Enhancing Prenatal Group Medical Visits with Mindfulness Skills: A Pragmatic Trial with Latina and BIPOC Pregnant Women Experiencing Multiple Forms of Structural Inequity

The hypothesis that CenteringPregnancy with Mindfulness Skills (CP +) would show greater benefits than CenteringPregnancy in reducing postpartum depression was supported. While controlling for baseline depression symptoms (CES-D) and family income, there was a statistically significant effect of group assignment on postpartum depression levels assessed with the EPDS (unstandardized coefficient B =  − 3.04, p < 0.05, standardized coefficient β =  − 0.31). Parents in CP + had lower EPDS postpartum depression (M = 4.90; SD = 4.06) than their peers in the CP program (M = 8.68; SD = 5.39), Cohen’s d = 0.80 (e.g., a large effect).

To provide an estimation of clinical significance in addition to statistical significance, we also examined clinical cut-points for the EPDS (≥ 10 and ≥ 11). Using the cut-point of 10 and above on the EPDS to maximize sensitivity and avoid false negatives (Levis et al., 2020), 10% of the CP + group (n = 2 of 20) and 42.1% of the CP group (n = 8 of 19) met the criteria for possible clinical depression. The chi-square test of group differences on the dichotomous indicator was significant (p < 0.05); however, when controlling for income and education as well as baseline CES-D score, there was a marginally significant effect of group condition on EPDS diagnosis outcome (cutoff ≥ 10), B =  − 6.185, p = 0.051, odds ratio = 0.002, 95% CI = [0.000, 1.018].

In addition, while controlling for baseline pregnancy anxiety and family income, there was a marginally significant effect of group assignment on anxiety measured at post-birth (B =  − 4.49, p = 0.085, β =  − 0.26). There was a trend showing that participants in CP + had lower anxiety post-birth (M = 35.80; SD = 7.81) than their peers in CP (M = 40.79; SD = 9.06), Cohen’s d = 0.59 (i.e., a medium effect) (see Table 3 for means and standard deviations of study variables).

The secondary hypotheses regarding group differences in dispositional mindfulness, depression symptoms, perceived stress, positive and negative emotions, and mind–body coping strategies, all measured in the 3rd trimester and post-birth, were not supported at either the 3rd trimester or post-birth time points (p-values > 0.05; e.g., when interaction terms were not included). Thus, moderation effects were further tested at post-birth that found several statistically significant group differences moderated by baseline levels (see Supplemental Materials with plots of the moderation effects). To summarize, parents in the CP + group showed greater improvements in dispositional mindfulness, depression symptoms, perceived stress, and negative emotions at post-birth than mothers in CP if they had higher baseline levels of mindfulness, or lower levels of depression symptoms, perceived stress, or negative emotions, respectively. This suggests that the CP + enhancements were more beneficial for people with better psychological well-being regarding these outcomes compared to CP.

As an indicator of feasibility and acceptability, we examined attrition across groups and found a difference of only 1 more participant lost to follow-up in the CP group compared to the CP + group, suggesting the CP + enhancements did not differentially influence retention. In the 3rd trimester and post-birth assessments, CP + participants were asked to rate on a 4-point Likert-type scale from Strongly Disagree to Strongly Agree whether each MBCP practice helped them feel less stress. On average, participants rated every practice as helpful across both timepoints with the highest rating for the in-session mindfulness practices in the 3rd trimester (M = 2.47, SD = 0.72) and the lowest (though still positive) rating for the body scan at post-birth (M = 2.07; SD = 0.59) (Table 4). Qualitative reports regarding what they found most helpful about the sessions matched the CP + enhancements, e.g., staying calmer and more present in body and mind during pregnancy, staying calm when their baby is crying, noticing baby’s subtle cues, and practicing self-compassion in parenting (a key facet of mindful parenting; Duncan et al., 2009) by recognizing “That there’s no perfect way to do it” (Table 4).

