Racial Differences in Nontraditional Risk Factors Associated with Cardiovascular Conditions in Pregnancy Among U.S. Women Veterans

Abstract

Background:

Pregnancy-related cardiovascular (CV) conditions are important predictors of future cardiovascular disease (CVD). Nontraditional factors, such as depression and chronic stress, have been associated with CVD, but their role in pregnancy-related CVD conditions (pCVD) remains unknown. To determine the association between nontraditional factors and CV conditions in pregnancy, and to explore if this risk varies by race.

Methods:

Using data from a prospective study of pregnant women within the veterans affairs health system (COMFORT study), we described the prevalence of nontraditional factors (e.g., depression, post-traumatic stress disorder [PTSD], chronic stress) and used logistic regression to determine the association between nontraditional factors and pregnancy-related CV conditions (pre-eclampsia/eclampsia, gestational hypertension, gestation diabetes, or preterm delivery). Analyses were then stratified by race.

Results:

Among 706 enrollees, 26% had pregnancy-related CV conditions. These women had significantly higher rates of depression (62% vs. 45%, p < 0.01), anxiety (50% vs. 37%, p = 0.01), PTSD (44% vs. 29%, p < 0.01), and high stress levels before pregnancy (22% vs. 16%, p = 0.05) compared with women with normal pregnancies. Overall, these factors were not associated with increased adjusted odds of pCVD. Overall, Black women had disproportionately higher rates of prepregnancy hypertension compared with White women (22% vs. 6%, p < 0.01).

Conclusions:

Women Veterans with pCVD are a high-risk group for future CVD, with disproportionately high rates of depression, anxiety, PTSD, and chronic stress. Racial disparities exist in pregnancy-related CV risk factors, which may further compound existing racial disparities in CVD among women Veterans.

Introduction

Pregnancy-related cardiovascular (CV) conditions are important predictors of future cardiovascular disease (CVD) in women. Having a hypertensive disorder of pregnancy or gestational diabetes carries a three- to fivefold increased risk of hypertension, stroke, myocardial infarction, and heart failure for women after pregnancy.^1

–5 Roughly 7%–9% of all pregnancies are complicated by one of these pregnancy-related CV conditions, with disproportionately higher rates among Black and non-Black women of color.^6

–10

Racial disparities in CVD outcomes later in life mirror those for pregnancy-related CV conditions, as Black women develop earlier CVD and have higher CVD-related mortality compared with White women.¹¹ Better understanding of the underlying risk factors for pregnancy-related CV conditions and how they vary by race is necessary to improve maternal health, and may also have a significant downstream impact on overall CV health for women across the life span.

While traditional risk factors for the development of pregnancy-related CV conditions have been well described, including maternal age, parity, prepregnancy obesity, hypertension, and diabetes, there is growing evidence that nontraditional factors, including psychosocial stress and mental health conditions, are associated with adverse CV outcomes.^12
–14 Large, epidemiologic studies have established independent links between depression, post-traumatic stress disorder (PTSD), chronic stress, and increased risk of CVD.^15,16 This risk may be accentuated among Black women, who face higher rates of chronic stress and systemic discrimination.^17,18 Despite this evidence, the contribution of these nontraditional factors to pregnancy-related CV conditions remains largely unknown.

Women Veterans are a fast-growing, diverse population with increased exposure to these nontraditional CVD risk factors before, during, and after military service.¹⁹ Women represent >10% of the total population receiving care through the veterans affairs (VA) health system, and almost half are of reproductive age.¹⁹ Non-Hispanic women of color represent a higher percentage of the Veteran population than the general civilian population, with Black women comprising 19% of the women Veteran population compared with 12% of women nationally.²⁰ Veteran women are susceptible to high rates of military sexual trauma, PTSD, and intimate partner violence.^19,21

In addition, women Veterans who were deployed in service of Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn were found to have higher risk of gestational diabetes and hypertensive disorders of pregnancy compared with U.S. civilian women.²² Thus, women Veterans represent a unique population to explore the link between nontraditional factors, pregnancy-related CV conditions, and race.

Using data from a longitudinal, prospective cohort study of pregnant women Veterans nationwide, we described the overall prevalence of nontraditional factors, including stress, depression, PTSD, and military sexual trauma, and examined whether these nontraditional factors were associated with an increased risk of pregnancy-related CV conditions. We further explored how the distribution of these nontraditional factors and risk associated with pregnancy-related CV conditions varied by race.

Methods

Study design and data collection

“COMFORT” is a prospective cohort study of pregnant Veterans from 15 different VA medical centers across the country that aims to understand access to non-VA community prenatal care and experiences with mental health care coordination for women Veterans. Full details of COMFORT recruitment and methods are described by Mattocks et al. and Kroll-Desrosiers et al.^21,23 In brief, pregnant Veterans using VA health care benefits who met eligibility criteria were enrolled and completed a baseline telephone survey (during the second or third trimester of pregnancy) and follow-up survey (at 12-week postpartum). Medical data were obtained from the VA electronic health record (EHR).

Our sample included women who were enrolled in COMFORT between January 28, 2016 and January 29, 2020. COMFORT was approved by the Veterans Administration Central Institutional Review Board. Study data were collected and managed using REDCap.

