Abstract
Objectives:
This study aimed to characterize cardiovascular risk profiles in an obstetric population using American Heart Association’s Life’s Essential 8 components and investigate which combinations are associated with hypertensive disorders of pregnancy (HDP).
Study Design:
Secondary analysis of a prospective observational cohort, Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be (nuMoM2b).
Main Outcome Measures:
We used latent class analysis to identify cardiovascular risk classes using Life’s Essential 8 indicators at 6–14 weeks’ gestation: blood pressure (BP), body mass index (BMI), and self-reported diet, sleep, smoke exposure, and physical activity. We assessed predictors of class membership and odds of HDP with logistic regression.
Results:
We identified a four-class solution with the following interpretative labels: (1) Healthy Behaviors/Normal BMI & BP (42%), (2) No Physical Activity/Normal BMI & BP (31%), (3) High Snoring/Elevated BMI (18%), and (4) Poor Behaviors/Normal BMI & BP (8%). Compared with the Healthy Behaviors/Normal BMI & BP class, individuals in the other risk profiles were more likely to be unmarried vs. married, Non-Hispanic Black vs. Non-Hispanic White, have government vs. commercial insurance, and report high vs. low periconceptual stress. Two risk profiles were associated with HDP: High Snoring/Elevated BMI had 2.4 times (95% CI 2.1–2.8) the odds and Poor Behaviors/Normal BMI & BP had 1.5 times (95% CI 1.2–1.8) the odds of HDP compared with Healthy Behaviors/Normal BMI & BP.
Conclusions:
Two cardiovascular risk profiles—High Snoring/Elevated BMI and Poor Behaviors/Normal BMI & BP—were linked to HDP, emphasizing the potential for personalized, multifactorial interventions targeting modifiable behaviors to reduce risk.
Keywords: BMI, healthy eating index, obesity, preeclampsia, snoring, physical activity
Introduction
Hypertensive disorders of pregnancy (HDP) complicate up to 15% of pregnancies, with about 1 in 12 women experiencing preeclampsia, one of the most severe forms [1]. The prevalence of HDP has nearly doubled, rising from 4.4% in 2010 to 9.4% in 2021 among pregnant individuals in the United States (US) [2]. These conditions pose both short and long-term risks to the mother and are associated with significant perinatal morbidity. Current therapeutic approaches to mitigate HDP focus on vigilant monitoring, antihypertensive therapy, and low dose aspirin [3, 4]. Growing evidence links the rising prevalence of HDP to lifestyle behaviors [5], alongside established risk factors like age, family history, and chronic disease [1], thus prioritizing lifestyle and wellness is essential for overall health and blood pressure management [5].
In 2022, the American Heart Association (AHA) introduced Life’s Essential 8 (LE8), a comprehensive framework developed to enhance cardiovascular health [6], comprised of eight components: diet, physical activity (PA), nicotine exposure, sleep health, body mass index (BMI), blood lipids, blood glucose, and blood pressure (BP). LE8 was designed to support cardiovascular health in the general population, but lifestyle and wellness guidelines tailored specifically for pregnancy remain limited. Several components of the framework, particularly the objective health measures– such as high BMI, chronic hypertension, and diabetes mellitus – are well-established independent risk factors for HDP [1]. Evidence connecting lifestyle behaviors during pregnancy to the development of HDP, however, is limited [7–9], especially in early pregnancy when prevention is paramount.
Cardiovascular risk factors are often studied in an isolated “variable-centered” as opposed to “person-centered” approach. Lifestyle and health variables often interact within an individual’s overall profile and rarely act alone. Examining co-occurring clinical and behavioral cardiovascular health patterns may reveal meaningful risk profiles, guiding personalized counseling and interventions for periconceptual and pregnant individuals. The objectives of this study were to: (1) identify lifestyle and clinical cardiovascular risk profiles based on LE8’s components measured in early pregnancy and (2) investigate which profiles are associated with HDP.
Methods
Data source
We conducted a secondary analysis of publicly available data from the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-To-Be (nuMoM2b) [10], a large, multicenter prospective cohort study conducted at 8 US medical centers from 2010 to 2013 [11]. Enrollment eligibility included pregnant women with a live singleton gestation, no previous pregnancy beyond 20 weeks’ gestation, and gestational age 6 weeks 0 days to 13 weeks 6 days at recruitment. Data were collected via multiple sources, including in-person interviews, self-reported surveys, and medical record review. Participants completed three study visits with trained research personnel, with visit 1 occurring between 6 weeks 0 days and 13 weeks 6 days of gestation. Trained and certified chart abstractors reviewed the medical records of all participants and recorded final birth outcomes. Full details of the study protocol have been published previously [11]. This study was deemed exempt from our Institutional Review Board approval as it involved de-identified data analysis of publicly available data with minimal risk to participants. All study procedures complied with ethical guidelines and regulatory requirements. The reporting of this study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cohort studies [12].
Study sample
We excluded participants with pre-existing diabetes mellitus and chronic hypertension, given their higher baseline risk for HDP [13]. We also excluded patients with a BP reading ≥140/90 mm Hg or missing BP data at study visit 1 to isolate cases of incident hypertension onset during pregnancy. We excluded patients whose gestational age at time of pregnancy outcome was <20 weeks or unknown, as HDP is a diagnosis only occurring ≥20 weeks’ gestation [13]. Lastly, we excluded all patients <20 years of age given differences in BMI classification for this age group [14].
Measures: Indicator variables
The indicator variables used to define latent class membership were derived from AHA’s LE8. A total of six components – BP, nicotine exposure, BMI, diet quality, sleep health, and PA – were selected. We considered diet quality, sleep health, PA, and nicotine exposure as a broad category of health behaviors (referred to as “behaviors”) and BP and BMI as objective health measures (referred to as “measures”). While blood lipids and glucose are components of LE8, these were not readily available across our population. Variables from study visit 1 were used to best approximate characteristics around the time of conception and early pregnancy.
BP was measured by trained study personnel at study visit 1. If a BP measurement ≥140/90 was obtained, the measurement was repeated after resting for 10 minutes, and the 2nd blood pressure was also recorded [11]. If more than one BP reading was reported, the highest value defined this measure. We dichotomized BP as “normal” (BP <120/<80 mm Hg) versus “elevated” (120–139/80–89 mm Hg) based on AHA guidelines for non-pregnant adults [15].
