Effects of March of Dimes Supportive Pregnancy Care on Social Support and Postpartum Depression

Abstract

Purpose

Group prenatal care, which integrates medical care with patient education and empowerment in a group setting, has the potential to build social support among pregnant women and reduce the risk of postpartum depression (PPD). Past research on group care’s effects on social support or PPD has produced inconsistent findings. Studies have tended to (1) examine direct effects on social support or PPD and (2) treat group care participation as a dichotomous variable. This study tests the hypothesis that group care has an indirect effect on PPD through its effect on social support. It uses both a dichotomous measure of group care participation and an ordinal measure of social contact with group members.

Method

This study used survey data from 199 women at two Tennessee health care sites who participated in traditional care (TC) or Supportive Pregnancy Care (SPC), a new group program developed by March of Dimes. Path analysis was employed, estimating standardized path coefficients with propensity score weighted multilevel modeling.

Results

The study found no evidence of an indirect relationship between SPC and PPD symptoms when modeling the dichotomous measure. Modeling the ordinal measure revealed more favorable PPD outcomes among women reporting high group member contact (compared with TC) and adverse outcomes among women reporting low group member contact.

Conclusion

SPC may be useful for preventing PPD symptoms among women socially engaged with other group care patients. Given the adverse effect among women who do not engage with other group members, group facilitators should encourage women to interact outside of scheduled sessions.

Keywords

group prenatal care postpartum depression social support supportive pregnancy care

Postpartum depression (PPD), or the onset of depression symptoms after childbirth, is a significant problem that can have serious effects on the well-being of mothers and their infants. A recent meta-analysis of research assessing PPD among healthy mothers (with no prior history of clinical or self-reported depression) indicted that the global prevalence of PPD is 17% (Shorey et al., 2018). In fact, PPD is believed to be the most common complication of childbirth (Robertson et al., 2004). The broad scope of PPD is concerning, considering that a recent review of the literature on PPD highlighted adverse consequences on mother’s well-being (e.g., anxiety, low quality of life, relationship difficulties, suicidal ideation, thoughts of infanticide), mother–child interactions (e.g., low mother–child bonding, low feelings of maternal competence, difficulty with infant care, inadequate use of infant healthcare services), and infant well-being (e.g., problematic sleep patterns and impaired cognitive, language, emotional, and social development; Slomian et al., 2019).

Although there are many risk factors for PPD (e.g., obesity, sleep disturbance, unhealthy diet, gestational diabetes, anxiety, violence/abuse, low marital/partner satisfaction, life stress, immigration/acculturation-related stressors, cesarean section delivery, multiple birth delivery, preterm birth, low birthweight, birth complications), experiencing low levels of social support stands out as one of the most well-established and mutable risk factors (Hutchens & Kearney, 2020; Xiao-hu & Zhi-hua, 2020). Social support can be defined as “the social resources that persons perceive to be available or that are actually provided to them by nonprofessionals in the context of both formal support groups and informal helping relationships” (Gottlieb & Bergen, 2010, p. 512). A recent meta-analysis synthesizing the research on social support and PPD found that the odds of mothers with weak social support exhibiting PPD symptoms were 2.64 times the odds of PPD symptoms among women with greater levels of social support (Ardiani et al., 2020). Group prenatal care, which integrates medical care with patient education and empowerment in a group setting (Walker & Worrell, 2008), may be a viable tool for building women’s social support networks during pregnancy and, in turn, reducing the risk of PPD.

The group prenatal care model is arguably synonymous with CenteringPregnancy, which was developed in the early 1990s by midwife Sharon Schindler Rising. As envisioned by Rising (1998), group prenatal care “unifies the components of prenatal care—risk assessment, education, and support within the group—and encourages women to take responsibility for their own health” (p. 46). Group prenatal care is relationship-centered by nature, and proponents have explained its intended social support benefits as follows: “meeting in a group setting with other women of the same gestational age who are experiencing similar physiological and psychological changes of pregnancy nurtures supportive relationships among patients” (Massey et al., 2006, pp. 286–287).