Data collection and measures: outcomes, nontraditional and traditional risk factors

Information about the index pregnancy, including all medical complications during the pregnancy, gestational age at delivery, and infant birth weight, was collected from participant report during the postpartum follow-up survey. All visits with a primary care physician (PCP) within the VA health system between the date of delivery through 9 months after delivery were captured to assess postpartum health care utilization for all women.

Our primary outcome was defined as any pregnancy-related CV condition during the index pregnancy, which included: pre-eclampsia/eclampsia, pregnancy-related hypertension, gestational diabetes, and/or preterm delivery (<37 weeks gestation). Nontraditional risk factors were determined a priori based on their association with CVD risk, and included depression, history of PTSD, military deployment, military sexual trauma, stress before pregnancy, and unemployment status. Depressive symptoms in pregnancy were assessed during the baseline survey using the 10-item Edinburgh Postnatal Depression Scale (EPDS).²⁴

History of military sexual trauma was evaluated using the validated VA Military Sexual Trauma Screening Tool and dichotomized as yes/any (>0) or no ( = 0).^25,26 Participants were asked to rank their perceived amount of daily stress in the 12 months before delivery on a Likert-type scale (“not stressful,” “somewhat stressful,” or “very stressful”). History of PTSD was obtained using International Classification of Diseases (ICD)-9/10 codes from VA EHR data at any point before pregnancy.

Traditional risk factors included maternal factors known to be associated with increased risk of hypertensive disorders of pregnancy and/or gestational diabetes. These included maternal age and parity at the time of pregnancy, multiple gestations (e.g., twins/triplets) in index pregnancy, cigarette use, and prepregnancy hypertension, hyperlipidemia, diabetes and obesity. Cigarette use was determined as reported during the baseline survey, and categorized as previous, current/active, or never. Prepregnancy hypertension, hyperlipidemia, and diabetes were obtained from the EHR, while prepregnancy obesity was defined as body mass index >30 kg/m², calculated from patient self-report of height and weight before pregnancy or a diagnosis of obesity in EHR at the time of pregnancy.

Race and ethnicity were defined by participant self-report during the baseline survey. Race and ethnicity were collected as separate variables, and each participant self-identified both racial category and ethnic group. Race was categorized as White, Black, and non-Black women of color (Asian, Native Hawaiian or Other Pacific Islander, American Indian or Alaska Native, or Other), and was not mutually exclusive. Ethnicity was categorized as “Hispanic/LatinX” or “Not Hispanic/LatinX.”

Statistical analyses

Demographics, nontraditional and traditional risk factors, and number of postpartum visits with a PCP within the VA were compared first between those with a pregnancy-related CV condition and those without. Baseline characteristics were then stratified by race (Black vs. White) and compared with respect to case status (pCVD vs. no pCVD). We used Pearson's chi-square (or Fisher's exact for cell sizes <10) tests and Student's t-test with Satterthwaite adjustment to compare characteristics between groups; the Cochrane-Armitage trend test was used for “perceived stress level” given its ordinal nature.

Complete-case multivariable logistic regression estimated the adjusted odds of having a pregnancy-related CV condition by nontraditional factors (depression [prenatal Edinburgh Postnatal Depression Scale score], history of PTSD [yes/no], history of military deployment [yes/no], history of military sexual trauma [yes/no], employment status [employed/unemployed], and perceived stress level before pregnancy [very/somewhat/not]). We assessed for multicollinearity using a correlation matrix of all potential included factors and condition indices before running our models, and found no evidence of collinearity.

Traditional risk factors were included as covariates. We included a fixed effect for VA site. Regression models were then stratified by race (White vs. Black). Given there are known differences in rates of hypertensive disorders of pregnancy and gestational diabetes between racial groups,^9,10 sensitivity analyses were conducted comparing results of the full model stratified by Black versus White versus non-Black women of color with the model stratified by White and Black only, and did not find any significant difference. Given the small sample size of non-Black women of color with pregnancy-related CV conditions resulted in small cell sizes (n < 10) in the model, we focused on the differences between White and Black women only. Women who self-identified as both White and Black were excluded.

The sample size for the final complete-case models was reduced by 74 (12.8%) due to missing values. There was no difference in missingness between race groups (Black vs. White) except for military sexual trauma, where Black women were more likely than White women to be missing these data (7.1% vs. 2.4%, p = 0.01). There was no difference in missingness by pregnancy-related CV condition (yes vs. no). Statistical significance was determined by p-value <0.05. All analyses were conducted in SAS version 9.2 (SAS Institute, Inc., Cary, NC).

Results

A total of 706 women completed both baseline and postpartum surveys, and were included in the analysis. One-quarter (26.3%, n = 186) of women had a pregnancy-related CV condition. Preterm birth was the most common condition affecting 14.2% all pregnancies, while 9% had pre-eclampsia or pregnancy-related hypertension, and 7% were complicated by gestational diabetes.

Table 1 shows the baseline characteristics of women by outcome, comparing those with a pregnancy-related CV condition with those without. The average age of women at the time of delivery was 32 years, and women with a pregnancy-related CV condition were slightly older than those with normal pregnancies (33.3 vs. 31.7 years, p < 0.01). Overall, 19.3% (n = 136) of women identified as Hispanic/2Latina, most of whom (57.4%, n = 78) were non-Black women of color, while 40% (n = 53) identified as White and only 6.4% (n = 5) identified as Black.