Nicotine exposure was self-reported at study visit 1. Current household exposure was assessed via self-reported answer to “In the three months prior to this pregnancy, how many people (excluding yourself) smoked cigarettes inside the home where you lived?” [11]. We categorized overall nicotine exposure into three groups: (1) current use or current household exposure, (2) former use with no current household exposure, and (3) never used and no current household exposure.
Pre-pregnancy body mass index (BMI) was calculated based on self-reported pre-pregnancy weight and height [11]. Pre-pregnancy BMI was categorized based on WHO classes: BMI <25 (underweight/normal weight), BMI 25–29.9 (overweight), BMI 30–39.9 (class I and II obesity), and BMI ≥40 (class III obesity) [14].
We used snoring as the indicator for sleep health given the association between sleep-disordered breathing and hypertension [9, 16]. Snoring status was dichotomized into “snorers” and “non-snorers,” which was assessed via question, “Do you snore?” [11].
Periconceptional dietary quality was assessed using the Healthy Eating Index-2010 (HEI-2010), a validated predictor of disease risk, which provides a summary score of overall diet quality [17]. Participants completed the validated Modified Block 2005 Food Frequency Questionnaire [18, 19], and the results determined the HEI-2010 score [17]. Higher scores indicate greater compliance to recommended intakes of food groups or nutrients. Participants were divided into quartiles based on the HEI-2010 scores, with the fourth quartile (Q4) indicating highest diet quality and first quartile (Q1) the poorest diet quality.
PA was evaluated through standardized questions adapted from the Behavioral Risk Factor Surveillance System (BRFSS) at study visit 1 [11]. Participants reported the frequency, duration and distance of the three activities they engaged in most frequently over the past month. Based on recommendations for exercise during pregnancy and postpartum [7], three categories were defined: (1) those engaging in 0 minutes of at least moderate-intensity PA per week (referred to as ‘no PA’), (2) those engaging in up to 150 minutes per week (‘inadequate’), and (3) those completing ≥150 minutes per week (‘adequate’).
Measures: Maternal demographic, psychosocial, and pregnancy characteristics
Baseline demographic and pregnancy characteristics were assessed at study visit 1 via detailed interview, including self-reported race/ethnicity, insurance, and federal poverty level. Medical history and comorbidities, defined based on the previously developed and validated Obstetric Comorbidity Index [20], were both self-reported at each study visit and confirmed via trained, certified chart reviewers at each study site [11].
Psychosocial measures were self-reported using validated scales for stress, depression, anxiety, and social support on visit 1 questionnaires [21–24]. Cohen 10-item Perceived Stress Scale (PSS10) [21] score defined low stress (0–13), moderate stress (14–26), and high stress (24–40). = Edinburgh Postnatal Depression Scale (EPDS) depression screen was considered positive if the total score was ≥10 or if the participant endorsed suicidal ideation [24, 25]. Spielberger Trait-Anxiety Inventory (STAI) [22] score defined no or low anxiety (20–37), moderate anxiety (38–44), and high anxiety (45–80). Social support was assessed with the Multidimensional Scale of Perceived Social Support (MSPSS) [23]. Because the MSPSS scores were highly skewed, participants were grouped into quartiles. The lowest quartile (Q1), comprising the bottom 25% of scores, represent individuals with the lowest perceived social support.
Measures: Primary Outcome
The primary outcome was HDP. For the purposes of this study, HDP included gestational hypertension, preeclampsia with and without severe features and eclampsia. Gestational hypertension and preeclampsia were defined clinically per ACOG recommendations: new onset maternal hypertension (≥140/90 mmHg) ≥20 weeks’ gestation with or without proteinuria, and/or symptomatic or laboratory evidence of end organ dysfunction [13].
Statistical Analysis
We compared baseline (study visit 1) maternal demographic, pregnancy characteristics and lifestyle factors including LE8 indicator variables between those who developed HDP and those who did not, using chi-square and Kruskal-Wallis tests where appropriate.
We used latent class analysis (LCA) to identify unobserved distinct cardiovascular health profiles based on LE8 components among pregnant individuals. This finite mixture modeling approach assigns each participant to the latent class they have the highest probability of membership [26], enabling the empirical delineation of cardiovascular health profiles prevalent in this population. LCA is particularly useful for this method due to the multidimensional nature of cardiovascular health, which is not easily captured by traditional single-dimensional models [27]. All LCA models were fit using poLCA in R [28], which allows individuals with missing indicator data to remain in the analysis using the expectation-maximization algorithm.
We used an iterative model fitting process to identify the optimal number of latent classes for the sample. Models with one through seven latent classes were fit to the six indicator variables. To determine the most appropriate representation of subgroup structure within the data, multiple model fit indices were compared: the loglikelihood, the G2 fit statistic, AIC, BIC, entropy. We also considered the relative size of latent classes (>3–5%), and separation (the extent each latent class is characterized by a unique response profile) [26]. The posterior probabilities for each class were extracted to quantify the likelihood of each participant’s class membership, with participants assigned to the class with the highest probability.
Second, to characterize each latent class, we examined the model’s item response probabilities (ρ), which represent the likelihood of a particular response to an observed characteristic (indicator variable), conditional on membership in a given class. We then assigned interpretative labels characterizing each latent class based on the overall pattern of item response probabilities across the indicator variables [26]. Additionally, gamma parameters (γ) were used to describe the relative size of each latent class to provide insights into the prevalence of each profile within the cohort.
Third, we used logistic regression to assess likelihood of class membership by participant baseline characteristics, as well as to examine the association between latent class membership and the primary outcome, HDP. Coefficients and corresponding 95% confidence intervals (95% CIs) are expressed as odds ratios (OR) representing the probability of class membership.
Finally, while our latent class approach accommodated missing data on indicator variables, we conducted a sensitivity analysis restricting the cohort to individuals with complete data to examine whether the optimal class structure differed among cases with complete data.
P values < 0.05 were considered significant. All statistical analyses were conducted using de-identified data in R (version 4.3.1) [28–30].
