Cardiovascular health and depression among postpartum women in the United States

Abstract

Background

The postpartum period is a critical time period for both cardiovascular health (CVH) and psychosocial health among women. While depression is a recognized cardiovascular risk factor in non-pregnant populations, the association between depression and CVH in the postpartum period has been understudied.

Objectives

To examine the association between depression and CVH among reproductive-age women and whether postpartum status modifies this association.

Methods

We analyzed cross-sectional data from non-pregnant women aged 20–44 years in the National Health and Nutrition Examination Survey (NHANES) 2007–2020. We evaluated whether CVH scores varied by postpartum status and depression status (Patient Health Questionnaire-9 [PHQ-9] scores ≥10). CVH was defined using the American Heart Association’s Life’s Essential 8 framework (blood pressure, glucose, lipids, body mass index [BMI], diet, physical activity, nicotine exposure, sleep; scores 0–100). We evaluated CVH scores by postpartum and depression status and tested for effect modification, adjusted for age and sociodemographics.

Results

The study population included 7208 women, representing 47,249,817 women at the population level (weighted n = 4253,164 postpartum; n = 42,996,653 not postpartum). Most postpartum women had an overall CVH score that was moderate (67%) or high (28%). Postpartum women had significantly worse CVH scores for physical activity (48.7 vs. 54.3, p-value 0.017) and BMI (57.3 vs. 63.3, p-value 0.014) compared to non-postpartum women. Among reproductive-age women, depression status was associated with lower scores for overall CVH and for most of the individual CVH metrics. Postpartum status modified the association between depression and CVH for sleep only (p-value 0.005).

Conclusion

Postpartum women had similar overall CVH compared to reproductive-age women who were not postpartum, but lower CVH scores for physical activity and BMI. Multi-component interventions that also target depression along with health behaviors may have a greater impact on CVH promotion in reproductive-age women.

Key points.

Question: What is the relationship between depression, postpartum status, and cardiovascular health among reproductive-age women?

Findings: In this national US-based study, postpartum women had significantly worse cardiovascular health scores for physical activity and body mass index. Depression was associated with lower overall cardiovascular health, and postpartum status modified the association between depression and sleep scores.

Meaning: Multi-component interventions that target depression along with health behaviors may have a greater impact on cardiovascular health promotion in reproductive-age women.

1. Introduction

Recent studies have demonstrated concerning trends of worse cardiovascular health (CVH) among young adults in the US with projections for worsening CVH particularly in reproductive-age women [[1], [2], [3], [4]]. During early adulthood, pregnancy and the transition to parenthood represent significant life events with the potential to profoundly influence CVH. The postpartum period has increasingly been recognized as a critical time to intervene to improve long-term CVH among women [5,6]. At the same time, the peripartum period is a vulnerable time when women are at increased risk of declining psychosocial health – including exacerbations of preexisting depression and new peripartum mood disorders [7,8]. Psychosocial health is relevant to CVH outside of pregnancy, [9] but whether the postpartum period potentiates this relationship is not known.

New parenthood and the challenges of the postpartum period may influence health behaviors. On the one hand, new parents may be motivated to engage in healthy behaviors for longevity and to model healthy behavior for their children. On the other hand, the time constraints, prioritization of parental responsibilities, and stresses associated with new parenthood may contribute to suboptimal health behaviors [10,11]. Several studies have documented worsening CVH behaviors in the postpartum period, including decreased physical activity, increased dietary intake, and weight retention [[12], [13], [14]]. However, no study to date has examined CVH among postpartum women in the US using a nationally representative sample. At the same time, women in the U.S. are having pregnancies at later ages, which also has important implications for future cardiovascular health [15].

