Posttraumatic
Content
Original Research
Posttraumatic Stress Disorder and Risk of
Spontaneous Preterm Birth
Jonathan G. Shaw, MD, MS, Steven M. Asch, MD, MPH, Rachel Kimerling, PhD, Susan M. Frayne, MD, MPH,
Kate A. Shaw, MD, MS, and Ciaran S. Phibbs, PhD
OBJECTIVE: To evaluate the association between antenatal posttraumatic stress disorder (PTSD) and spontaneous preterm delivery.
METHODS: We identified antenatal PTSD status and
spontaneous preterm delivery in a retrospective cohort of
16,334 deliveries covered by the Veterans Health Administration from 2000 to 2012. We divided mothers with
PTSD into those with diagnoses present the year before
delivery (active PTSD) and those only with earlier diagnoses
(historical PTSD). We identified spontaneous preterm birth
and potential confounders including age, race, military
deployment, twins, hypertension, substance use, depression, and results of military sexual trauma screening and
then performed multivariate regression to estimate
adjusted odds ratio (OR) of spontaneous preterm delivery
as a function of PTSD status.
From the Center for Innovation to Implementation, the National Center for
Posttraumatic Stress Disorder, the Women’s Health Section, and the Health Economics Resource Center, Department of Veterans Affairs, Palo Alto Health Care System,
Palo Alto, and the March of Dimes Center for Prematurity Research at Stanford, the
Centers for Health Policy/Primary Care & Outcomes Research, the Division of General Medical Disciplines, and the Departments of Obstetrics & Gynecology and Pediatrics, Stanford University School of Medicine, Stanford, California.
Dr. Shaw was supported in part by the VA Office of Academic Affairs and
Health Services Research & Development funds. Development of the Women’s
Health Evaluation Initiative (WHEI) data used in this study was supported by
VA Women’s Health Services.
The authors thank Laurie Zephyrin, MD, MPH, MBA, of Women’s Health
Services in the VA Central Office, and Rita Popat, PhD, of Stanford University
for thoughtful review of the manuscript, and VA Palo Alto Staff, Lakshmi
Ananth, MS, Vidhya Balasubramanian, MS, Eric Berg, MS, Sarah Friedman,
MSPH, Fay Saechao, MPH, Meghan Saweikis, MS, JD, and Susan Schmitt,
PhD, for invaluable consultation and technical advice.
The findings and conclusions in this report are those of the authors and do not
necessarily represent the official position of the Department of Veterans Affairs.
Corresponding author: Jonathan G. Shaw, MD, MS, 117 Encina Commons,
Room 206, Stanford, CA 94305; e-mail: [email protected].
Financial Disclosure
The authors did not report any potential conflicts of interest.
© 2014 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/14
VOL. 124, NO. 6, DECEMBER 2014
RESULTS: Of 16,334 births, 3,049 (19%) were to mothers
with PTSD diagnoses, of whom 1,921 (12%) had active
PTSD. Spontaneous preterm delivery was higher in those
with active PTSD (9.2%, n5176) than those with historical
(8.0%, n590) or no PTSD (7.4%, n5982) before adjustment
(P5.02). The association between PTSD and preterm birth
persisted, when adjusting for covariates, only in those with
active PTSD (adjusted OR 1.35, 95% confidence interval
[CI] 1.14–1.61). Analyses adjusting for comorbid psychiatric
and medical diagnoses revealed the association with active
PTSD to be robust.
CONCLUSION: In this cohort, containing an unprecedented number of PTSD-affected pregnancies, mothers with
active PTSD were significantly more likely to suffer spontaneous preterm birth with an attributable two excess preterm
births per 100 deliveries (95% CI 1–4). Posttraumatic stress
disorder’s health effects may extend, through birth outcomes, into the next generation.
(Obstet Gynecol 2014;124:1111–9)
DOI: 10.1097/AOG.0000000000000542
LEVEL OF EVIDENCE: II
P
reterm birth is a leading cause of infant morbidity
and mortality.1 In the United States 12% of deliveries are preterm, and roughly half are spontaneous as
opposed to medically indicated.2 Although certain risk
factors for spontaneous preterm birth have been clearly
identified, including demographic characteristics, substance use, and multiple gestations, the etiology remains
poorly understood; efforts to reduce the preterm birth
rate have made little progress over the past two decades.1–4 A growing number of studies suggest a role for
psychosocial factors such as maternal stress5–7 and
depression.8–10 However, the effect of posttraumatic
stress disorder (PTSD) remains unclear.
Posttraumatic stress disorder is a complex of
disruptive symptoms arising from a traumatic experience (eg, violence, disaster). Its prevalence varies
substantially between populations based on exposure.11 U.S. surveys show women affected at higher
OBSTETRICS & GYNECOLOGY
1111
rates than men, and estimate lifetime prevalence of
PTSD among females to be 10–14%.12,13
Posttraumatic stress disorder could affect preterm
delivery directly through biological pathways or
indirectly through risky health behaviors and poor
self-care (eg, attendance to medical care, nutrition,
and sleep). Previous studies have been limited by
inadequate sample size, heterogeneity in diagnostic
criteria, and generalizability concerns.14–23
We examined whether PTSD is associated with
spontaneous preterm birth by evaluating deliveries
covered by the Veterans Health Administration. With
standardized mandatory PTSD screening,24 a high
PTSD prevalence (13–21% in reproductive-aged
women),25,26 and centralized national data, the Veterans Health Administration provides an ideal setting in
which to examine the association between PTSD and
preterm birth in a cohort of unprecedented size.
MATERIALS AND METHODS
Using data from national clinical and administrative
databases for Veterans Health Administration-based
(nonobstetric) care and outsourced (obstetric) Veterans Health Administration care, we conducted a retrospective cohort analysis of all Veterans Health
Administration-reimbursed deliveries in fiscal years
2000–2012 and evaluated associations between antepartum PTSD and spontaneous preterm delivery:
16,477 deliveries were identified using a validated
algorithm.27 Our main Veterans Health Administration data set is derived from standardized hospital
discharge abstracts, entered by professional coders,
and diagnosis data from such discharge abstracts have
previously been shown to be reliably coded.28 We
excluded 102 women without Veterans Health
Administration encounters before delivery (and thus
without antecedent Veterans Health Administration
PTSD screening or clinical data) and 41 additional
deliveries resulting from irreconcilable data.
Our primary outcome was spontaneous preterm
birth. Spontaneous cases were deemed most relevant to
evaluating the direct contribution of PTSD to premature
delivery. Cases were identified by International Classification of Diseases, 9th Revision, Clinical Modification
(ICD-9-CM) diagnosis code 644.2 “spontaneous onset of
delivery before 37 weeks” in the delivery claim, which in
external validation corresponds to a median gestational
age of 35 weeks (see Appendix 1, available online at
http://links.lww.com/AOG/A571, for a description of
validation).
The Veterans Health Administration routinely
screens for PTSD within its primary care system,29
and prior Veterans Health Administration studies
1112
Shaw et al
confirm the reliability of ICD-9-CM 309.81 to identify
PTSD.30,31 We used Veterans Health Administration
encounters from 1997 to 2012 (ensuring a minimum
3-year look-back for our cohort, which begins with
deliveries in 2000) to identify PTSD in any encounter
before delivery. For each delivery, we further distinguished two mutually exclusive categories of PTSD
status in pregnancy: “active PTSD” and “historical
PTSD.” This distinction was invoked based on prior
research that observed poorer birth outcomes in
women with current PTSD symptoms during pregnancy, not past (recovered) PTSD.32 We categorized
as “active PTSD” those cases in which a PTSD diagnosis was documented in any encounter(s) within 365
days before the day of delivery, presuming these represented pregnancies with clinically relevant PTSD symptoms in the prenatal period; PTSD cases not meeting
this criterion were labeled “historical PTSD.” In choosing a 1-year window to define active PTSD, we aimed to
maximize overlap with the pregnancy and ensure likelihood of a Veterans Health Administration encounter for
all participants in that time period. (A more restrictive
9-month window was explored in a sensitivity analysis.)
By our inclusion criteria, all women had Veterans
Health Administration encounters before delivery, and
more than 99% had an encounter within 1 year before
delivery and thus the opportunity for active PTSD to be
documented.
We collected data on two potential trauma exposures, military sexual trauma and recent military deployment, both associated with high rates of PTSD within the
Veterans Health Administration.25,33 Mandatory onetime screening for military sexual trauma has been in
place since 2002 through a brief validated instrument33
(see Appendix 1, http://links.lww.com/AOG/A571, for
screening questions). We used the Department of
Defense Roster34 to identify veterans with recent deployment in support of operations in Afghanistan or Iraq.
We obtained demographic covariates from Veterans Health Administration administrative data, including maternal age, race, and marital status. We identified
obstetric risk factors including twins or higher-order
gestations and prior cesarean delivery from obstetric
hospitalizations using validated ICD-9-CM codes.35,36
We extracted medical comorbidities from previously
developed Veterans Health Administration chronic
condition data sets.37 Specifically, we recorded those
conditions that were both prevalent in our cohort
(2% or more) and have been suggested as risk factors
for preterm birth3: hypertension, diabetes, and asthma.
To identify mental health diagnoses other than
PTSD, we used a modified version38 of the Agency for
Healthcare Research and Quality Clinical Classifications
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
Software’s categories of mental illness.39 Applying the
same criteria as for PTSD, we created three-level indicator variables (active, historical, none) for the most
prevalent disorders. We identified active drug and alcohol abuse and tobacco dependence in the antenatal
period by the presence of diagnostic codes in Veterans
Health Administration encounters within 1 year before
delivery or within reimbursed prenatal encounters
(where screening for substances should be standard of
care) to maximize sensitivity.
Delivery was our unit of analysis and we used
statistical methods accounting for repeated measures
adjusting effect estimates to take into account a withinperson correlation for women with repeat deliveries in
our cohort. We performed unadjusted x2 bivariate analysis, then performed adjusted multivariate logistic
regression (SAS 9.2) using generalized estimating equations modeling (clustered by individual) to determine
adjusted odds ratio (OR) of spontaneous preterm delivery by PTSD status as a three-level predictor comparing
“active PTSD” and “historical PTSD” with none.
We performed regressions in an iterative (additive) fashion, exploring demographic and obstetric
factors, and potential trauma (military sexual trauma,
deployment) as possible confounders. For parsimony, covariates were only retained in subsequent
models if they altered the b coefficient of the primary
predictor, active PTSD, by 10% or more. All models
were adjusted for multiple gestations given their
undisputed association with spontaneous preterm
birth. Our resulting primary model adjusted for
age, race, multiple gestation, and deployment history. We examined interaction terms for PTSD*military sexual trauma and PTSD*deployment history.