The extant research has produced inconsistent evidence of group prenatal care’s effects on social support and PPD symptoms. Regarding social support, one early randomized controlled trial found no effects (Ickovics et al., 2011), whereas two recent studies, one a randomized controlled trial (Mazzoni et al., 2020) and the other an observational study (Chae et al., 2017), did find significant favorable effects of group prenatal care on social support. Regarding the effects of group prenatal care on PPD symptoms, a recent systematic review of the extant research unveiled limited evidence of benefits (Buultjens et al., 2020). Five of the nine studies included in the review yielded nonsignificant effects on PPD, three indicated significant effects for specific subpopulations of prenatal patients (e.g., adolescents, women at greater psychosocial risk for PPD), and one study yielded significant effects among a general sample of prenatal care patients. Interestingly, two studies included in this review measured effects of group prenatal care on both social support and PPD outcomes. Both of these studies indicated that women participating in group prenatal care reported significantly greater social support compared with women in traditional care (TC); however, neither study demonstrated significant effects of group care on PPD symptoms (Chae et al., 2017; Mazzoni et al., 2020).

The limited evidence of group prenatal care’s effects on social support and PPD may be explained by two prevalent themes in the extant research. First, aside from one noteworthy exception (see Chae et al., 2017), research on the effects of group prenatal care on social support and/or PPD has examined direct effects on one or both of these outcomes. However, group prenatal care may have an indirect effect on PPD symptoms through its relationship with social support. That is, participation in group prenatal care may lead to higher measures of social support, which, in turn, may lead to lower measures of PPD symptoms.

Second, the extant research examining effects of group prenatal care on psychosocial outcomes such as social support and PPD symptoms has operationalized group prenatal care participation as a binary measure whereby women who participate in group prenatal care are compared with women who participate in traditional one-on-one prenatal care. The underlying logic of the group prenatal care model relies on the assumption that women will develop relationships with one another (Massey et al., 2006; Rising, 1998). Surprisingly, there is a dearth of research examining effects of group prenatal care on psychosocial outcomes among women with differing levels of social engagement with other group members.

The present study examines the effects of Supportive Pregnancy Care (SPC), a new group prenatal care program developed by March of Dimes (MOD), on social support and PPD symptoms. It specifically tests the following hypotheses:

Hypothesis 1: Participation in SPC will not have a significant, direct effect on PPD symptoms.

Hypothesis 2: Participation in SPC will have a significant positive effect on social support, which, in turn, will have a significant negative effect on PPD symptoms.

Hypothesis 3: The positive effect of SPC participation on social support, and the indirect negative effect on PPD symptoms, will be more pronounced among women who report greater levels of contact with SPC group members.

Materials and Method

In 2016, MOD selected six pilot sites across the state of Tennessee to begin implementing SPC. Patients at each of the study pilot sites could self-select into SPC or traditional one-on-one prenatal care. Similar to most group prenatal care programs, SPC promotes social support, health education, and skill-building in a group setting. SPC entails 8 to 12 pregnant women with similar due dates meeting together for 10 sessions of approximately 90 to 120 minutes each. During each session, a licensed obstetric provider meets individually with each woman to perform a physical assessment in a semiprivate area within the group space. The provider and a cofacilitator then meet with the group to discuss topics related to pregnancy, labor and delivery, social support, postpartum care, and infant care (MOD, 2017).

The study authors were contracted by MOD from 2016 to 2019 to conduct a pilot evaluation of SPC. The evaluation employed a quasi-experimental research design and relied on two sources of data: (1) de-identified medical record data from the population of women who delivered at each of the pilot sites during the study timeline and (2) survey data from a subsample of women receiving prenatal care at each of the pilot sites. Medical record findings from this pilot study are reported elsewhere (Kettrey & Steinka-Fry, 2020). The present report relays survey findings relevant to social support and PPD symptoms.