Results
Participant characteristics
The publicly available dataset of the nuMoM2b cohort included 9,289 participants, of whom 6,304 (67.9%) were eligible for inclusion in this analysis (Figure 1). Maternal demographic, psychosocial, and pregnancy characteristics are shown in Table 1. Overall, 24.5% (n=1,542) of our analytic dataset developed HDP, 20.9% (n=1,320) had gestational hypertension, 3.5% (n=219) had preeclampsia with and without severe features, and 0.05% had eclampsia (n=3). Patients with HDP were more likely to be older than 35 years at study visit 1, be non-Hispanic Black, unmarried, and have a maximum education level of high school (all p <0.05) relative to those who did not develop HDP. Patients who developed HDP were also more likely to have excessive gestational weight gain, as well as a low or moderate STAI score (all p<0.05).
Figure 1.
Study cohort flow chart. BMI: Body mass index
Table 1.
Baseline demographic and pregnancy characteristics and their associations with HDP, n=6,304
| No hypertensive disorders of pregnancy (N=4762, 75.5%) | Hypertensive disorders of pregnancy (N=1542, 24.5%) | P-Value | |||||
|---|---|---|---|---|---|---|---|
| N | Median (IQR) | Range | N | Median (IQR) | Range | ||
| Gestational age (weeks) at screening | 4762 | 12.0 (11.0–13.0) | 5.0–13.0 | 1542 | 12.0 (11.0–13.0) | 6.0–13.0 | 0.011 |
| Pre-pregnancy body mass index | 4620 | 23.5 (21.3–27.3) | 11.5–64.7 | 1505 | 25.8 (22.6–30.9) | 16.5–58.2 | < 0.001 |
| Mean (SD) | Mean (SD) | ||||||
| Maternal age (years) at study visit 1 | 28.0 (4.9) | 28.2 (5.2) | 0.200 | ||||
| No. (%) | No. (%) | ||||||
| Age ≥ 35 years at study visit 1 | 454 (9.5%) | 186 (12.1%) | 0.004 | ||||
| Gestational weight gain | < 0.001 | ||||||
| Inadequate | 807 (17.9%) | 159 (10.7%) | |||||
| Adequate | 1203 (26.6%) | 263 (17.6%) | |||||
| Excessive | 2510 (55.5%) | 1069 (71.7%) | |||||
| Race/Ethnicity | < 0.001 | ||||||
| Non-Hispanic White | 3158 (66.3%) | 1021 (66.2%) | |||||
| Non-Hispanic Black | 443 (9.3%) | 225 (14.6%) | |||||
| Hispanic/Latina | 782 (16.4%) | 169 (11.0%) | |||||
| Asian | 182 (3.8%) | 52 (3.4%) | |||||
| Other | 197 (4.1%) | 75 (4.9%) | |||||
| Insurance Type | 0.10 | ||||||
| Commercial | 3519 (74.3%) | 1125 (73.4%) | |||||
| Government supported | 1044 (22.0%) | 365 (23.8%) | |||||
| Self/other | 173 (3.7%) | 42 (2.7%) | |||||
| Federal Poverty Category | 0.89 | ||||||
| >200% of federal poverty level | 3076 (74.8%) | 968 (74.5% | |||||
| 100–200% of federal poverty level | 556 (13.5%) | 174 (13.4%) | |||||
| < 100% of federal poverty level | 479 (11.7%) | 158 (12.2%) | |||||
| Marital status | 0.010 | ||||||
| Married | 3239 (68.1%) | 993 (64.6%) | |||||
| Single/Divorced/Widowed | 1517 (31.9%) | 545 (35.4%) | |||||
| Maternal Education Status | 0.035 | ||||||
| No more than high school | 561 (11.8%) | 213 (13.8%) | |||||
| More than high school | 4196 (88.2%) | 1328 (86.2%) | |||||
| Maternal medical comorbidity | 1914 (40.4%) | 669 (43.4%) | 0.033 | ||||
| PSS10 Categories | 0.48 | ||||||
| Low perceived stress | 2926 (61.9%) | 946 (61.7%) | |||||
| Moderate perceived stress | 1691 (35.8%) | 542 (35.4%) | |||||
| High perceived stress | 110 (2.3%) | 44 (2.9%) | |||||
| EPDS category | 0.12 | ||||||
| Negative screening | 3997 (86.7%) | 1334 (88.3%) | |||||
| Positive screening | 611 (13.3%) | 177 (11.7%) | |||||
| MSPSS quartile | 0.09 | ||||||
| Low perceived social support (Q1) | 897 (20.8%) | 329 (23.8%) | |||||
| Low-moderate perceived social support (Q2) | 1012 (23.4%) | 325 (23.6%) | |||||
| High-moderate perceived social support (Q3) | 1050 (24.3%) | 312 (22.6%) | |||||
| High perceived social support (Q4) | 1357 (31.4%) | 414 (30.0%) | |||||
| STAI category | 0.007 | ||||||
| Low anxiety | 104 (2.4%) | 53 (3.9%) | |||||
| Moderate anxiety | 1784 (42.0%) | 592 (43.6%) | |||||
| High anxiety | 2357 (55.5%) | 713 (52.5%) | |||||
IQR: Interquartile Range; SD: Standard Deviation; EPDS: Edinburgh Postnatal Depression Scale; MSPSS: Multidimensional Scale of Perceived Social Support; PSS10: Perceived Stress Scale-10; STAI: Spielberger Trait-Anxiety Inventory
Table 2 shows LE8 indicator variables among individuals who did and did not develop HDP. Characteristics associated with increased HDP incidence included increasing pre-pregnancy BMI, elevated BP, poorer diet quality, snoring, and tobacco exposure (p<0.001 for all). Adequate PA was not associated with HDP risk (p=0.669).
Table 2.
