Revista Brasileira de Psiquiatria.

2015;37:203210
& 2015 Associacao Brasileira de Psiquiatria
doi:10.1590/1516-4446-2014-1479

ORIGINAL ARTICLE

Childhood abuse increases the risk of depressive and

anxiety symptoms and history of suicidal behavior in
Mexican pregnant women
Ma. Asuncion Lara,1 Laura Navarrete,1 Lourdes Nieto,1 Huynh-Nhu Le2
1
Instituto Nacional de Psiquiatra Ramon de la Fuente Muniz, Ciudad de Mexico, DF, Mexico. 2Department of Psychology, George Washington
University, Washington, DC, USA.

Objective: To explore the relationship between individual and co-occurring childhood sexual, physical,
and verbal abuse, prenatal depressive (PDS) and anxiety symptoms (PAS), and history of suicidal
behavior (HSB) among Mexican pregnant women at risk of depression.
Methods: A sample of 357 women screened for PDS was interviewed using the Childhood Experience of
Care and Abuse Questionnaire (CECA-Q), the Beck Depression Inventory (BDI-II), the anxiety subscale of
the Hopkins Symptoms Checklist (SCL-90), and specific questions on verbal abuse and HSB.
Results: Logistic regression analyses showed that women who had experienced childhood sexual
abuse (CSA) were 2.60 times more likely to develop PDS, 2.58 times more likely to develop PAS, and
3.71 times more likely to have HSB. Childhood physical abuse (CPA) increased the risk of PAS
(odds ratio [OR] = 2.51) and HSB (OR = 2.62), while childhood verbal abuse (CVA) increased PDS
(OR = 1.92). Experiencing multiple abuses increased the risk of PDS (OR = 3.01), PAS (OR = 3.73),
and HSB (OR = 13.73).
Conclusions: Childhood sexual, physical, and verbal abuse, especially when they co-occur, have an
impact on PDS and PAS and lifetime HSB. These findings suggest that pregnant women at risk for
depression should also be screened for trauma as a risk factor for perinatal psychopathology.
Keywords: Child abuse; pregnancy; prenatal depression; prenatal anxiety; suicide

Introduction
It is well documented that individuals abused as children
experience a variety of psychological problems, including
depression, anxiety, and suicide attempts.1-3 Early adverse
experiences, in concert with hereditable factors, alter
brain and behavior (cognition, emotional regulation, social
interaction styles) of children at a young age.2 According to
Grassi-Oliveira et al.,4 child abuse can be perceived as an
agent for neurodevelopmental disruption and, depending
on when it occurs, may cause serious neurological scars
in certain structures, which may make some individuals
vulnerable to certain types of psychopathology.
Pregnancy can be particularly challenging for women who
were victims of childhood abuse. Pregnancy is a time of
stress requiring new or enhanced coping skills. During this
period, women experience many physical changes, such as
sensations of the fetus moving within their bodies or frequent
pelvic examinations, which may be more difficult for those
who have experienced abuse.5 Furthermore, pregnant
women are at increased risk of psychopathology, and those
with a history of abuse may be particularly vulnerable.6

Correspondence: Ma. Asuncion Lara, Ramon de la Fuente, National

Institute of Psychiatry, Calzada Mexico-Xochimilco, 101, San Lorenzo
Huipulco, 14370, Tlalpan, DF, Mexico.
E-mail: [email protected]
Submitted May 27 2014, accepted Nov 04 2014.

Research has examined the relationship between a

history of childhood abuse and psychopathology during
pregnancy. As a whole, this work has found that having a
history of childhood physical abuse (CPA) and/or sexual
abuse (CSA) is related to suicidal ideation and/or suicide
attempts during pregnancy.1,7 However, findings have
been mixed regarding the relationship between childhood
abuse and increased risk of prenatal depression or
anxiety. Leeners et al.6 conducted a systematic literature
review in 2006 and concluded that pregnant women with a
history of CSA were more likely to experience depression
and anxiety. However, subsequent studies yielded conflicting findings. While some studies found a history of
CSA and/or CPA to be associated with depression during
pregnancy,7-10 others have failed to detect a link.11,12 To
our knowledge, only one study examined the effect of a
history of childhood verbal abuse (CVA), but did not find
significant associations with depression.9 Overall, the emerging evidence suggests that, with some exceptions,11,12 a
history of abuse, particularly sexual and physical8-10 but not
CVA,9 is associated with several negative mental health
consequences in pregnant women.6,8-9
Psychopathology during pregnancy is of particular
importance because it has well-documented negative
health effects on both mother and child. Depression
during pregnancy adversely affects eating habits, which,
in turn, leads to poor prenatal care; depression in this
period also increases a womans risk for postpartum

204

MA Lara et al.