The survey sample was recruited by the first study author between July 2017 and January 2018 through two means: (1) circulating a flyer informing patients about a “prenatal care study” at each pilot site and (2) visiting pilot sites to invite prenatal care patients to participate in a “prenatal care study.” All prenatal care patients over the age of 18 years were eligible to participate. Patients who provided informed consent completed an initial survey on recruitment and were sent a postdelivery survey 7 weeks after their due date. The postdelivery survey consisted of 80 items asking about prenatal care participation (e.g., SPC or TC participation, level of contact SPC participants had with other SPC group members), pregnancy outcomes (e.g., weight gain, gestational age at delivery), health-related behaviors (e.g., smoking, alcohol/drug use, breastfeeding, future birth control plan), and psychosocial outcomes (e.g., PPD symptoms, social support, health literacy, maternal self-efficacy). Surveys took 10 to 20 minutes to complete, and participants were given the option of completing surveys online or returning printed surveys via mail, with approximately half selecting each of these two options. As an incentive, participants received a $20 gift card for completing the initial survey and a $30 gift card for the postdelivery survey.

Participants

A total of 300 prenatal care patients across the six pilot sites were recruited into the survey sample. Of these 300 participants, 251 completed the postdelivery survey. The analytic sample consisted of survey participants from two of the six pilot sites. The remaining four sites were dropped due to low recruitment (n = 3 across two sites) or because dissimilarities between the SPC and TC groups were so pronounced that they could not be adjusted with propensity score methods (n = 38 across two sites). Regarding the latter, participants in the SPC and TC groups were so different on the collective variables used to calculate propensity scores that scores representing SPC participants’ probability of selecting into SPC all approached 1.0 and scores representing TC participants’ probability of selecting into SPC all approached 0 (variables used to calculate propensity scores are described below with Data Analysis methods).

Thus, the survey sample consisted of 210 participants who completed the postdelivery survey at two sites, which we identify as Site 1 and Site 2 (n = 164 and n = 46, respectively). The retention rate at these two sites was 88.24% (i.e., 238 completed the initial survey). Site 1 is a private practice clinic in a rural setting serving a patient population that is mostly White and privately insured. Site 2 is a community clinic associated with an acute care hospital in an urban setting serving a patient population that is mostly African American and covered by public insurance. Characteristics of survey participants are summarized in Table 1. Pertinent to external validity, there were a few notable differences between the women in the survey sample and the population of women delivering at Sites 1 and 2 throughout the duration of this study.

Table 1.

Characteristics of Final Analytic Sample of Prenatal Care Participants by Site and Prenatal Care Format, N = 199.

Characteristics	Site 1		Site 2		Aggregate sample
Characteristics	SPC (n = 33)	TC (n = 125)	SPC (n = 23)	TC (n = 18)	SPC (n = 56)	TC (n = 143)
Participant characteristics used for propensity scores, %
Age, years, M (SD)	29.85 (5.26)	28.19 (5.27)	24.83 (4.74)	22.67 (3.71)	27.27 (5.60)	27.50 (5.41)
White	100.00	96.80	17.39	16.67	66.07	86.71
Black/African American	0.00	1.60	78.26	77.78	32.14	11.19
Hispanic	3.03	4.80	0.00	0.00	1.80	4.20
Spanish language	0.00	0.00	0.00	0.00	0.00	0.00
Married	75.76	76.00	30.44	5.56	57.14	67.13
Public insurance	24.24	33.60	82.61	88.89	48.21	40.56
Highest education attained
Less than high school	0.00	0.80	17.39	11.11	7.14	2.10
High school graduate	0.00	9.60	26.09	66.67	10.71	16.78
Some college, technical, or associate’s degree	33.33	38.40	47.83	16.67	39.29	35.67
Bachelor’s degree	45.45	28.00	4.35	5.56	28.57	25.17
Postgraduate	12.12	23.20	4.35	0.00	14.29	20.28
Previous births, M (SD)	0.00 (1.00)	1.00 (1.13)	1.00 (1.11)	1.00 (1.58)	0.00 (1.05)	1.00 (1.21)
Planned pregnancy	78.79	70.40	26.09	22.22	57.14	64.34
Gestational age at first visit
Less than 6 weeks	33.33	24.00	8.70	33.33	23.21	25.17
6 to 12 weeks	60.60	71.20	52.17	50.00	57.14	68.53
13 to 18 weeks	6.06	3.20	21.74	16.67	12.50	4.90
19 to 24 weeks	0.00	0.80	17.39	0.00	7.14	0.70
25 to 30 weeks	0.00	0.00	0.00	0.00	0.00	0.00
31 to 36 weeks	0.00	0.80	0.00	0.00	0.00	0.70
More than 36 weeks	0.00	0.00	0.00	0.00	0.00	0.00
Attended non-SPC class	12.12	17.60	17.39	38.89	19.64	20.28
Contact with SPC group members
Never	87.88	NA	69.57	NA	80.36	NA
Rarely	3.03	NA	8.70	NA	5.36	NA
Sometimes	9.10	NA	17.39	NA	12.5	NA
Often	0.00	NA	4.35	NA	1.79	NA
Survey outcomes, M (SD)
FSSQ
Confidant support	4.88 (1.34)	4.75 (1.47)	4.74 (1.51)	3.82 (1.46)	4.82 (1.40)	4.63 (1.57)
Affective support	5.27 (1.27)	5.11 (1.47)	5.07 (1.46)	4.09 (2.03)	5.19 (1.34)	4.98 (1.58)
EPDS	3.55 (5.12)	3.18 (3.55)	2.65 (3.98)	6.17 (5.98)	3.18 (4.67)	3.56 (4.03)