Prevalence of latent class indicator variables according to hypertensive disorders of pregnancy, n=6,304
| No hypertensive disorders of pregnancy (N=4762, 75.5%) No. % |
Hypertensive disorders of pregnancy (N=1542, 24.5%) No. % |
P-Value | |
|---|---|---|---|
| Prepregnancy body mass index | < 0.001 | ||
| Normal/underweight (<25.0) | 2851 (61.7%) | 682 (45.3%) | |
| Overweight (25.0–29.9) | 1040 (22.5%) | 388 (25.8%) | |
| Class I&II Obesity (30.0 – 39.9) | 421 (9.1%) | 230 (15.3%) | |
| Class III Obesity (≥40) | 308 (6.7%) | 205 (13.6%) | |
| Blood pressure | < 0.001 | ||
| Normal | 3673 (78.8%) | 1003 (65.9%) | |
| Elevated | 987 (21.2%) | 520 (34.1%) | |
| Healthy Eating Index-2010 quartile | 0.012 | ||
| Least healthy diet (Q1) | 981 (24.1%) | 369 (27.8%) | |
| Low-moderate healthy diet (Q2) | 1012 (24.8%) | 337 (25.4%) | |
| High-moderate healthy diet (Q3) | 1026 (25.2%) | 324 (24.4%) | |
| Most healthy diet (Q4) | 1054 (25.9%) | 296 (22.3%) | |
| Physical activity | 0.67 | ||
| None | 1858 (45.9%) | 628 (47.3%) | |
| Inadequate | 501 (12.4%) | 158 (11.9%) | |
| Adequate | 1692 (41.8%) | 543 (40.9%) | |
| Snoring status | < 0.001 | ||
| Snorer | 882 (29.2%) | 415 (42.2%) | |
| Non-snorer | 2138 (70.8%) | 568 (57.8%) | |
| Nicotine use and household exposure | < 0.001 | ||
| Current use or exposure | 433 (9.1%) | 195 (12.7%) | |
| Current use only | 149 (3.1%) | 57 (3.7%) | |
| Household exposure only | 208 (4.4%) | 105 (6.8%) | |
| Both current use and exposure | 76 (1.6%) | 33 (2.1%) | |
| Former use and no exposure | 1511 (31.8%) | 570 (37.0%) | |
| Never use and no exposure | 2808 (59.1%) | 774 (50.3%) |
Latent class models
Fit indices for models ranging from 1 to 7 classes are presented in Supplementary Table 1. The 4-class model emerged as the most optimal solution, indicated by its comparably low BIC value despite variation in other fit indices across metrics. Additionally, this model offered a balance between statistical fit, parsimony, class separation, relative size, and clinical interpretability.
Latent Class Membership
The four classes identified by LCA are shown in Figure 2, with item response probabilities (ρ) for observed indicator variables within each latent class. The largest class, Healthy Behaviors/Normal BMI & BP (42.9%) was characterized by normal BP and BMI along with overall healthy behaviors, including the highest probability of adequate PA (77.5%) and non-snorers (77.1%). The second-largest class, No PA/Normal BMI & BP (30.8%), also exhibited predominantly normal BP and BMI but was distinguished by the highest percentage of no PA (87.3%), moderate diet quality, and no current nicotine use or exposure (0.2%). The High Snoring/Elevated BMI class (17.8%) was defined by the highest percentage of snorers (68.5%) and no normal/underweight BMI (0%). Finally, the smallest class, Poor Behaviors/Normal BMI & BP (8.4%), combines poor health behaviors—including current nicotine use or exposure (71.9%), limited PA, and poor diet quality—but with objective health measures (BP, BMI) classified mostly within the normal range.
Figure 2.
Item response probabilities for observed indicator variables within latent classes, n=6,304. Footnote: Bolded horizontal lines represent each of the four latent classes across the six indicator variables. Heat map shading represents the item response probabilities (i.e., the likelihood an individual assigned to the latent class had a given attribute), with the darker the shaded representing a higher likelihood of participant endorsement. The n; % reported for each latent class on the vertical axis indicates the size and prevalence of each latent class in the sample. BMI: body mass index; BP: blood pressure; HEI-2010: Healthy Eating Index-2010 (Q1: Least healthy diet, Q2: Low-moderate healthy diet, Q3: High-moderate healthy diet, Q4: Most healthy diet); PA: Physical Activity
Predictors of Latent Class Membership
Table 3 illustrates the influence of baseline sociodemographic, medical, and psychosocial non-indicator characteristics on the likelihood of class membership relative to the Healthy Behaviors/Normal BMI & BP class. Compared with this reference class, individuals in the other three classes had higher odds of being non-Hispanic-Black, live below 100% of federal poverty level, have government insurance, be unmarried, and have maximum education of high school (Table 3). Furthermore, individuals in the non-reference classes were more likely to have high perceived stress, a positive depression screen, and low social support.
Table 3.
