MOMs Chat & Care Study: Rationale and design of a pragmatic randomized clinical trial to prevent severe maternal morbidity among Black birthing people

Abstract

Background:

Black birthing people are disproportionately affected by severe maternal morbidity (SMM). The MOMs Chat & Care Study (R01NR021134) is a pragmatic, randomized clinical trial designed to test the effectiveness of an integrated care model to facilitate timely, appropriate care for high-risk Black birthing people and reduce the risk for SMM.

Methods:

We will recruit 674 adult, English and Spanish-speaking Black birthing people who are less than 17 weeks gestational age, considered high risk based on the Obstetrics-Comorbidity Index and/or history of preeclampsia, and receive care at a Northwell Health obstetric practice. Participants will be randomized to either MOMs High Touch or Low Touch. In both intervention arms participants will receive close monitoring via chatbot technology and navigation to timely care and services by the MOMs team throughout the prenatal and postpartum periods, Fitbit to track physical activity, and bi-weekly postpartum telehealth visits up to 6-weeks postpartum. MOMs High Touch will also receive 12 bi-weekly self-management support telehealth visits during pregnancy and a home blood pressure monitor. The two arms will be compared on incidence of SMM at labor and delivery (Aim 1), SMM-related hospitalizations at 1-month and 1-year postpartum (Aim 1a), time to preeclampsia diagnosis and treatment (Aim 2), perceived social support (Aim 3), and physical activity trajectories (exploratory Aim 4). Mixed methods will be used to examine facilitators and barriers to intervention implementation (Aim 5).

Conclusion:

Findings from this study will inform how to feasibly implement an effective and sustainable integrated care approach to address SMM disparities.

Severe maternal morbidity (SMM) – “unexpected outcomes of labor and delivery that result in significant short- or long-term consequences to a birthing person’s health¹” – is a health equity priority as it disproportionately affects Black birthing people leading to increased risk for rehospitalization post-delivery,² disability,³ and mortality.¹ Black birthing people have the highest rate of SMM throughout the birthing continuum⁴ and are twice as likely to experience SMM compared to non-Hispanic White birthing people.⁵ These racial/ethnic disparities persist even when adjusting for economic background and medical coverage.^6,7 Hypertensive disorders during pregnancy, notably preeclampsia,^8–12 as well as physical inactivity,^13–17 obesity,^18,19 poor mental health,^20–22 lack of social support,²³ and the experience of health-related social needs (e.g., food insecurity, unstable housing, lack of transportation)^24–27 increase risks for SMM and exacerbate disparities. However, most cases of SMM are avoidable if timely, appropriate care is provided.¹

In 2020, we launched the MOMs (Maternal OutcoMes) Navigation Program at Northwell Health, the largest healthcare provider in New York State. Applying evidence-based protocols for transitional care management after hospital discharge,^28–30 MOMs is a patient-centered, integrated care model approach that consists of access to weekly chats via a chatbot that provides education and resources throughout the birthing continuum and a team of care management coordinators (CMCs – non-clinical staff), registered nurses (RNs), and a nurse practitioner who identify and support high-risk birthing people for up to six weeks post-delivery. From 2020 to 2021, there were 2500 racially and ethnically diverse, high-risk birthing people enrolled in the MOMs Program.³¹ Participation in MOMs was associated with an overall 56% reduction in 30-day post-delivery rehospitalizations related to indicators of SMM. Among Black birthing people, there was a 77% reduction in SMM-related hospitalizations. These preliminary findings are promising; however, the long-term effectiveness of the MOMs Navigation Program needs to be established. Furthermore, because SMM can occur throughout the birthing continuum, examining the feasibility and effectiveness of extending the MOMs Navigation Program to the prenatal period is warranted.

The MOMs Chat & Care Study (R01NR021134) is a five-year, pragmatic, randomized clinical trial funded in September 2023 designed to test the effectiveness of an integrated care model (i.e., MOMs Navigation Program) at two different levels of intensity (High Touch [telehealth visits during prenatal period + original MOMs Navigation Program] vs. Low Touch [original MOMs Navigation Program]) to facilitate timely, appropriate care for high-risk Black birthing people and reduce the risk for SMM. The aims of the trial are to:

1. Compare MOMs High Touch (MOMs-HT) to MOMs Low Touch (MOMs-LT) on the incidence of SMM at the time of labor and delivery.