Note. SPC = Supportive Pregnancy Care; TC = traditional care; FSSQ = Functional Social Support Questionnaire; EPDS = Edinburgh Postnatal Depression Scale; n = sample size.

At Site 1, women in the survey sample were slightly older (Site 1 survey sample M age = 28.53 and population M age = 27.69) but were similar in racial demographics (Site 1 survey sample 97.47% White and population = 96.92% White) and similar in percent publicly insured (Site 1 survey sample 31.65% publicly insured and population = 35.21% publicly insured). At Site 2, women in the survey sample were younger (Site 2 survey sample M age = 23.88 and population M age = 25.60) and were dissimilar in racial demographics (Site 2 survey sample 17.07% White and population 0.5% White) as well as dissimilar in the percent publicly insured (Site 2 survey sample 85.37% publicly insured and population 35.1% publicly insured). Full demographic profiles for populations at Sites 1 and 2 are described elsewhere (Kettrey & Steinka-Fry, 2020).

Exogenous Variables

To explore the effects of both SPC participation and contact with other SPC group members on the outcome variables, this study used two main predictors: (1) a binary measure of SPC participation and (2) an ordinal measure of the extent to which SPC participants had contact with other group members outside of scheduled sessions. The binary measure comes from a survey item asking study participants, “Did you participate in the March of Dimes Supportive Pregnancy Care (SPC) program? This program provided prenatal care in a group setting with other pregnant women.” Possible responses were yes and no. Analyses compare respondents who indicated “yes” (SPC) with those who indicated “no” (TC).

The ordinal measure comes from a survey item asking those study participants who indicated that they participated in SPC, “How often did you meet with other Supportive Pregnancy Care (SPC) group members in person outside of group meetings?” Possible responses were never, rarely, sometimes, and often. Analyses compare each of these ordinal categories with TC.

Endogenous Variables

The main outcomes of this analysis were social support and PPD. Social support was measured using the Duke-UNC Functional Social Support Questionnaire (Broadhead et al., 1988). The Functional Social Support Questionnaire consists of eight items measured on a 6-point scale, ranging from “as much as I would like” to “much less than I would like.” Items load onto two social support factors: confidant support (five items) and affective support (three items). Confidant support “reflects primarily a confidant relationship where important matters in life are discussed and shared,” whereas affective support “reflects a more emotional form of support or caring” (Broadhead et al., 1988, p. 715). Example questionnaire items are “I get chances to talk to someone I trust about my personal and family problems” (confidant support) and “I get love and affection” (affective support). Within each factor, mean scores were calculated, with possible scores ranging from 1 to 6. In the current study, Cronbach’s alphas for confidant support were .95 for Site 1 and .94 for Site 2. Cronbach’s alphas for affective support were .93 for Site 1 and .94 for Site 2.

PPD symptoms were measured with the Edinburgh Postnatal Depression Scale (EPDS; Cox et al., 1987). The EPDS consists of 10 items measured on a 4-point scale, with scale category labels (scored 0 to 3) differing between items (some items are reverse coded). Items ask respondents to indicate how they have felt within the past 7 days in reaction to prompts such as “I have been anxious or worried for no good reason” and “I have looked forward with enjoyment to things.” Mean scores were calculated, with possible scores ranging from 0 to 30 and higher scores indicating higher depressive symptomology. Cronbach’s alpha for the EPDS in the present study were .89 for Site 1 and .89 for Site 2.