Likelihood of class assignment based on participant baseline characteristics, expressed as odds ratios, n=6,304
| Poor behaviors/Normal BMI & BP (n= 532, 8.4%) | High snoring/Elevated BMI (n=1,123, 17.8%) | No PA/Normal BMI & BP (n=1,944, 30.8 %) | Healthy behaviors/Normal BMI & BP (n=2,705, 42.9%) | |
|---|---|---|---|---|
| Age at study visit 1 | ||||
| < 35 years | ref | ref | ref | ref |
| ≥ 35 years | 0.4 (0.3, 0.6) | 0.9 (0.8, 1.2) | 0.7 (0.5, 0.8) | 1.0 |
| Race/Ethnicity | ||||
| Non-Hispanic White | ref | ref | ref | ref |
| Non-Hispanic Black | 8.6 (6.5, 11.5) | 6.3 (4.9, 8.1) | 4.3 (3.4, 5.5) | 1.0 |
| Hispanic/Latina | 1.4 (1.0, 2.0) | 2.5 (2.0, 3.1) | 3.8 (3.2, 4.5) | 1.0 |
| Asian | 0.2 (0.1, 0.6) | 0.6 (0.4, 0.9) | 1.5 (1.1, 2.0) | 1.0 |
| Other | 1.7 (1.1, 2.6) | 1.6 (1.2, 2.2) | 1.0 (0.8, 1.4) | 1.0 |
| Insurance Type | ||||
| Commercial | ref | ref | ref | ref |
| Government/Military | 14.0 (11.2, 17.5) | 3.7 (3.1, 4.5) | 3.8 (3.3, 4.5) | 1.0 |
| Self/other | 3.8 (2.4, 6.0) | 1.4 (1.0, 2.1) | 1.3 (1.0, 1.9) | 1.0 |
| Federal Poverty Category | ||||
| >200% of federal poverty level | ref | ref | ref | ref |
| 100–200% of federal poverty level | 7.1 (5.3, 9.7) | 3.1 (2.5, 3.9) | 2.4 (2.0, 2.9) | 1.0 |
| < 100% of federal poverty level | 18.1 (13.4, 24.4) | 3.8 (2.9, 4.9) | 3.8 (3.0, 4.8) | 1.0 |
| Marital status | ||||
| Married | ref | ref | ref | ref |
| Single/Divorced/Widowed | 21.6 (17.1, 27.3) | 4.4 (3.8, 5.1) | 3.3 (2.9, 3.8) | 1.0 |
| Maternal Education Status | ||||
| No more than high school | 20.9 (15.8, 27.6) | 6.2 (4.7, 8.0) | 5.4 (4.2, 6.9) | 1.0 |
| More than high school | ref | ref | ref | ref |
| Maternal medical comorbidity | ||||
| no | ref | ref | ref | ref |
| yes | 1.2 (1.0, 1.4) | 1.2 (1.1, 1.4) | 0.9 (0.8, 1.0) | 1.0 |
| PSS10 Categories | ||||
| Low perceived stress | ref | ref | ref | ref |
| Moderate perceived stress | 3.8 (3.1, 4.6) | 1.9 (1.7, 2.3) | 1.7 (1.5, 2.0) | 1.0 |
| High perceived stress | 12.4 (7.5, 20.4) | 4.5 (2.8, 7.2) | 2.4 (1.5, 3.8) | 1.0 |
| EPDS category | ||||
| Negative screening | ref | ref | ref | ref |
| Positive screening | 3.8 (3.0, 4.8) | 2.1 (1.7, 2.6) | 1.8 (1.5, 2.1) | 1.0 |
| MSPSS quartile | ||||
| Low perceived social support (Q1) | 3.8 (2.9, 4.9) | 2.3 (1.8, 2.8) | 2.0 (1.6, 2.3) | 1.0 |
| Low-moderate perceived social support (Q2) | 1.5 (1.1, 2.0) | 1.4 (1.2, 1.7) | 1.2 (1.0, 1.4) | 1.0 |
| High-moderate perceived social support (Q3) | 1.1 (0.8, 1.4) | 1.3 (1.0, 1.6) | 1.1 (0.9, 1.3) | 1.0 |
| High perceived social support (Q4) | ref | ref | ref | ref |
| STAI category | ||||
| Moderate | 0.3 (0.2, 0.4) | 0.4 (0.2, 0.6) | 0.4 (0.3, 0.7) | 1.0 |
| High | 0.2 (0.1, 0.3) | 0.3 (0.2, 0.5) | 0.4 (0.2, 0.6) | 1.0 |
BMI: body mass index; BP: blood pressure; EPDS: Edinburgh Postnatal Depression Scale; MSPSS: Multidimensional Scale of Perceived Social Support; PA: physical activity; PSS10: Perceived Stress Scale-10; STAI: Spielberger Trait-Anxiety Inventory
Association with HDP,
Finally, Table 4 shows the odds of HDP by cardiovascular risk class. The High Snoring/Elevated BMI (OR 2.4, 95% CI 2.1–2.8) class and Poor Behaviors/Normal BMI & BP (OR 1.5, 95% CI 1.2–1.8) class had 2.4 and 1.5 times the odds of HDP relative to the Healthy Behaviors/Normal BMI & BP class, respectively. The No PA/Normal BMI & BP class was not associated with HDP (OR 1.0, 95% CI 0.8–1.1).
Table 4.
Risk of developing hypertensive disorders of pregnancy by class, n=6,304
| N, % | Odds ratio (95% confidence interval) | |
|---|---|---|
| Healthy behaviors/Normal BMI & BP (n=2,705, 42.9%) | 565 (20.9%) | Reference |
| No PA/Normal BMI & BP (n=1,944, 30.8 %) | 392 (20.2%) | 1.0 (0.8, 1.1) |
| High snoring/Elevated BMI (n=1,123, 17.8%) | 435 (38.7%) | 2.4 (2.1, 2.8) |
| Poor Behaviors/Normal BMI & BP (n= 532, 8.4%) | 150 (28.2%) | 1.5 (1.2, 1.8) |
BMI: body mass index; BP: blood pressure; PA: physical activity
Sensitivity analysis for missing data
As shown in Supplementary Tables 2 and 3, the prevalences of baseline characteristics and indicator variables were similar between the two cohorts. Similarly, the four classes identified in the repeated LCA were consistent with those from the primary analysis, suggesting model identification was robust to missing data. (See Supplementary Figure 1).
Discussion
Drawing on the AHA LE8 framework, we identified four distinct cardiovascular health profiles of lifestyle behaviors and health measures in early pregnancy. The Poor Behaviors/Normal BMI & BP and High Snoring/Elevated BMI profiles demonstrated significantly higher risk of HDP compared with the Healthy Behaviors/Normal BMI & BP class (OR 1.5, 95% CI 1.2–1.8 and OR 2.4, 95% CI 2.1–2.8 respectively), while the No PA/Normal BMI & BP profile was not associated with HDP (OR 1.0, 95% CI 0.8–1.1).
We selected the AHA’s LE8 as indicator variables for our model because they have been identified as actionable targets for improving cardiovascular health in the general population [6]. Recognizing that a single variable approach may not fully capture the multifactorial risk of HDP, some studies have examined combinations of variables. One common approach is the use of composite risk scores. Studies exploring the additive effect of various metabolic health factors (i.e. BMI, BP, plasma glucose, cholesterol, smoking) found that the risk of preeclampsia increased with increasing number of risk factors [31, 32]. However, the use of composite risk scores has limitations – they provide little insight into the contribution of individual risk factors, the impact of specific concurrent risk factor combinations, and potential correlations between variables.
Unlike composite scores, LCA allows for person-level classification and provides a more nuanced understanding of health and disease risk. Outside of the US, two studies applied LCA to identify subgroups of pregnant populations and their associations with HDP [33, 34]. A Dutch perinatal registry study identified four subgroups based on behavioral, psychosocial, and socioeconomic factors but did not include objective health metrics [33]. Indicators across socioeconomic and household characteristics, lifestyle factors, psychosocial characteristics, and healthcare expenditures identified five latent classes of vulnerability among pregnant people in the Netherlands [34]. While some overlapping indicators with exist, this study focused on domains of vulnerability, whereas ours specifically examined cardiovascular risk profiles.