The current study addressed the primary aim of a small trial intended to pilot a mindfulness enhancement to an existing model of group prenatal care, CenteringPregnancy, that was developed with input from CP providers. It was conducted in a community-based partnership with a safety net public hospital and a community-based organization serving pregnant people experiencing homelessness and housing instability. In this context, CenteringPregnancy was compared with a CP + version augmented with content adapted from Mindfulness-Based Childbirth and Parenting with involvement of the developers of both evidence-based intervention models. In a test of group differences on postpartum depression (PPD), participation in the CP + group was shown to be more beneficial, with lower rates of PPD than the CP group, with a large effect size. Thus, support was found for the primary self-report outcome of the trial specified in advance. One of the most robust outcomes of MBPs, including MBCP and other prenatal MBPs, is a reduction in depression (Leng et al., 2023; Yan et al., 2022). CP is also a robust model with available evidence showing many benefits, including some indication of potential impact on depression among pregnant adolescents (Felder et al., 2017). Contrasting the two versions in a time and attention-matched comparison was intended to isolate the additive benefit of the enhancement of GMVs with MBCP skills training. The lower rate of postpartum depression we found for CP + at post-birth follow-up combined with a trend toward lower postpartum anxiety partially supported study hypotheses regarding impacts on mental health. Trend level effects were expected with our limited sample size and require future investigation in larger trials.

Rates of PPD across conditions in our trial were similar to those found by Lönnberg et al. (2021) in an n = 193 RCT of MBCP vs. Lamaze in Sweden that yielded a 9% rate of PPD in the MBCP group vs. a 29% rate in the Lamaze group at 3 months postpartum. Reducing postpartum mood disturbance is of critical importance for early parenting. Depressed mothers show decreased neural activation in response to their own infant’s cries (aligned with behavioral inhibition of appropriate responses to infant cues) compared to non-depressed mothers (Laurent & Ablow, 2012). What is more, PPD is linked with long-term negative effects on child developmental outcomes (Pearlstein et al., 2009). The MBCP RCT in Sweden (Lönnberg et al., 2021) included parenting and child outcomes. They found a significant effect of MBCP on parent-reported child social-emotional development compared to the Lamaze group. Future directions for CP/CP + include longitudinal follow-up of families.

We found a trend toward lower postpartum anxiety in the CP + group when controlling for baseline prenatal anxiety. A recent noninferiority trial of MBSR suggests that mindfulness training can be as effective as psychotropic medication for the treatment of anxiety disorders (Hoge et al., 2023). Given uncertainty about the effects of medications on fetal and child development, and potential for overestimating risks of medication use (Nordeng et al., 2010), pregnant and postpartum people may wish to avoid pharmacotherapy. During pregnancy, for example, exposure to selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors (SNRIs), and second-generation antipsychotics (SGAs) has been shown to be related to elevated risk for poor neonatal adaptation syndrome (Viguera et al., 2023). Earlier studies showed adverse effects of fluoxetine, citalopram, doxepin, bupropion, and nefazodone in infants who were breastfeeding, yet untreated PPD has comparable adverse effects and other risks such as suicide and infanticide (Pearlstein, et al., 2009). Avoiding medication requires effective alternative treatment options; evidence is accumulating that mindfulness training may be a viable alternative for some mental health conditions.

Other hypothesized secondary outcomes on psychological well-being were only partially supported, with statistically significant differences found between the two groups only for people with better baseline levels of functioning on mindfulness, depression symptoms, or negative emotion. The sample size here was quite small for moderation analyses and should be interpreted with caution. These findings contrast with a systematic review and meta-analysis of perinatal MBPs for pregnant women with and without mental health issues that showed greater improvement for those with worse mental health (Yan et al., 2022). It may be that receipt of care occurred too late in pregnancy to reduce the risk presented by poor prenatal psychological well-being. The average gestational age in weeks of participants at baseline assessment was 19.67 weeks (SD = 3.85). There may also have been limitations to our measurement approach. Although we used measures validated in Spanish, interviewer notes from the study visits indicate that our conceptualization of stress and coping was a mismatch for some participants’ beliefs and cultural norms around not naming pregnancy-related anxiety or stress and instead focusing on a positive, faith-based approach to viewing pregnancy as a blessing. This contrasts with provider reports of prenatal anxiety in our pre-pilot survey.

The MBCP additions to CP + are also low-dose compared to the full 9-week course, with short in-session practices, and gentle invitations to practice mindfulness outside of group when possible, using guided audio instructions ranging in duration from approximately 6 min of breath awareness practice to 27 min of mindful movement practice. In contrast, MBCP participants are encouraged to practice mindfulness meditation for 30 min per day, 6 days per week for 9 weeks, similar to the 8-week MBSR and MBCT programs. Other low-dose mindfulness programs have shown benefit on depression (Virgili, 2015; Xia et al., 2022), as we saw here, but it may be that more practice is needed to shift the other indicators. One meta-analysis examining the dose–response relationship between MBPs and psychological outcomes showed MBPs were beneficial, yet dose did not robustly influence outcomes for depression, anxiety, and stress, whereas MBP doses related to program intensity, actual program use, and facilitator contact were key to improvements in mindfulness (Strohmaier, 2020). In our study, although participants were receptive to the mindfulness content during their group sessions, they often reported substantial barriers to practicing mindfulness outside of their CP + groups, e.g., due to the lack of access to a private and quiet space to practice. Unfortunately, we do not have sufficient data on participant group session attendance or home practice to report those rates or conduct dose–response analyses. These gaps in our data represent an important limitation of the study and are areas for future investigation, particularly when adopting a fully pragmatic trial design vs. the intent-to-treat analysis approach taken here.