depression.13 Prenatal depression has been associated

with premature birth and low birth weight.13 Anxiety during
pregnancy is associated with prematurity,14 while suicide
attempts during pregnancy are associated with higher
rates of maternal and perinatal morbidity and, in some
cases, with perinatal mortality.15
In short, there is a need for further studies on whether
pregnant women with a history of CSA, CPA, and/or
CVA are at increased risk of psychopathology. Such
research has important implications for the health of both
women and children. Additionally, there is a dearth of
research in Latin America on the relationship between a
womans prior experience of childhood abuse and
perinatal psychopathology. We found no such studies
conducted in Latin America, despite the fact that an
estimated 12.3% of Mexican women report clinical
depression during pregnancy16 and 23.3% of Brazilian
women reported significant depressive symptoms during
this period.17
The aim of this secondary data analysis is to explore
the relationship between CPA, CSA, or CVA and prenatal
depressive symptoms (PDS), prenatal anxiety symptoms
(PAS), and a history of suicidal behavior (HSB) in a
sample of Mexican pregnant women at risk of depression.
It should be noted that suicidal behavior during pregnancy
was not our outcome variable, because women with a
suicide attempt in the past 6 months were excluded from
the primary randomized controlled trial (RCT) and
referred to treatment for ethical reasons. Nevertheless,
this sample provides an opportunity to study the effects of
the experiences of trauma in women who were screened
for depressive symptoms, as studies often find that
women with depression tend to report higher rates of
sexual abuse than their non-depressed counterparts.18
The effect of abuse is studied within the context of the
type(s) of abuse and mediating (social support, partner
relationship quality, and history of depression) and
confounding (marital status, income, and education)
factors. There is evidence that these factors are related
to an increased risk of depressive symptoms during
pregnancy.19

Methods
Sample
This study is based on secondary analysis of data
collected as part of an RCT evaluating a psychoeducational intervention to prevent postpartum depression.20 In
the primary study, a convenience sample was drawn from
pregnant women seeking attention at three health
facilities in Mexico City: 1) a hospital setting that provides
care for women with high-risk pregnancies; 2) a womens
clinic that provides obstetric and gynecological services
for women in the Armed Forces as well as for the wives of
men in the Armed Forces; and 3) a community health care
center providing prenatal care and other comprehensive
medical care for local women. The inclusion criteria were:
1) pregnant women aged over 18 years; 2) up to 26
weeks of gestation; 3) at least a minimum reading ability
(having completed primary school); 4) living in the
Rev Bras Psiquiatr. 2015;37(3)

Metropolitan area of Mexico City; 5) no indication of

bipolar disorder, substance abuse, or reported suicide
attempts during the previous 6 months; and 6) high
depressive symptomatology (score X 16 on the Center
for Epidemiological Studies Depression Scale [CES-D]21
and/or previous history of depression, but not currently in
psychological/psychiatric treatment) (Figure 1). Three
hundred and fifty-seven women completed baseline
information and were therefore included in this secondary
analysis. No additional data were collected for these
analyses. Interviews were conducted by trained psychologists from the research team in the waiting rooms of the
institutions. The original study was approved by the
Institutional Review Board of the Ramon de la Fuente
National Institute of Psychiatry.
Instruments
Sociodemographic data
The variables included in the study were: age, education,
family income, and partner status (with partner: married or
living with partner; without partner: single, divorced, or
widowed).
Abuse
CSA was measured using three questions from the
Childhood Experience of Care and Abuse Questionnaire
(CECA-Q)22: 1) When you were a child or teenager
did you ever have any unwanted sexual experiences?; 2)
Did anyone force you or persuade you to have
sexual intercourse against your wishes before age 17?;
and 3) Can you think of any upsetting sexual experiences
before age 17 with a related adult or someone in
authority, e.g., a teacher? A positive answer to any one
of these questions (yes/no) was indicative of CSA.
Severity was defined22 as a positive response to any of
the following situations: the perpetrator was a) someone
she knew, b) a relative, or c) a member of her household,
or the incident(s) described involved d) more than one
encounter, e) the touching of her private parts, f) the
participant being forced to touch the perpetrators private
parts, or g) sexual intercourse. CPA was also measured
using one of the CECA-Q questions: When you were a
child or teenager were you ever hit repeatedly with an
implement (such as a belt or a stick) or punched, kicked,
or burnt by someone in the household? A positive answer
to this question (yes/no) was indicative of CPA. CVA was
measured after abuse studies23,24 using a similar format
to CPA: When you were a child or teenager were you ever
threatened or insulted by someone in the household? and
added: Did it occur very frequently? A positive answer
(yes/no) to both questions was indicative of CVA. Similar
abuse questions were used in a previous study with
Mexican women at risk of depression, which resulted in
an efficient measure of CSA, CPA, and CVA for this
population.18
Depressive symptoms
The Beck Depression Inventory, Second Edition (BDI-II)25
evaluates depressive symptoms for the 2 weeks prior to

Childhood abuse in pregnant women

205

Figure 1 Recruitment of study participants.

its administration. It consists of 21 items scored on a

four-point scale ranging from 0 to 3. The cutoff score is
14; thus, a BDI-II 4 14 indicates a high risk of clinical
depression.25 In Mexico, one study found that the
reliability of this measure was excellent (Cronbachs
a = 0.87).26
History of depression
This aspect was assessed by the mood disorders module
of the Structured Clinical Interview (SCID)27 based on the
DSM-IV criteria.