Data Analysis

Main statistical analyses were conducted in R 3.5.0 (R Core Team, 2018). Due to the relatively small survey sample size, propensity score weighting, rather than matching, was used to correct for self-selection bias (Guo & Fraser, 2015). Weighting permitted the retention of all participants for whom a propensity score could be calculated. Propensity scores were estimated at the site level using binary logistic regression to estimate the probability of participants selecting into SPC. Variables used to calculate propensity scores included age, race, language, marital status, type of insurance, highest education attained, previous number of births, planned/unplanned pregnancy, gestational age at first prenatal care visit, and participation in any non-SPC childbirth class. Weights were calculated as 1/(probability of being in the SPC group) for SPC and 1/(1 − probability of being in the SPC group) for TC. After calculating weights, site-level trimming was performed such that any weight exceeding the 99th percentile was replaced with the value that fell exactly at the 99th percentile to reduce bias from influential cases (Lee et al., 2011).

Missing data were rare and, thus, were not imputed. Two participants were dropped from Site 1 and three from Site 2 for failing to provide a response to one of the variables used for propensity score calculation (reducing the sample from 210 to 205). Additionally, to retain a consistent sample between analyses, missing data were handled with listwise deletion, resulting in a final sample of 199. In all, a total of 5.24% of cases (11 of 210) were dropped due to missing data.

A power analysis was conducted using Optimal Design Software version 3.01 (Spybrook et al., 2011) assuming a two-tailed alpha of .05 and an R² value of .30. The power analysis indicated that the following sample sizes were required for .80 power in detecting small-to-medium effects: social support (n = 177) and PPD symptoms (n = 147). Site-level sample sizes were not adequate for detecting the effects on these outcomes; however, the total sample size from the aggregated sites was adequate. Thus, to increase statistical power, analyses were conducted on the aggregate dataset using methods that account for within-site clustering.

To test the hypothesized relationships between SPC, social support, and PPD symptoms, path analysis methods were employed (Asher, 1983). The specific relationships that were modeled using the binary exogenous variable are depicted in Figure 1, and relationships modeled using the ordinal exogenous variables are depicted in Figure 2. Figure 1 posits a direct positive relationship between SPC participation and both forms of social support (i.e., confidant and affective support). It also posits a direct negative relationship between each form of social support and PPD symptoms. Vertical arrows entering the social support and PPD variables represent residual variance, or the amount of variance in these variables that cannot be explained by the proposed model. A double-headed arrow connects the residual variance of confidant and affective support, indicating that the unexplained variance in each of these variables is correlated. The paths depicted in Figure 2 are similar to those depicted in Figure 1, except that separate paths are modeled from each of the ordinal categories to the two forms of social support.

Figure 1.

Path model depicting the hypothesized relationship between SPC, social support, and postpartum depression symptoms.

Figure 2.

Path model depicting the hypothesized relationship between contact with SPC group members, social support, and postpartum depression symptoms.

Standardized path coefficients demonstrating relationships between variables were estimated using propensity score weighted multilevel modeling that included site as a grouping variable. These analyses were conducted using the lavaan package in R, which allows researchers to conduct path analysis while applying weights and accounting for clustering of multilevel data (Rosseel, 2021).

Results

To test Hypothesis 1, PPD symptoms were regressed on SPC (vs. TC) using propensity score weighted multilevel modeling. Findings indicate that SPC participation is not significantly related to PPD symptoms (β = −.07, p > .10). Thus, Hypothesis 1 was supported.

To test Hypothesis 2, standardized coefficients were estimated for each path specified in Figure 1. As the results summarized in Figure 3 show, this model fit the data well, according to Hu and Bentler’s (1999) standards (e.g., RMSEA [root mean square error of approximation] < .06; SRMR [standardized root mean square residual] < .08; CFI [comparative fit index] > .95; TLI [Tucker-Lewis index] > .95). Although paths from both forms of social support to PPD symptoms were significant, neither the path from SPC to confidant support nor the path from SPC to affective support was significant. Thus, Hypothesis 2, which posited an indirect relationship between SPC and PPD symptoms that runs through social support, was not supported.