   1. Compare the two study arms on the incident rate of SMM-related hospital admissions at 1-month and 1-year postpartum.

2. Compare the two study arms on time to documented preeclampsia diagnosis and initiation of treatment (low-dose aspirin).

3. Examine the effect of the two study arms on perceived social support domains (informational, emotional, and tangible) from enrollment to 1-month and 1-year postpartum.

4. Explore the effect of each study arm on patterns of engagement in physical activity from study enrollment to 1-year postpartum and subsequent association with maternal health outcomes.

5. Examine implementation determinants and outcomes using a mixed methods approach.

This manuscript describes the study design and the protocols for recruitment, intervention delivery, data collection, and analyses.

METHODS

Design

MOMs Chat & Care is a two-arm, pragmatic randomized clinical trial with a qualitative component to examine facilitators and barriers to intervention implementation and sustainability. This trial has been approved by the Northwell Health Institutional Review Board (IRB). The overall trial design is illustrated in Figure 1.

Figure 1. MOMs Chat & Care Study Design.

MOMs Chat & Care is a two-arm, pragmatic, randomized clinical trial. Potentially eligible patients are identified using the Northwell MOMs Navigation Program Maternal Health Registry. Eligible patients are randomized to either MOMs High Touch (MOMs-HT) or MOMs Low Touch (MOMs-LT) during the prenatal period up to 17 weeks gestational age. Participants will receive their assigned intervention up to delivery and will then receive bi-weekly postpartum check-in calls for the first six weeks of the postpartum period. Participants will be followed up to one-year postpartum.

Setting

The trial is taking place at Northwell Health, which serves over 10 million people residing in urban, suburban, and rural communities throughout Long Island, New York City, and Westchester County and represents a broad spectrum of racial, ethnic, and socioeconomic diversity. Northwell Health is comprised of 21 hospitals and 900 ambulatory sites across New York, including over 50 OB/GYN practice sites. In 2022, the prevalence of SMM among deliveries at Northwell was over three times the national average (4.3% vs. 1.4%⁵) with the highest prevalence among Black birthing people (9.7%).

Participants

During the first month of recruitment (June 2024), study participants needed to meet the following inclusion criteria: 1) aged 18 years or older; 2) self-identify as Black or African American; 3) pregnant with gestational age of less than 15 weeks; 4) Obstetrics-Comorbidity Index (OB-CMI)³² risk score ≥3 and/or history of preeclampsia; 5) English or Spanish as primary language; and 6) receive care at one of 16 Northwell Health Physician Partners (NHPP) obstetrics (OB) practice sites. Note, these 16 sites were selected based on having the highest number of deliveries in 2022. However, it was determined that some of these inclusion criteria were too restrictive and limited the recruitment pool. Thus, in July 2024, we received IRB approval to modify the following inclusion criteria: pregnant with gestational age of less than 17 weeks; OB-CMI risk score ≥ 2 and/or history of preeclampsia; and receive care at one of 18 NHPP OB practice sites. The other inclusion criteria remain the same. Participants are ineligible if any of the above criteria are not met, if they are enrolled in another maternal health study, and if they are not able to provide informed consent due to cognitive or psychiatric impairment.

Procedures

Recruitment.

This study is currently recruiting participants. We are leveraging the existing Northwell MOMs Navigation Program Maternal Health Registry to identify potentially eligible participants for the MOMs Chat & Care Study. Study staff are emailing or mailing all potentially eligible participants a recruitment letter and a study brochure that provides an overview of the study. After 2 days (email invitation) or 5 days (mailed invitation letter), study staff contact the potential participant by telephone to further describe the study and confirm eligibility. Study staff obtain verbal informed consent from eligible and interested participants and proceed with enrollment. Recruitment will take place over 27–30 months with a goal to recruit and enroll 25 participants per month. REDCap is being used to track participant recruitment, enrollment, and completion of questionnaires.

Randomization.

Eligible participants are randomly assigned using a computer-generated algorithm employing a 1:1 ratio, to either the MOMs-HT or MOMs-LT intervention arm. To ensure equal allocation of eligible participants across the 18 recruitment sites, we apply block randomization with varying block sizes. The randomization scheme is generated by the biostatistician prior to enrolling patients with a set seed of the random number generator for reproducibility and auditability. The PI and most Co-Is (with the exception of the biostatistician and the Co-I overseeing intervention training and supervision) are masked to randomization assignment. Participants, their clinical providers, and study staff are not masked to treatment assignment.