History of suicidal behavior

This construct was measured using questions from both
Medina-Mora et al.30 and Gonzalez-Forteza et al.31 The
questions on this topic were: Have you ever been about
to take your own life?30 and Have you ever intentionally
hurt, cut, poisoned, or harmed yourself in order to take
your own life?31 The percentage of women responding
affirmatively to these questions was 24.1% and 17.1%,
respectively. HSB was defined as an affirmative answer
to either of the above questions.
Social support

Anxiety symptoms
The study used the anxiety subscale of the Hopkins
Symptoms Checklist (SCL-90) to measure anxiety symptoms.28 This 10-item scale evaluates the degree of
unease felt within the past 2 weeks on a five-point Likert
scale (none = 0; extreme = 4). Its internal consistency
in the Mexican population is moderate (Cronbachs
a = 0.80).29 A score of 4 18 is considered an indicator
of high anxiety symptomatology.28

This variable was measured using the Social Support

Apgar (SSA)32 scale, which assesses the womans perceptions of and satisfaction with various sources of
support. The scale consists of 30 questions with responses
indicating degree of satisfaction (1 = almost never; 2 =
sometimes; 3 = almost always). The Spanish version,
evaluated with Latino women who live in the United States,
demonstrated excellent internal consistency (Cronbachs
a = 0.94).33 Since the scale had not previously been used
Rev Bras Psiquiatr. 2015;37(3)

206

MA Lara et al.

in Mexico, its internal consistency was estimated for this

same sample and was also excellent (Cronbachs a = 0.90
at 6 weeks and a = 0.91 at 4-6 months postpartum). The
cutoff score o 20 reflects a low level of satisfaction with
ones social support.32
Partner relationship quality
This variable was measured using the Abbreviated
Version of Dyadic Adjustment Scale (A-DAS). This
instrument consists of seven of the original 32 items
developed by Spanier.34 The combined score ranges
from 0-30, with higher scores indicating greater levels of
marital/partner adjustment. The A-DAS has been found to
have good psychometric properties among ethnic minorities, including Latinos. Modeled after the work of
Sharpley and Cross, a cutoff point of o 15 was used to
indicate a low level of relationship quality.35
Overview of data analyses
Frequencies were calculated for categorical variables and
means, medians, and standard deviations for continuous
variables. Three logistic regression analyses were conducted to assess the impact of CSA, CPA, and CVA on each
dependent variable: PDS, PAS, and HSB. These analyses
yielded maximum likelihood estimates for combined relative
odds with 95% confidence intervals (95%CI). Additional
logistic regressions were conducted to adjust for confounding (partner status, income, and education) and mediating
(history of major depression, partner relationship quality, and
social support) factors. Age was not included in the model
because it is not a significant predictor of the dependent
variables, according to a literature review18; age was also not
significantly correlated with the abuse variables in the
present study (with CSA, r = 0.04; with CPA, r = 0.04; with
CVA, r = 0.02). Partner status, income, and education were
not significantly related to dependent variables in this study,
but were included because they have been shown to be
significant predictors in other studies.18
Before conducting the logistic regression analyses, a
correlation matrix was constructed to assess the strength
and direction of the relationship between the variables
using point-biserial correlations.
Three additional logistic regression analyses assessed
the impact of the number of abuses on psychopathology,
adjusting for the same mediating and confounding factors.
The independent variable number of abuses was constructed by adding each independent type of abuse (CSA,
CPA, CVA). Analyses were conducted using STATA
software version 11.

Table 1 Types and number of childhood abuses (n=357)

Type of abuses

n (%)

CSA
CPA
CVA

117 (32.8)
196 (54.9)
87 (24.4)

Number of abuses
One type
CSA
CPA
CVA
Two types
CSA & CPA
CSA & CVA
CPA & CVA
Three types

105 (41.7)
28 (11.1)
59 (23.5)
18 (7.1)
90 (35.7)
22 (8.7)
11 (4.4)
57 (22.6)
57 (22.6)

CPA = child physical abuse; CSA = child sexual abuse; CVA = child
verbal abuse.

severe, with women reporting at least one of the following:

the perpetrator was someone she knew (92.3%), a
relative (65.8%), lived in same household (42.7%),
abused her on more than one occasion (62.4%), touched
her private parts (74.4%), forced her to touch his (or her)
private parts (43.6%), or made her engage in sexual
intercourse (25.6%). Approximately half (54.9%) of the
women reported a history of CPA, and 24.4%, a history of
CVA. As to the number of abuses experienced, 41.7%
experienced one, 35.7% two, and 22.6% three.
Over one-third of the sample (37.5%) displayed significant depressive symptoms (BDI-II 4 14) and 14.8%
reported significant anxiety symptoms (SCL-90 4 18),
while a quarter (26.3%) reported HSB. Respondents
experienced the mediating variables as follows: 27.7%
lacked social support (APGAR o 20), 17.1% had low
partner adjustment (DAS o 15), and 36.7% had a history
of depression (SCID).
Influence of type of abuse on psychopathology
The results of the point-biserial correlations between the
variables involved are given in Table 2. The main finding
is that sociodemographic variables were not related to
abuse. Table 3 shows the results of the logistic regression
analyses of type of abuse (CSA, CPA, CVA) on each
dependent variable separately (PDS, PAS, and HSB).
The odds ratios (ORs) are reflected in column 2 for the
crude models, column 3 for the adjusted models with
mediating factors, and column 4 for the adjusted models
with possible confounders.