Postpartum depression has emerged as a factor associated with subsequent development of CVD. Several studies from the US, Canada, and Europe have demonstrated an independent association between perinatal mood disorders and incident CVD [[16], [17], [18], [19], [20], [21], [22]]. However, the pathways underlying this association are unknown. One potential pathway is that individuals with postpartum depression may engage in suboptimal health behaviors that affect their risk of CVD. Health behaviors have been found to be an important mechanism explaining the association between depressive symptoms and CVD in the general population but have not been well studied among women with postpartum depression [23,24]. A study using data from UK Biobank found that body mass index (BMI), but not other lifestyle factors, partially mediated the association between postpartum depression and subsequent chronic disease [17]. However, they did not examine mediators of the association between postpartum depression and cardiovascular disease. Adverse pregnancy outcomes such as preeclampsia and gestational diabetes may also play a role in the association between postpartum depression and subsequent CVD. There is growing evidence of a bidirectional relationship between peripartum depression and adverse pregnancy outcomes, which in turn are associated with future CVD risk [[25], [26], [27], [28]].

We sought to assess CVH among reproductive-age women by postpartum status using the American Heart Association’s (AHA) Life’s Essential 8 (LE8) framework and data from the National Health and Nutrition Examination Survey (NHANES) 2007–2020. We also sought to evaluate whether postpartum status modifies the association between depression and CVH among reproductive-age women. We hypothesized that postpartum women would have lower CVH behavior scores than non-postpartum women, and that postpartum status would modify the association between depression and CVH.

2. Methods

2.1. Study population

We combined cross-sectional data from the following NHANES cycles: 2007–2008, 2009–2010, 2011–2012, 2013–2014, 2015–2016, and 2017-March 2020. Women aged 20 to 44 years were included in the analysis; urine pregnancy test results are publicly released by NHANES for women aged 20–44 years only. We excluded women who were currently pregnant (confirmed by urine pregnancy test, n = 391), those missing pregnancy status (n = 467), and those who were 0–5 months postpartum (n = 226) (Supplementary Figure 1). Postpartum status was determined based on the question “How many months ago did you have your baby?” We restricted the analysis to those who were 6–24 months postpartum based on self-report at the time of the survey due to the challenges of pregnancy-related impact on CVH metrics (glucose, lipids, BMI) during the first 6 months postpartum. Hemoglobin A1c is generally not recommended for screening during the first 3 months postpartum because it is affected by increased turnover of red blood cells [29]. Cholesterol levels are also affected by pregnancy, returning to baseline levels by about 3 months postpartum [6]. While fasting glucose and triglycerides may be used [29], only a subset of participants in NHANES undergo fasting labs. Although lactation may affect lipid levels, both the AHA and National Lipid Association recommend that screening of lipids can be considered after about 12 weeks postpartum despite breastfeeding status [29]. Additionally, standard BMI cutoffs are not applicable to weight in the early postpartum period. We chose 6 months postpartum because the American College of Obstetricians and Gynecologists (ACOG) recommends reaching prepregnancy weight by 6–12 months postpartum [30].

2.2. Outcomes

We used the AHA’s LE8 to evaluate CVH metrics [29,31]. LE8 includes four CVH factors (blood pressure, glucose, lipids, BMI) and four CVH behaviors (diet, physical activity, nicotine exposure, sleep). Since all participants included in the analysis were at least 6 months postpartum, we used the LE8 CVH scoring algorithms for the general non-pregnant population (Supplementary Table 1) [31]. If any individual CVH metric was missing (such as diet or physical activity), the overall CVH score was considered missing for that participant.

2.3. Exposures and covariates

The primary exposures of interest were postpartum status and depression, which was defined as a Patient Health Questionnaire-9 (PHQ-9) score of 10 or higher [32,33]. Sociodemographic characteristics and pregnancy history were based on self-report: age, race, ethnicity, US born, education level, marital status, family income to poverty ratio, health insurance, gravidity, number of livebirths, macrosomia (baby 9 lb or more at birth), breastfeeding status, and gestational diabetes (in any previous pregnancy). Participants could select more than one type of insurance. Participants who selected Medicaid, any public insurance (including military, Indian Health service, state-sponsored health plan), a single-service plan, or a combination of public and private insurance, were categorized as public insurance. Participants who selected only private insurance were categorized as private insurance.