Retention strategies.

To retain study participants, we will use effective approaches for retaining racial/ethnic minoritized participants in research.³³ Specifically, study staff and interventionists will maintain regular contact with participants via chatbot, phone, email, and/or telehealth visits, and when possible, study staff and interventionists will match the study population in terms of race, ethnicity, cultural background, and/or language. Participants in both intervention arms will receive $25 for completing the baseline, 1-month, and 1-year postpartum questionnaires ($75 in total), and they will be able to keep self-monitoring devices, issued as part of the study protocol, after the study ends.

Interventions

The integrated care model being tested in this study is informed by the “Three Delays Model.”³⁴ This model theorizes that severe maternal morbidity and mortality occurs because of delays in three phases: Delay 1) delay in birthing person and/or their family recognizing they need to seek care; Delay 2) delay in birthing person reaching a healthcare facility with clinicians and staff trained in obstetrics and/or perinatal care; and Delay 3) delay in birthing person receiving timely and appropriate evidence-based care. Maternal morbidity cannot be attributed to just one of these delays, but a combination of these delays along with other factors.^34,35,36 Furthermore, these delays are interrelated in that a delay in the birthing person recognizing that they need to seek care (delay 1) can result in further delay of them reaching the healthcare facility (delay 2) and receiving timely, appropriate care (delay 3), but also obstacles that lead to delays 2 and 3 (e.g., lack of medical coverage, unreliable transportation) can lead to a birthing person making a decision to not seek care when they need to.³⁴ Thus, the intervention components across both study arms are designed to prevent delays that occur at the patient and health system levels and address clinical and social needs immediately (Figure 2).

Figure 2. Three Delays Model with MOMs-CC Intervention Components.

The MOMs Chat & Care intervention components are informed by the Three Delays Model. To address the delay in the birthing person recognizing they need to seek care, self-management education and support is offered via the Northwell Pregnancy & Peds Chats (i.e., chatbot) and the prenatal telehealth visits. Through the chatbot and navigation to medical and social services provided by the MOMs team, birthing persons will receive support in reaching healthcare facilities and receiving timely, appropriate care.

Participants in both intervention arms will have access to the Northwell Pregnancy & Peds Chats and a wearable device (Fitbit) to track physical activity. Northwell Pregnancy and Peds Chats is a personalized care digital conversation chatbot delivered via the Amwell/Conversa platform. All birthing people are automatically enrolled in the Chats when they initiate prenatal care with a Northwell Health obstetrician. The chatbot is accessible on any electronic device (i.e., smartphone, tablet, or computer) and will be available to participants from the time of enrollment to 1-year postpartum. Chats are not generated by artificial intelligence but programmed with a set of questions and specific actions based on participant responses. During the prenatal and postpartum periods, study participants will receive a weekly chat (in English or Spanish) via text message or email that says, “Your Northwell Health Chat is ready!” with a link to the platform. Each chat begins by asking if there are any updates the patient wants to share regarding their pregnancy. The chats provide timely, brief educational tools and resources to promote a healthy pregnancy and birth; encouragement to engage in physical activity daily and eat a healthy diet; personalized birthing phase recommendations; information about what to expect each week during the birthing continuum and at appointments; baby care during the first year; and the ability to track blood pressure, physical activity, and other measurements.

In addition, the chatbot is designed to capture clinical, behavioral health, and social risks (i.e., housing instability, food insecurity, lack of transportation, low level of support, interpersonal violence) information. If the birthing person responds with any clinical or behavioral health concerns, enters an elevated blood pressure value, and/or reports a social risk, the chatbot is designed to escalate this concern. During normal business hours (Monday through Friday, 8a-4p), a nurse within the MOMs Navigation team responds to alerts, and outside of business hours, alerts route to a 24/7 nurse-led call center. A nurse will follow up via phone within one day for moderate concerns/ ‘yellow flags’ (e.g., lack of transportation to obstetrics appointment), or immediately for severe concerns/ ‘red flags’ (e.g., feeling unsafe at home) and connect the participant to clinical, behavioral health, and/or social services as needed. Participants can also initiate a chat on their own outside of the scheduled weekly chat to request assistance with scheduling appointments or to report any concerns, which will be escalated to the MOMs Navigation team.

MOMs Low Touch (MOMs-LT).