Our partner sites—the midwifery service of the safety net public hospital and the community-based organization—were focused on linking participants with needed services around housing, food security, legal support, and healthcare. Ninety-two percent of our sample received Medicaid vs. a national average of 41% of women giving birth in 2021 (Osterman et al., 2023). Both groups (CP and CP +) received equivalent access and support for utilizing those services. We saw group differences at baseline in family income with the CP + group reporting higher income than the CP group. We controlled for both family income and baseline levels of outcomes in our analyses per clinical trial analysis guidelines, yet substantially lower family income likely has qualitatively different detrimental impacts on multiple dimensions, particularly in the very high cost of living area of the study, that are unaccounted for by simple indicators. Ongoing research is testing the provision of cash transfers (e.g., child tax credit or universal basic income) to families with young children and their impact on child and family well-being long-term. A natural experiment with n = 1266 participants demonstrated better outcomes, including mental health, physical health, and financial well-being, in adulthood 20 years later for people whose families had received cash transfers when they were children (Copeland et al., 2022). Until systems can provide everyone with the basic levels of resources needed for sustaining a healthy and safe life for their new babies, it is to be expected that experiencing very low income during the prenatal period would lead to distress that group care and mindfulness are sorely insufficient to ameliorate. That said, we did hear from participants that mindfulness helped them to cope with stressful life events in ways that may have made accessing other services easier.

There were indications of receptivity to the mindfulness additions evidenced in the participants’ rating of the MBCP elements and in their open-ended responses. Retention was nearly equivalent in the two conditions, i.e., CP +  = 20/25 (80%) and CP = 19/24 (79.2%), with slightly lower rates of attrition than found in a systematic review of mindfulness-based perinatal interventions (Leng et al., 2023) that calculated average attrition across studies as 22.5% for the experimental groups and 23.2% for the control groups. In our study, satisfaction with care and birth experiences was high across the two conditions. Participants particularly appreciated that their labor and delivery experiences involved a low level of medical intervention (Liu et al., 2017). It was a consideration when planning the study that the mindfulness components may not be well-received by participants, and these indications suggest that it was feasible and did not undermine satisfaction with care offered through CP in a midwifery care context. Thus, the study met at least one criterion of non-inferiority that is desirable when adapting an evidence-based program that could be further tested in an equivalence or non-inferiority trial. The safety net public hospital setting was somewhat unique in that it prioritizes midwifery care and CP as options for pregnant people in an effective partnership with obstetrics, which could serve as a model for prenatal care more broadly (Hutchison et al., 2011).

Although further research on CP + is certainly needed following this small pilot, when considering dissemination of this model, provider training is a central factor. MBCP facilitator training requires a depth of personal mindfulness practice (in alignment with certification criteria for other major MBPs) that was not required in the current study. It is unknown if facilitating lower dose MBPs, such as CP + , with fidelity requires the same level of professional training and personal mindfulness meditation practice as MBCP, MBSR, or MBCT. CP providers are trained by the Centering Healthcare Institute in “facilitative leadership,” an essential element of the CP model that is in alignment with mindful listening. One study of CP demonstrated that process fidelity to using a facilitative leadership style contributed more to CP effects on birth outcomes than content adherence to discussion topics (Novick et al., 2013). Full-dose MBCP professional training may be needed to prepare to deliver CP + most effectively and may lead to better CP fidelity (even in the absence of offering MBCP content for participants) through strengthening the facilitative group process dimension. It may be that facilitative leadership was the active mechanism of change in our trial, leading to some better outcomes in the CP + group due to provider mindfulness training. We were, however, unable to assess this.

Following completion of study data collection, we provided a daylong mindfulness training for CP providers and group co-leaders from our partner sites. In an evaluation of the training, they reported their enthusiasm for adopting the CP + model and for more training to do so. When asked about key takeaways, several respondents noted things like, “I need to prioritize my own mindfulness practice.” They were also far more interested in learning about how to implement the specific MBCP practices than about the research behind the approach. Since that time, CHI has added some mindfulness elements to the CP curriculum and our partner sites are continuing to implement some aspects of CP + . The midwifery service has also increased the number of CNM providers who identify as Latina and has begun CP groups for Black-identified pregnant people offered by Black-identified CNMs, which offers more racial/ethnic concordance between BIPOC patients and their prenatal care providers.