2.4. Analysis

We report mean CVH scores (out of 100) with 95% confidence intervals (CI) for postpartum women and non-postpartum women. We also report frequencies of high (80–100), moderate (50–79), and low CVH (0–49). We compared continuous CVH scores for postpartum women compared to women who were not postpartum using multivariable linear regression. We adjusted for sociodemographic factors that were selected a priori: age, race, ethnicity, education level, marital status, and insurance status. Due to missing data and the small number of postpartum women, we limited the covariates to key sociodemographic factors that have been demonstrated to be associated with CVH. We then looked at CVH by depression, adjusting for postpartum status, age, race, ethnicity, education level, marital status, and insurance status. To assess whether the association between depression status and CVH differed by postpartum status, we included a multiplicative interaction term between depression and postpartum status. When interaction terms were significant, we stratified the models by postpartum status. In a sensitivity analysis, we used multivariable linear regression to examine the association of CVH with depressive symptoms (continuous PHQ-9 score) and postpartum status. In another sensitivity analysis, we restricted the postpartum women to those beyond one year postpartum.

All analyses were survey-weighted to account for NHANES sampling. P-values < 0.05 were considered significant. A publicly available SAS macro was used to calculate the Healthy Eating Index scores for the diet CVH metric. Stata (version 19.5) was used for the analyses. This study utilizes publicly available, de-identified data from NHANES. Because this research involves only the analysis of existing publicly available, de-identified data that cannot reasonably be used to identify individual participants, it qualifies for exemption from Institutional Review Board review.

3. Results

Our study population included 7208 women (n = 625 postpartum; n = 6583 not postpartum) representing 47,249,817 women at the population level (weighted n = 4253,164 postpartum; n = 42,996,653 not postpartum) (Supplementary Figure 1). Among the total population, the average age was 32.1 years (standard deviation [SD] 7.4 years), 19.1% were Hispanic, 13.6% were non-Hispanic Black, 58.2% were non-Hispanic White, 81.4% were born in the US, 67.7% had at least some college education, 57.6% were married or living with a partner, 29.5% had a family income to poverty ratio of 1.30 or lower, 53.5% had private insurance, and 24.0% had public insurance (Table 1). Depression was present in 11.1%, and the mean PHQ-9 score was 3.7 (SD 4.4). Among the postpartum women, 47.8% were 6–12 months postpartum at the time of the survey, 31.0% were 13–18 months postpartum, and 21.1% were 19–24 months. A total of 18.8% postpartum women were currently breastfeeding.

Table 1.

Baseline characteristics of non-pregnant women aged 20–44 years in NHANES.