We explored three possible explanatory pathways:
preselected chronic comorbidities (hypertension,
diabetes, asthma), behavioral risks (drug, alcohol,
or tobacco use), and other mental health disorders
frequently codiagnosed with PTSD (those disorders
with prevalence 2% or more in our cohort). We did not
include these as confounders in our primary analysis
because they could arguably represent intermediate
steps in the causal pathway (ie, if the pathophysiology
of chronic PTSD predisposes women to the selected
comorbidities). In evaluating the potential explanatory
role of codiagnosed mental health conditions, we
placed these variables in regression models, both
alongside PTSD and in place of PTSD. Lastly, we
tested our model assumptions in a series of sensitivity
analyses, including models restricted to first deliveries or singleton deliveries, models adjusted for
obstetric history (eg, prior spontaneous preterm
birth, excluding multiples), and models applying
VOL. 124, NO. 6, DECEMBER 2014
more restrictive timeframes (see Appendix 2, available online at http://links.lww.com/AOG/A571, for
a detailed description).
This research was approved by Stanford University’s institutional review board as part of the Women’s Health Evaluation Initiative.
RESULTS
The cohort included 16,334 deliveries among 14,047
women. There were 1,248 (7.6%) spontaneous preterm deliveries. Of the 16,334 deliveries, 3,049 (19%)
were to women who carried an antepartum diagnosis
of PTSD, two thirds of whom (1,921 [12%]) had active
PTSD. Nearly one third of the deliveries (4,948) were
to women with recent deployment (Afghanistan or
Iraq); 3,568 (23%) deliveries were to women reporting
a history of military sexual trauma.
Table 1 presents descriptive characteristics by
PTSD status. Those with active PTSD were significantly more likely to have been deployed (45%)
than were those with historical (32%) or no PTSD
(28%). Those with active and historical PTSD
were more likely to report military sexual trauma
(57% and 46%) than those without PTSD (16%) and
carried a significantly higher burden of active
comorbid mental health, drug-, and alcoholrelated diagnoses (P,.001 for all). The unadjusted
proportion of spontaneous preterm birth (Table 1)
was significantly higher in those with active PTSD
(9.2%) than those with historical (8.0%) or no PTSD
(7.4%) (P5.02).
In our primary adjusted model (Table 2), active
PTSD remained associated with spontaneous preterm
delivery (adjusted OR 1.35 [1.14–1.61]), whereas historical PTSD’s association was nonsignificant
(adjusted OR 1.06 [0.84–1.34]).
Although deployment history was associated with
a lower risk of spontaneous preterm birth, adjusted OR
0.71 (0.61–0.81)—consistent with our expectation that
the selection process for deployment results in a subpopulation of healthier, lower-risk individuals (the “healthy
warrior effect”)40—we found no statistical support for our
a priori hypothesis that deployment modifies the effect
of PTSD (P5.42 for interaction term active PTSD*deployment added to our base model). In contrast, military
sexual trauma showed no association with spontaneous
preterm birth (unadjusted OR 1.08 [0.94–1.24]; adjusted
OR 0.99 [0.84–1.15]; see Appendix 3, available online
at http://links.lww.com/AOG/A571, Table A1).
However, as shown in Table 3, when the interaction
of military sexual trauma and PTSD was explored, we
observed that those with both active PTSD and military sexual trauma carried the greatest, and most
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1113
Table 1. Characteristics of Deliveries Covered by the Veterans Administration (2000–2012), by
Posttraumatic Stress Disorder Status
Deliveries by PTSD Status
Characteristic
Spontaneous preterm birth
Demographics
Maternal age (y)
19–24
25–29
30–34
35–39
40–48
Race
White
African American or black
Asian
Native Hawaiian or other Pacific
Islander
American Indian or Alaskan Native
Missing or declined to answer
Married†
Potential trauma exposure
Military sexual trauma‡
Previously deployed (U.S. Operations
Enduring Freedom, Iraqi
Freedom, and New Dawn)
Obstetric history
Twins or higher-order gestation
Prior cesarean delivery
Parity§
1
2
3 or more (maximum 5)
Prior spontaneous preterm delivery§
Chronic medical conditions (within 3 y
antepartum)
Hypertension
Diabetes
Asthma
Substance abuse or dependence diagnoses
(within 1 y antepartum)
Drug
Alcohol
Tobacco
Active mental health comorbidities (within
1 y antepartum)
Depressive disorder
Anxiety disorder (other than PTSD)
Adjustment disorder
Bipolar disorder
Personality disorder
Active PTSD Diagnosis
(Within 365 Days
Antepartum) (n51,921)
Historical PTSD Diagnosis
(More Than 365 Days
Antepartum) (n51,128)
176 (9.2)
90 (8.0)
None
(n513,285)
982 (7.4)
P*
.02
,.001
273
787
515
269
77
(14.2)
(41.0)
(26.8)
(14.0)
(4.0)
65
429
375
198
61
(5.8)
(38.0)
(33.2)
(17.6)
(5.4)
2,290
5,257
3,689
1,614
435
(17.2)
(40.0)
(27.8)
(12.2)
(3.3)
1,344
371
26
36
(70.0)
(19.3)
(1.4)
(1.9)
776
243
11
9
(68.8)
(21.5)
(1.0)
(0.8)
8,142
3,059
219
188
(61.3)
(23.0)
(1.7)
(1.4)
,.001
20 (1.0)
124 (6.5)
824 (43.2)
8 (0.7)
81 (7.2)
475 (43.0)
102 (0.8)
1,575 (11.9)
6,187 (47.5) ,.001
1,085 (57.7)
863 (44.9)
513 (46.2)
361 (32.0)
1,970 (15.7) ,.001
3,760 (28.3) ,.001
41 (2.1)
275 (14.3)
31 (2.8)
195 (17.3)
276 (2.1)
1,847 (13.9)
1,677
220
24
30
(87.3)
(11.5)
(1.3)
(1.6)
863
223
42
25
(76.5)
(19.8)
(3.7)
(2.2)
11,494
1,574
217
132
.3
.007
,.001
(86.5)
(11.9)
(1.6)
(1.0) ,.001
169 (8.8)
90 (4.7)
206 (10.7)
91 (8.1)
36 (3.2)
138 (12.2)
835 (6.3)
414 (3.1)
990 (7.5)
,.001
.002
,.001
217 (11.3)
165 (8.6)
351 (18.3)
50 (4.4)
26 (2.3)
143 (12.7)
387 (2.9)
175 (1.3)
1,219 (9.2)
,.001
,.001
,.001
327
148
45
62
29
2,089
1,019
553
342
178
,.001
,.001
,.001
,.001
,.001
1,190
552
188
181
172
(62.0)
(28.7)
(9.8)
(9.4)
(9.0)
(29.0)
(13.1)
(4.0)
(5.5)
(2.6)
(15.7)
(7.7)
(4.2)
(2.6)
(1.3)
PTSD, posttraumatic stress disorder.
Data are n (%) unless otherwise specified.
* x2.
†
Missing for 307 (2%).
‡
Missing for 814 (5%).
§
Estimate; data unavailable for deliveries before 2000 and non–Veterans Health Administration deliveries.
1114
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
Table 2. Association of Posttraumatic Stress Disorder Status and Spontaneous Preterm Birth: Unadjusted
and Primary Adjusted Model (N516,334; Preterm Delivery Events51,248)
Unadjusted Model
n
OR
95% CI
P
aOR
95% CI
P
13,285
1,128
1,921
1
1.09
1.26
Reference
0.87–1.36
1.07–1.50
0.5
0.006
1
1.06
1.35
Reference
0.84–1.34
1.14–1.61
.6
,.001
2,628
6,473
4,579
2,081
573
1
0.99
1.15
1.19
1.21
Reference
0.83–1.19
0.96–1.40
0.95–1.49
0.87–1.69
.9
.1
.1
.3
10,262
3,673
256
187
130
1,780
348
4,984
1
1.49
1.27
1.99
1.35
1.10
7.15
0.71
Reference
1.29–1.71
0.82–1.96
1.15–3.45
0.85–2.13
0.90–1.35
5.66–9.03
0.61–0.81
,.001
.3
.01
.2
.3
,.001
,.001
Parameter
PTSD
None
Historical (more than 365 d antepartum)
Active (within 365 d antepartum)
Maternal age (y)
19–25
25–29
30–34
35–39
40 or older
Race
White
Black or African American
Asian
American Indian or Alaskan Native
Hawaiian or Pacific Islander
Missing or not reported
Twins or higher-order gestation
Deployed (U.S. Operations Enduring Freedom, Iraqi
Freedom, and New Dawn)
Multivariate Model*
OR, odds ratio; CI, confidence interval; aOR, adjusted odds ratio; PTSD, posttraumatic stress disorder.
* Generalized estimating equations with logit linkage and clustered by unique individual using exchangeable correlation matrix for repeat
deliveries. Independent variables are those listed. Model fit: C-statistic (in logistic regression model without clustering)50.62 in adjusted
model as compared with 0.52 in unadjusted model; 58,536.6 in the adjusted model as compared with 8,816.0 in the unadjusted model.
significant, risk (adjusted OR 1.43 [1.15–1.77]),
although the interaction term was not statistically significant (P5.25 for interaction term active PTSD*military sexual trauma).
Table 3. Interaction of Posttraumatic Stress
Disorder Status and Military Sexual
Trauma History as Predictors of
Spontaneous Preterm Birth*
Multilevel Variable
Combining PTSD
and Military Sexual
Trauma Status
PTSD
Active
Active
Historical
Historical
None
None
Military
Sexual
Trauma
n
OR
+
2
+
2
+
2
1,085
797
598
513
1,970
10,557
1.43
1.19
0.93
1.16
0.96
1
95% CI
P
1.15–1.77
.001
0.89–1.58
.2
0.66–1.33
.7
0.85–1.57
.3
0.79–1.16
.7
Reference Reference
OR, odds ratio; CI, confidence interval; PTSD, posttraumatic stress
disorder; +, positive screen; 2, negative screen.
* In model adjusted for age, race, twins or higher-order gestations,
and deployment history.