Figure 3.

Standardized path coefficients for the relationship between SPC, social support, and postpartum depression symptoms, N = 199.

To test Hypothesis 3, standardized coefficients were estimated for each path specified in Figure 2. As the results summarized in Figure 4 show, this model fit the data well, according to Hu and Bentler’s (1999) standards. The paths between Never contacting SPC group members outside of sessions and the two social support variables were nonsignificant. Women who Rarely contacted SPC group members outside of sessions exhibited PPD scores that were 0.37 standard deviations greater than the scores of women in TC [((−1.75)(−0.67)) + ((−2.04)(0.39))]. Women who Sometimes contacted SPC group members outside of sessions exhibited PPD scores that were 0.31 standard deviations lower than scores of women in TC [((0.92)(−0.67)) + ((0.79)(0.39))]. Women who Often contacted SPC group members outside of sessions exhibited PPD scores that were 0.36 standard deviations lower than scores of women in TC [((1.0)(−0.67)) + ((0.79)(0.39))].

Figure 4.

Standardized path coefficients for the relationship between contact with SPC group members, social support, and postpartum depression symptoms.

Findings from this path analysis provide general support for Hypothesis 3. Greater levels of contact with SPC group members are related to decreased PPD symptoms through a relationship with affective and confidant support. However, some unexpected findings did emerge. First, affective support was related to increases (rather than decreases) in PPD symptoms. Second, an unfavorable effect was observed for women who reported Rarely contacting SPC group members outside sessions. Finally, despite the noteworthy magnitude of significant effects and good model fit, the residual variance for the two social support variables was high, especially for affective support.

Discussion and Conclusions

The present study tested the hypothesis that group prenatal care has an indirect effect on PPD symptoms, which is facilitated through a direct effect of group care on two specific forms of emotional social support (i.e., confidant and affective support). Diverging from previous research on psychosocial outcomes of group prenatal care, this study conceptualized group care as both (1) a dichotomous variable comparing group care with TC and (2) an ordinal variable comparing different levels of engagement with group care members with TC. This latter conceptualization is important considering that the underlying logic of the group prenatal care model relies on the assumption that women will develop relationships with one another (Massey et al., 2006; Rising, 1998). However, not all women will develop such relationships and, thus, may not experience all the intended benefits of group prenatal care.

Path analysis results provided no evidence of an indirect relationship between SPC and PPD symptoms when modeling the dichotomous group care variable. However, modeling the ordinal variable revealed more favorable PPD outcomes among women reporting greater contact with group members. Importantly, there was a sizable unfavorable effect on PPD symptoms among women who reported rarely contacting other group members outside sessions. These women experienced significantly greater PPD symptoms than the women in TC. This has important implications for practice, as it suggests psychosocial benefits of group prenatal care may only be exhibited by women who make connections with other group members. Women who do not forge relationships with other group members may actually experience deleterious effects, perhaps as a result of unfulfilled expectations for social connection. Thus, it may be worthwhile for group prenatal care providers to encourage the women in their groups to build relationships and socialize with each other outside of group sessions.

It is important to note that, although the model with the ordinal variable fit the data well, the residual variance for social support was high, especially for affective support. This indicates that contact with group members explained minimal variation in affective support. Additionally, affective support was related to higher (rather than lower) measures of PPD symptoms. As previously discussed, confidant support “reflects primarily a confidant relationship where important matters in life are discussed and shared,” whereas affective support “reflects a more emotional form of support or caring” (Broadhead et al., 1988, p. 715). Perhaps confidant support (e.g., discussing important matters of life) has more practical utility than affective support (e.g., getting love and affection) during life changes associated with the postpartum transition. Future qualitative work should explore this possibility.

Findings from this study must be interpreted within the context of some key limitations. Women in this study self-selected into group prenatal care or TC. Although selection bias was accounted for with propensity score weighting, future research would benefit from randomly assigning women to groups. Additionally, this study relied on samples from two sites with very distinct demographic profiles. To increase statistical power, these samples were combined and analyzed together using multilevel modeling to account for clustering within site. Future research should use larger samples to permit analyses of subsamples of women in order to determine any differences in relationships between group prenatal care, social support, and PPD symptoms based on demographic profiles.