Results

Prenatal depressive symptoms

Mean age was 27.055.98 years, with 11.393.11 years

of education and a monthly income of $5457.964490.50
pesos (13 pesos equals approximately one U.S. dollar).
The vast majority of participants, 86.6%, lived with a
partner.
Frequency of abuse is shown in Table 1. One third of
the sample (32.8%) reported a history of CSA. Based on
the CECA-Q, 94% of the cases of CSA were classified as

The OR for CSA remained significant in the three models,

showing that women with CSA were 2.60 (crude model),
2.49 (adjusting for mediating factors), and 1.85 (adjusting
for confounding factors) more likely to have PDS than
women without a history of CSA. CPA had no effect on
PDS in any of the models, while CVA increased the risk of
PDS 1.92 times in the crude model and 2.53 times when
adjusting for mediating factors.

Rev Bras Psiquiatr. 2015;37(3)

Childhood abuse in pregnant women

Prenatal anxiety symptoms

In this model, CSA and CPA remained significant.
Women with a history of CSA had an OR of 2.58 (crude
model), 2.13 (adjusting for mediating factors) and 2.17
(adjusting for possible confounders). For those with CPA,
the risk of PAS increased 2.51, 2.44, and 2.46-fold in
each model respectively.
History of suicidal behavior
Results of the logistic regression analyses for HSB
(dependent variable) showed that CSA and CPA were
significant in all three models. CSA increased the risk of a
HSB 3.71, 3.61, and 3.68 times, respectively. CPA also

Table 2 Correlations of the abuse variables and other

variables included in the regression models (n=357)

CSA
CPA
CVA
PDS
PAS
HSB
History of depression
Poor partner relationship
Low social support
Marital status
Education
Income

CSA

CPA

CVA

1
0.17*
0.24*
0.33w
0.27w
0.27*
0.14*
0.15w
0.27w
0.05
-0.07
-0.03

1
0.41*
0.10w
0.26*
0.20*
0.12w
0.12w
0.15*
0.02
-0.03
-0.08

1
0.14*
0.21w
0.20*
0.18*
0.08
0.24w
0.02
-0.00
-0.04

CPA = child physical abuse; CSA = child sexual abuse; CVA = child
verbal abuse; HSB = history of suicidal behavior; PAS = prenatal
anxiety symptoms; PDS = prenatal depressive symptoms.
* p o 0.001; w p o 0.05.

207

increased the odds of having a HSB 2.62, 2.57 and 2.64

times, respectively. In contrast, CVA was not significantly
related to HSB.

Influence of number of abuses on psychopathology

Prenatal depressive symptoms
Table 4 shows the results of the logistic regression
analyses for number of abuses (CSA, CPA) on each
dependent variable separately. As in the previous
table, the OR is reflected in column 2 for the crude
models, column 3 for the adjusted models with mediating
factors, and column 4 for the adjusted models with
possible confounders. The OR for PDS remained significant only when both abuses co-occurred and only in the
unadjusted model and the adjusted model with possible
confounders. These models showed that women who
experienced three types of abuses were 3.01 (crude
model) times more likely to have PDS than women
without any abuse. Conversely, women who suffered one
or two abuses were no more likely to have PDS than nonabused women.
Prenatal anxiety symptoms
Although experiencing one type of abuse did not increase
the risk of having elevated PAS, women who experienced
two and three types of abuse had an increased risk of
PAS of 5.63 and 3.73, respectively, in the crude model,
with similar values for the adjusted models. The small
number of women who had experienced three abuses
and had PAS (n=5) may explain why experiencing three
types of abuse seemed to have less impact than
experiencing two.

Table 3 Odds ratios (95%CI) for type of childhood abuse and prenatal depressive and anxiety symptoms and history of suicide
behavior (n=357)
Crude model

Type of abuse

Adjusted for history of depression, poor

partner relations, and low social support

Adjusted for marital status,

income, and education

OR

95%CI

OR

95%CI

OR

95%CI

PDS (BDI-II)*
CSA
CPA
CVA

2.60y
1.10
1.92||

1.93-4.75
0.62-1.82
1.01-3.65

2.49||
0.98
2.53||

1.86-4.61
0.58-1.80
1.79-4.26

1.85||
0.95
2.50

1.38-3.69
0.56-1.61
1.79-4.04

PAS (SCL-90)w
CSA
CPA
CVA

2.58||
2.51||
1.15

1.78-4.18
1.15-5.44
0.73-2.20

2.13||
2.44||
1.39

1.56-4.64
1.12-5.31
0.65-2.00

2.17||
2.46y
1.39

1.55-5.25
1.12-5.38
0.65-2.00

HSB (ideation/attempt)=
CSA
CPA
CVA

3.71y
2.62||
1.75

1.87-7.35
1.75-4.48
0.83-2.67

3.61y
2.57||
1.70

1.81-7.20
1.77-4.39
0.80-2.70

3.68||
2.64||
1.68

1.82-7.44
1.80-4.61
0.77-2.65

95%CI = 95% confidence interval; BDI-II = Beck Depression Inventory; CPA = child physical abuse; CSA = child sexual abuse; CVA = child
verbal abuse; HSB = history of suicidal behavior; OR = odds ratio; PAS = prenatal anxiety symptoms; PDS = prenatal depressive symptoms;
SCL-90 = Hopkins Symptoms Checklist.
* Crude model: w2 = 11.11, p = 0.01; first model: w2 = 25.43, p = 0.00; second model: w2 = 26.84, p = 0.00.
w
Crude model: w2 = 11.43, p = 0.01; first model: w2 = 12.31, p = 0.04; second model: w2 = 14.03, p = 0.01.
=
Crude model: w2 = 26.95, p = 0.00; first model: w2 = 30.19, p = 0.00; second model: w2 = 36.75; p = 0.00.
y
p p 0.01; || p p 0.05.
Rev Bras Psiquiatr. 2015;37(3)

208

MA Lara et al.

