPD and facilitate early intervention.
Additional research could evaluate if these results vary
by the type, perpetrator or timing of abuse. Also it would
be interesting to examine if other psychosocial factors,
such as poor social support and marked social conflict
impact the association of abuse-PPD. Lastly, given the
rather homogeneous population in this study further
work could assess this relationship in more diverse
populations.

123

Matern Child Health J (2012) 16:929935

Acknowledgments Funding sources that support this research
include the NIH RO3 HD048865 and NIH Womens Reproductive
Health Research Program K12 HD001259.

References
1. Ross, L. E., & Dennis, C. L. (2009). The prevalence of postpartum depression among women with substance use, an abuse
history, or chronic illness: A systematic review. Journal of
Womens Health (Larchmt), 18(4), 475486.
2. Lang, A. J., Rodgers, C. S., & Lebeck, M. M. (2006). Associations between maternal childhood maltreatment and psychopathology and aggression during pregnancy and postpartum. Child
Abuse and Neglect, 30(1), 1725.
3. Nayak, M. B., & Al-Yattama, M. (1999). Assault victim history
as a factor in depression during pregnancy. Obstetrics and
Gynecology, 94(2), 204208.
4. OHara, M. W., & Swain, A. M. (1996). Rates and risk of
postpartum depressiona meta-analysis. International Review of
Psychiatry, 8, 3754.
5. Kendler, K. S., Bulik, C. M., Silberg, J., Hettema, J. M., Myers,
J., & Prescott, C. A. (2000). Childhood sexual abuse and adult
psychiatric and substance use disorders in women: An epidemiological and cotwin control analysis. Archives of General Psychiatry, 57(10), 953959.
6. Cohen, M. M., Schei, B., Ansara, D., Gallop, R., Stuckless, N., &
Stewart, D. E. (2002). A history of personal violence and postpartum depression: Is there a link? Archives of Womens Mental
Health, 4, 8392.
7. Records, K., & Rice, M. J. (2005). A comparative study of
postpartum depression in abused and non-abused women.
Archives of Psychiatric Nursing, 19(6), 281290.
8. OHara, M. W., Rehm, L. P., & Campbell, S. B. (1983). Postpartum
depression. A role for social network and life stress variables. The
Journal of Nervous and Mental Disease, 171(6), 336341.
9. Howell, E. A., Mora, P. A., DiBonaventura, M. D., & Leventhal,
H. (2009). Modifiable factors associated with changes in postpartum depressive symptoms. Archives of Womens Mental
Health, 12(2), 113120.
10. Bunevicius, R., Kusminskas, L., Bunevicius, A., Nadisauskiene,
R. J., Jureniene, K., & Pop, V. J. (2009). Psychosocial risk factors
for depression during pregnancy. Acta Obstetricia et Gynecologica Scandinavica, 88(5), 599605.
11. Weiss, E. L., Longhurst, J. G., & Mazure, C. M. (1999). Childhood sexual abuse as a risk factor for depression in women:
Psychosocial and neurobiological correlates. American Journal of
Psychiatry, 156(6), 816828.
12. McCutcheon, V. V., Heath, A. C., Nelson, E. C., Bucholz, K. K.,
Madden, P. A., & Martin, N. G. (2009). Accumulation of trauma
over time and risk for depression in a twin sample. Psychological
Medicine, 39(3), 431441.
13. Hammen, C., Kim, E. Y., Eberhart, N. K., & Brennan, P. A.
(2009). Chronic and acute stress and the prediction of major
depression in women. Depression and Anxiety, 26(8), 718723.
14. LaCoursiere, D. Y. (2010). The association between prepregnancy obesity and screening positive for postpartum depression.
British Journal of Obstetrics and Gynaecology, 117, 10111018.
15. Ahluwalia, I. B., Merritt, R., Beck, L. F., & Rogers, M. (2001).
Multiple lifestyle and psychosocial risks and delivery of small for
gestational age infants. Obstetrics and Gynaecology, 97(5 Pt 1),
649656.
16. Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of
postnatal depression. Development of the 10-item edinburgh

Matern Child Health J (2012) 16:929935

postnatal depression scale. The British Journal of Psychiatry,
150, 782786.
17. Cox, J. L., & Holden, J. M. (2003). Perinatal mental health: A
guide to the Edinburgh postnatal depression scale. London:
Gaskell, Royal College of Psychiatrists.

935
18. Schisterman, E. F., Cole, S. R., & Platt, R. W. (2009). Overadjustment bias and unnecessary adjustment in epidemiologic
studies. Epidemiology, 20(4), 488495.

123

Copyright of Maternal & Child Health Journal is the property of Springer Science & Business Media B.V. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.