Unlike these prior studies that incorporated socioeconomic variables to define their classes [33, 34], our study intentionally excluded them. Instead, we focused on clinically modifiable risk factors in early pregnancy, aligning with the AHA’s LE8: we first derived latent classes from modifiable behaviors and metrics, then examined how a person’s sociodemographic and psychosocial characteristics influenced the probability of their membership to a given cardiovascular risk profile. Our findings reflect well-established associations between structural and social deprivation and poor health outcomes [6]. Participants in the two high-HDP risk classes were more likely to experience social, economic and psychological vulnerability compared with those in the lower HDP risk profile, Healthy Behaviors/Normal BMI & BP, highlighting how one’s social environment and economic stability may impact the ability to incorporate healthy behaviors, especially during pregnancy.
Notably, neither of the prior studies discussed found associations between their latent classes and HDP [33, 34]. In contrast, our study demonstrated that the High Snoring/Elevated BMI and Poor Behaviors/Normal BMI & BP classes were both associated with HDP when compared with Healthy Behaviors/Normal BMI & BP. This finding underscores the multifactorial nature of HDP, suggesting that the interplay of modifiable health behaviors and objective health measures may be more predictive than broader combinations that include less modifiable socioeconomic factors.
Our study emphasizes the importance of early identification of risk factors and accompanying intervention at the first prenatal visit, a critical period when these factors may be amenable to change and prior to the development of adverse outcomes. We describe two profiles associated with HDP risk: one composed primarily of poor health behaviors (nicotine exposure, low-quality diet, minimal PA) despite normal BMI and BP, and another associated with elevated BMI and snoring, which often co-occur and may reflect underlying sleep-disordered breathing [35]. Indeed, sleep-disordered breathing, including obstructive sleep apnea, is linked to increased risk of HDP [36], and emerging evidence suggests treatment with continuous positive airway pressure may reduce this risk [37]. The 2-fold increased HDP risk among the High Snoring/Elevated BMI class underscores a clinically actionable opportunity to improve prenatal risk stratification. Incorporating screening for habitual snoring into routine pregnancy care may enable earlier identification and treatment of at-risk individuals, potentially mitigating HDP risk. This profile also had the greatest proportion of elevated BP and the strongest association with HDP. While the AHA has reduced the threshold for hypertension diagnosis in nonpregnant adults, defining stage I hypertension as 130–139/80–89 mm Hg [15], this adjustment has not yet been adapted in pregnancy. Our findings, along with prior research, suggest that use of nonpregnant hypertension definitions is associated with increased risk for HDP [38–40].
Although two risk profiles share the same disease endpoint, their distinct characteristics emphasize the need for tailored clinical management. Given the multifactorial nature of HDP, addressing all risk factors is challenging. Our framework may help clinicians prioritize impactful, modifiable changes, reinforcing the need for tailored lifestyle interventions to improve maternal health.
While the AHA’s LE8 promotes behaviors to improve cardiovascular wellbeing in the general population, no pregnancy-specific framework exists, despite the fact that less than 1 in 10 of pregnant individuals experience optimal cardiovascular health [6]. Given the unique inflammatory and metabolic changes of pregnancy, research is increasingly focused on developing tailored cardiometabolic frameworks [41]. Studying modifiable risk factors throughout pregnancy could improve HDP risk stratification and prevention. Additionally, implementation research is needed to create practical, pregnancy-specific guidelines. For instance, while both pregnant and non-pregnant adults are advised to get 150 minutes of weekly physical activity [42], many pregnant individuals struggle to meet this goal, highlighting the need for more achievable recommendations [43, 44].
When assessing sociodemographic variables, many appear to be significantly associated with class membership. Disparities in both HDP and long-term cardiovascular health risk exist across multiple domains of social strata, further influencing ability to access healthy foods, areas for PA and access to healthcare. For instance, Black women are disproportionately affected by HDP and more likely to develop long-term adverse sequelae after HDP history [45]. The extent to which lived experiences contribute to the increased risk of HDP in Black communities remains uncertain. However, it is well established that social determinants of health are fundamental drivers of health inequities. Disproportionate exposure to individual, household, and neighborhood disadvantages among Black individuals continues to fuel persistent disparities in maternal and infant outcomes [45, 46]. In order to address this impact, researchers should formulate efforts to optimize obstetric care in a sociodemographically vulnerable population to diminish maternal health disparities.
Our study has notable strengths. We intentionally selected indicator variables for our LCA model that are modifiable risk factors based on the AHA’s LE8, emphasizing actionable health metrics that can be targeted for intervention. Secondly, all variables were assessed early in pregnancy (between 6 weeks 0 days and 13 weeks 6 days of gestation), ensuring that all exposures were measured before the onset of HDP. Next, we used LCA to identify distinct health profiles described by combinations of risk factors, moving beyond single-variable approaches. We had access to detailed diet, activity, sleep, and psychosocial variables, as well as obstetric characteristics adjudicated by trained medical chart abstractors, from a large and geographically diverse prospective cohort. In contrast to the relatively homogenous Dutch sample [33, 34], our multi-center cohort across the US may be more generalizable to diverse patient populations.
However, our study is not without limitations. While our cohort was recruited from 8 academic medical centers within the US, our findings may not be generalizable to other health care settings. Additionally, the AHA’s LE8 framework has not been validated specifically for use in pregnant populations, although poor cardiovascular health as defined by these variables has been associated with increased cardiovascular risk in the general population [47, 48]. Our cohort also predates the 2014 U.S Preventive Services Task Force’s expansion of indications for aspirin use [49], limiting the applicability of our findings to more contemporary populations.
In this secondary analysis of nuMoM2b, we characterized distinct cardiovascular risk profiles in early pregnancy using the AHA’s LE8 framework. Two profiles – Poor Behaviors/Normal BMI & BP and High Snoring/Elevated BMI – were significantly associated with HDP, suggesting distinct pathways to disease risk. Identifying these high-risk profiles suggest avenues for multifactorial, personalized interventions targeting modifiable behaviors to mitigate HDP risk.