Table 4 Odds ratios (95%CI) for number of childhood abuses (sexual, physical, and verbal) and prenatal depressive and
anxiety symptoms and history of suicide behavior (n=357)
Crude model

Number of abuses

Adjusted for history of depression, poor

partner relations, and low social support

Adjusted for marital status,

income, and education

OR

95%CI

OR

95%CI

OR

95%CI

PDS (BDI-II)*
One type
Two types
Three types

1.16
1.82
3.01y

0.64-2.09
0.95-2.48
1.36-6.64

1.02
1.34
1.12

0.55-1.88
0.67-1.66
0.92-2.85

1.00
1.32
1.12

0.54-1.85
0.66-1.68
0.92-1.87

PAS (SCL-90)w
One type
Two types
Three types

1.52
5.63||
3.73y

0.85-1.72
1.97-16.09
1.17-11.87

1.41
5.15y
3.28y

0.81-2.02
1.76-15.08
1.09-10.80

1.43
4.98y
3.37y

0.81-1.79
1.69-14.64
1.01-11.14

HSB (ideation/attempt)=
One type
Two types
Three types

3.89y
8.24||
13.73||

1.04-14.76
2.29-29.61
3.75-50.25

3.77y
7.43y
13.06||

1.00-14.11
2.03-27.24
3.45-49.36

1.41
7.15y
12.75||

0.90-1.9
1.93-26.43
3.35-48.47

95%CI = 95% confidence interval; BDI-II = Beck Depression Inventory; HSB = history of suicidal behavior; OR = odds ratio; PAS = prenatal
anxiety symptoms; PDS = prenatal depressive symptoms; SCL-90 = Hopkins Symptoms Checklist.
* Crude model: w2 = 9.98, p = 0.01; first model: w2 = 24.66, p = 0.00; second model: w2 = 28.80, p = 0.00.
w
Crude model: w2 = 13.32, p = 0.00; first model: w2 = 14.21, p = 0.02; second model: w2 = 15.45, p = 0.04.
=
Crude model: w2 = 25.79, p = 0.00; first model: w2 = 29.12, p = 0.00; second model: w2 = 35.30, p = 0.00.
y
p p 0.05; || p p 0.01.

History of suicidal behavior

The confluence of abuse had the greatest impact on the
HSB. The ORs for HSB were 3.89 in women who suffered
one type of abuse, 8.24 in women who suffered two types
of abuses, and 13.73 in women who suffered three types of
abuses (crude model). These results were similar when
adjusting for mediating and confounding variables.

Discussion
The purpose of this study was to examine the relationship
between childhood sexual, physical, and verbal abuse
and psychopathology (depression, anxiety, and HSB) in a
sample of Mexican pregnant women at risk for depression. To our knowledge, this was the first study in this
population to examine these relationships.
One of the main findings of this study is that a third of
the pregnant women at risk for depression (32.8%) had
suffered CSA. Disturbingly, most (94%) of them reported
experiencing severe forms of CSA. Compared to other
studies of pregnant women, the rate of CSA in this study
was within the range (7-62%) of that reported in a
systematic international literature review.6 This figure is
also similar to the CSA rate (34.6%) of Mexican nonpregnant women screened for depression,18 but much
higher than the CSA rate (7.3%) of non-pregnant, notat-risk Mexican women over 14 years of age.36
CPA was found in more than half of the sample (54.9%),
and almost one-quarter (24.4%) endorsed a history of CVA.
These rates are consistent with previous studies in the United
States (52% and 27%, respectively)9 but higher than those of
CPA found in non-pregnant Mexican women (42.2%).36
Overall, these data suggest that early trauma is at least as
frequent among pregnant Mexican women as among women
from other countries, and may be more frequent than in nonpregnant Mexican females who are not at risk of depression.
Rev Bras Psiquiatr. 2015;37(3)

As in other studies,7 we found that CSA was associated

with depressive (OR = 2.60) and anxiety symptoms (OR =
2.58). These associations remained significant even after
adjusting for mediating and confounding factors. Compared with previous research, the effect on depressive
symptoms appears to be higher in this sample than that
reported by Rich-Edwards et al.10 (OR = 1.23) and Chung
et al.9 (OR = 1.94), possibly because of differences in the
risk of depression and social disadvantages, as described
below. On the other hand, CSA had the greatest impact on
HSB (OR = 3.71). Farber et al.1 also found that a history of
CSA was related to suicidal ideation during pregnancy and
to suicide attempts before pregnancy in a sample of
psychiatric obstetric patients.
Consistent with other studies, we failed to find an
association between CPA and depressive symptoms,11,12
but found that women who suffered from CPA were at
an increased risk of developing anxiety symptoms
(OR = 2.5), a variable not often examined in previous
research.11 CPA also increased the risk of HSB
(OR = 2.62), which persisted after controlling for mediating and confounding factors. This finding is contrary to a
study conducted by Farber et al.,1 who did not find a
significant association between physical abuse alone and
a history of suicide attempts. This could be due to the fact
that a smaller proportion in their sample (8.7%) reported
CPA compared to our sample (54.9%).
CVA has been the least studied of the three types of
abuse in perinatal women. Our results showed that its
only effect was on PDS (OR = 1.92), contrary to Chung
et al.,9 who did not find an effect of CVA on such
symptoms. In retrospect, these findings are not surprising, given that of the three types of abuse, CVA may be
considered the least severe.
Those who reported three types of abuse were more
likely to experience PDS (OR = 3.01) than those who
reported one or two types. Having suffered two or three

