Postpartum depression (PPD) is depression that occurs after delivery [1]. Postpartum depression symptoms (PDS), which vary by individual, affect up to one in eight women [2]. When left untreated, PPD can negatively affect one’s ability to care for themselves and their infant and can negatively impact long-term bonding and child development [3].

Social support helps alleviate the stress of a new baby and allows for better adjustment to life changes [4]; perceived social support from partners is associated with decreased maternal and infant distress [5]. Social support is described as functions of sharing information (informational), providing hands-on services (instrumental), emotive sharing of experiences (emotional), and offering approval to each other (appraisal) [6]. Multiple types of social support are important for the transition to parenthood and healthy infant development. In qualitative studies, postpartum women commonly express a need for social support [7] and how it is important to physical and emotional recovery [8]. A systematic review examining 10 qualitative studies focused on postpartum social support found that women benefit from all forms of social support, including support from other mothers [9]. Studies have found a relationship between low social support and PPD [10] and between low social support and PDS [11–14] in a variety of populations, including various race and age groups in the United States (U.S.).

The Pregnancy Risk Assessment Monitoring System (PRAMS) is an ongoing collaboration between participating jurisdictions and the Centers for Disease Control and Prevention (CDC) to collect data on maternal attitudes and experiences before, during, and after birth [15], and is used to monitor PDS². PRAMS data are not routinely used to study postpartum social support, because social support questions are not included on the core questionnaire. Jurisdictions may include questions from the PRAMS standard (optional) question set or develop their own state-specific questions. Beginning in 2016, Illinois added one standard question to assess instrumental support and one state-specific question to assess partner emotional support.

Few studies have assessed the relationship of social support and PDS/PPD using PRAMS. One study using PRAMS data from New York City found that partner social support was not a significant predictor of PPD for any racial/ethnic group [16]. Another study using Ohio Pregnancy Assessment Survey data, a PRAMS-like survey, found that low social support (both instrumental support and emotional support) was associated with increased PDS in crude analyses, but this association did not maintain statistical significance in adjusted models [17]. The objective of this study was to assess the association of social support and PDS among postpartum Illinois women with a recent live birth.

Maternal characteristics of the study population are shown in Table 1. Approximately 41.0% had a college degree or higher and 31.3% were 30–34 years old. More than half of Illinois postpartum women were NH White (55.9%), did not have Medicaid insurance (58.2%), were married (61.3%), and had births resulting from intended pregnancies (59.1%).

Among Illinois postpartum women, 77.0% had high instrumental support and 79.3% had high partner emotional support (Table 2). There were statistically significant differences in the distributions of both types of support for each maternal characteristic (p < 0.0001). High levels of both support types were most common among women who: had a college degree or higher, were 30–34 years old, did not have Medicaid insurance, were married, and had an intended pregnancy. High instrumental support was highest among NH White women (84.7%) and lowest among NH Asian women (58.7%). In contrast, high partner emotional support was highest among NH Asian women (88.4%) and lowest among NH Black women (66.1%).

Among Illinois postpartum women, 63.5% had high composite support, 29.3% had moderate composite support, and 7.1% had low composite support (Table 3). There were statistically significant differences in the distribution of the composite support measure for each maternal characteristic (p < 0.0001). Groups with high levels of both support types also had high composite support. Low composite support was most common among women who: had less than a high school education, were 24 years old or younger, were NH Black, had Medicaid insurance, were not married, and had an unintended pregnancy.

Approximately 9.6% of Illinois postpartum women experienced PDS, and there were statistically significant (p < 0.0001) differences in PDS by all maternal characteristics (Table 4). The highest prevalence of PDS was reported by women who had less than a high school degree, were 24 years and younger, had Medicaid insurance, were not married, and had an unintended pregnancy. When comparing PDS by race/ethnicity, NH White women had the lowest PDS prevalence (8.5%) and NH Black women had the highest PDS prevalence (12.9%).