Characteristics	Reproductive-age women not recently postpartum (Weighted N = 42,996,653)	Reproductive-age women recently postpartum (6–24 months after delivery) (Weighted N = 4253,164)	Total (Weighted N = 47,249,817)	p-value
PHQ-9 score, mean (standard deviation)	3.7 (4.5)	3.4 (4.1)	3.7 (4.4)	0.161
Depression *, n (%)	4439,414 (11.3%)	425,378 (10.0%)	4864,792 (11.1%)	0.507
Age, mean (standard deviation) years	32.4 (7.4)	29.4 (6.0)	32.1 (7.4)	<0.001
Race and ethnicity, n (%)				0.025
Hispanic	7990,626 (18.6%)	1020,397 (24.0%)	9011,023 (19.1%)
Non-Hispanic Black	5977,381 (13.9%)	461,526 (10.9%)	6438,907 (13.6%)
Non-Hispanic White	25,082,273 (58.3%)	2440,057 (57.4%)	27,522,330 (58.2%)
Other or multiple races	3946,374 (9.2%)	331,183 (7.8%)	4277,557 (9.1%)
Born outside the US, n (%)	7897,627 (18.4%)	877,141 (20.6%)	8774,768 (18.6%)	0.252
Education, n (%)				0.260
<9th grade	1540,365 (3.6%)	180,352 (4.2%)	1720,717 (3.6%)
9th-11th grade, no diploma	3941,306 (9.2%)	468,880 (11.0%)	4410,186 (9.3%)
High school graduate or GED	8162,077 (19.0%)	964,456 (22.7%)	9126,533 (19.3%)
Some college or associate’s degree	15,452,059 (35.9%)	1401,476 (33.0%)	16,853,535 (35.7%)
College grad or above	13,887,163 (32.3%)	1238,000 (29.1%)	15,125,164 (32.0%)
Marital status, n (%)				<0.001
Married or living with partner	23,933,171 (55.7%)	3289,293 (77.3%)	27,222,464 (57.6%)
Widowed, divorced or separated	4890,296 (11.4%)	203,619 (4.8%)	5093,916 (10.8%)
Never married	14,170,713 (33.0%)	760,251 (17.9%)	14,930,963 (31.6%)
Family income to poverty ratio, n (%)				<0.001
≤ 1.30	11,394,088 (28.5%)	1546,251 (39.0%)	12,940,339 (29.5%)
1.31 – 1.85	4752,279 (11.9%)	507,672 (12.8%)	5259,951 (12.0%)
1.86 – 3.50	9348,137 (23.4%)	922,075 (23.2%)	10,270,212 (23.4%)
> 3.50	14,452,526 (36.2%)	993,395 (25.0%)	15,445,921 (35.2%)
Health insurance, n (%) ⁎⁎				<0.001
Private	23,251,346 (54.5%)	1840,378 (43.4%)	25,091,723 (53.5%)
Public	9855,519 (23.1%)	1396,729 (32.9%)	11,252,248 (24.0%)
None	9559,327 (22.4%)	1006,293 (23.7%)	10,565,620 (22.5%)
Gravidity, n (%)				<0.001
0	15,104,102 (38.3%)	0 (0.0%)	15,104,102 (34.6%)
1	5109,186 (13.0%)	1114,063 (26.2%)	6223,249 (14.3%)
2	6801,393 (17.3%)	1235,310 (29.0%)	8036,703 (18.4%)
3	5589,160 (14.2%)	798,058 (18.8%)	6387,218 (14.6%)
4	3440,226 (8.7%)	528,794 (12.4%)	3969,020 (9.1%)
5 or more	3357,631 (8.5%)	576,939 (13.6%)	3934,570 (9.0%)
Number of livebirths, n (%)				<0.001
0	15,248,253 (41.3%)	22,065 (0.5%)	15,270,318 (37.1%)
1	5785,738 (15.7%)	1600,883 (37.6%)	7386,621 (17.9%)
2	8403,093 (22.7%)	1489,453 (35.0%)	9892,546 (24.0%)
3	4821,166 (13.0%)	647,809 (15.2%)	5468,974 (13.3%)
4	1976,214 (5.3%)	315,496 (7.4%)	2291,710 (5.6%)
5 or more	710,823 (1.9%)	177,458 (4.2%)	888,281 (2.2%)
Macrosomia, n (%)	3322,466 (9.0%)	534,562 (12.6%)	3857,028 (9.4%)	0.052
History of gestational diabetes in any pregnancy, n (%)	2395,967 (6.1%)	540,779 (12.7%)	2936,746 (6.7%)	<0.001
Number of months postpartum, n (%)				–
6–12	–	2033,905 (47.8%)	–
13–18	–	1319,745 (31.0%)	–
19–24	–	899,513 (21.1%)	–
Survey cycle, n (%)				<0.001
2007–2008	6695,826 (15.6%)	526,212 (12.4%)	7222,038 (15.3%)
2009–2010	6711,631 (15.6%)	483,900 (11.4%)	7195,531 (15.2%)
2011–2012	6511,878 (15.1%)	352,206 (8.3%)	6864,085 (14.5%)
2013–2014	6366,274 (14.8%)	750,568 (17.6%)	7116,842 (15.1%)
2015–2016	6288,475 (14.6%)	837,853 (19.7%)	7126,327 (15.1%)
2017-prepandemic 2020	10,422,570 (24.2%)	1302,424 (30.6%)	11,724,994 (24.8%)

GED: general educational development.

^⁎Depression was defined as a PHQ-9 score of 10 or higher.