In general, findings from this study indicate that group prenatal care can improve social support and reduce risk of developing PPD symptoms among pregnant women, but only among those who actively engage with other group members. A large amount of unexplained variance in the social support variables suggests that social support (and, consequently, PPD) may be influenced by a number of variables that were not modeled in this study. Thus, although group prenatal care may have some psychosocial benefits for women who engage with group members, it is important for group prenatal care providers to address a range of PPD risk factors in addition to low social support (e.g., obesity, sleep disturbance, unhealthy diet, gestational diabetes, anxiety, violence/abuse, low marital/partner satisfaction, life stress, immigration/acculturation-related stressors, cesarean section delivery, multiple birth delivery, preterm birth, low birthweight, birth complications; Hutchens & Kearney, 2020; Xiao-hu & Zhi-hua, 2020). Additionally, it is vital that group prenatal care providers be prepared to refer high-risk patients to services that have demonstrated effectiveness at preventing or treating PPD symptoms, such as psycho-pharmacological and psycho-therapeutic approaches (Cuijpers et al., 2008; Huang et al., 2020; Zhou et al., 2020). Ultimately, group prenatal care may play an important role in preventing PPD symptoms when implemented as one aspect of a holistic approach to care.

Footnotes

Acknowledgements

The authors would like to thank Emily Tanner-Smith for her methodological guidance during the early phases of this study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by March of Dimes. The opinions expressed by the study authors do not necessarily reflect the opinions of March of Dimes or of March of Dimes funders.

ORCID iDs

Heather Hensman Kettrey

Katarzyna T. Steinka-Fry

References

Ardiani

Soemanto

R. B.

Murti

(2020). Meta-analysis: The association between social support and postpartum depression. Journal of Maternal and Child Health, 5(6), 641–650. https://doi.org/10.26911/thejmch.2020.05.06.04

Asher

H. B.

(1983). Causal modeling (2nd ed.). Sage.

Broadhead

W. E.

Gehlbach

S. H.

de Gruy

F. V.

Kaplan

B. H.

(1988). The Duke-UNC functional social support questionnaire: Measurement of social support in family medicine patients. Medical Care, 26(7), 709–723. https://doi.org/10.1097/00005650-198807000-00006

Buultjens

Farouque

Karimi

Whitby

Milgrom

Erbas

(2020). The contribution of group prenatal care to maternal psychological health outcomes: A systematic review. Women and Birth. Advance online publication. https://doi.org/10.1016/j.wombi.2020.12.004

Chae

S. Y.

Chae

M. H.

Kandula

Winter

R. O.

(2017). Promoting improved social support and quality of life with the CenteringPregnancy group model of prenatal care. Archives of Women’s Mental Health, 20, 209–220. https://doi.org/10.1007/s00737-016-0698-1

Cox

J. L.

Holden

J. M.

Sagovsky

(1987). Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry, 150(6), 782–786.

Cuijpers

Brannmark

J. G.

van Straten

(2008). Psychological treatment of postpartum depression: A meta-analysis. Journal of Clinical Psychology, 64(1), 103–118. https://doi.org/10.1002/jclp.20432

Gottlieb

B. H.

Bergen

A. E.

(2010). Social support concepts and measures. Journal of Psychosomatic Research, 69(5), 511–520. https://doi.org/10.1016/j.jpsychores.2009.10.001

Guo

Fraser

M. W.

(2015). Propensity score analysis: Statistical methods and applications (2nd ed.). Sage.

10.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariate structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

11.

Huang

Yang

Lei

Yan

Tian

Huang

Lei

(2020). The short- and long-term effectiveness of mother-infant psychotherapy on postpartum depression: A systematic review and meta-analysis. Journal of Affective Disorders, 260(January), 670–679. https://doi.org/10.1016/j.jad.2019.09.056

12.

Hutchens

B. F.

Kearney

(2020). Risk factors for postpartum depression: An umbrella review. Journal of Midwifery & Women’s Health, 65(1), 96–108. https://doi.org/10.1111/jmwh.13067

13.