Supplementary Material
Highlights.
Four distinct cardiovascular risk profiles based upon Life’s Essential 8 components collected in early pregnancy were identified.
Two risk profiles, High Snoring/Elevated BMI and Poor Behaviors/Normal BMI & BP, were linked to increased odds of hypertensive disorders of pregnancy.
Examining early pregnancy health behaviors and measures underscores the potential for personalized interventions targeting modifiable factors to reduce the risk of hypertensive disorders of pregnancy.
Acknowledgements
The parent study was supported by grant funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD): U10 HD063036; U10 HD063072; U10 HD063047; U10 HD063037; U10 HD063041; U10 HD063020; U10 HD063046; U10 HD063048; and U10 HD063053. In addition, support was provided by Clinical and Translational Science Institutes: UL1TR001108 and UL1TR000153. We acknowledge NICHD DASH for providing the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be data that was used for this research.
Funding
This work was supported by the National Institutes of Health, Agency for Healthcare Research and Quality, and Icahn School of Medicine at Mount Sinai, K01HS029525 and UL1TR004419.
Glossary
- ACOG
American College of Obstetricians and Gynecologists
- AHA
American Heart Association
- AIC
Akaike information criterion
- BIC
Bayesian Information Criterion
- BMI
body mass index
- BP
Blood Pressure
- CI
confidence interval
- EPDS
Edinburgh Postnatal Depression Scale
- HDP
Hypertensive disorders of pregnancy
- LCA
latent class analysis
- LE8
Life’s Essential 8
- MSPSS
Multidimensional Scale of Perceived Social Support
- nuMoM2b
Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be
- OR
Odds ratio
- PA
Physical activity
- PSS10
10-item Perceived Stress Scale
- STAI
Spielberger Trait-Anxiety Inventory
- US
United States
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Ford ND, Cox S, Ko JY, et al. Hypertensive Disorders in Pregnancy and Mortality at Delivery Hospitalization - United States, 2017–2019. MMWR Morb Mortal Wkly Rep 2022;71(17):585–591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ayyash MK, McLaren R Jr., Shaman M, et al. Trends in Preeclampsia Risk Factors in the US From 2010 to 2021. Jama. 2024;332(2):167–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Burton GJ, Redman CW, Roberts JM, et al. Pre-eclampsia: pathophysiology and clinical implications. BMJ 2019;366:l2381. [DOI] [PubMed] [Google Scholar]
- 4.Force USPST, Davidson KW, Barry MJ, et al. Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: US Preventive Services Task Force Recommendation Statement. JAMA. 2021;326(12):1186–1191. [DOI] [PubMed] [Google Scholar]
- 5.Sharma G, Grandhi GR, Acquah I, et al. Social Determinants of Suboptimal Cardiovascular Health Among Pregnant Women in the United States. J Am Heart Assoc 2022;11(2):e022837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lloyd-Jones DM, Allen NB, Anderson CAM, et al. Life’s Essential 8: Updating and Enhancing the American Heart Association’s Construct of Cardiovascular Health: A Presidential Advisory From the American Heart Association. Circulation. 2022;146(5):e18–e43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Physical Activity and Exercise During Pregnancy and the Postpartum Period: ACOG Committee Opinion, Number 804. Obstet Gynecol 2020;135(4):e178–e188. [DOI] [PubMed] [Google Scholar]
- 8.Yee LM, Silver RM, Haas DM, et al. Quality of periconceptional dietary intake and maternal and neonatal outcomes. Am J Obstet Gynecol 2020;223(1):121.e121–121.e128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Facco FL, Grobman WA, Kramer J, et al. Self-reported short sleep duration and frequent snoring in pregnancy: impact on glucose metabolism. Am J Obstet Gynecol. 2010;203(2):142.e141–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Silver R Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be (Version 1). NICHD Data and Specimen Hub 2019. [Google Scholar]
- 11.Haas DM, Parker CB, Wing DA, et al. A description of the methods of the Nulliparous Pregnancy Outcomes Study: monitoring mothers-to-be (nuMoM2b). Am J Obstet Gynecol 2015;212(4):539.e531–539.e524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61(4):344–349. [DOI] [PubMed] [Google Scholar]
- 13.Gestational Hypertension and Preeclampsia: ACOG Practice Bulletin, Number 222. Obstet Gynecol 2020;135(6):e237–e260. [DOI] [PubMed] [Google Scholar]
- 14.Zierle-Ghosh A, Jan A. Physiology, Body Mass Index. StatPearls. Treasure Island (FL): StatPearls Publishing Copyright © 2025, StatPearls Publishing LLC. 2025. [PubMed] [Google Scholar]
- 15.Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13–e115. [DOI] [PubMed] [Google Scholar]
- 16.Hawkins MS, Conlon RK, Donofry S, et al. Sleep characteristics modify the associations of physical activity during pregnancy and gestational weight gain. Arch Gynecol Obstet 2023;308(1):101–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Guenther PM, Casavale KO, Reedy J, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet 2013;113(4):569–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Block G, Woods M, Potosky A, et al. Validation of a self-administered diet history questionnaire using multiple diet records. J Clin Epidemiol 1990;43(12):1327–1335. [DOI] [PubMed] [Google Scholar]
- 19.Bodnar LM, Simhan HN, Parker CB, et al. Racial or Ethnic and Socioeconomic Inequalities in Adherence to National Dietary Guidance in a Large Cohort of US Pregnant Women. J Acad Nutr Diet 2017;117(6):867–877 e863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Leonard SA, Kennedy CJ, Carmichael SL, et al. An Expanded Obstetric Comorbidity Scoring System for Predicting Severe Maternal Morbidity. Obstetrics & Gynecology 2020;136(3):440–449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cohen S, Kamarck T, Mermelstein R. A Global Measure of Perceived Stress. Journal of Health and Social Behavior. 1983;24(4):385–396. [PubMed] [Google Scholar]