^⁎⁎Participants could report more than one type of insurance

Missing values: US born (0.06%), education (0.03%), marital status (0.01%), income ratio (7.1%), health insurance (0.70%), gravidity (7.6%), livebirths (12.8%), macrosomia (13.0%), history of gestational diabetes (7.6%), PHQ-9 and depression (7.6%).

Compared to women who were not postpartum, postpartum women were on average younger and more likely to be Hispanic, to be married or living with a partner, have a lower income level, and have public insurance (Table 1). There was no significant difference in depression or PHQ-9 score between postpartum women and non-postpartum women.

The mean adjusted overall CVH score (out of 100) among postpartum women was 70.4 (95% confidence interval [CI] 68.7–72.0). The majority (94.7%) of postpartum women had an overall CVH score that was moderate (67.0%) or high (27.7%) (Fig. 1). Among postpartum women, the highest mean CVH scores (out of 100) were for blood glucose (94.0, 95% CI 92.3–95.6), followed by blood pressure (88.4, 95% CI 85.9–90.9), sleep (81.9, 95% CI 79.3–84.7), lipids (79.6, 95% CI 75.6–83.6), and nicotine exposure (75.9, 95% CI 71.8–80.1). The lowest CVH scores were for diet (36.2, 95% CI 34.5–37.9), physical activity (48.7, 95% CI 43.9–53.6), and BMI (57.3, 95% CI 52.5–62.1). Only 1.3% of postpartum women had a high CVH score for diet (Fig. 1).

Fig. 1.

Cardiovascular health among postpartum and non-postpartum women aged 20–44 years.

Panel A: Postpartum women; Panel B: Non-postpartum women, CV: cardiovascular, CVH: cardiovascular health, Missing values: overall cardiovascular health (8.1%), blood pressure (3.9%), glucose (2.9%), cholesterol (3.7%), diet (4.4%), physical activity (0.1%), smoking (0.08%), sleep (0.2%).

The mean adjusted overall CVH score among non-postpartum women was 71.5 (95% CI 70.9–72.0) compared to 70.4 (95% CI 68.7–72.0) among postpartum women. Overall CVH scores did not vary significantly between postpartum women and those who were not postpartum, adjusting for sociodemographics (Fig. 2). However, there were differences by postpartum status in scores for two of the individual CVH measures. Compared to women who were not postpartum, postpartum women had significantly lower CVH scores for physical activity (48.7 vs. 54.3, p-value 0.017) and BMI (57.3 vs. 63.3, p-value 0.014).

Fig. 2.

Cardiovascular health scores by postpartum status in non-pregnant women aged 20–44 years.

Adjusted for age, race, ethnicity, education level, marital status, and insurance.

In the whole study population, depression was associated with worse overall CVH and worse CVH scores for lipids, BMI, diet, physical activity, nicotine exposure, and sleep, adjusting for age and other sociodemographic factors (Supplementary Table 2). Postpartum status modified the association between depression and CVH for sleep only (p-value for interaction term 0.005) (Fig. 3). In stratified analyses, depression was associated with lower CVH scores for sleep among women who were not postpartum (beta −14.00, 95% CI −17.65, −10.35) (Supplementary Table 3). However, among postpartum women, depression was not significantly associated with CVH scores for sleep (beta −1.50, 95% CI −9.24, 6.23).

Fig. 3.

Association between depression and cardiovascular health scores in non-pregnant women aged 20–44 years, stratified by postpartum status.

Beta coefficient represents the association between depression status (as a binary variable) and cardiovascular health score (as a linear measure). Models were adjusted for age, race/ethnicity, education level, marital status, insurance. Models for postpartum women additionally adjusted for number of months postpartum.

Sensitivity analyses that used PHQ-9 scores instead of the binary depression variable demonstrated an association between PHQ-9 and each of the CVH metrics (Supplementary Table 4). In the sensitivity analyses, postpartum status did not modify the association between PHQ-9 and any of the CVH metrics. In a sensitivity analysis excluding postpartum individuals who were within 1 year postpartum, there was no significant difference in physical activity and BMI LE8 scores between postpartum women beyond 1 year postpartum and non-postpartum individuals (Supplementary Table 5). However, postpartum individuals who were beyond 1 year postpartum had higher cholesterol scores compared to non-postpartum individuals (83.2 vs 78.3, p-value 0.005).