Ickovics

J. R.

Reed

Magriples

Westdahl

Rising

S. S.

Kershaw

T. C.

(2011). Effects of group prenatal care on psychosocial risk in pregnancy: Results from a randomised controlled trial. Psychology & Health, 26(2), 235–250. https://doi.org/10.1080/08870446.2011.531577

14.

Kettrey

H. H.

Steinka-Fry

(2020). Effects of March of Dimes Supportive Pregnancy Care on maternal and infant health across diverse patient populations: A quasi-experimental multi-site pilot study. Prevention Science, 21, 293–307. https://doi.org/10.1007/s11121-019-01083-5

15.

Lee

B. K.

Lessler

Stuart

E. A.

(2011). Weight trimming and propensity score weighting. PLoS One, 6(3), e18174. https://doi.org/10.1371/journal.pone.0018174

16.

March of Dimes. (2017, June 8). Pregnancy loves company: How Supportive Pregnancy Care benefits moms & babies. https://medium.com/@marchofdimesfnd/pregnancy-loves-company-how-supportive-pregnancy-care-benefits-moms-babies-894b7c02ea47

17.

Massey

Rising

S. S.

Ickovics

(2006). CenteringPregnancy group prenatal care: Promoting relationship-centered care. Journal of Obstetric, Gynecologic, & Neonatal Nursing, 35(2), 286–294. https://doi.org/10.1111/j.1552-6909.2006.00040.x

18.

Mazzoni

Hill

Briggs

Barbier

Cahill

Macones

Colditz

Tuuli

Carter

(2020). The effect of group prenatal care for women with diabetes on social support and depressive symptoms: A pilot randomized trial. Journal of Maternal-Fetal & Neonatal Medicine, 33(9), 1505–1510. https://doi.org/10.1080/14767058.2018.1520832

19.

R Core Team. (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing.

20.

Rising

S. S.

(1998). Centering pregnancy: An interdisciplinary model of empowerment. Journal of Nurse-Midwifery, 43(1), 46–54. https://doi.org/10.1016/S0091-2182(97)00117-1

21.

Robertson

Grace

Wallington

Stewart

D. E.

(2004). Antenatal risk factors for postpartum depression: A synthesis of recent literature. General Hospital Psychiatry, 26(4), 289–295. https://doi.org/10.1016/j.genhosppsych.2004.02.006

22.

Rosseel

(2021). The lavaan tutorial. Ghent University.

23.

Shorey

Chee

C. Y. I.

E. D.

Chan

Y. H.

Tam

W. W. S.

Chong

Y. S.

(2018). Prevalence and incidence of postpartum depression among healthy mothers: A systematic review and meta-analysis. Journal of Psychiatric Research, 104(September), 235–248. https://doi.org/10.1016/j.jpsychires.2018.08.001

24.

Slomian

Honvo

Emonts

Reginster

J. Y.

Bruyere

(2019). Consequences of maternal postpartum depression: A systematic review of maternal and infant outcomes. Women’s Health, 15, 1–55. https://doi.org/10.1177%2F1745506519844044

25.

Spybrook

Bloom

Congdon

Hill

Martinez

Raudenbush

S. W.

(2011). Optimal design plus empirical evidence: Documentation for the “Optimal Design” software Version 3.0. William T. Grant Foundation.

26.

Walker

D. S.

Worrell

(2008). Promoting healthy pregnancies through perinatal groups: A comparison of CenteringPregnancy® group prenatal care and childbirth education classes. Journal of Perinatal Education, 17(1), 27–34. https://doi.org/10.1624%2F105812408X267934

27.

Xiao-hu

Zhi-hua

(2020). Risk factors for postpartum depression: An evidence-based systematic review of systematic reviews. Asian Journal of Psychiatry, 53(October), 102353, https://doi.org/10.1016/j.ajp.2020.102353

28.

Zhou

Wang

Zhu

Feng

Xue

Yang

(2020). The effectiveness of mHealth interventions on postpartum depression: A systematic review and meta-analysis. Journal of Telemedicine and Telecare. Advance online publication. https://doi.org/10.1177%2F1357633X20917816