Table 1.

Baseline Characteristics of Women Veterans by Pregnancy-Related Cardiovascular Condition^a

	Total (n = 706)	Pregnancy-related CV condition (n = 186, 26.3%)	No pregnancy-related CV condition (n = 520, 73.7%)	p
Demographics
Maternal age, mean ± SD (range)	32.1 ± 4.6 (20.0–49.0)	33.3 ± 4.0 (23.0–42.0)	31.7 ± 4.7 (20.0–49.0)	<0.01
Ethnicity, n (%)^b
Hispanic/Latina	136 (19.3)	37 (19.9)	99 (19.0)	0.76
Not Hispanic/Latina	567 (80.3)	147 (79.0)	420 (80.8)	0.76
Place of residence, n (%)
Urban	507 (71.8)	128 (68.8)	379 (72.9)	0.26
Rural	158 (22.4)	47 (25.3)	111 (21.3)	0.26
Currently employed, n (%)	382 (54.1)	110 (59.1)	272 (52.3)	0.09
Type of health insurance, n (%)
Private	222 (31.4)	57 (30.6)	165 (31.7)	0.67
Government	205 (29)	57 (30.6)	148 (28.5)	0.62
Service-connected disability, n (%)	520 (73.7)	142 (76.3)	378 (72.7)	0.87
History of deployment, n (%)	436 (61.8)	118 (63.4)	318 (61.2)	0.49
Postpartum health care utilization
>1 PCP visit within 9 months after delivery, n (%)	558 (79.0)	149 (80.1)	409 (78.7)	0.47
Mean number of PCP visits within 9 months after delivery, mean ± SD (range)^c	2.5 ± 2.5 (0–23)	2.7 ± 2.7 (0–23)	2.4 ± 2.4 (0–20)	0.22
Traditional risk factors
Parity, mean ± SD (range)	1 ± 1 (0–6)	1 ± 1 (0–6)	1 ± 1 (0–5)	0.59
Multiple gestations, n (%)	18 (2.5)	11 (5.9)	7 (1.3)	<0.01
Prepregnancy BMI, mean ± SD (range)	28 ± 6 (14–53)	29 ± 6 (14–49)	27 ± 5 (15–54)	<0.01
Prepregnancy diagnoses, n (%)
Obesity	268 (38.0)	98 (52.7)	170 (32.7)	<0.01
Hypertension	74 (10.5)	42 (22.6)	32 (6.2)	<0.01
Dyslipidemia	133 (18.8)	38 (20.4)	95 (18.3)	0.47
Diabetes	29 (4.1)	15 (8.1)	14 (2.7)	<0.01
Smoking status, n (%)
Active	34 (4.8)	7 (3.8)	27 (5.2)	0.55
Never	415 (58.8)	106 (57.0)	309 (59.4)
Past (quit)	251 (35.6)	71 (38.2)	180 (34.6)
Nontraditional risk factors
Prepregnancy mental health conditions
Depression	404 (57.2)	124 (66.7)	280 (53.8)	<0.01
Anxiety	332 (47.0)	99 (53.2)	233 (44.8)	0.03
PTSD	277 (39.2)	89 (47.8)	188 (36.2)	<0.01
Military sexual trauma, n (%)	378 (53.5)	109 (58.6)	269 (51.7)	0.12
Daily stress level during pregnancy, n (%)
Not stressful	235 (33.3)	49 (26.3)	186 (35.8)	0.02
Somewhat stressful	335 (47.5)	92 (49.5)	243 (46.7)
Very stressful	125 (17.7)	40 (21.5)	85 (16.3)

Bolded values are statistically significant with p-value < 0.05.

Pregnancy-related CVD condition includes self-reported pre-eclampsia or eclampsia, gestational hypertension, gestational diabetes, and/or preterm delivery. Chi-square (or Fisher's exact when cell size <10) tests were used for categorical variables and Student's t-test with Satterthwaite adjustment was used to compare continuous variables. p-Values for “Daily stress level during pregnancy” from Cochrane-Armitage trend tests.

Missing ethnicity data for three participants: two with pregnancy-related CVD and one without.

Excludes n = 1 outlier with >30 postnatal clinic visits in 9 months after childbirth.

BMI, body mass index; CV, cardiovascular; PCP, primary care physician; PTSD, post-traumatic stress disorder; SD, standard deviation.

There were no statistically significant differences between groups by ethnicity, place of residence, employment status, type of health insurance, or percentage on service-connected disability. Overall, ∼80% of all women had at least one follow-up visit with a PCP in the VA system by 9 months after delivery, with no significant difference in rates between those with a pregnancy-related CV condition versus those without (80% vs. 79%, p = 0.47).

Nontraditional factors were common among both groups (Table 1). Rates of prepregnancy mental health conditions were significantly higher among women with pregnancy-related CV conditions than those with normal pregnancies; over half (67%) had depression (vs. 54%, p < 0.01), 53% had anxiety (vs. 45%, p = 0.03), and 48% had a history of PTSD (vs. 36%, p < 0.01). Women with pregnancy-related CV conditions also reported higher stress levels before pregnancy compared with those with normal pregnancies (22% reported “very stressful” daily levels vs. 16% in normal pregnancies, p = 0.02). There was no statistically significant difference between groups in rates of military sexual trauma before pregnancy.