Childhood abuse in pregnant women

types of abuse (OR = 5.63 and 3.73) increased the risk of

PAS more than suffering a single type. Experiencing three
types of abuse resulted in an even greater impact on HSB
than each individual type of abuse, increasing the risk to
13.73 from 3.89 (when reporting one) and 8.24 (when
reporting two). Similarly, studies conducted by Holzman et
al.8 and Chung et al.9 found a positive link between a
background involving multiple types of abuse and the risk of
depressive symptoms. The confluence of the different types
of abuse may result in a qualitatively different experience
and impact than that derived from one type of abuse.
In summary, CSA, CPA, and CVA, especially when they
co-occurred, had a significant impact on PDS, PAS, and
lifetime HSB in pregnant Mexican women.37 In particular,
women who experienced multiple types of abuse were
more likely to report HSB. In light of this finding, it is striking
that so few studies have examined HSB in pregnant
women, with even fewer considering childhood abuse as a
risk factor. A history of suicide attempts has been
associated with significant depressive symptoms in Latino
women living in the United States and in pregnant Mexican
women.38 A possible explanation for this relationship is
that early traumatic life experiences (such as abuse)
subsequently affect how the brain responds to stressors
throughout adult life, influencing the aging process and
increasing susceptibility to the psychopathology, including
mood and anxiety disorders.2 This connection occurs in
part because the nervous system regulates and responds
to systemic processes via the neuroendocrine, autonomic,
and immune systems. Social factors have also been shown
to have a powerful impact on brain development, structure,
and function throughout the life course, thereby also
affecting mental health outcomes.39
From a cultural perspective, the findings of this study
decrease the likelihood of significant differences between
the impact of abuse on pregnant women living in Mexico
and those living in other countries.8-10 Nevertheless,
cultural differences may be significant in other areas. For
example, there are marked challenges in Mexico regarding access to suitable mental health services.40 From a
healthcare perspective, these findings are significant
because prenatal depression and anxiety have been
shown to negatively affect both mother and infant health
in the United States and Europe.13,14 Moreover, it has
been documented that CSA and CPA are risk factors for
later partner violence during pregnancy.37 Given the
health consequences, there is a need for additional
studies on psychopathology in pregnant women with a
history of childhood abuse, particularly focusing on
Latin America, to confirm the extent to which cultural
differences should be considered.
Several limitations in our study should be taken into
account. First, these results were based on secondary data
analyses of a sample of pregnant women who participated in
an RCT to evaluate an intervention to prevent postpartum
depression.20 These participants were included in the study
because they met the criteria for being at high risk for
depression. This risk profile may have increased the rates of
depressive and anxiety symptoms and HSB among this
sample, thereby potentially overestimating the link between

209

childhood abuse and mental health outcomes. This same

limitation was cited in other studies of childhood abuse and
psychopathology involving clinical or high-risk samples.41
However, for ethical reasons, pregnant women with a suicide
attempt in the 6 months prior to the baseline interview and
with other severe pathologies (major depression, bipolar
disorder, substance abuse, or currently in psychological/
psychiatric treatment) were barred from participation in the
RCT. Excluding these cases may have decreased the rates
of depressive symptoms, anxiety symptoms, and HSB in
comparison to a sample including the more severe cases.
Nevertheless, there is evidence that the frequency of actual
suicidal behavior in pregnant Mexican women is extremely
low (2.6 and 0.9%).42 Second, our main outcome variables
(PDS, PAS, and HSB) were measured concurrently with
childhood abuse. Therefore, it is likely that women experiencing these high levels of emotional distress are more
sensitive to recalling adversity. Third, the fact that questionnaires were administered in person as opposed to in a
written format may have influenced womens responses,
since most people have difficulty talking about their childhood abuse without first establishing proper rapport with the
interviewer.43 We attempted to compensate for this limitation
by using psychologists who were highly trained to create an
emotionally safe environment for participants as interviewers
in the study. Finally, it is important to keep in mind the
limitations of data collected through retrospective childhood
trauma instruments.44 As previously noted, the description of
events may be biased, affecting estimates of abuse
prevalence and their link with clinical variables. Recollection
may also vary at different developmental stages: very early
trauma events (those occurring prior to 6 years of age) may
not be recalled as frequently or accurately as those occurring
at later stages, leading to nonsystematic underreporting and
a greater possibility of distortion.
The results of this study add to the small but growing
body of research on perinatal psychopathology in women
in developing countries, particularly Latin America.
Further research is required to determine the scope of
these findings in the region, particularly in regard to the
effects of childhood maltreatment on suicidality before
and during pregnancy. The fact that a high number of
pregnant women reported untreated symptoms of depression and anxiety, a history of suicidality, and sexual and
physical abuse suggests the need for more comprehensive screening in prenatal care. Such screening, coupled
with feasible, cost-effective interventions which have yet
to be developed would help reduce negative consequences for the mother and their effect on the child.
Educating perinatal women and health care providers
about major risk factors for depression, anxiety, and
suicidality during pregnancy is an essential starting point
to promote the mental health of mothers and infants.