VOL. 124, NO. 6, DECEMBER 2014
In analysis of possible explanatory pathways (Table 4;
see Appendix 3, http://links.lww.com/AOG/A571,
Table A2 for full details), adjusting for comorbid
hypertension, diabetes, and asthma, made little difference—suggesting that the higher prevalence of these
chronic conditions observed in PTSD-affected women
did not explain the increased risk of spontaneous preterm birth. Only hypertension was a significant predictor of spontaneous preterm birth (adjusted OR 1.75
[1.43–2.13]) and its addition to the model did not significantly change the effect of active PTSD (from
adjusted OR 1.35 to adjusted OR 1.33 [1.12–1.59]).
Adjusting for drug, alcohol, and tobacco use mildly
attenuated the independent risk estimate for active
PTSD (adjusted OR 1.29 [1.08–1.55]); among the three
substance categories, only active drug dependence or
abuse was itself a significant predictor (adjusted OR
1.38 [1.03–1.85]). Lastly, adjusting for co-occurring psychiatric disorders, we found no evidence that an alternative comorbid mental health condition better
explained the effect observed—no other active mental
disorder was predictive of spontaneous preterm birth at
a statistically significant level and, when all were
included in the model, the effect of active PTSD, independent of co-occurring mental disorders, remained
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1115
Table 4. Key Sensitivity Analyses of the Association of Active Posttraumatic Stress Disorder and
Spontaneous Preterm Birth: Exploration of Potential Intermediaries
Active PTSD*: aOR of Preterm Birth
(Reference5No PTSD)
Model
Covariates Included
1 (base model) PTSD status*+age+race+twins or higher order+deployed†
2
Model 1+chronic disease indicators‡
3
Model 1+substance abuse§
4
Model 1+other psychiatric diagnosesk
aOR
95% CI
P
1.35
1.32
1.29
1.30
1.14–1.61
1.11–1.58
1.08–1.55
1.07–1.58
,.001
.002
.005
.004
PTSD, posttraumatic stress disorder; aOR, adjusted odds ratio; CI, confidence interval.
All models use generalized estimating equations with logit linkage, clustered by unique individual for repeat deliveries, except Model 6,
which had no repeat deliveries and thus used standard logistic regression. N516,334 in all models unless otherwise specified.
* PTSD status modeled as three-level variable (active, historical, or none), but for simplicity, only active PTSD results shown; see Appendix 3
(http://links.lww.com/AOG/A571) for full results.
†
Previously deployed in service of U.S. Operations Enduring freedom, Iraqi Freedom, or New Dawn.
‡
Chronic disease indicators are hypertension, diabetes, and asthma.
§
Substance abuse is drug dependence or abuse, alcohol dependence or abuse, and tobacco use diagnoses in the prenatal period.
k
Psychiatric diagnoses adjusted for are depressive, anxiety, adjustment, bipolar, and personality disorders.
similar and significant (adjusted OR 1.30 [1.07–1.58]).
Numerous additional sensitivity analyses of our model
were explored—including adjusting for prior spontaneous
preterm birth, restricting cohort to first deliveries, reinclusion of dropped demographic covariates, and alternative timeframes and definitions for active PTSD—and the
effect size for active PTSD remained robust (Table 5; see
Appendix 3, http://links.lww.com/AOG/A571, Tables
A3–A5 for full details).
DISCUSSION
We analyzed an unprecedented14–23 number of parturients with active PTSD (nearly 2,000) within a health
care system that routinely screens for PTSD and affirmed it is associated with increased risk of spontaneous preterm delivery—a finding robust to adjustment
for other known risk factors. The 35% increased odds
of spontaneous preterm delivery in those with active
PTSD is clinically relevant (two excess preterm births
per 100 affected deliveries) and on par with risks such
as advanced maternal age (older than 35 years)41 and,
within our cohort, only slightly smaller than the wellestablished risk factor of African American race.
Our findings build on suggestive previous studies
too small to detect this association.14–23 We benefit from
a design that confirms PTSD temporally preceded the
delivery, adding support for a causal relationship. The
Veterans Health Administration has mandatory PTSD
screening, using a validated instrument24 built into
the electronic medical record,42 and our reliance on
clinician-entered encounter diagnoses is supported by
prior Veterans Health Administration studies confirming the ICD-9-CM diagnosis reliably predicts PTSD.30,31
Several limitations are noteworthy. First, we were
unable to measure degree of prematurity and do not
account for medically indicated (eg, induced) preterm
births; however, the preterm births that we focused on—
spontaneous—disproportionately account for very preterm births3 and are most relevant to understanding the
role stress plays in early-onset delivery. Second,
deployment and military sexual trauma were the only
two trauma exposures examined; neither demonstrated
significant interaction with PTSD, but they are unlikely
Table 5. Key Sensitivity Analyses of the Association of Active Posttraumatic Stress Disorder and
Spontaneous Preterm Birth: Robustness Checks of Model Structure
Model
5
6
7
8
Structural Assumptions Modified
Model
Model
Model
Model
1
1
1
1
with
with
with
with
adjustment for prior spontaneous preterm birth
cohort restricted to first deliveries (n514,034)
cohort restricted to singleton deliveries (n515,986)
cohort restricted to deliveries after year 2007* (n59,932)
aOR
95% CI
P
1.35
1.37
1.32
1.36
1.13–1.60
1.14–1.65
1.10–1.59
1.10–1.69
,.001
,.001
.003
.004
aOR, adjusted odds ratio; CI, confidence interval.
All models use generalized estimating equations with logit linkage, clustered by unique individual for repeat deliveries, except Model 6,
which had no repeat deliveries and thus used standard logistic regression. N516,334 in all models unless otherwise specified.
* Year in which an electronic reminder for PTSD screening implemented in the Veterans Health Administration.
1116
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
the only sources of PTSD in our sample, and we cannot
comment on the role of premilitary trauma. Third, the
generalizability of our findings to non-Veterans Health
Administration populations is uncertain. However, the
association is not unique to combat veterans because
the majority of our cohort was nondeployed. The spontaneous preterm birthrate in our cohort (7.6%) is similar
to the 8.1% observed nationally43; the 12% prevalence
of active PTSD in our veteran cohort is disturbingly
high but comparable to the 7–14% reported in populations receiving maternity care in urban, public-payer
clinics.32,44 Military women experience diverse traumas, yet their most common antecedent of PTSD
remains sexual trauma45—the same as for women in
the general population.46
Although we cannot rule out unmeasured or
residual confounding, the robustness of our findings
to adjustments for known risk factors is reassuring.
The stark difference observed for active compared
with historical PTSD suggests we are not confounded
by the shared, unobserved, sociodemographic characteristics predisposing these mothers to PTSD. One
potential confounder is antidepressant use. Although
prior observation links antidepressants to preterm
birth,47 it remains uncertain whether the effect is from
antidepressants or the underlying depression9; adjusting for antidepressant use may introduce additional
confounding (by indication) rather than resolving
it.48 Given this, and limitations of Veterans Health
Administration pharmacy data, we did not adjust for
psychiatric medication use. However, post hoc analysis confirms that women with active antenatal depression were more likely prescribed antidepressants
than women with active PTSD. If antidepressant use
explained the increased spontaneous preterm risk
associated with PTSD, we would see equal or greater
increased risk among the more than 3,500 deliveries
to women with active depression; but, in our cohort,
we detected no such signal—thus, it is improbable that
antidepressant use is the pathway from PTSD to spontaneous preterm birth.
Our study adds to the nascent understanding of
the relationship between stress and preterm labor,
suggesting the abnormal stress response imparted by
PTSD49 might contribute to premature delivery. It
also identifies PTSD-affected patients as an important
clinical population in which to focus efforts to elucidate, and hopefully interrupt, the pathway from stress
to preterm birth. Plausible biologic mechanisms associating PTSD with spontaneous preterm birth include
neuroendocrine, inflammatory, and cardiovascular
alterations—all of which have been implicated in
our incomplete understanding of premature labor.3
VOL. 124, NO. 6, DECEMBER 2014
Among the indirect mechanisms that we could examine, drug abuse and hypertension only slightly attenuated the relationship between active PTSD and
spontaneous preterm delivery, suggesting direct effects over and above these factors.
Identifying PTSD-affected pregnancies as high risk
is clinically important and widely relevant. One in 20
U.S. pregnancies is likely in women affected by PTSD12
and one in five among women veterans returning from
military duty in Iraq and Afghanistan.26
Regardless of setting or population, obstetric and
primary care providers will inevitably find themselves caring for women with active PTSD in
pregnancy and preconception and need to be aware
of it as a risk factor. Brief, effective screening tools
for PTSD exist24 and could feasibly be included in
prenatal care, especially in populations with high
prevalence. Our study highlights the importance of
ensuring women with PTSD are connected to appropriate mental health care in the prenatal period
not only to address stress dysregulation, but also
the potential maladaptive behaviors that too often
accompany untreated PTSD, and raises hope that
appropriate treatment will not only improve maternal
well-being, but may well improve infant outcomes. If
future clinical trials determine that PTSD treatment
reduces risk for preterm delivery, we will have a blueprint for how to prevent the invisible wounds of
trauma from extending into the next generation.
REFERENCES
1. Althabe F, Howson CP, Kinney M, Lawn J; World Health
Organization. Born too soon: the global action report on preterm birth. 2012 Available at: http://www.who.int/pmnch/media/
news/2012/201204_borntoosoon-report.pdf. Retrieved December
18, 2013.
2. Ananth CV, Vintzileos AM. Epidemiology of preterm birth
and its clinical subtypes. J Matern Fetal Neonatal Med 2006;
19:773–82.
3. U.S. Institute of Medicine. Preterm birth: causes, consequences,
and prevention. Washington (DC): National Academies Press;
2007:772.
4. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84.
5. Class QA, Lichtenstein P, Långström N, D’Onofrio BM. Timing of prenatal maternal exposure to severe life events and
adverse pregnancy outcomes: a population study of 2.6 million
pregnancies. Psychosom Med 2011;73:234–41.
6. Holzman C, Senagore P, Tian Y, Bullen B, Devos E, Leece C,
et al. Maternal catecholamine levels in midpregnancy and risk
of preterm delivery. Am J Epidemiol 2009;170:1014–24.
7. Copper RL, Goldenberg RL, Das A, Elder N, Swain M,
Norman G, et al. The preterm prediction study: maternal stress
is associated with spontaneous preterm birth at less than thirtyfive weeks’ gestation. National Institute of Child Health and
Human Development Maternal-Fetal Medicine Units Network.
Am J Obstet Gynecol 1996;175:1286–92.
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1117
8. Straub H, Adams M, Kim JJ, Silver RK. Antenatal depressive
symptoms increase the likelihood of preterm birth. Am J Obstet
Gynecol 2012;207:329.e1–4.
veterans returning from Iraq and Afghanistan seen at Department of Veterans Affairs facilities. Arch Intern Med 2007;167:
476–82.
9. Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar S,
Katon WJ. A meta-analysis of depression during pregnancy
and the risk of preterm birth, low birth weight, and intrauterine
growth restriction. Arch Gen Psychiatry 2010;67:1012–24.
26. Mattocks KM, Skanderson M, Goulet JL, Brandt C, Womack J,
Krebs E, et al. Pregnancy and mental health among women
veterans returning from Iraq and Afghanistan. J Womens
Health (Larchmt) 2010;19:2159–66.
10. Li D, Liu L, Odouli R. Presence of depressive symptoms during
early pregnancy and the risk of preterm delivery: a prospective
cohort study. Hum Reprod 2009;24:146–53.
27. Kuklina EV, Whiteman MK, Hillis SD, Jamieson DJ,
Meikle SF, Posner SF, et al. An enhanced method for identifying obstetric deliveries: implications for estimating maternal
morbidity. Matern Child Health J 2008;12:469–77.
11. Reed RV, Fazel M, Goldring L. Post-traumatic stress disorder.
BMJ 2012;344:e3790.
12. National Comorbidity Survey. NCS-R appendix tables: Table 1.
Lifetime prevalence of DSM-IV/WMH-CIDI disorders by sex
and cohort. Table 2. Twelve-month prevalence of DSMIV/WMH-CIDI disorders by sex and cohort. 2005. Available at:
http://www.hcp.med.harvard.edu/ncs/publications.php.
Retrieved December 12, 2012.
13. Kilpatrick DG, Resnick HS, Milanak ME, Miller MW,
Keyes KM, Friedman MJ. National estimates of exposure to
traumatic events and PTSD prevalence using DSM-IV and
DSM-5 criteria. J Trauma Stress 2013;26:537–47.
14. Seng JS, Oakley DJ, Sampselle CM, Killion C, GrahamBermann S, Liberzon I. Posttraumatic stress disorder and pregnancy complications. Obstet Gynecol 2001;97:17–22.
15. Chang HL, Chang TC, Lin TY, Kuo SS. Psychiatric morbidity
and pregnancy outcome in a disaster area of Taiwan 921 earthquake. Psychiatry Clin Neurosci 2002;56:139–44.
28. Meux E, Stith S, Zach A. Report of results from the OSHPD
reabstracting project: an evaluation of the reliability of selected
patient discharge data, July through 1988. Sacramento (CA):
Office of Statewide Health Planning and Development, California Department of Health Services; 1990.
29. VA/DoD clinical practice guideline for the management of posttrauma disorders. Available at: http://www.healthquality.va.
gov/ptsd/cpg_PTSD-FULL-201011612.pdf. Retrieved August
27, 2013.
30. Gravely AA, Cutting A, Nugent S, Grill J, Carlson K,
Spoont M. Validity of PTSD diagnoses in VA administrative
data: comparison of VA administrative PTSD diagnoses to selfreported PTSD Checklist scores. J Rehabil Res Dev 2011;48:
21–30.
31. Frayne SM, Miller DR, Sharkansky EJ, Jackson VW, Wang F,
Halanych JH, et al. Using administrative data to identify mental
Illness: what approach is best? Am J Med Qual 2009;25:42–50.
16. Berkowitz GS, Wolff MS, Janevic TM, Holzman IR, Yehuda R,
Landrigan PJ. The World Trade Center disaster and intrauterine growth restriction. JAMA 2003;290:595–6.
32. Seng JS, Low LK, Sperlich M, Ronis DL, Liberzon I. Prevalence, trauma history, and risk for posttraumatic stress disorder
among nulliparous women in maternity care. Obstet Gynecol
2009;114:839–47.
17. Engel SM, Berkowitz GS, Wolff MS, Yehuda R. Psychological
trauma associated with the World Trade Center attacks and its
effect on pregnancy outcome. Paediatr Perinat Epidemiol 2005;
19:334–41.
33. Kimerling R, Gima K, Smith MW, Street A, Frayne S. The
Veterans Health Administration and military sexual trauma.
Am J Public Health 2007;97:2160–6.
18. Rogal SS, Poschman K, Belanger K, Howell HB, Smith MV,
Medina J, et al. Effects of posttraumatic stress disorder on pregnancy outcomes. J Affect Disord 2007;102:137–43.
19. Rosen D, Seng JS, Tolman RM, Mallinger G. Intimate partner
violence, depression, and posttraumatic stress disorder as additional predictors of low birth weight infants among low-income
mothers. J Interpers Violence 2007;22:1305–14.
20. Morland L, Goebert D, Onoye J, Frattarelli L, Derauf C,
Herbst M, et al. Posttraumatic stress disorder and pregnancy
health: preliminary update and implications. Psychosomatics
2007;48:304–8.
34. Epidemiology Program, Post Deployment Health Strategic
Healthcare Group, Office of Public Health, Veterans Health
Administration, Department of Veterans Affairs. Analysis of
VA health care utilization among operation enduring freedom, operation Iraqi freedom, and operation New Dawn Veterans, from 1st Qtr FY 2002 through 1st Qtr FY 2012. Washington
(DC); 2012. Available at: http://www.publichealth.va.gov/docs/
epidemiology/healthcare-utilization-report-fy2012-qtr1.pdf.
Retrieved April 7, 2014.
35. Zhang J, Troendle J, Reddy UM, Laughon SK, Branch DW,
Burkman R, et al. Contemporary cesarean delivery practice in
the United States. Am J Obstet Gynecol 2010;203:326.e1–10.
21. Xiong X, Harville EW, Mattison DR, Elkind-Hirsch K,
Pridjian G, Buekens P. Exposure to Hurricane Katrina, posttraumatic stress disorder and birth outcomes. Am J Med Sci
2008;336:111–5.
36. Lydon-Rochelle MT, Holt VL, Nelson JC, Cárdenas V, Gardella C,
Easterling TR, et al. Accuracy of reporting maternal in-hospital
diagnoses and intrapartum procedures in Washington State linked
birth records. Paediatr Perinat Epidemiol 2005;19:460–71.
22. Lipkind HS, Curry AE, Huynh M, Thorpe LE, Matte T. Birth
outcomes among offspring of women exposed to the September
11, 2001, terrorist attacks. Obstet Gynecol 2010;116:917–25.
37. Yoon J, Scott JY, Phibbs CS, Wagner TH. Recent trends in
Veterans Affairs chronic condition spending. Popul Health
Manag 2011;14:293–8.
23. Seng J, Low L, Sperlich M, Ronis D, Liberzon I. Post-traumatic
stress disorder, child abuse history, birthweight and gestational
age: a prospective cohort study. BJOG 2011;118:1329–39.
38. Turchik JA, Pavao J, Hyun J, Mark H, Kimerling R. Utilization
and intensity of outpatient care related to military sexual
trauma for veterans from Afghanistan and Iraq. J Behav Health
Serv Res 2012;39:220–33.
24. Prins A, Ouimette P, Kimerling R, Camerond RP, Hugelshofer DS,
Shaw-Hegwer J, et al. The primary care PTSD screen (PC–PTSD):
development and operating characteristics. Prim Care Psychiatry
2004;9:9–14.
25. Seal KH, Bertenthal D, Miner CR, Sen S, Marmar C. Bringing
the war back home: mental health disorders among 103,788 US
1118
Shaw et al
39. Healthcare cost and utilization project (HCUP) clinical classifications software (CCS). Rockville (MD): Agency for Healthcare
Quality and Research; 2012.
40. Larson GE, Highfill-McRoy RM, Booth-Kewley S. Psychiatric
diagnoses in historic and contemporary military cohorts: combat
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
deployment and the healthy Warrior effect. Am J Epidemiol 2008;
167:1269–76.
three federally qualified health care centers. Matern Child
Health J 2014;18:1056–65.
41. Cnattingius S, Forman MR, Berendes HW, Isotalo L. Delayed
childbearing and risk of adverse perinatal outcome. A
population-based study. JAMA 1992;268:886–90.
45. Suris A. Sexual assault in women veterans: an examination of
PTSD risk, health care utilization, and cost of care. Psychosom
Med 2004;66:749–56.
42. Health Provider Systems Office of Information & Technology.
VistA Clinical Reminders User Manual, Version 2.0 Patch
PXRM*2*6. Department of Veterans Affairs; 2007. Available
at: http://www.va.gov/vdl/documents/Clinical/CPRS-Clinical_
Reminders/pxrm_2_6_um.pdf. Retrieved July 10, 2013.
46. Kessler RC. Posttraumatic stress disorder in the national comorbidity survey. Arch Gen Psychiatry 1995;52:1048–60.
43. Elixhauser A, Wier LM. Complicating conditions of pregnancy
and childbirth, 2008: statistical brief #113. Healthcare cost and
utilization project (HCUP) statistical briefs. Rockville (MD):
Agency for Health Care Policy and Research (US); 2006. Available
at: http://www.ncbi.nlm.nih.gov/books/NBK56037/. Retrieved
May 17, 2013.
44. Kim HG, Harrison PA, Godecker AL, Muzyka CN. Posttraumatic stress disorder among women receiving prenatal care at
47. Ross LE, Grigoriadis S, Mamisashvili L, Vonderporten EH,
Roerecke M, Rehm J, et al. Selected pregnancy and delivery outcomes after exposure to antidepressant medication: a systematic
review and meta-analysis. JAMA Psychiatry 2013;70:436–43.
48. Grzeskowiak LE, Gilbert AL, Morrison JL. Investigating outcomes following the use of selective serotonin reuptake inhibitors for treating depression in pregnancy: a focus on
methodological issues. Drug Saf 2011;34:1027–48.
49. Sherin JE, Nemeroff CB. Post-traumatic stress disorder: the
neurobiological impact of psychological trauma. Dialogues Clin
Neurosci 2011;13:263–78.
Maintenance of Certiļ¬cation Articles
Maintenance of Certiļ¬cation is a program of the American Board of Obstetrics
and Gynecology (ABOG; http://www.abog.org). All articles from the reading
lists for the current year will be listed on the Obstetrics & Gynecology web site
at the beginning of January, May, and August.
To access the lists, go to www.greenjournal.org and click on the “ABOG MOC II” tab.