Similarly, women with pregnancy-related CV conditions had significantly higher rates of traditional risk factors, including a higher proportion of multiple gestations (5.9% vs. 1.3%, p < 0.01) and higher rates of prepregnancy obesity (53% vs. 33%, p < 0.01), hypertension (23% vs. 6%, p < 0.01) and diabetes (8% vs. 3%, p < 0.01) compared with women with normal pregnancies (Table 1).

Traditional and nontraditional factors by race

Rates of pregnancy-related conditions were similar across racial groups (26% overall, 25% of White women, 28% of Black women, and 28% of non-Black women of color), though the types of conditions differed by race (Table 2). Among Black women, those with pCVD had significantly higher rates of prepregnancy obesity (55% vs. 37%, p < 0.03), hypertension (43% vs. 14%, p < 0.01), diabetes (14% vs. 3%, p = 0.01), history of smoking (30% vs. 11%, p = 0.01), and depression (70% vs. 56%, p = 0.04) compared with those without pCVD (Table 3). There were no differences in rates of anxiety, PTSD, military sexual trauma, and daily stress levels between Black women with pCVD and Black women with normal pregnancies.

Table 2.

Distribution of Pregnancy-Related Cardiovascular Conditions by Race

Outcome variables, n (%)	All (n = 706)	White (n = 422, 59.8%)	Black (n = 155, 21.9%)	Non-Black women of color ^b (n = 129, 18.3%)
No pregnancy-related CV condition	520 (73.7)	316 (74.9)	111 (71.6)	93 (72.1)
Any pregnancy-related CV condition^a	186 (26.3)	106 (25.1)	44 (28.4)	36 (27.9)
Preterm birth	100 (14.2)	55 (13.0)	26 (16.8)	19 (14.7)
Pre-eclampsia and/or pregnancy-related HTN	67 (9.5)	35 (8.3)	17 (11.0)	15 (11.6)
Gestational DM	49 (6.9)	30 (7.1)	6 (3.9)	13 (10.1)
Pregnancy-related HTN	46 (6.5)	25 (5.9)	12 (7.7)	9 (7.0)
Pre-eclampsia	35 (5.0)	15 (3.6)	10 (6.5)	10 (7.8)

Pregnancy-related cardiovascular conditions are not mutually exclusive.

Includes Asian, Native Hawaiian or Other Pacific Islander, American Indian or Alaska Native, or other.

DM, diabetes mellitus; HTN, hypertension.

Table 3.

Baseline Characteristics of Women Veterans by Pregnancy-Related Cardiovascular Disease Condition and Race^a

	Black women (n = 155)			White women (n = 422)
	pCVD n = 44 (28.4%)	No pCVD n = 111 (71.6%)	p	pCVD n = 106 (25.1%)	No pCVD n = 316 (74.9%)	p
Demographics
Maternal age, mean ± SD (range)	34.1 ± 3.8 (25.0–42.0)	31.9 ± 4.9 (22.0–43.0)	0.01	32.7 ± 4.1 (23.0–41.0)	31.8 ± 4.6 (20.0–49.0)	0.07
Ethnicity, n (%)
Hispanic/Latina	0 (0.0)	5 (4.5)	0.33	13 (12.3)	40 (12.7)	0.94
Not Hispanic/Latina	43 (97.7)	106 (95.5)	0.33	92 (86.8)	276 (87.3)	0.94
Place of residence, n (%)
Urban	38 (86.4)	94 (84.7)	0.28	66 (62.3)	211 (66.8)	0.41
Rural	6 (13.6)	8 (7.2)	0.28	33 (31.1)	86 (27.2)	0.41
Currently employed, n (%)	26 (59.1)	63 (56.8)	0.68	65 (61.3)	178 (56.3)	0.40
Type of health insurance, n (%)
Private	9 (20.5)	25 (22.5)	0.70	38 (35.8)	117 (37.0)	0.76
Government	12 (27.3)	38 (34.2)	0.37	36 (34.0)	86 (27.2)	0.19
Service-connected disability, n (%)	34 (77.3)	79 (71.2)	0.44	83 (78.3)	229 (72.5)	0.24
History of deployment, n (%)	25 (56.8)	62 (55.9)	0.84	71 (67.0)	197 (62.3)	0.31
Postpartum health care utilization
>1 PCP visit within 9 months after delivery, n (%)	36 (81.8)	90 (81.1)	0.49	84 (79.2)	250 (79.1)	0.98
Mean number of PCP visits within 9 months after delivery, mean ± SD (range)†	2.4 ± 1.8 (0–7)	2.5 ± 2.1 (0–11)	0.84	2.6 ± 2.5 (0–11)	2.3 ± 2.3 (0–16)	0.23
Traditional risk factors
Parity, mean ± SD (range)	1.0 ± 1.1 (0–4)	1.1 ± 1.1 (0–5)	0.72	1.0 ± 1.1 (0–6)	0.9 ± 1 (0–5)	0.59
Multiple gestations, n (%)	5 (11.4)	0 (0.0)	<0.01	4 (3.8)	5 (1.6)	0.23
Prepregnancy BMI, mean ± SD (range)	29.7 ± 5.2 (18.5–40.7)	28.6 ± 6.0 (14.5–52.5)	0.31	28.4 ± 6.2 (13.8–48.5)	26.9 ± 5.1 (18.2–43.6)	0.01
Prepregnancy diagnoses, n (%)
Obesity	24 (54.6)	41 (36.9)	0.03	53 (50.0)	99 (31.3)	<0.01
Hypertension	19 (43.2)	15 (13.5)	<0.01	13 (12.3)	12 (3.8)	<0.01
Dyslipidemia	10 (22.7)	21 (18.9)	0.53	18 (17.0)	56 (17.7)	0.90
Diabetes	6 (13.6)	3 (2.7)	0.01	5 (4.7)	6 (1.9)	0.15
Smoking status, n (%)
Active	1 (2.3)	3 (2.7)	0.01	5 (4.7)	19 (6.0)	0.71
Never	29 (65.9)	96 (86.5)		59 (55.7)	160 (50.6)
Past (quit)	13 (29.6)	12 (10.8)		41 (38.7)	134 (42.4)
Nontraditional risk factors
Prepregnancy mental health conditions, n (%)
Depression	31 (70.4)	62 (55.9)	0.04	69 (65.1)	170 (53.8)	0.02
Anxiety	18 (40.9)	38 (34.2)	0.35	64 (60.4)	155 (49.0)	0.03
PTSD	20 (45.5)	46 (41.4)	0.53	49 (46.2)	102 (32.3)	0.01
Military sexual trauma, n (%)	25 (56.8)	51 (45.9)	0.30	61 (57.6)	167 (52.8)	0.43
Daily stress level during pregnancy, n (%)
Not stressful	12 (27.3)	33 (29.7)	0.85	28 (26.4)	118 (37.3)	0.10
Somewhat stressful	20 (45.4)	52 (46.9)		58 (54.7)	146 (46.2)
Very stressful	10 (22.7)	25 (22.5)		18 (17.0)	48 (15.2)