4. Discussion

Central Illustration.

Abbreviations: BMI (body mass index), CVH (cardiovascular health), NHANES (National Health and Nutrition Examination Survey).

In a national US population-based sample, depression status was strongly associated with worse overall CVH among reproductive-age women, regardless of postpartum status. Postpartum status did not significantly modify the association between depression and CVH scores, with the exception of CVH scores for sleep.

Lower CVH among reproductive-age women with depression was driven primarily by differences in health behaviors and BMI. Depression is a known risk factor for CVD in the general population [23,24], and there is emerging research that postpartum depression is associated with increased risk of subsequent development of CVD [[18], [19], [20], [21], [22]]. However, the pathways underlying the association between postpartum depression and CVD have not been well studied. Postpartum depression could potentially lead to suboptimal CVH behaviors, including increased smoking, worse diet, worse sleep, and postpartum weight retention, which in turn could lead to increased risk of CVD. In fact, in the general population, health behaviors largely explain the association between depressive symptoms and adverse CV events [24]. At the same time, research has demonstrated that positive psychological health (such as optimism, sense of purpose, gratitude, and positive affect) is associated with healthier behaviors, greater engagement in preventive screening, and decreased incidence of CVD [9]. More research is needed to assess whether CVH behaviors explain the association between postpartum depression and CVD. Beyond health behaviors, postpartum depression is associated with hormonal, endothelial, and inflammatory changes that may also contribute to CVD risk [[34], [35], [36]]. Prior research has shown that depression leads to hyperactivity of the hypothalamic–pituitary–adrenal axis and increased cortisol, autonomic dysfunction, increased platelet reactivity, insulin resistance, and increased inflammatory cytokines and acute phase reactants [37,38].

We found that depression was associated with worse sleep scores in non-postpartum women but not in postpartum women. We hypothesize this is because sleep disruption is common in the postpartum period, driven by infant care demands. The relationships between sleep, physical activity, and BMI are likely interrelated and reinforcing. Prior research has shown depression and sleep quality are associated in a bidirectional manner [39,40]. In the postpartum period, physical activity may be a modifiable factor linking sleep, depression, and weight retention. Postpartum exercise interventions have been shown to reduce the severity of depressive symptoms and lower the odds of postpartum depression [41]. Poor postpartum sleep quality in turn has been linked to increased weight retention and higher BMI, among both young adults and postpartum individuals [[42], [43], [44]]. Poor maternal sleep quality has been linked to behaviors like physical inactivity, decreased fruit and vegetable intake, and more emotional eating, which may in turn be related to more gestational weight retention [45].

Regarding CVH during the postpartum period, we found that most postpartum women had an overall CVH score that was moderate or high. The CVH scores among postpartum women varied markedly across the metrics studied, and the pattern of variation in CVH scores is similar to what has been reported in the general population of women in NHANES. We further clarify these findings by demonstrating that postpartum women had worse CVH scores for physical activity and BMI, compared to women who were not postpartum. Postpartum women had physical activity and BMI scores that were on average 6 points lower. A prior study found that each 10-point higher LE8 score was associated with 22–40% lower risk of CVD and 17–21% lower risk of all-cause mortality [46], suggesting that even modest differences in LE8 scores translate into clinically meaningful differences. Another study found that the physical activity LE8 score was the single most impactful LE8 component for all-cause and CVD-specific mortality [47]. The postpartum period presents challenges to healthy lifestyle maintenance, as the transition to parenthood brings major life changes, shifting priorities, and new demands on time and energy for new parents. Previous studies have found declines in physical activity and diet quality during the postpartum period in cohorts in the US and other countries [12,[48], [49], [50], [51], [52], [53], [54], [55], [56], [57]]. Sleep generally worsens postpartum, and poor sleep may contribute to or be exacerbated by physical inactivity and poor diet [58,59]. Physical inactivity and suboptimal diet may contribute to postpartum weight retention, which has been found to be associated with worse cardiometabolic outcomes and increased risk of adverse pregnancy outcomes in subsequent pregnancies [13,14,48,[60], [61], [62], [63], [64]]. When we restricted the sample to those who were beyond 1 year postpartum, we found no difference in BMI and physical activity scores between postpartum and non-postpartum women. This may reflect in part the smaller sample size, but also suggests that differences in BMI and physical activity in the first year postpartum largely reflect gestational weight gain and early postpartum challenges in participating in regular exercise, which may improve in the later postpartum period.