Acknowledgements
This study was supported by Consejo Nacional de
Ciencia y Tecnologa (CONACyT, Salud-2003-C01.021).
We are grateful to Centro de Salud Jose Castro
Villagrana, Hospital de Especialidades de la Mujer de la

Rev Bras Psiquiatr. 2015;37(3)

210

MA Lara et al.

SEDENA, and Instituto Nacional de Perinatologa Isidro

Espinosa de los Reyes. Thanks are also due to research
team members Cecilia Mattei, Teresa Garca Hubard,
Claudia Navarro, Gabriela Letechipa, Erika Trujillo,
Yazmn Quintero, Sofa Arce, Alejandra Guerrero, Araceli
Aguilar, and Brenda Jimenez; and to Nicole DeSario
Hedrick for her assistance in manuscript edits.

Disclosure
The authors report no conflicts of interest.

References
1 Farber EW, Herbert SE, Reviere SL. Childhood abuse and suicidality
in obstetrics patients in an hospital-based urban prenatal clinic. Gen
Hosp Psychiatry. 1996;18:56-60.
2 Hassel S, McKinnon MC, Cusi AM, MacQueen GM. An overview of
psychological and neurobiological mechanisms by which early
negative experiences increase risk of mood disorders. J Can Acad
Child Adolesc Psychiatry. 2011;20:277-88.
3 Carr CP, Martins CM, Stingel AM, Lemgruber VB, Juruena MF. The
role of early life stress in adult psychiatric disorders: a systematic
review according to childhood trauma subtypes. J Nerv Ment Dis.
2013;201:1007-20.
4 Grassi-Oliveira R, Ashy M, Stein LM. Psychobiology of childhood
maltreatment: effects of allostatic load? Rev Bras Psiquiatr.
2008;30:60-8.
5 Lang AJ, Rodgers CS, Lebeck MM. Associations between maternal
childhood maltreatment and psychopathology and aggression during
pregnancy and postpartum. Child Abuse Negl. 2006;30:17-25.
6 Leeners B, Richter-Appelt H, Imthurn B, Rath W. Influence of childhood sexual abuse on pregnancy, delivery, and the early postpartum
period in adult women. J Psychosom Res. 2006;61:139-51.
7 Leeners B, Rath W, Block E, Gorres G, Tschudin S. Risk factors for
unfavorable pregnancy outcome in women with adverse childhood
experiences. J Perinat Med. 2014;42:171-8.
8 Holzman C, Eyster J, Tiedje LB, Roman LA, Seagull E, Rahbar MH. A
life course perspective on depressive symptoms in mid-pregnancy.
Matern Child Health J. 2006;10:127-38.
9 Chung EK, Mathew L, Elo IT, Coyne JC, Culhane JF. Depressive
symptoms in disadvantaged women receiving prenatal care: the
influence of adverse and positive childhood experiences. Ambul
Pediatr. 2008;8:109-16.
10 Rich-Edwards JW, James-Todd T, Mohllajee A, Kleinman K, Burke A,
Gillman MW, et al. Lifetime maternal experiences of abuse and risk of
pre-natal depression in two demographically distinct populations in
Boston. Int J Epidemiol. 2011;40:375-84.
11 Edwards B, Galletly C, Semmler-Booth T, Dekker G. Antenatal psychosocial risk factors and depression among women living in socioeconomically disadvantaged suburbs in Adelaide, South Australia.
Aust N Z J Psychiatry. 2008;42:45-50.
12 Dayan J, Creveuil C, Dreyfus M, Herlicoviez M, Baleyte JM, OKeane V.
Developmental model of depression applied to prenatal depression:
role of present and past life events, past emotional disorders and
pregnancy stress. PLoS One. 2010;5:e12942.
13 Bennett HA, Einarson A, Taddio A, Koren G, Einarson TR. Prevalence of depression during pregnancy: systematic review. Obstet
Gynecol. 2004;103:698-709.
14 Ross LE, McLean LM. Anxiety disorders during pregnancy and the
postpartum period: A systematic review. J Clin Psychiatry.
2006;67:1285-98.
15 Gandhi SG, Gilbert WM, McElvy SS, Kady D, Danielson B, Xing G,
et al. Maternal and neonatal outcomes after attempted suicide.
Obstet Gynecol. 2006;107:984-90.
16 Ocampo R, Heinze G, Ontiveros MP. Deteccion de depresion postparto en el Instituto Nacional de Perinatologa. Psiquiatra. 2007;23:
18-22.
17 Melo EF Jr, Cecatti JG, Pacagnella RC, Leite DF, Vulcani DE, Makuch
MY. The prevalence of perinatal depression and its associated factors
in two different settings in Brazil. J Affect Disord. 2012;136:1204-8.