Bolded values are statistically significant with p-value < 0.05.

Total n = 577. Non-Black women of color (n = 129) were not included. Pregnancy-related CVD condition includes self-reported pre-eclampsia or eclampsia, gestational hypertension, gestational diabetes, and/or preterm delivery. Chi-square (or Fisher's exact when cell size <10) tests were used for categorical variables, and Student's t-test with Satterthwaite adjustment was used to compare continuous variables. p-Values for “Daily stress level during pregnancy” from Cochrane-Armitage trend tests.

Similarly, White women with pCVD had higher rates of prepregnancy obesity (50% vs. 31%, p < 0.01), hypertension (12% vs. 4%, p < 0.01), and depression (65% vs. 54%, p = 0.02) than those without pCVD, while rates of anxiety and PTSD were significantly higher among White women with pCVD than among White women with normal pregnancies (60% vs. 50%, p = 0.03; 46% vs. 33%, p = 0.01, respectively). Overall, Black women had disproportionately higher rates of prepregnancy hypertension compared with White women (22% vs. 6%, p < 0.01).

Results of the multivariable regression are shown in Table 4. Overall, there were no significant associations between nontraditional factors (depression, PTSD, history of deployment, military sexual trauma, employment status, or perceived daily stress) and adjusted odds of having a pregnancy-related CV condition in the overall population (model 1). Further, the adjusted association between nontraditional factors and pregnancy-related CV conditions did not differ substantially by race (models 2 and 3).

Table 4.

Association Between Nontraditional and Traditional Factors and Pregnancy-Related Cardiovascular Conditions by Race^a

	Model 1: Black and White women (n = 503 ^b )			Model 2: Black women (n = 128)			Model 3: White women (n = 375)
	OR	95% LCI	95% UCI	OR	95% LCI	95% UCI	OR	95% LCI	95% UCI
Nontraditional factors
Race (Black vs. White)	0.63	0.36	1.10
Depression (EPDS score, continuous)	1.03	0.98	1.07	1.10	1.01	1.20	1.01	0.96	1.06
History of PTSD	1.64	0.98	2.76	1.02	0.30	3.49	2.02	1.13	3.62
History of deployment	1.14	0.71	1.83	1.35	0.49	3.68	1.10	0.64	1.90
Military sexual trauma	0.76	0.46	1.26	0.80	0.26	2.52	0.77	0.43	1.36
Employed	0.98	0.61	1.57	0.63	0.23	1.74	1.08	0.62	1.86
Perceived stress during pregnancy
Very stressful	1.31	0.78	2.19	0.48	0.15	1.52	1.70	0.94	3.08
Somewhat stressful	1.34	0.68	2.67	0.52	0.14	1.93	1.57	0.69	3.57
Traditional factors
Maternal age (continuous)	1.05	0.99	1.10	1.10	0.98	1.24	1.05	0.98	1.11
Parity (continuous)	0.94	0.75	1.18	0.69	0.41	1.16	1.00	0.77	1.29
Multiple gestation in current pregnancy	7.70	2.24	26.48				3.74	0.90	15.62
Smoking status
Active smoking	0.56	0.18	1.70				0.48	0.14	1.61
Previous smoking (quit)	0.90	0.56	1.47				0.61	0.36	1.06
Prepregnancy hypertension	4.33	2.21	8.50	4.26	1.51	12.02	3.71	1.48	9.31
Prepregnancy dyslipidemia	0.79	0.44	1.44	1.50	0.49	4.61	0.69	0.34	1.38
Prepregnancy diabetes mellitus	3.54	1.08	11.55				1.56	0.36	6.72
Prepregnancy obesity	2.40	1.51	3.83	3.33	1.17	9.50	2.43	1.42	4.16

Bolded values are statistically significant with p-value < 0.05.