In the current study, the White investigators and CNMs lacked racial/ethnic concordance with the predominantly Latina BIPOC participants who made up most of the sample, which was only 10% White. Study staff and group co-leaders, some of whom identify as Latina, had greater concordance and ability to meet linguistic needs; however, no providers or study staff identified as Black or Indigenous. Some of the Black women in the SF Bay Area study emphasized their strong preference for receiving care from BIPOC providers (Kemet et al., 2022). Two participants identified as LGBTQ + , a population that commonly experiences deficits in competent and respectful prenatal care (Kukura, 2022); our partner site is known for providing high-quality care for LGBTQ + people. For both CP + and MBCP, additional linguistic and cultural adaptation is also likely needed both for intervention content and measurement tools. For example, when the study was being planned, we used “atención plena” as the Spanish-language translation of the term “mindfulness.” More recent work suggests that “atención consciente” or “conscious attention” provides a better translation as it captures the awareness dimension of the construct (Ibinarriaga Soltero, 2021; Ibinarriaga Soltero et al., 2023). Recent recommendations for culturally responsive MBPs for African Americans include using African American facilitators, culturally familiar settings, and incorporating cultural values and terminology (Watson-Singleton et al., 2019). Across healthcare in the U.S., much work is needed to address inequities and as Carter, the EleVATE Women Collaborative, and Mazonni (2021; p. 108) have pointed out, “group prenatal care alone will not dismantle structural racism. Healthcare institutions need to name ‘racism’; change discriminatory policies; disaggregate data by race and ethnicity to identify inequities; and reallocate time, money and personnel in pursuit of this effort.”

On the continuum of pragmatic trials to explanatory researcher-controlled trials, our study was closer to the real-world pragmatic end (Patsopoulos, 2022); however, it could have been greatly strengthened using community-based participatory (CBPR) or participatory action research (PAR) frameworks. We gathered CP provider input through the survey conducted prior to developing, implementing, and testing CP + ; however, we did not adequately engage affected communities in voicing their needs and desires regarding integrating prenatal healthcare and mindfulness training beyond the pre-pilot groups conducted in English. Ongoing and pending collaborative work is oriented toward following the lead of Latina and other BIPOC community leaders, community health workers, and doulas advancing maternal-child health equity using principles from liberation psychology.

CP is currently available in 44 states and territories through nearly 500 sites (CHI, 2023). Thus, if CP + proves effective in larger trials, it has a ready dissemination network. There also may be avenues for testing further cultural adaptation and dissemination of the full MBCP program under BIPOC leadership with CBPR and PAR methods. Childbirth education holds great public health potential as it is widely accepted and accessed, with reports of half of pregnant people attending some form of childbirth education annually in the U.S. (Declercq et al., 2013). Childbirth education has shown some positive effects on obstetric-related outcomes in recent research yet does not appear to improve perinatal mental health (Vanderlaan et al., 2023). A strength of the MBCP approach is that it can replace traditional hospital and community-based childbirth education if adequate resources are allocated. An abbreviated version of MBCP is being examined for delivery within the National Health Service in the U.K. with promising preliminary results (Warriner et al., 2018).

Adapting evidence-based interventions must be undertaken with care. For mindfulness programs, Loucks et al. (2022) have provided guidelines for doing so with integrity, with which our approach aligns. Ultimately, if the mindfulness field is to move toward greater inclusivity at minimum, and liberatory practice in the face of systemic injustice as a visionary intention, we must follow the leadership of BIPOC scholars such as Michael Yellowbird on “Decolonizing Mindfulness,” (Yellow Bird, 2016), Rhonda Magee on “The Inner Work of Racial Justice” (Magee, 2019), and Natalie Watson-Singleton, Angela Rose Black, and Mindfulness for the People on mindfulness by and for Black women (Watson-Singleton & Black, 2022), among others. Their work has informed our interpretation of the results of this small study and a reimagining of how our future directions may better serve expectant parents and the babies yet to be born, to whom we offer the wishes from the CP + lovingkindness meditation:

“May you be safe and protected, Que esté a salvo y protegidx,

May you be happy, Que sea feliz,

Maybe you be healthy, Que esté sanx,

May you live with calm and in peace, Que viva con tranquilidad y en paz.”