Rev Bras Psiquiatr. 2015;37(3)

18 Navarrete L, Navarro C, Lara MA. Abuso sexual en la infancia en

mujeres con sintomatologa depresiva. Psiquiatra. 2007;23:8-12.
19 Lancaster CA, Gold KJ, Flynn HA, Yoo H, Marcus SM, Davis MM.
Risk factors for depressive symptoms during pregnancy: a systematic
review. Am J Obstet Gynecol. 2010;202:5-14.
20 Lara MA, Navarro C, Navarrete L. Outcome results of a psychoeducational intervention in pregnancy to prevent PPD: a randomized
control trial. J Affect Disord. 2010;122:109-17.
21 Radloff LS. The CES-D scale: a self-report depression scale for research
in the general population. Appl Psychol Meas. 1977;1:385-401.
22 Bifulco A, Bernazzani O, Moran PM, Jacobs C. The childhood
experience of care and abuse questionnaire (CECA.Q): validation in
a community series. Br J Clin Psychol. 2005;44:563-81.
23 Bellis MA, Hughes K, Leckenby N, Perkins C, Lowey H. National
household survey of adverse childhood experiences and their relationship with resilience to health-harming behaviors in England. BMC
Med. 2014;12:72.
24 Yeoman K, Safranek T, Buss B, Cadwell BL, Mannino D. Adverse
childhood experiences and adult smoking, Nebraska, 2011. Prev
Chronic Dis. 2013;10:E159.
25 Beck A, Steer R, Brown G. Manual for Beck Depression Inventory.
San Antonio: Psychological Corporation; 1996.
26 Jurado S, Villegas ME, Mendez L, Rodrguez F, Loperena V, Varela R.
La estandarizacion del inventario de depresion de Beck para los
residentes de la ciudad de Mexico. Salud Ment. 1998;21:26-31.
27 First MB, Spitzer RL, Gibbon M, Williams JB. Structured Clinical
Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV).
Washington: American Psychiatric Press; 1996.
28 Derogatis LR. Administration, Scoring and Procedures Manual-I for
the Revised Version SCL-90R. Baltimore: Johns Hopkins; 1977.
29 Lara MC, Espinosa SI, Cardenas ML, Focil M, Cavazos J. Confiabilidad y validez de la SCL-90 en la evaluacion de psicopatologa
en mujeres. Salud Ment. 2005;28:42-50.
30 Medina-Mora ME, Rascon ML, Tapia R, Marino MC, Juarez F,
Villatoro J, et al. Trastornos emocionales en poblacion urbana mexicana: resultados de un estudio nacional. An Inst Mex Psiquiatr.
1992;3:48-55.
31 Gonzalez C, Villatoro J, Alcantar I, Medina ME, Fleiz C, Bermudez P,
et al. Prevalencia de intento suicida en estudiantes adolescentes de
la ciudad de Mexico: 1997 y 2000. Salud Ment. 2002;25:1-12.
32 Norwood SL. The Social Support Apgar: instrument development and
testing. Res Nurs Health. 1996;19:143-52.
33 Diaz MA, Le HN, Cooper BA, Munoz RF. Interpersonal factors and
perinatal depressive symptomatology in a low-income Latina sample.
Cultur Divers Ethnic Minor Psychol. 2007;13:328-36.
34 Spanier GB. Measuring dyadic adjustment: new scales of assessing
the quality of marriage and similar dyads. J Marriage Fam.
1976;38:15-28.
35 Sharpley CF, Cross DG. A psychometric evaluation of the Spanier
Dyadic Adjustment Scale. J Marriage Fam. 1982;44:739-42.
36 Rivera-Rivera L, Allen B, Chavez-Ayala R, Avila-Burgos L. Abuso fsico
y sexual durante la ninez y revictimizacion de las mujeres mexicanas
durante la edad adulta. Salud Publica Mex. 2006;48:268-78.
37 Cuevas S, Blanco J, Juarez C, Palma O, Valdez-Santiago R.
Violencia y embarazo en usuarios del sector salud en estados de alta
marginacion en Mexico. Salud Publica Mex. 2006;48:239-49.
38 Lara MA, Le HN, Letechipia G, Hochhausen L. Prenatal depression in
Latinas in the U.S. and Mexico. Matern Child Health J. 2009;13:567-76.
39 McEwen BS. Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev. 2007;87:873-904.
40 Borges G, Medina-Mora ME, Wang PS, Lara C, Berglund P, Walters
E. Treatment and adequacy of treatment for mental disorder among
respondents to the Mexico National Comorbidity Survey.
Am J Psychiatry. 2006;163:1371-8.
41 Lacelle C, Hebert M, Lavoie F, Vitaro F, Tremblay RE. Sexual health
in women reporting a history of child sexual abuse. Child Abuse Negl.
2012;36:247-59.
42 Lara MA, Letechipia G. Ideacion y comportamiento suicida en
embarazadas. Salud Ment. 2009;32:381-7.
43 Goldman JD, Padayachi UK. Some methodological problems in
estimating incidence and prevalence in child sexual abuse research.
J Sex Res. 2000;37:305-14.
44 Roy CA, Perry JC. Instruments for the assessment of childhood
trauma in adults. J Nerv Ment Dis. 2004;192:343-51.