All models predict the presence of pregnancy-related CV condition using logistic regression. For each model, VA site was also included as a fixed effect, and was not found to be significant (ORs not reported here). All other variables included in the model are presented here.

The sample size for the final complete-case models was n = 503, which was reduced by 74 (12.8% of 577) due to missing values for one or more of the following variables: history of PTSD (n = 35); history of deployment (n = 6); military sexual trauma (n = 21); employed (n = 3); perceived stress during pregnancy (n = 9); age (n = 30); smoking status (n = 5); prepregnancy hypertension (n = 35); prepregnancy dyslipidemia (n = 35); prepregnancy diabetes mellitus (n = 35); and prepregnancy obesity (n = 23). There was no difference in missingness between race groups (Black vs. White) except for military sexual trauma, where Black women were more likely than White women to be missing these data (7.1% vs. 2.4%, p = 0.01). There was no difference in missingness by pregnancy-related CV condition (yes vs. no).

EPDS, Edinburgh Postnatal Depression Scale; LCI, lower confidence interval; OR, odds ratio; UCI, upper confidence interval; VA, veterans affairs.

Among Black women, depression was associated with a slightly increased odds of pregnancy-related CV condition (odds ratio [OR] 1.1, 95% confidence interval [CI] 1.01–1.2), while history of PTSD was associated with an increased odds of having a pregnancy-related CV condition only among White women (OR 2.02, 95% CI 1.13–3.62). Prepregnancy hypertension and obesity were both associated with increased odds of pregnancy-related CV conditions among both Black and White women, carrying a two- to fourfold increased odds of having a pregnancy-related CV condition.

Discussion

Using data from a nationwide, prospective cohort study of pregnant women Veterans, we found that 26% experienced a pregnancy-related CV condition, most commonly preterm delivery and pre-eclampsia/pregnancy-related hypertension. Depression, anxiety, PTSD, and high stress levels were more common among women with these conditions. Racial disparities exist in prevalence of CV risk factors among women Veterans, as Black women with pregnancy-related CV conditions experience high rates of prepregnancy obesity, hypertension, diabetes, depression, and PTSD.

There is a large and growing burden of CVD among U.S. women Veterans.²⁷ While CVD rates are highest among women >65 years old, our results highlight a looming burden of CV risk factors among younger women Veterans in their 20–40 seconds. One-quarter of all women in our study had a pregnancy-related CV condition, which can be considered a “positive stress test” equivalent, indicating an increased risk of future CVD.²⁸ On top of that, rates of traditional CVD risk factors were twice as high in this group of women before their pregnancies, with over half having obesity and one-fifth having hypertension and hyperlipidemia.

Recent work by Haskell et al. found that young Veterans are accumulating CVD risk factors quickly, with 50% of women developing at least one CVD risk factor within 5 years of separation from the military, and Black women had a greater risk of accumulating CVD risk factors than White women.²⁹ Without intervention, the burden of CV conditions during pregnancy and across the life course will likely increase.

Women with pregnancy-related CV conditions also had the highest burden of chronic stressors. While our results did not find a clear association between nontraditional factors and increased risk of CV conditions in pregnancy among all women, the sheer prevalence of depression, PTSD, history of military sexual trauma, and high self-perceived stress levels among women with pregnancy-related conditions is striking. The cumulative exposure of these psychosocial stressors likely still has a significant long-term impact on CV health of these women through increased allostatic load.^{13,15,17,30,31}

Our results highlight that there is a potential intersection between pregnancy-related CV conditions and chronic stressors, which may further compound future CV risk. The pregnancy period may be too early to see the direct effects of these factors on pregnancy-related CV conditions specifically, though the high rates of these chronic stressors and trauma, coupled with other traditional CV risk factors of obesity and hypertension, are a perfect storm of CV risk in this already high-risk population.

A component of this intersection between chronic stress and pregnancy-related CV conditions may be attributable to racial disparities in prepregnancy risk factors driven by structural racism and underlying systemic discrimination. While we did not find any major differences in the association between nontraditional risk factors and adjusted odds of pregnancy-related CV condition by race in the stratified models, Black women had disproportionately higher rates of prepregnancy hypertension than White women overall. This illustrates that racial disparities in traditional CV risk factors already exist among young women Veterans before pregnancy.

A systematic review by the VA Women's Health Disparities Research working group in 2016 found large gaps in women's health care as a whole in the VA system, with no studies focusing on racial differences in pregnancy or CV care for women.³² Thus, there is a critical need to recognize the role of these different CV risk factors for women of color, including pregnancy-related CV conditions, to begin to address them early and prevent persistent racial disparities in CVD among women.