The MOMs-LT intervention arm is similarly designed like the original Northwell MOMs Navigation Program. Once randomized and enrolled, study staff will ensure participants in the MOMs-LT arm are enrolled in the Northwell Pregnancy and Peds Chats to use throughout the prenatal and postpartum periods. They will also mail out the Fitbit within one week of enrollment. Participants will be asked to wear the Fitbit daily during the entire study period. Close to delivery, a MOMs care management coordinator (CMC) or registered nurse (RN - for birthing people with severe high risk such as diabetes with insulin treatment) will be assigned to MOMs-LT participants to conduct clinical, behavioral health, and social risk check-in calls 24 and 72 hours after labor and delivery. The CMC or RN will continue with telehealth visits via phone or videoconferencing every two weeks for the first six weeks of the postpartum period. Navigation to clinical, behavioral health, and social services (e.g., WIC, SNAP) will be provided as needed. Also, CMCs and RNs may remind participants about upcoming appointments, blood pressure monitoring, and taking medications.

MOMs High Touch (MOMs-HT).

For participants randomized to MOMs-HT, study staff will ensure they are enrolled in the Northwell Pregnancy and Peds Chats and will also mail out a Fitbit (to be worn daily during study period) and a home blood pressure monitor (including written and video instructions on self-monitoring blood pressure) within one week of enrollment. Participants will be encouraged to measure their blood pressure twice a day if they have chronic hypertension or preeclampsia, or twice a week if they do not, during the prenatal period. During postpartum, participants will be told to follow provider recommendations on how often they should measure their blood pressure. They will be encouraged to enter their most recent blood pressure reading in the chatbot platform when they receive a weekly chat.

Participants in the MOMs-HT arm will also be assigned a CMC or RN who will deliver self-management support every two weeks for a total of 12 telehealth visits via videoconferencing or telephone (depending on the participant’s preference) during the prenatal period. See Table 1 for a list of self-management support topics that will be covered during the telehealth visits. These prenatal telehealth visits are aligned with ACOG care guidelines and, thus, considered a quality improvement component of this study.³⁷ The first telehealth visit will be 30–45 minutes in duration and follow-up visits will be 15–20 minutes. These telehealth visits will involve brief clinical and behavioral health check-ins, navigation to clinical or social services as needed, follow-up on referrals to resources to address social risks, maternal health education, and self-management support. Clinical check-ins will consist of asking the participant if they have experienced any signs or symptoms of concern since the last check-in (e.g., pain, discomfort, headaches, bleeding), screening for depression and anxiety, and reminders about upcoming medical appointments.

Table 1.

Prenatal Telehealth Visit Session Topics

Signs and Symptoms: Preeclampsia, High Blood Pressure, High Blood Sugar

Target Numbers During and After Pregnancy: Blood Pressure, Blood Sugar, and Weight

Self-monitoring Clinical Measures: Blood Pressure, Blood Sugar, Weight

Taking Action on Elevated Clinical Numbers

Healthy Nutrition for Birthing Parent and Baby

Physical Activity During and After Pregnancy

Weight Management During and After Pregnancy

Managing Stress (include Postpartum Depression and Baby Blues)

Managing Gestational Diabetes

Breastfeeding/ Lactation

Building a Support Network for Behavior Change/ Communicating Needs

Sexual Health/Contraception

Antepartum Vaccinations

Birth Planning

The CMC or RN will assess progress with engagement in self-management behaviors during each visit, including asking about minutes and type of physical activity (via data from the Fitbit), self-monitoring (blood pressure, blood sugar, weight), and taking medications. CMCs and RNs will facilitate goal setting and problem solving with participants to set realistic goals related to their physical activity and other self-management behaviors. Then, they will create an action plan to put their goal(s) into action, while also identifying and addressing barriers (including health-related social needs) to engaging in behavior change.

In the postpartum period, a CMC or RN will conduct a clinical check-in (with navigation as needed) by telephone 24 and 72 hours after labor and delivery. The CMC or RN will continue with telehealth visits via phone or videoconferencing every two weeks for the first six weeks of the postpartum period. Telehealth visits during the postpartum period will be 15–20 minutes in duration. CMCs or RNs may provide self-management support as needed, but the primary focus of these visits will be to assess clinical, behavioral health, and social risk concerns.