Overall the findings highlight the need for additional support for postpartum women in the US to engage in heart-healthy lifestyle behavior. Ideally such interventions would begin to engage women preconception, during pregnancy, and throughout the postpartum period. Postpartum women face substantial structural and social barriers to engaging in healthy behaviors, including competing caregiving demands, lack of childcare support, and limited social support [11,65]. In the US, these challenges may be exacerbated by short maternity leave, lack of federally mandated paid parental leave, and high childcare costs relative to other high-income countries [66,67]. Policies that expand parental leave and childcare access could meaningfully improve postpartum women's ability to engage in lifestyle modification to improve their long-term CVH. Previous studies have found an association between increased maternity leave duration and improvements in both physical and mental health [[67], [68], [69], [70]]. Other potential interventions include postpartum peer support groups, behavioral health coaches, and digital interventions that promote lifestyle modification [6,[71], [72], [73], [74]]. Another promising intervention is postpartum transitions clinics that address CVH, particularly for patients at increased risk for CVD from adverse pregnancy outcomes. These clinics incorporate cardiovascular risk screening, counseling, and lifestyle interventions in the postpartum period and have been shown to increase screening rates and longitudinal follow-up in the first year postpartum [75,76].

Our study highlights the importance of integrating psychosocial health into lifestyle modification interventions in all reproductive-age women, including during the postpartum period. Our findings provide evidence for declining CVH in the postpartum period, particularly related to physical activity and BMI. A key barrier to optimal CVH is depression [9]. Therefore, multi-component interventions that target depression along with health behaviors may have a greater impact on CVH promotion. Recommended interventions to treat and manage perinatal depression include pharmacologic treatment, psychotherapy, cognitive behavioral therapy, stress management, mindfulness techniques, expressive writing, and addressing contributing factors like social isolation [27]. Integrating such interventions with lifestyle modification may benefit postpartum women who are undergoing a major life transition and establishing new routines and habits that may affect their CVH. Pregnancy is a powerful motivator for tobacco cessation [77], but relapse is common postpartum [78]. Given the strong association we found between depression and nicotine exposure, interventions to help women maintain tobacco cessation efforts in the postpartum period need to screen for and address postpartum depression. Depression was significantly associated with physical inactivity in our study. Moderate-intensity physical activity has been shown to reduce the symptoms of postpartum depression [65]. At the same time, diagnosing and treating depression may have the potential to help reproductive-age women achieve physical activity goals.