Rates of postpartum follow-up with a PCP within the VA system were higher among Veteran women with pregnancy-related CV conditions, particularly among Black women, than in previously reported studies of civilian women. Previous non-Veteran studies have reported that only 57%–63% of women with high-risk pregnancies have a visit with a PCP by 12 months postpartum, thus representing a missed opportunity for early chronic disease prevention.^33
–35

Over three-quarters of women with a pregnancy-related CV condition in our study had at least one VA primary care follow-up visit within 9 months of delivery. This represents a critical window for policymakers and providers to focus on retaining these high-risk women in longitudinal care for early CV risk mitigation. The VA has a unique opportunity to provide innovative postpartum care that addresses women CVD health and social determinants of health. Given all Veteran women receive their pregnancy-related care outside of the VA in community-based practices, VA providers have a clear window to recognize and intervene in these pregnancy-related CV conditions early in the peripartum period when women return to the VA system.

Unique interventions that address how to integrate women's health services across the health system need to be explored, such as integrating maternal health care into existing postpartum services like lactation consultations and pregnancy-centering groups, buttressing support for mental health and psychosocial needs, and expanding telehealth options for contraception counseling and interpregnancy planning, need to be explored to ultimately reduce impending CV risk and improve racial disparities in CVD outcomes among young women.

Limitations

We used data from the largest cohort of pregnant women Veterans across the United States, yet we had a relatively small sample size of women with the outcome in our sample, which may have led to underpowering of our outcomes, particularly with respect to racial differences. The use of a composite outcome could have obscured the associations of interest, as the four conditions included in the outcome have heterogeneous risk factors (e.g., cigarette smoking decreases the risk of pre-eclampsia but increases the risk of preterm delivery) and racial/ethnic predispositions. The outcome variable and risk factors included in our analyses were collected directly from participants, and may be susceptible to self-report and recall biases.

While we recognize that race is a proxy variable for social, political, or marginalized experiences in the United States, as opposed to a biologic or genetic measurement, we present it as a method for understanding health inequities by pregnancy outcome status, so that we can begin to understand our population subgroups and address the complex challenges hidden within this variable. Despite these limitations, this study is the first to investigate the role of nontraditional factors in pregnancy-related CV conditions among women Veterans and included a racially diverse population representative of the overall women Veteran population.

Conclusions

One-quarter of women Veterans develop a CV condition during pregnancy, and have disproportionately high rates of depression, anxiety, PTSD, and chronic stress. Notable racial disparities exist in the distribution of traditional CV risk factors before pregnancy among women Veterans, which may further compound existing racial disparities in both pregnancy-related CV outcomes and future CVD among women Veterans. High rates of early postpartum follow-up in the VA system indicate an indelible opportunity for innovative approaches to individualize and optimize CVD risk among young women Veterans going forward which may be critical to reducing the existing racial disparities in CVD risk among the fast-growing population of women Veterans.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Dr. Shepherd-Banigan was supported by the U.S. Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development a VA HSR&D Career Development Award (CDA 17-006). The remaining authors have no financial conflicts of interest.

References

Behrens

, Basit

, Lykke

, et al. Association between hypertensive disorders of pregnancy and later risk of cardiomyopathy. JAMA, 2016; 315:1026–1033.

Hermes

, Tamsma

, Grootendorst

, et al. Cardiovascular risk estimation in women with a history of hypertensive pregnancy disorders at term: A longitudinal follow-up study. BMC Pregnancy Childbirth, 2013; 13:126.

Ray

, Vermeulen

, Schull

, Redelmeier

. Cardiovascular health after maternal placental syndromes (CHAMPS): Population-based retrospective cohort study. Lancet, 2005; 366:1797–1803.

Stuart

, Tanz

, Missmer

, et al. Hypertensive disorders of pregnancy and maternal cardiovascular disease risk factor development: An observational cohort study. Ann Intern Med, 2018; 169:224–232.

Kramer

, Campbell

, Retnakaran

. Gestational diabetes and the risk of cardiovascular disease in women: A systematic review and meta-analysis. Diabetologia, 2019; 62:905–914.

Breathett

, Muhlestein

, Foraker

, Gulati

. Differences in preeclampsia rates between African American and Caucasian women: Trends from the National Hospital Discharge Survey. J Womens Health (2002), 2014; 23:886–893.

Fingar

, Mabry-Hernandez

, Ngo-Metzger

, Wolff

, Steiner

, Elixahuser

. Delivery hospitalizations involving preeclampsia and eclampsia, 2005–2014. HCUP statistical brief #222. Rockville, MD: Agency for Healthcare Research and Quality, 2017.

Petersen

, Davis

, Goodman

, et al. Racial/ethnic disparities in pregnancy-related deaths—United States, 2007–2016. Morb Mortal Wkly Rep, 2019; 68:762–765.

Casagrande

, Linder

, Cowie

. Prevalence of gestational diabetes and subsequent Type 2 diabetes among U.S. women. Diabetes Res Clin Pract, 2018; 141:200–208.

10.

Dabelea

, Snell-Bergeon

, Hartsfield

, Bischoff

, Hamman

, McDuffie

. Increasing prevalence of gestational diabetes mellitus (GDM) over time and by birth cohort. Diabetes Care, 2005; 28:579.