Intervention Training and Fidelity

As part of their role on the MOMs Navigation team, all CMCs and RNs are trained in transitional care management and motivational interviewing. For this study, CMCs, RNs, and the MOMs Program Supervisor have also received a 1-week training on behavior change principles, chronic disease self-management, and self-management support co-led by the study PI (Fitzpatrick) and Co-I (Vrany). Booster trainings will be offered annually. Furthermore, CMCs and RNs have participated in a 6-week intensive introductory doula training program delivered by a partnering organization. The doula training included live virtual training modules on pregnancy, childbirth, and postpartum care (including breastfeeding, newborn care, and recovery from childbirth) and assignments that CMCs and RNs completed on their own time to further develop knowledge and skills needed to support birthing people and their families throughout the birthing continuum.

To monitor fidelity to the intervention, the CMCs and RNs will participate in bi-weekly to monthly group supervision/case management with the MOMs Program Supervisor and Dr. Vrany (Co-I). Completion/attendance at prenatal telehealth visits and postpartum check-ins will be tracked and reviewed based on CMC and RN documentation in the EHR.

Clinical and Community Partner Engagement

To support relevancy and sustainability of this study, we formed a Steering Committee and a Community Advisory Board (CAB). The Steering Committee consists of Northwell Health clinicians (e.g., OB/GYN, cardiology), leadership of the Northwell Center for Maternal Health, and leaders and representatives from Health Management (Northwell’s care management entity). This committee meets quarterly and provides input on study design and implementation. The CAB consists of representatives from partnering organizations (Mama Glow, Birth Justice Warriors, Black Coalition for Safe Motherhood, and Women’s Diversity Network) as well as two patient partners that received the MOMs Program in the past. The CAB also meets quarterly and provides input on areas such as recruitment and retention strategies, intervention sustainability, and policy impact.

Study Measures (See Table 2)

Table 2.

Study Outcomes and Data Sources

Outcomes	Data Source	Time Points
		Baseline	Antepartum	Labor & Delivery	1-Month Postpartum	1-Year Postpartum
Primary Outcome
Severe Maternal Morbidity (SMM)a	EHR			X	X	X
Secondary Outcomes
Preeclampsia Diagnosisb	EHR		X
Use of low-dose aspirin	Q, EHR		X	X	X
Perceived Social Support (PROMIS Informational and Emotional Support), Tangible Support	Q	X			X	X
Exploratory
Physical Activity
Minutes of activity per week	Fitbit		X	X	X	X
Barriers to Exercise	Q	X			X	X
Implementation Barriers, Facilitators, and Outcomes	Qual		X	X	X	X
Covariates
Demographics	EHR, Q	X
Pregnancy and Birth History	EHR, Q	X			X
Gestational Age at Delivery	EHR			X
Type of Delivery (vaginal or c-section)	EHR			X
OB-CMI Score	EHR			X
Physical and Mental Health Status
Comorbid Health Conditions	EHR		X	X	X	X
Depression (PHQ-2) and Anxiety (GAD-7)	NWH Chats		X		X	X
Social Risks	NWH Chats		X		X	X
Healthcare Utilization (Outpatient, Inpatient, ED)
Frequency, diagnosis, prognosis	EHR		X	X	X	X

^aSMM defined as having a diagnosis of one or more of the CDC 21 SMM indicators.

^bPreeclampsia diagnosis using ICD10 codes O14.1, O14.10, O14.12, O14.13, O14.14, or O14.15.

Abbreviations: EHR=electronic Health record; Q=Questionnaire; NWH Chats=Northwell Health Pregnancy and Peds Chats; Qual=Qualitative Interviews/Mixed Methods

Severe maternal morbidity (SMM).

The primary outcome for this trial is the incidence of SMM at the time of labor and delivery. SMM (yes or no) is defined as having ≥ 1 ICD-10 diagnosis or procedure codes that correspond to the Centers for Disease Control & Prevention (CDC) 21 SMM indicators,³⁸ which will be extracted from the EHR. This is the current practice used by healthcare systems in the US and will allow for comparison to state and national prevalence. Additionally, the incident rate of SMM-related hospital admissions at 1-month and 1-year postpartum will be evaluated.

Secondary outcomes.