- 22.Littleton HL, Breitkopf CR, Berenson AB. Correlates of anxiety symptoms during pregnancy and association with perinatal outcomes: a meta-analysis. Am J Obstet Gynecol 2007;196(5):424–432. [DOI] [PubMed] [Google Scholar]
- 23.Zimet GD, Powell SS, Farley GK, et al. Psychometric characteristics of the Multidimensional Scale of Perceived Social Support. J Pers Assess 1990;55(3–4):610–617. [DOI] [PubMed] [Google Scholar]
- 24.Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry. 1987;150:782–786. [DOI] [PubMed] [Google Scholar]
- 25.Screening and Diagnosis of Mental Health Conditions During Pregnancy and Postpartum: ACOG Clinical Practice Guideline No. 4. Obstet Gynecol 2023;141(6):1232–1261. [DOI] [PubMed] [Google Scholar]
- 26.Collins LM, Lanza ST. Latent class and latent transition analysis for applications in the social, behavioral, and health science: Wiley 2016. [Google Scholar]
- 27.Jahangiry L, Abbasalizad Farhangi M, Najafi M, et al. Clusters of the Risk Markers and the Pattern of Premature Coronary Heart Disease: An Application of the Latent Class Analysis. Front Cardiovasc Med 2021;8:707070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Linzer DA, Lewis JB. poLCA: An R Package for Polytomous Variable Latent Class Analysis. Journal of Statistical Software. 2011;42(10):1 – 29. [Google Scholar]
- 29.Zeileis A HT. Diagnostic Checking in Regression Relationships. News. 2022;2(3):7–10. [Google Scholar]
- 30.Robinson D HA, Couch S. broom: Convert Statistical Objects into Tidy Tibbles Available: https://CRAN.R-project.org/package=broom. [Google Scholar]
- 31.Lei Q, Niu J, Lv L, et al. Clustering of metabolic risk factors and adverse pregnancy outcomes: a prospective cohort study. Diabetes/Metabolism Research and Reviews. 2016;32(8):835–842. [DOI] [PubMed] [Google Scholar]
- 32.Perak AM, Lancki N, Kuang A, et al. Associations of gestational cardiovascular health with pregnancy outcomes: the Hyperglycemia and Adverse Pregnancy Outcome study. Am J Obstet Gynecol 2021;224(2):210.e211–210.e217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Crone MR, Luurssen-Masurel N, Bruinsma-van Zwicht BS, et al. Pregnant women at increased risk of adverse perinatal outcomes: A combination of less healthy behaviors and adverse psychosocial and socio-economic circumstances. Preventive medicine. 2019;127:105817–105817. [DOI] [PubMed] [Google Scholar]
- 34.Molenaar JM, van der Meer L, Bertens LCM, et al. Defining vulnerability subgroups among pregnant women using pre-pregnancy information: a latent class analysis. European Journal of Public Health. 2022;33(1):25–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Peppard PE, Young T, Palta M, et al. Longitudinal Study of Moderate Weight Change and Sleep-Disordered Breathing. JAMA. 2000;284(23):3015–3021. [DOI] [PubMed] [Google Scholar]
- 36.Facco FL, Parker CB, Reddy UM, et al. Association Between Sleep-Disordered Breathing and Hypertensive Disorders of Pregnancy and Gestational Diabetes Mellitus. Obstet Gynecol 2017;129(1):31–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Tantrakul V, Ingsathit A, Liamsombut S, et al. Treatment of obstructive sleep apnea in high risk pregnancy: a multicenter randomized controlled trial. Respir Res 2023;24(1):171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ishii K, Baba S, Ikehara S, et al. Impact of stage 1 hypertension in the first and second trimesters on adverse pregnancy outcomes: The Japan Environment and Children’s study (JECS). Pregnancy Hypertens. 2022;30:232–237. [DOI] [PubMed] [Google Scholar]
- 39.Xiao Y, Liu J, Teng H, et al. Stage 1 hypertension defined by the 2017 ACC/AHA guidelines and neonatal outcomes: Systematic review and meta-analysis. Pregnancy Hypertens. 2021;25:204–212. [DOI] [PubMed] [Google Scholar]
- 40.Greenberg VR, Silasi M, Lundsberg LS, et al. Perinatal outcomes in women with elevated blood pressure and stage 1 hypertension. Am J Obstet Gynecol 2021;224(5):521.e521–521.e511. [DOI] [PubMed] [Google Scholar]
- 41.Killeen SL, Yelverton CA, Geraghty AA, et al. The Edmonton Obesity Staging System and pregnancy outcomes in women with overweight or obesity: A secondary analysis of a randomized controlled trial. Clin Obes 2022;12(3):e12510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Piercy KL, Troiano RP, Ballard RM, et al. The Physical Activity Guidelines for Americans. Jama. 2018;320(19):2020–2028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Ussery EN, Hyde ET, Bombard JM, et al. Physical Activity Before and During Pregnancy, Colorado Pregnancy Risk Assessment Monitoring System, 2012–2015. Prev Chronic Dis 2020;17:E55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Hesketh KR, Evenson KR. Prevalence of U.S. Pregnant Women Meeting 2015 ACOG Physical Activity Guidelines. Am J Prev Med 2016;51(3):e87–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Johnson JD, Louis JM. Does race or ethnicity play a role in the origin, pathophysiology, and outcomes of preeclampsia? An expert review of the literature. Am J Obstet Gynecol 2022;226(2S):S876–S885. [DOI] [PubMed] [Google Scholar]
- 46.Tiako MJN, McCarthy C, Meisel ZF, et al. Association between Low Urban Neighborhood Greenness and Hypertensive Disorders of Pregnancy. Am J Perinatol 2023;40(11):1185–1192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Sebastian SA, Shah Y, Paul H, et al. Life’s Essential 8 and the risk of cardiovascular disease: a systematic review and meta-analysis. European Journal of Preventive Cardiology. 2024. [DOI] [PubMed] [Google Scholar]
- 48.Jin C, Li J, Liu F, et al. Life’s Essential 8 and 10-Year and Lifetime Risk of Atherosclerotic Cardiovascular Disease in China. Am J Prev Med 2023;64(6):927–935. [DOI] [PubMed] [Google Scholar]
- 49.LeFevre ML. Low-dose aspirin use for the prevention of morbidity and mortality from preeclampsia: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2014;161(11):819–826. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.