11.

Benjamin

, Virani

, Callaway

, et al. Heart disease and stroke statistics-2018 update: A report from the American Heart Association. Circulation, 2018; 137:e67–e492.

12.

Bairey Merz

, Dwyer

, Nordstrom

, Walton

, Salerno

, Schneider

. Psychosocial stress and cardiovascular disease: Pathophysiological links. Behav Med, 2002; 27:141–147.

13.

Gebreab

, Diez Roux

, Brenner

, et al. The impact of lifecourse socioeconomic position on cardiovascular disease events in African Americans: The Jackson Heart Study. J Am Heart Assoc, 2015; 4:e001553.

14.

Rozanski

, Blumenthal

, Kaplan

. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation, 1999; 99:2192–2217.

15.

Cohen

, Edmondson

, Kronish

. State of the art review: depression, stress, anxiety, and cardiovascular disease. Am J Hypertens, 2015; 28:1295–1302.

16.

Remch

, Laskaris

, Flory

, Mora-McLaughlin

, Morabia

. Post-traumatic stress disorder and cardiovascular diseases. Circ Cardiovasc Qual Outcomes, 2018; 11:e004572.

17.

Kalinowski

, Taylor

, Spruill

. Why are young black women at high risk for cardiovascular disease?. Circulation, 2019; 139:1003–1004.

18.

Felix

, Lehman

, Nolan

, et al. Stress, resilience, and cardiovascular disease risk among black women. Circ Cardiovasc Qual Outcomes, 2019; 12:e005284.

19.

Frayne

, Phibbs

, Saechao

, et al. Sourcebook: Women Veterans in the Veterans Health Administration. Volume 4: Longitudinal Trends in Sociodemographics, Utilization, Health Profile, and Geographic Distribution. Washington DC: Women's Health Evaluation Initiative, Women's Health Services, Veterans Health Administration, Department of Veterans Affairs, February. 2018.

20.

Women Veterans Report: The Past, Present, and Future of Women Veterans. Wasington, DC: National Center for Veterans Analysis and Statistics. Department of Veterans Affairs, February 2017.

21.

Kroll-Desrosiers

, Crawford

, Moore Simas

, Clark

, Bastian

, Mattocks

. Rates and correlates of depression symptoms in a sample of pregnant veterans receiving veterans health administration care. Womens Health Issues, 2019; 29:333–340.

22.

Katon

, Mattocks

, Zephyrin

, et al. Gestational diabetes and hypertensive disorders of pregnancy among women veterans deployed in service of operations in Afghanistan and Iraq. J Womens Health (2002), 2014; 23:792–800.

23.

Mattocks

, Baldor

, Bean-Mayberry

, et al. Factors impacting perceived access to early prenatal care among pregnant veterans enrolled in the department of veterans affairs. Womens Health Issues, 2019; 29:56–63.

24.

Cox

, Holden

, Sagovsky

. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry, 1987; 150:782–786.

25.

Kimerling

, Gima

, Smith

, Street

, Frayne

. The Veterans Health Administration and military sexual trauma. Am J Public Health, 2007; 97:2160–2166.

26.

McIntyre

, Butterfield

, Nanda

, et al. Validation of a trauma questionnaire in veteran women. J Gen Intern Med, 1999; 14:186–189.

27.

Han

, Yano

, Watson

, Ebrahimi

. Cardiovascular care in women veterans. Circulation, 2019; 139:1102–1109.

28.

Brown

, Warner

, Gianos

, et al. Promoting risk identification and reduction of cardiovascular disease in women through collaboration with obstetricians and gynecologists: A presidential advisory from the American Heart Association and the American College of Obstetricians and Gynecologists. Circulation, 2018; 137:e843–e852.

29.

Haskell

, Brandt

, Burg

, et al. Incident cardiovascular risk factors among men and women veterans after return from deployment. Med Care, 2017; 55:948–955.

30.

Wamala

, Lynch

, Kaplan

. Women's exposure to early and later life socioeconomic disadvantage and coronary heart disease risk: The Stockholm Female Coronary Risk Study. Int J Epidemiol, 2001; 30:275–284.

31.

O'Rand

. Cumulative processes in the life course. In: Elder

Jr. and Giele

, eds. The craft of life course research. Washington, DC: The Guilford Press, 2009:121–140.

32.

Carter

, Borrero

, Wessel

, Washington

, Bean-Mayberry

, Corbelli

. Racial and ethnic health care disparities among women in the veterans affairs healthcare system: A systematic review. Womens Health Issues, 2016; 26:401–409.

33.

Levine

, Nkonde-Price

, Limaye

, Srinivas

. Factors associated with postpartum follow-up and persistent hypertension among women with severe preeclampsia. J Perinatol, 2016; 36:1079–1082.

34.

Bennett

, Chang

, Levine

, et al. Utilization of primary and obstetric care after medically complicated pregnancies: An analysis of medical claims data. J Gen Intern Med, 2014; 29:636–645.

35.

Bernstein

, Quinn

, Ameli

, et al. Follow-up after gestational diabetes: A fixable gap in women's preventive healthcare. BMJ Open Diabetes Res Care, 2017; 5:e000445.