Links to content are provided, as well as an indication of the article’s status (ie,
available by subscription only or open access).
rev 12/2014
VOL. 124, NO. 6, DECEMBER 2014
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1119
Posttraumatic Stress Disorder and Risk of
Spontaneous Preterm Birth
Jonathan G. Shaw, MD, MS, Steven M. Asch, MD, MPH, Rachel Kimerling, PhD, Susan M. Frayne, MD, MPH,
Kate A. Shaw, MD, MS, and Ciaran S. Phibbs, PhD
OBJECTIVE: To evaluate the association between antenatal posttraumatic stress disorder (PTSD) and spontaneous preterm delivery.
METHODS: We identified antenatal PTSD status and
spontaneous preterm delivery in a retrospective cohort of
16,334 deliveries covered by the Veterans Health Administration from 2000 to 2012. We divided mothers with
PTSD into those with diagnoses present the year before
delivery (active PTSD) and those only with earlier diagnoses
(historical PTSD). We identified spontaneous preterm birth
and potential confounders including age, race, military
deployment, twins, hypertension, substance use, depression, and results of military sexual trauma screening and
then performed multivariate regression to estimate
adjusted odds ratio (OR) of spontaneous preterm delivery
as a function of PTSD status.
From the Center for Innovation to Implementation, the National Center for
Posttraumatic Stress Disorder, the Women’s Health Section, and the Health Economics Resource Center, Department of Veterans Affairs, Palo Alto Health Care System,
Palo Alto, and the March of Dimes Center for Prematurity Research at Stanford, the
Centers for Health Policy/Primary Care & Outcomes Research, the Division of General Medical Disciplines, and the Departments of Obstetrics & Gynecology and Pediatrics, Stanford University School of Medicine, Stanford, California.
Dr. Shaw was supported in part by the VA Office of Academic Affairs and
Health Services Research & Development funds. Development of the Women’s
Health Evaluation Initiative (WHEI) data used in this study was supported by
VA Women’s Health Services.
The authors thank Laurie Zephyrin, MD, MPH, MBA, of Women’s Health
Services in the VA Central Office, and Rita Popat, PhD, of Stanford University
for thoughtful review of the manuscript, and VA Palo Alto Staff, Lakshmi
Ananth, MS, Vidhya Balasubramanian, MS, Eric Berg, MS, Sarah Friedman,
MSPH, Fay Saechao, MPH, Meghan Saweikis, MS, JD, and Susan Schmitt,
PhD, for invaluable consultation and technical advice.
The findings and conclusions in this report are those of the authors and do not
necessarily represent the official position of the Department of Veterans Affairs.
Corresponding author: Jonathan G. Shaw, MD, MS, 117 Encina Commons,
Room 206, Stanford, CA 94305; e-mail: [email protected].
Financial Disclosure
The authors did not report any potential conflicts of interest.
© 2014 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/14
VOL. 124, NO. 6, DECEMBER 2014
RESULTS: Of 16,334 births, 3,049 (19%) were to mothers
with PTSD diagnoses, of whom 1,921 (12%) had active
PTSD. Spontaneous preterm delivery was higher in those
with active PTSD (9.2%, n5176) than those with historical
(8.0%, n590) or no PTSD (7.4%, n5982) before adjustment
(P5.02). The association between PTSD and preterm birth
persisted, when adjusting for covariates, only in those with
active PTSD (adjusted OR 1.35, 95% confidence interval
[CI] 1.14–1.61). Analyses adjusting for comorbid psychiatric
and medical diagnoses revealed the association with active
PTSD to be robust.
CONCLUSION: In this cohort, containing an unprecedented number of PTSD-affected pregnancies, mothers with
active PTSD were significantly more likely to suffer spontaneous preterm birth with an attributable two excess preterm
births per 100 deliveries (95% CI 1–4). Posttraumatic stress
disorder’s health effects may extend, through birth outcomes, into the next generation.
(Obstet Gynecol 2014;124:1111–9)
DOI: 10.1097/AOG.0000000000000542
LEVEL OF EVIDENCE: II
P
reterm birth is a leading cause of infant morbidity
and mortality.1 In the United States 12% of deliveries are preterm, and roughly half are spontaneous as
opposed to medically indicated.2 Although certain risk
factors for spontaneous preterm birth have been clearly
identified, including demographic characteristics, substance use, and multiple gestations, the etiology remains
poorly understood; efforts to reduce the preterm birth
rate have made little progress over the past two decades.1–4 A growing number of studies suggest a role for
psychosocial factors such as maternal stress5–7 and
depression.8–10 However, the effect of posttraumatic
stress disorder (PTSD) remains unclear.
Posttraumatic stress disorder is a complex of
disruptive symptoms arising from a traumatic experience (eg, violence, disaster). Its prevalence varies
substantially between populations based on exposure.11 U.S. surveys show women affected at higher
OBSTETRICS & GYNECOLOGY
1111
rates than men, and estimate lifetime prevalence of
PTSD among females to be 10–14%.12,13
Posttraumatic stress disorder could affect preterm
delivery directly through biological pathways or
indirectly through risky health behaviors and poor
self-care (eg, attendance to medical care, nutrition,
and sleep). Previous studies have been limited by
inadequate sample size, heterogeneity in diagnostic
criteria, and generalizability concerns.14–23
We examined whether PTSD is associated with
spontaneous preterm birth by evaluating deliveries
covered by the Veterans Health Administration. With
standardized mandatory PTSD screening,24 a high
PTSD prevalence (13–21% in reproductive-aged
women),25,26 and centralized national data, the Veterans Health Administration provides an ideal setting in
which to examine the association between PTSD and
preterm birth in a cohort of unprecedented size.
MATERIALS AND METHODS
Using data from national clinical and administrative
databases for Veterans Health Administration-based
(nonobstetric) care and outsourced (obstetric) Veterans Health Administration care, we conducted a retrospective cohort analysis of all Veterans Health
Administration-reimbursed deliveries in fiscal years
2000–2012 and evaluated associations between antepartum PTSD and spontaneous preterm delivery:
16,477 deliveries were identified using a validated
algorithm.27 Our main Veterans Health Administration data set is derived from standardized hospital
discharge abstracts, entered by professional coders,
and diagnosis data from such discharge abstracts have
previously been shown to be reliably coded.28 We
excluded 102 women without Veterans Health
Administration encounters before delivery (and thus
without antecedent Veterans Health Administration
PTSD screening or clinical data) and 41 additional
deliveries resulting from irreconcilable data.
Our primary outcome was spontaneous preterm
birth. Spontaneous cases were deemed most relevant to
evaluating the direct contribution of PTSD to premature
delivery. Cases were identified by International Classification of Diseases, 9th Revision, Clinical Modification
(ICD-9-CM) diagnosis code 644.2 “spontaneous onset of
delivery before 37 weeks” in the delivery claim, which in
external validation corresponds to a median gestational
age of 35 weeks (see Appendix 1, available online at
http://links.lww.com/AOG/A571, for a description of
validation).
The Veterans Health Administration routinely
screens for PTSD within its primary care system,29
and prior Veterans Health Administration studies
1112
Shaw et al
confirm the reliability of ICD-9-CM 309.81 to identify
PTSD.30,31 We used Veterans Health Administration
encounters from 1997 to 2012 (ensuring a minimum
3-year look-back for our cohort, which begins with
deliveries in 2000) to identify PTSD in any encounter
before delivery. For each delivery, we further distinguished two mutually exclusive categories of PTSD
status in pregnancy: “active PTSD” and “historical
PTSD.” This distinction was invoked based on prior
research that observed poorer birth outcomes in
women with current PTSD symptoms during pregnancy, not past (recovered) PTSD.32 We categorized
as “active PTSD” those cases in which a PTSD diagnosis was documented in any encounter(s) within 365
days before the day of delivery, presuming these represented pregnancies with clinically relevant PTSD symptoms in the prenatal period; PTSD cases not meeting
this criterion were labeled “historical PTSD.” In choosing a 1-year window to define active PTSD, we aimed to
maximize overlap with the pregnancy and ensure likelihood of a Veterans Health Administration encounter for
all participants in that time period. (A more restrictive
9-month window was explored in a sensitivity analysis.)
By our inclusion criteria, all women had Veterans
Health Administration encounters before delivery, and
more than 99% had an encounter within 1 year before
delivery and thus the opportunity for active PTSD to be
documented.
We collected data on two potential trauma exposures, military sexual trauma and recent military deployment, both associated with high rates of PTSD within the
Veterans Health Administration.25,33 Mandatory onetime screening for military sexual trauma has been in
place since 2002 through a brief validated instrument33
(see Appendix 1, http://links.lww.com/AOG/A571, for
screening questions). We used the Department of
Defense Roster34 to identify veterans with recent deployment in support of operations in Afghanistan or Iraq.
We obtained demographic covariates from Veterans Health Administration administrative data, including maternal age, race, and marital status. We identified
obstetric risk factors including twins or higher-order
gestations and prior cesarean delivery from obstetric
hospitalizations using validated ICD-9-CM codes.35,36
We extracted medical comorbidities from previously
developed Veterans Health Administration chronic
condition data sets.37 Specifically, we recorded those
conditions that were both prevalent in our cohort
(2% or more) and have been suggested as risk factors
for preterm birth3: hypertension, diabetes, and asthma.
To identify mental health diagnoses other than
PTSD, we used a modified version38 of the Agency for
Healthcare Research and Quality Clinical Classifications
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
Software’s categories of mental illness.39 Applying the
same criteria as for PTSD, we created three-level indicator variables (active, historical, none) for the most
prevalent disorders. We identified active drug and alcohol abuse and tobacco dependence in the antenatal
period by the presence of diagnostic codes in Veterans
Health Administration encounters within 1 year before
delivery or within reimbursed prenatal encounters
(where screening for substances should be standard of
care) to maximize sensitivity.
Delivery was our unit of analysis and we used
statistical methods accounting for repeated measures
adjusting effect estimates to take into account a withinperson correlation for women with repeat deliveries in
our cohort. We performed unadjusted x2 bivariate analysis, then performed adjusted multivariate logistic
regression (SAS 9.2) using generalized estimating equations modeling (clustered by individual) to determine
adjusted odds ratio (OR) of spontaneous preterm delivery by PTSD status as a three-level predictor comparing
“active PTSD” and “historical PTSD” with none.
We performed regressions in an iterative (additive) fashion, exploring demographic and obstetric
factors, and potential trauma (military sexual trauma,
deployment) as possible confounders. For parsimony, covariates were only retained in subsequent
models if they altered the b coefficient of the primary
predictor, active PTSD, by 10% or more. All models
were adjusted for multiple gestations given their
undisputed association with spontaneous preterm
birth. Our resulting primary model adjusted for
age, race, multiple gestation, and deployment history. We examined interaction terms for PTSD*military sexual trauma and PTSD*deployment history.