The prevalence and crude and adjusted OR for PDS for each type of social support are shown in Table 5. PDS was reported by 6.7% of women with high instrumental support compared to 19.2% of women with low instrumental support (crude OR = 3.3, 95% CI = 2.7–4.0). After adjusting for maternal characteristics, those with low instrumental support had approximately three times the odds of PDS compared to those with high instrumental support (adjusted OR = 3.1, 95% CI = 2.5–3.8). Similarly, PDS was reported by 7.3% of women with high partner emotional support compared to 18.4% of women with low partner emotional support (crude OR = 2.9, 95% CI = 2.3–3.5). After adjusting for maternal characteristics, those with low partner emotional support had approximately two- and- a- half times the odds of PDS compared to those with high partner emotional support (adjusted OR = 2.5, 95% CI = 2.1–3.2).

The three levels of the composite support measure showed a dose–response relationship with PDS (Table 5). PDS was reported by 5.4% of women with high composite support, compared to 14.1% of women with moderate composite support (crude OR = 2.9, 95% CI = 2.3–3.6), and 28.4% of women with low composite support (crude OR = 7.0, 95% CI = 5.2–9.3). After adjusting for maternal characteristics, compared to those with high composite support, women with moderate composite support had approximately three times the odds of PDS (adjusted OR=2.7, 95% CI=2.2–3.4) and women with low composite support had approximately six times the odds of PDS (adjusted OR = 6.1, 95% CI = 4.5–8.2).

In the model that included an interaction term for instrumental support and social support, the beta coefficient for the interaction term was not statistically significant (p = 0.40), demonstrating no statistically significant multiplicative interaction for the effects of instrumental support and partner emotional support on PDS (results not shown).

The sensitivity analyses that used alternate definitions for each social support type (instrumental, partner emotional and a composite support measure) did not substantially change the direction or magnitude of the results, demonstrating that the findings were robust to the coding decisions; finally, excluding 2020 deliveries from our adjusted models did not substantively change the findings (results not shown).

This study showed PDS were associated with instrumental support, partner emotional support, and a composite support measure among Illinois postpartum women who delivered a live birth in 2016–2020. These associations persisted after adjusting for maternal characteristics. The composite support measure showed a dose–response relationship with PDS, with increasing PDS prevalence as social support decreased across three levels. This is consistent with another cross-sectional study which found that pregnant women who had more social support exhibited lower psychological stress and displayed fewer depressive symptoms, compared to pregnant women with less social support [19]. Our study also demonstrated that instrumental support, partner emotional support, and a composite support measure significantly differed by maternal education, age at delivery, race/ethnicity, Medicaid insurance status, marital status, and pregnancy intention.

The prevalence of instrumental support and partner emotional support had similar patterns across all maternal characteristics, except for race/ethnicity. Specifically, NH Asian women had the lowest prevalence of high instrumental support and the highest prevalence of high partner emotional support. This demonstrates the importance of assessing multiple types of social support, as evaluating one may not fully represent social support experiences and could result in under- or over-estimation of the relationship with PDS. Our study did not specifically examine the role of race/ethnicity in the relationship between social support and PDS. One qualitative study showed there are racial/ethnic differences in the way women ask for help from their social networks, underscoring the dynamic nature of how cultural perceptions of social support can influence PPD [8]. Another study looking at a racially and ethnically diverse group of women found that higher levels of social support had a strong protective association against PPD for all races/ethnicities; however, social support was more protective against PPD for White women than for Black or Latina women [10].

This research is important for healthcare professionals who can leverage clinical interactions to assess social support needs and connect new mothers and families to resources as a potential strategy to mitigate PDS. Healthcare professionals are uniquely qualified to be key players in supporting first-time mothers in the early postpartum period and “fostering emotional functional social support from significant others within mothers’ social network.” [6] The American College of Obstetrics and Gynecologists (ACOG) acknowledges that many women lack formal or informal support in the first 6 weeks postpartum and suggests that providers work with their patients during prenatal care to identify sources of postpartum social and material support [20]. In a recent survey of U.S. postpartum care providers, most respondents identified transitioning to parenthood, which included evaluating sources of social support, as an important priority for their patients [21]. Healthcare providers realize the importance of social support, but likely need more resources to engage and support patients with their social support needs.

This study is unique because it utilizes two PRAMS questions to assess multiple aspects of social support, instrumental support and partner emotional support, separately and in combination. The two previous studies using PRAMS or PRAMS-like data did not assess partner emotional support in combination with other types of support, so our approach is innovative and unique [16, 17]. Our study proposes a composite support measure that could be replicated in other PRAMS jurisdictions that use these questions.