This study had several important limitations. Since the study used cross-sectional data, we are unable to establish causality. NHANES is not designed to survey postpartum women. Given the small numbers of postpartum women in the study, there were substantially wider confidence intervals observed among postpartum participants, which may have affected our ability to detect differences between postpartum and non-postpartum participants. We excluded women in the early postpartum period, who are at increased risk of postpartum depression and likely have unique CVH challenges [79,80]. NHANES only asks about postpartum status up to 24 months, but women who are beyond 24 months postpartum may continue to experience lifestyle and psychosocial changes related to motherhood that influence both depression and CVH. Women with a history of adverse pregnancy outcomes are at higher risk of worse cardiovascular outcomes. However, NHANES does not ask about several key adverse pregnancy outcomes such as hypertensive disorders of pregnancy and preterm birth, so we could not control for adverse pregnancy outcomes. This is an important limitation given the association between adverse pregnancy outcomes and both depression and long-term cardiovascular risk [[25], [26], [27], [28]]. Another limitation is potential for residual confounding given the lack of adjustment for gravidity, parity, pre-pregnancy BMI, gestational weight gain, anti-depressant use, and duration of breastfeeding. Breastfeeding status represents a potentially important confounder given its association with postpartum depression and reduced risk of long-term cardiovascular disease [81,82]. Several studies have shown that exclusive breastfeeding is associated with lower likelihood of postpartum depression [83,84]. At the same time, there is evidence of a complex and bidirectional association: depressive symptoms predict early cessation of breastfeeding, and breastfeeding challenges predict the development of postpartum depression [82]. Although we controlled for education, income, health insurance, and nativity, we did not control for other social determinants of health, such as social connectivity and neighborhood built environment, which are associated with both depression and cardiovascular health [[85], [86], [87]]. Although BMI is used for AHA’s Life’s Essential 8, it does not account for differences in body fat and lean mass, or fat distribution (peripheral versus central fat), which have important implications for cardiovascular risk [[88], [89], [90]]. Additionally, postpartum depression is defined by ACOG as depressive episodes that begin within 12 months after delivery [91], but in our analysis we included women up to 24 months postpartum. The Edinburgh Postnatal Depression Scale demonstrates both content validity and internal consistency for screening for postpartum depression [92]. However, in our analysis depression was defined using the PHQ-9, which may lack sufficient internal consistency and may have affected our findings of no difference in depression between postpartum and non-postpartum women [92]. Finally, we could not determine timing of depression onset, and the analysis does not account for pre-pregnancy or prenatal depression, or depression treatment history.

This study also has several notable strengths. Although AHA’s Life’s Essential 8 framework has been recommended for optimizing CVH in the postpartum period [6], this is one of the first applications of Life’s Essential 8 focused on the postpartum population [93,94]. Another strength is that we used a nationally representative sample of the US population to evaluate CVH among postpartum women compared with women who were not postpartum. And we extend the prior research on postpartum depression and CVD by examining whether postpartum status modifies the association between depression and CVH. Future research priorities include prospective studies to evaluate adverse pregnancy outcomes, postpartum depression, and CVH trajectories and investigate the underlying mechanisms. There is a need for trials that evaluate integrated interventions designed to improve both mental health and CVH outcomes in reproductive-age women.

The postpartum period has increasingly been recognized by both cardiology and obstetric professional societies as a key time for declining CVH and psychosocial health among women [6,[95], [96], [97], [98], [99]]. In this analysis, we highlight several targets for behavioral intervention in the postpartum period, most notably physical activity and weight management. Additionally, we draw attention to the importance of postpartum depression screening to identify individuals who may be at even greater need of counseling, support, and lifestyle modification postpartum to improve CVH.

Disclosures

During the preparation of this work the author(s) used Claude (Anthropic) in order to proofread the manuscript and ChatGPT to create the central illustration. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article.

CRediT authorship contribution statement

Megan M. McLaughlin: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization. Catherine Lee: Methodology, Formal analysis. Xiaoning Huang: Writing – review & editing, Methodology. Nisha I. Parikh: Writing – review & editing. Kristina M. Deligiannidis: Writing – review & editing, Methodology. Sadiya S. Khan: Writing – review & editing, Methodology. Alexis L. Beatty: Writing – review & editing, Supervision, Methodology.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Megan McLaughlin reports financial support was provided by National Institutes of Health. Funding and Acknowledgements: Dr. McLaughlin receives support from the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under Award Number (K12 AR084219). Dr. Deligiannidis receives support from the National Institutes of Health (R01MH120313).

Disclosures: Dr. Deligiannidis serves as a consultant to Sage Therapeutics, Brii Biosciences, Gerbera Therapeutics, GH Research, Neuroscience Software and Reunion Neuroscience. Dr. Deligiannidis served as a study principal investigator for contracted research awarded to the Feinstein Institutes for Medical Research from Sage Therapeutics and Premier Healthcare. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.