We explored three possible explanatory pathways:
preselected chronic comorbidities (hypertension,
diabetes, asthma), behavioral risks (drug, alcohol,
or tobacco use), and other mental health disorders
frequently codiagnosed with PTSD (those disorders
with prevalence 2% or more in our cohort). We did not
include these as confounders in our primary analysis
because they could arguably represent intermediate
steps in the causal pathway (ie, if the pathophysiology
of chronic PTSD predisposes women to the selected
comorbidities). In evaluating the potential explanatory
role of codiagnosed mental health conditions, we
placed these variables in regression models, both
alongside PTSD and in place of PTSD. Lastly, we
tested our model assumptions in a series of sensitivity
analyses, including models restricted to first deliveries or singleton deliveries, models adjusted for
obstetric history (eg, prior spontaneous preterm
birth, excluding multiples), and models applying
VOL. 124, NO. 6, DECEMBER 2014
more restrictive timeframes (see Appendix 2, available online at http://links.lww.com/AOG/A571, for
a detailed description).
This research was approved by Stanford University’s institutional review board as part of the Women’s Health Evaluation Initiative.
RESULTS
The cohort included 16,334 deliveries among 14,047
women. There were 1,248 (7.6%) spontaneous preterm deliveries. Of the 16,334 deliveries, 3,049 (19%)
were to women who carried an antepartum diagnosis
of PTSD, two thirds of whom (1,921 [12%]) had active
PTSD. Nearly one third of the deliveries (4,948) were
to women with recent deployment (Afghanistan or
Iraq); 3,568 (23%) deliveries were to women reporting
a history of military sexual trauma.
Table 1 presents descriptive characteristics by
PTSD status. Those with active PTSD were significantly more likely to have been deployed (45%)
than were those with historical (32%) or no PTSD
(28%). Those with active and historical PTSD
were more likely to report military sexual trauma
(57% and 46%) than those without PTSD (16%) and
carried a significantly higher burden of active
comorbid mental health, drug-, and alcoholrelated diagnoses (P,.001 for all). The unadjusted
proportion of spontaneous preterm birth (Table 1)
was significantly higher in those with active PTSD
(9.2%) than those with historical (8.0%) or no PTSD
(7.4%) (P5.02).
In our primary adjusted model (Table 2), active
PTSD remained associated with spontaneous preterm
delivery (adjusted OR 1.35 [1.14–1.61]), whereas historical PTSD’s association was nonsignificant
(adjusted OR 1.06 [0.84–1.34]).
Although deployment history was associated with
a lower risk of spontaneous preterm birth, adjusted OR
0.71 (0.61–0.81)—consistent with our expectation that
the selection process for deployment results in a subpopulation of healthier, lower-risk individuals (the “healthy
warrior effect”)40—we found no statistical support for our
a priori hypothesis that deployment modifies the effect
of PTSD (P5.42 for interaction term active PTSD*deployment added to our base model). In contrast, military
sexual trauma showed no association with spontaneous
preterm birth (unadjusted OR 1.08 [0.94–1.24]; adjusted
OR 0.99 [0.84–1.15]; see Appendix 3, available online
at http://links.lww.com/AOG/A571, Table A1).
However, as shown in Table 3, when the interaction
of military sexual trauma and PTSD was explored, we
observed that those with both active PTSD and military sexual trauma carried the greatest, and most
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1113
Table 1. Characteristics of Deliveries Covered by the Veterans Administration (2000–2012), by
Posttraumatic Stress Disorder Status
Deliveries by PTSD Status
Characteristic
Spontaneous preterm birth
Demographics
Maternal age (y)
19–24
25–29
30–34
35–39
40–48
Race
White
African American or black
Asian
Native Hawaiian or other Pacific
Islander
American Indian or Alaskan Native
Missing or declined to answer
Married†
Potential trauma exposure
Military sexual trauma‡
Previously deployed (U.S. Operations
Enduring Freedom, Iraqi
Freedom, and New Dawn)
Obstetric history
Twins or higher-order gestation
Prior cesarean delivery
Parity§
1
2
3 or more (maximum 5)
Prior spontaneous preterm delivery§
Chronic medical conditions (within 3 y
antepartum)
Hypertension
Diabetes
Asthma
Substance abuse or dependence diagnoses
(within 1 y antepartum)
Drug
Alcohol
Tobacco
Active mental health comorbidities (within
1 y antepartum)
Depressive disorder
Anxiety disorder (other than PTSD)
Adjustment disorder
Bipolar disorder
Personality disorder
Active PTSD Diagnosis
(Within 365 Days
Antepartum) (n51,921)
Historical PTSD Diagnosis
(More Than 365 Days
Antepartum) (n51,128)
176 (9.2)
90 (8.0)
None
(n513,285)
982 (7.4)
P*
.02
,.001
273
787
515
269
77
(14.2)
(41.0)
(26.8)
(14.0)
(4.0)
65
429
375
198
61
(5.8)
(38.0)
(33.2)
(17.6)
(5.4)
2,290
5,257
3,689
1,614
435
(17.2)
(40.0)
(27.8)
(12.2)
(3.3)
1,344
371
26
36
(70.0)
(19.3)
(1.4)
(1.9)
776
243
11
9
(68.8)
(21.5)
(1.0)
(0.8)
8,142
3,059
219
188
(61.3)
(23.0)
(1.7)
(1.4)
,.001
20 (1.0)
124 (6.5)
824 (43.2)
8 (0.7)
81 (7.2)
475 (43.0)
102 (0.8)
1,575 (11.9)
6,187 (47.5) ,.001
1,085 (57.7)
863 (44.9)
513 (46.2)
361 (32.0)
1,970 (15.7) ,.001
3,760 (28.3) ,.001
41 (2.1)
275 (14.3)
31 (2.8)
195 (17.3)
276 (2.1)
1,847 (13.9)
1,677
220
24
30
(87.3)
(11.5)
(1.3)
(1.6)
863
223
42
25
(76.5)
(19.8)
(3.7)
(2.2)
11,494
1,574
217
132
.3
.007
,.001
(86.5)
(11.9)
(1.6)
(1.0) ,.001
169 (8.8)
90 (4.7)
206 (10.7)
91 (8.1)
36 (3.2)
138 (12.2)
835 (6.3)
414 (3.1)
990 (7.5)
,.001
.002
,.001
217 (11.3)
165 (8.6)
351 (18.3)
50 (4.4)
26 (2.3)
143 (12.7)
387 (2.9)
175 (1.3)
1,219 (9.2)
,.001
,.001
,.001
327
148
45
62
29
2,089
1,019
553
342
178
,.001
,.001
,.001
,.001
,.001
1,190
552
188
181
172
(62.0)
(28.7)
(9.8)
(9.4)
(9.0)
(29.0)
(13.1)
(4.0)
(5.5)
(2.6)
(15.7)
(7.7)
(4.2)
(2.6)
(1.3)
PTSD, posttraumatic stress disorder.
Data are n (%) unless otherwise specified.
* x2.
†
Missing for 307 (2%).
‡
Missing for 814 (5%).
§
Estimate; data unavailable for deliveries before 2000 and non–Veterans Health Administration deliveries.
1114
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
Table 2. Association of Posttraumatic Stress Disorder Status and Spontaneous Preterm Birth: Unadjusted
and Primary Adjusted Model (N516,334; Preterm Delivery Events51,248)
Unadjusted Model
n
OR
95% CI
P
aOR
95% CI
P
13,285
1,128
1,921
1
1.09
1.26
Reference
0.87–1.36
1.07–1.50
0.5
0.006
1
1.06
1.35
Reference
0.84–1.34
1.14–1.61
.6
,.001
2,628
6,473
4,579
2,081
573
1
0.99
1.15
1.19
1.21
Reference
0.83–1.19
0.96–1.40
0.95–1.49
0.87–1.69
.9
.1
.1
.3
10,262
3,673
256
187
130
1,780
348
4,984
1
1.49
1.27
1.99
1.35
1.10
7.15
0.71
Reference
1.29–1.71
0.82–1.96
1.15–3.45
0.85–2.13
0.90–1.35
5.66–9.03
0.61–0.81
,.001
.3
.01
.2
.3
,.001
,.001
Parameter
PTSD
None
Historical (more than 365 d antepartum)
Active (within 365 d antepartum)
Maternal age (y)
19–25
25–29
30–34
35–39
40 or older
Race
White
Black or African American
Asian
American Indian or Alaskan Native
Hawaiian or Pacific Islander
Missing or not reported
Twins or higher-order gestation
Deployed (U.S. Operations Enduring Freedom, Iraqi
Freedom, and New Dawn)
Multivariate Model*
OR, odds ratio; CI, confidence interval; aOR, adjusted odds ratio; PTSD, posttraumatic stress disorder.
* Generalized estimating equations with logit linkage and clustered by unique individual using exchangeable correlation matrix for repeat
deliveries. Independent variables are those listed. Model fit: C-statistic (in logistic regression model without clustering)50.62 in adjusted
model as compared with 0.52 in unadjusted model; 58,536.6 in the adjusted model as compared with 8,816.0 in the unadjusted model.
significant, risk (adjusted OR 1.43 [1.15–1.77]),
although the interaction term was not statistically significant (P5.25 for interaction term active PTSD*military sexual trauma).
Table 3. Interaction of Posttraumatic Stress
Disorder Status and Military Sexual
Trauma History as Predictors of
Spontaneous Preterm Birth*
Multilevel Variable
Combining PTSD
and Military Sexual
Trauma Status
PTSD
Active
Active
Historical
Historical
None
None
Military
Sexual
Trauma
n
OR
+
2
+
2
+
2
1,085
797
598
513
1,970
10,557
1.43
1.19
0.93
1.16
0.96
1
95% CI
P
1.15–1.77
.001
0.89–1.58
.2
0.66–1.33
.7
0.85–1.57
.3
0.79–1.16
.7
Reference Reference
OR, odds ratio; CI, confidence interval; PTSD, posttraumatic stress
disorder; +, positive screen; 2, negative screen.
* In model adjusted for age, race, twins or higher-order gestations,
and deployment history.
VOL. 124, NO. 6, DECEMBER 2014
In analysis of possible explanatory pathways (Table 4;
see Appendix 3, http://links.lww.com/AOG/A571,
Table A2 for full details), adjusting for comorbid
hypertension, diabetes, and asthma, made little difference—suggesting that the higher prevalence of these
chronic conditions observed in PTSD-affected women
did not explain the increased risk of spontaneous preterm birth. Only hypertension was a significant predictor of spontaneous preterm birth (adjusted OR 1.75
[1.43–2.13]) and its addition to the model did not significantly change the effect of active PTSD (from
adjusted OR 1.35 to adjusted OR 1.33 [1.12–1.59]).