This study has several limitations inherent to the PRAMS design. First, PRAMS data are cross-sectional, so a causal relationship between PDS and social support cannot be determined. However, longitudinal studies demonstrated the overall protective influence of social support on PDS at 1 year postpartum among adolescent women [12], and that low perceived social support at 1 month postpartum predicted depressive symptoms at 6 months among first-time mothers [22]. In our study, the PDS and social support questions of interest all began with the phrase “since your new baby was born,” so were not specific to a certain point in the postpartum period. The association of PDS and social support could vary based on the timing of survey response, but we did not explore response timing in this study. Due to question phrasing, we could also not assess social support during pregnancy, which has also been shown to be associated with PDS [23]. Finally, PRAMS data may be subject to reporting and recall biases. This study relied on self-reported symptoms that are common with PPD, but a report of PDS is not indicative of a clinical diagnosis of PPD. Self-reported PDS may be underreported due to social desirability bias, but this misclassification would likely lead our study to underestimate the true relationship between PDS and social support. Additionally, these symptoms may be related to other mental health conditions and are not specific to PPD.

There are also limitations to the measurement of social support in this study. First, we could not assess emotional support from non-partners, informational support, or appraisal support, due to lack of Illinois PRAMS questions; thus, our study was limited to instrumental support and partner emotional support. When social support is assessed in clinical practice, it is standard to use an instrument that asks about multiple types of social support and weighs each type of support equally, such as the Multidimensional Scale of Perceived Social Support [24]. We measured instrumental support by the respondent’s perceived ability to receive five specific supports from her social network and did not reflect actual support received, nor the source. The question measuring partner emotional support asks only about support from a “husband or partner,” with no mention of other relationships. As a result, this question may not be applicable to all intimate relationships and does not specifically assess other sources of emotional support (e.g., family and friends). This study was also not able to assess the complexities of nonmarital relationships nor how the woman’s current living/housing situation may affect partner emotional support. One study found that both intimate partner support and friend/family support were independently associated with a significantly lower risk of PPD for all women, but intimate partner support had a greater impact on PPD than friend/family support for married and cohabiting women, and was equal in magnitude to friend/family support for single women [25]. This reiterates the nuanced complexity of how partner emotional support could impact PPD.

While this study examined and controlled for factors that were determined to confound the relationship between PDS and social support, possible residual confounding exists. Namely, we could not control for history of other mental health conditions, though prepregnancy depression and anxiety has been associated with increased risk of PPD [26]. Medicaid enrollment was used as a proxy for socioeconomic status, but this may not sufficiently address confounding. Unmeasured confounders could affect the magnitude of the association of PDS and social support in either direction.

It is worth noting that this study time frame included mostly births prior to the SARS-CoV-2 pandemic but did include 2020 births. Normal social support patterns during 2020 and beyond may have been disrupted due to the restriction of social gatherings and quarantine measures. Additionally, rates of depression symptoms among U.S. adults increased during the pandemic, particularly for those with low household income and who were unmarried [27]. However, a sensitivity analysis excluding 2020 births from our adjusted models did not substantively change our findings. Future analyses could compare the patterns of social support and PDS pre- and postpandemic for postpartum women to assess how support patterns may have shifted.

This study contributes to an ongoing body of U.S. research that demonstrates a protective association between maternal social support and PDS. Our findings reiterate the importance of addressing social support for postpartum individuals within clinical and community health settings. These findings support efforts such as ACOG’s Guidelines for Perinatal Care which calls for creating a care team inclusive of postpartum family and friend support [28]. The routine assessment of social support during perinatal care might also offer important context for the screening, prevention, and clinical management of PPD and could prompt connections to supportive services. Public health and clinical care entities could consider new programs to increase social support for new parents or ways to enhance social support elements within existing programs. Research has demonstrated that those at high risk for PPD may benefit from maternal supports such as home visitation and peer support [29]. Studies from various countries suggest that interventions such as couples counseling, interpersonal psychotherapy, and follow-u p calls from an advanced practice nurse may be effective in improving social support among postpartum individuals [30]. Ultimately, acknowledging social support as a powerful asset during the postpartum period, and working to bolster that support, could improve postpartum mental health for mothers.