Adjusting for drug, alcohol, and tobacco use mildly
attenuated the independent risk estimate for active
PTSD (adjusted OR 1.29 [1.08–1.55]); among the three
substance categories, only active drug dependence or
abuse was itself a significant predictor (adjusted OR
1.38 [1.03–1.85]). Lastly, adjusting for co-occurring psychiatric disorders, we found no evidence that an alternative comorbid mental health condition better
explained the effect observed—no other active mental
disorder was predictive of spontaneous preterm birth at
a statistically significant level and, when all were
included in the model, the effect of active PTSD, independent of co-occurring mental disorders, remained
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1115
Table 4. Key Sensitivity Analyses of the Association of Active Posttraumatic Stress Disorder and
Spontaneous Preterm Birth: Exploration of Potential Intermediaries
Active PTSD*: aOR of Preterm Birth
(Reference5No PTSD)
Model
Covariates Included
1 (base model) PTSD status*+age+race+twins or higher order+deployed†
2
Model 1+chronic disease indicators‡
3
Model 1+substance abuse§
4
Model 1+other psychiatric diagnosesk
aOR
95% CI
P
1.35
1.32
1.29
1.30
1.14–1.61
1.11–1.58
1.08–1.55
1.07–1.58
,.001
.002
.005
.004
PTSD, posttraumatic stress disorder; aOR, adjusted odds ratio; CI, confidence interval.
All models use generalized estimating equations with logit linkage, clustered by unique individual for repeat deliveries, except Model 6,
which had no repeat deliveries and thus used standard logistic regression. N516,334 in all models unless otherwise specified.
* PTSD status modeled as three-level variable (active, historical, or none), but for simplicity, only active PTSD results shown; see Appendix 3
(http://links.lww.com/AOG/A571) for full results.
†
Previously deployed in service of U.S. Operations Enduring freedom, Iraqi Freedom, or New Dawn.
‡
Chronic disease indicators are hypertension, diabetes, and asthma.
§
Substance abuse is drug dependence or abuse, alcohol dependence or abuse, and tobacco use diagnoses in the prenatal period.
k
Psychiatric diagnoses adjusted for are depressive, anxiety, adjustment, bipolar, and personality disorders.
similar and significant (adjusted OR 1.30 [1.07–1.58]).
Numerous additional sensitivity analyses of our model
were explored—including adjusting for prior spontaneous
preterm birth, restricting cohort to first deliveries, reinclusion of dropped demographic covariates, and alternative timeframes and definitions for active PTSD—and the
effect size for active PTSD remained robust (Table 5; see
Appendix 3, http://links.lww.com/AOG/A571, Tables
A3–A5 for full details).
DISCUSSION
We analyzed an unprecedented14–23 number of parturients with active PTSD (nearly 2,000) within a health
care system that routinely screens for PTSD and affirmed it is associated with increased risk of spontaneous preterm delivery—a finding robust to adjustment
for other known risk factors. The 35% increased odds
of spontaneous preterm delivery in those with active
PTSD is clinically relevant (two excess preterm births
per 100 affected deliveries) and on par with risks such
as advanced maternal age (older than 35 years)41 and,
within our cohort, only slightly smaller than the wellestablished risk factor of African American race.
Our findings build on suggestive previous studies
too small to detect this association.14–23 We benefit from
a design that confirms PTSD temporally preceded the
delivery, adding support for a causal relationship. The
Veterans Health Administration has mandatory PTSD
screening, using a validated instrument24 built into
the electronic medical record,42 and our reliance on
clinician-entered encounter diagnoses is supported by
prior Veterans Health Administration studies confirming the ICD-9-CM diagnosis reliably predicts PTSD.30,31
Several limitations are noteworthy. First, we were
unable to measure degree of prematurity and do not
account for medically indicated (eg, induced) preterm
births; however, the preterm births that we focused on—
spontaneous—disproportionately account for very preterm births3 and are most relevant to understanding the
role stress plays in early-onset delivery. Second,
deployment and military sexual trauma were the only
two trauma exposures examined; neither demonstrated
significant interaction with PTSD, but they are unlikely
Table 5. Key Sensitivity Analyses of the Association of Active Posttraumatic Stress Disorder and
Spontaneous Preterm Birth: Robustness Checks of Model Structure
Model
5
6
7
8
Structural Assumptions Modified
Model
Model
Model
Model
1
1
1
1
with
with
with
with
adjustment for prior spontaneous preterm birth
cohort restricted to first deliveries (n514,034)
cohort restricted to singleton deliveries (n515,986)
cohort restricted to deliveries after year 2007* (n59,932)
aOR
95% CI
P
1.35
1.37
1.32
1.36
1.13–1.60
1.14–1.65
1.10–1.59
1.10–1.69
,.001
,.001
.003
.004
aOR, adjusted odds ratio; CI, confidence interval.
All models use generalized estimating equations with logit linkage, clustered by unique individual for repeat deliveries, except Model 6,
which had no repeat deliveries and thus used standard logistic regression. N516,334 in all models unless otherwise specified.
* Year in which an electronic reminder for PTSD screening implemented in the Veterans Health Administration.
1116
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
the only sources of PTSD in our sample, and we cannot
comment on the role of premilitary trauma. Third, the
generalizability of our findings to non-Veterans Health
Administration populations is uncertain. However, the
association is not unique to combat veterans because
the majority of our cohort was nondeployed. The spontaneous preterm birthrate in our cohort (7.6%) is similar
to the 8.1% observed nationally43; the 12% prevalence
of active PTSD in our veteran cohort is disturbingly
high but comparable to the 7–14% reported in populations receiving maternity care in urban, public-payer
clinics.32,44 Military women experience diverse traumas, yet their most common antecedent of PTSD
remains sexual trauma45—the same as for women in
the general population.46
Although we cannot rule out unmeasured or
residual confounding, the robustness of our findings
to adjustments for known risk factors is reassuring.
The stark difference observed for active compared
with historical PTSD suggests we are not confounded
by the shared, unobserved, sociodemographic characteristics predisposing these mothers to PTSD. One
potential confounder is antidepressant use. Although
prior observation links antidepressants to preterm
birth,47 it remains uncertain whether the effect is from
antidepressants or the underlying depression9; adjusting for antidepressant use may introduce additional
confounding (by indication) rather than resolving
it.48 Given this, and limitations of Veterans Health
Administration pharmacy data, we did not adjust for
psychiatric medication use. However, post hoc analysis confirms that women with active antenatal depression were more likely prescribed antidepressants
than women with active PTSD. If antidepressant use
explained the increased spontaneous preterm risk
associated with PTSD, we would see equal or greater
increased risk among the more than 3,500 deliveries
to women with active depression; but, in our cohort,
we detected no such signal—thus, it is improbable that
antidepressant use is the pathway from PTSD to spontaneous preterm birth.
Our study adds to the nascent understanding of
the relationship between stress and preterm labor,
suggesting the abnormal stress response imparted by
PTSD49 might contribute to premature delivery. It
also identifies PTSD-affected patients as an important
clinical population in which to focus efforts to elucidate, and hopefully interrupt, the pathway from stress
to preterm birth. Plausible biologic mechanisms associating PTSD with spontaneous preterm birth include
neuroendocrine, inflammatory, and cardiovascular
alterations—all of which have been implicated in
our incomplete understanding of premature labor.3
VOL. 124, NO. 6, DECEMBER 2014
Among the indirect mechanisms that we could examine, drug abuse and hypertension only slightly attenuated the relationship between active PTSD and
spontaneous preterm delivery, suggesting direct effects over and above these factors.
Identifying PTSD-affected pregnancies as high risk
is clinically important and widely relevant. One in 20
U.S. pregnancies is likely in women affected by PTSD12
and one in five among women veterans returning from
military duty in Iraq and Afghanistan.26
Regardless of setting or population, obstetric and
primary care providers will inevitably find themselves caring for women with active PTSD in
pregnancy and preconception and need to be aware
of it as a risk factor. Brief, effective screening tools
for PTSD exist24 and could feasibly be included in
prenatal care, especially in populations with high
prevalence. Our study highlights the importance of
ensuring women with PTSD are connected to appropriate mental health care in the prenatal period
not only to address stress dysregulation, but also
the potential maladaptive behaviors that too often
accompany untreated PTSD, and raises hope that
appropriate treatment will not only improve maternal
well-being, but may well improve infant outcomes. If
future clinical trials determine that PTSD treatment
reduces risk for preterm delivery, we will have a blueprint for how to prevent the invisible wounds of
trauma from extending into the next generation.
REFERENCES
1. Althabe F, Howson CP, Kinney M, Lawn J; World Health
Organization. Born too soon: the global action report on preterm birth. 2012 Available at: http://www.who.int/pmnch/media/
news/2012/201204_borntoosoon-report.pdf. Retrieved December
18, 2013.
2. Ananth CV, Vintzileos AM. Epidemiology of preterm birth
and its clinical subtypes. J Matern Fetal Neonatal Med 2006;
19:773–82.
3. U.S. Institute of Medicine. Preterm birth: causes, consequences,
and prevention. Washington (DC): National Academies Press;
2007:772.
4. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84.
5. Class QA, Lichtenstein P, Långström N, D’Onofrio BM. Timing of prenatal maternal exposure to severe life events and
adverse pregnancy outcomes: a population study of 2.6 million
pregnancies. Psychosom Med 2011;73:234–41.
6. Holzman C, Senagore P, Tian Y, Bullen B, Devos E, Leece C,
et al. Maternal catecholamine levels in midpregnancy and risk
of preterm delivery. Am J Epidemiol 2009;170:1014–24.
7. Copper RL, Goldenberg RL, Das A, Elder N, Swain M,
Norman G, et al. The preterm prediction study: maternal stress
is associated with spontaneous preterm birth at less than thirtyfive weeks’ gestation. National Institute of Child Health and
Human Development Maternal-Fetal Medicine Units Network.
Am J Obstet Gynecol 1996;175:1286–92.
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1117
8. Straub H, Adams M, Kim JJ, Silver RK. Antenatal depressive
symptoms increase the likelihood of preterm birth. Am J Obstet
Gynecol 2012;207:329.e1–4.
veterans returning from Iraq and Afghanistan seen at Department of Veterans Affairs facilities. Arch Intern Med 2007;167:
476–82.
9. Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar S,
Katon WJ. A meta-analysis of depression during pregnancy
and the risk of preterm birth, low birth weight, and intrauterine
growth restriction. Arch Gen Psychiatry 2010;67:1012–24.
26. Mattocks KM, Skanderson M, Goulet JL, Brandt C, Womack J,
Krebs E, et al. Pregnancy and mental health among women
veterans returning from Iraq and Afghanistan. J Womens
Health (Larchmt) 2010;19:2159–66.
10. Li D, Liu L, Odouli R. Presence of depressive symptoms during
early pregnancy and the risk of preterm delivery: a prospective
cohort study. Hum Reprod 2009;24:146–53.
27. Kuklina EV, Whiteman MK, Hillis SD, Jamieson DJ,
Meikle SF, Posner SF, et al. An enhanced method for identifying obstetric deliveries: implications for estimating maternal
morbidity. Matern Child Health J 2008;12:469–77.
11. Reed RV, Fazel M, Goldring L. Post-traumatic stress disorder.
BMJ 2012;344:e3790.
12. National Comorbidity Survey. NCS-R appendix tables: Table 1.
Lifetime prevalence of DSM-IV/WMH-CIDI disorders by sex
and cohort. Table 2. Twelve-month prevalence of DSMIV/WMH-CIDI disorders by sex and cohort. 2005. Available at:
http://www.hcp.med.harvard.edu/ncs/publications.php.
Retrieved December 12, 2012.
13. Kilpatrick DG, Resnick HS, Milanak ME, Miller MW,
Keyes KM, Friedman MJ. National estimates of exposure to
traumatic events and PTSD prevalence using DSM-IV and
DSM-5 criteria. J Trauma Stress 2013;26:537–47.
14. Seng JS, Oakley DJ, Sampselle CM, Killion C, GrahamBermann S, Liberzon I. Posttraumatic stress disorder and pregnancy complications. Obstet Gynecol 2001;97:17–22.
15. Chang HL, Chang TC, Lin TY, Kuo SS. Psychiatric morbidity
and pregnancy outcome in a disaster area of Taiwan 921 earthquake. Psychiatry Clin Neurosci 2002;56:139–44.
28. Meux E, Stith S, Zach A. Report of results from the OSHPD
reabstracting project: an evaluation of the reliability of selected
patient discharge data, July through 1988. Sacramento (CA):
Office of Statewide Health Planning and Development, California Department of Health Services; 1990.
29. VA/DoD clinical practice guideline for the management of posttrauma disorders. Available at: http://www.healthquality.va.
gov/ptsd/cpg_PTSD-FULL-201011612.pdf. Retrieved August
27, 2013.
30. Gravely AA, Cutting A, Nugent S, Grill J, Carlson K,
Spoont M. Validity of PTSD diagnoses in VA administrative
data: comparison of VA administrative PTSD diagnoses to selfreported PTSD Checklist scores. J Rehabil Res Dev 2011;48:
21–30.
31. Frayne SM, Miller DR, Sharkansky EJ, Jackson VW, Wang F,
Halanych JH, et al. Using administrative data to identify mental
Illness: what approach is best? Am J Med Qual 2009;25:42–50.
16. Berkowitz GS, Wolff MS, Janevic TM, Holzman IR, Yehuda R,
Landrigan PJ. The World Trade Center disaster and intrauterine growth restriction. JAMA 2003;290:595–6.
32. Seng JS, Low LK, Sperlich M, Ronis DL, Liberzon I. Prevalence, trauma history, and risk for posttraumatic stress disorder
among nulliparous women in maternity care. Obstet Gynecol
2009;114:839–47.
17. Engel SM, Berkowitz GS, Wolff MS, Yehuda R. Psychological
trauma associated with the World Trade Center attacks and its
effect on pregnancy outcome. Paediatr Perinat Epidemiol 2005;
19:334–41.
33. Kimerling R, Gima K, Smith MW, Street A, Frayne S. The
Veterans Health Administration and military sexual trauma.
Am J Public Health 2007;97:2160–6.
18. Rogal SS, Poschman K, Belanger K, Howell HB, Smith MV,
Medina J, et al. Effects of posttraumatic stress disorder on pregnancy outcomes. J Affect Disord 2007;102:137–43.
19. Rosen D, Seng JS, Tolman RM, Mallinger G. Intimate partner
violence, depression, and posttraumatic stress disorder as additional predictors of low birth weight infants among low-income
mothers. J Interpers Violence 2007;22:1305–14.
20. Morland L, Goebert D, Onoye J, Frattarelli L, Derauf C,
Herbst M, et al. Posttraumatic stress disorder and pregnancy
health: preliminary update and implications. Psychosomatics
2007;48:304–8.
34. Epidemiology Program, Post Deployment Health Strategic
Healthcare Group, Office of Public Health, Veterans Health
Administration, Department of Veterans Affairs. Analysis of
VA health care utilization among operation enduring freedom, operation Iraqi freedom, and operation New Dawn Veterans, from 1st Qtr FY 2002 through 1st Qtr FY 2012. Washington
(DC); 2012. Available at: http://www.publichealth.va.gov/docs/
epidemiology/healthcare-utilization-report-fy2012-qtr1.pdf.
Retrieved April 7, 2014.
35. Zhang J, Troendle J, Reddy UM, Laughon SK, Branch DW,
Burkman R, et al. Contemporary cesarean delivery practice in
the United States. Am J Obstet Gynecol 2010;203:326.e1–10.
21. Xiong X, Harville EW, Mattison DR, Elkind-Hirsch K,
Pridjian G, Buekens P. Exposure to Hurricane Katrina, posttraumatic stress disorder and birth outcomes. Am J Med Sci
2008;336:111–5.
36. Lydon-Rochelle MT, Holt VL, Nelson JC, Cárdenas V, Gardella C,
Easterling TR, et al. Accuracy of reporting maternal in-hospital
diagnoses and intrapartum procedures in Washington State linked
birth records. Paediatr Perinat Epidemiol 2005;19:460–71.
22. Lipkind HS, Curry AE, Huynh M, Thorpe LE, Matte T. Birth
outcomes among offspring of women exposed to the September
11, 2001, terrorist attacks. Obstet Gynecol 2010;116:917–25.
37. Yoon J, Scott JY, Phibbs CS, Wagner TH. Recent trends in
Veterans Affairs chronic condition spending. Popul Health
Manag 2011;14:293–8.
23. Seng J, Low L, Sperlich M, Ronis D, Liberzon I. Post-traumatic
stress disorder, child abuse history, birthweight and gestational
age: a prospective cohort study. BJOG 2011;118:1329–39.
38. Turchik JA, Pavao J, Hyun J, Mark H, Kimerling R. Utilization
and intensity of outpatient care related to military sexual
trauma for veterans from Afghanistan and Iraq. J Behav Health
Serv Res 2012;39:220–33.
24. Prins A, Ouimette P, Kimerling R, Camerond RP, Hugelshofer DS,
Shaw-Hegwer J, et al. The primary care PTSD screen (PC–PTSD):
development and operating characteristics. Prim Care Psychiatry
2004;9:9–14.
25. Seal KH, Bertenthal D, Miner CR, Sen S, Marmar C. Bringing
the war back home: mental health disorders among 103,788 US
1118
Shaw et al
39. Healthcare cost and utilization project (HCUP) clinical classifications software (CCS). Rockville (MD): Agency for Healthcare
Quality and Research; 2012.
40. Larson GE, Highfill-McRoy RM, Booth-Kewley S. Psychiatric
diagnoses in historic and contemporary military cohorts: combat
PTSD and Risk of Spontaneous Preterm Birth
OBSTETRICS & GYNECOLOGY
deployment and the healthy Warrior effect. Am J Epidemiol 2008;
167:1269–76.
three federally qualified health care centers. Matern Child
Health J 2014;18:1056–65.
41. Cnattingius S, Forman MR, Berendes HW, Isotalo L. Delayed
childbearing and risk of adverse perinatal outcome. A
population-based study. JAMA 1992;268:886–90.
45. Suris A. Sexual assault in women veterans: an examination of
PTSD risk, health care utilization, and cost of care. Psychosom
Med 2004;66:749–56.
42. Health Provider Systems Office of Information & Technology.
VistA Clinical Reminders User Manual, Version 2.0 Patch
PXRM*2*6. Department of Veterans Affairs; 2007. Available
at: http://www.va.gov/vdl/documents/Clinical/CPRS-Clinical_
Reminders/pxrm_2_6_um.pdf. Retrieved July 10, 2013.
46. Kessler RC. Posttraumatic stress disorder in the national comorbidity survey. Arch Gen Psychiatry 1995;52:1048–60.
43. Elixhauser A, Wier LM. Complicating conditions of pregnancy
and childbirth, 2008: statistical brief #113. Healthcare cost and
utilization project (HCUP) statistical briefs. Rockville (MD):
Agency for Health Care Policy and Research (US); 2006. Available
at: http://www.ncbi.nlm.nih.gov/books/NBK56037/. Retrieved
May 17, 2013.
44. Kim HG, Harrison PA, Godecker AL, Muzyka CN. Posttraumatic stress disorder among women receiving prenatal care at
47. Ross LE, Grigoriadis S, Mamisashvili L, Vonderporten EH,
Roerecke M, Rehm J, et al. Selected pregnancy and delivery outcomes after exposure to antidepressant medication: a systematic
review and meta-analysis. JAMA Psychiatry 2013;70:436–43.
48. Grzeskowiak LE, Gilbert AL, Morrison JL. Investigating outcomes following the use of selective serotonin reuptake inhibitors for treating depression in pregnancy: a focus on
methodological issues. Drug Saf 2011;34:1027–48.
49. Sherin JE, Nemeroff CB. Post-traumatic stress disorder: the
neurobiological impact of psychological trauma. Dialogues Clin
Neurosci 2011;13:263–78.
Maintenance of Certiļ¬cation Articles
Maintenance of Certiļ¬cation is a program of the American Board of Obstetrics
and Gynecology (ABOG; http://www.abog.org). All articles from the reading
lists for the current year will be listed on the Obstetrics & Gynecology web site
at the beginning of January, May, and August.
To access the lists, go to www.greenjournal.org and click on the “ABOG MOC II” tab.
Links to content are provided, as well as an indication of the article’s status (ie,
available by subscription only or open access).
rev 12/2014
VOL. 124, NO. 6, DECEMBER 2014
Shaw et al
PTSD and Risk of Spontaneous Preterm Birth
1119