We will also use data from the EHR, chatbot, and questionnaires to capture preeclampsia diagnosis and initiation of low-dose aspirin as treatment during the prenatal period. The following ICD-10 codes will be used to determine preeclampsia diagnosis: O14.1, O14.10, O14.12, O14.13, O14.14, or O14.15. Furthermore, we will assess informational, emotional, and tangible perceived social support. We will administer the 8-item Informational Support and the 8-item Emotional Support scales from the Patient Reported Outcomes Measurement Information System (PROMIS) to all participants over the phone at baseline, and via emailed REDCap link at 1-month and 1-year postpartum. Both measures are on a 5-point Likert scale (ranging from ‘never’ to ‘always’) and have been found to be valid and reliable measures of informational and emotional support in both English and Spanish.³⁹ We will use our newly developed 8-item tangible support measure that assesses the degree to which patients feel supported in accessing clinical care, medications, social services, and other resources to address barriers to care and self-management.

Exploratory Outcomes.

Using data from the Fitbit, we will examine participants’ minutes of physical activity per week from baseline to 1-year postpartum. Participant barriers to participating in exercise will also be assessed using the 14-item Barriers to Exercise Scale,⁴⁰ which will be administered over the phone at baseline and via an emailed REDCap link at 1-month and 1-year postpartum.

Process measures.

We will also extract and examine several process measures. Social risks will be assessed via the chatbot at 14- and 28-weeks’ gestation, 3-days postpartum, 5- and 12-weeks postpartum, and 5, 7, 9, and 11 months postpartum. Health-related social needs identified, referrals to social services and/or community-based resources, and outcome of those referrals will also be documented by CMCs in the EHR and extracted. Participant engagement with the weekly chats from the chatbot, responses to screeners, and viewing of educational tools and resources on the chatbot will also be tracked and extracted. Yellow and red flags from the chatbot, as well as navigation to care and services, maternal healthcare utilization, and diagnosis codes associated with hypertensive disorders or other pregnancy-related complications documented in the EHR will also be captured.

Qualitative data.

We will assess barriers and facilitators to implementing the MOMs-HT and MOMs-LT interventions using a mixed-methods approach. Specific implementation outcomes are feasibility, acceptability, fidelity, and sustainability of the interventions. To examine these implementation outcomes, we will conduct semi-structured interviews with 50 study participants (25 in each intervention arm) and narrative interviews with 10 participants who experience a pregnancy complication (5 in each intervention arm) around 1–2 months postpartum; a focus group with the MOMs Navigation team during Year 3; and collect field notes from Steering Committee and CAB meetings throughout the study period. All qualitative data collection instruments (interview and focus group guides, and structured note templates) were developed for this study and informed by our implementation outcomes and the Three Delays Model.³⁴

Adverse events.

Miscarriages and maternal death will be tracked using the EHR or self-report from participants or family members as adverse events.

Statistical and qualitative analysis plan

Before we carry out analyses, we will audit the data for quality and completeness. We will examine variable distributions for outliers and assess them to ensure that they meet the assumptions of the planned analysis. Baseline characteristics will be compared among those with and without missing data. For missing data, we will use multivariate imputation by chained equation (MICE).⁴¹ Depending on the amount of missing data we will perform sensitivity analyses of different multiple imputation (MI) strategies including MI-predictive mean matching (PMM) and log MI-PMM.⁴² All analyses will be carried out according to the intent-to-treat principle (ITT), i.e., participant data will be analyzed according to the assigned intervention arm. Sensitivity analyses will be performed to compare results from complete case analysis and MI to assess if imputation induced any bias. All analyses will be performed at 0.05 significance level and completed with R statistical software, version 4.4.0 (R Core Team, 2024).

SMM.

We will provide point estimates and 95% confidence intervals (CIs) for the incidence of SMM at labor and delivery in each study arm: MOMs-HT and MOMs-LT. To estimate the cumulative unadjusted incidence risk ratio (RR) for the association between study arms (MOMs-LT as the reference arm) and SMM (yes/no), we will employ a log-binomial regression model with maximum likelihood estimation. The same methodology will be used to estimate the adjusted RR in a multivariable model, adjusting for potential confounders including participant age at enrollment, OB-CMI score, type of medical coverage, body mass index (BMI), and neighborhood deprivation index (NDI).^43,44 Similar to previous studies,^45,46 analyses will be conducted separately for SMM including and excluding transfusions. As an exploratory analysis, we will compare the frequencies of the 21 SMM indicators between the two arms.

Sample Size Calculation.

Based on the prevalence of SMM among Black birthing people who delivered at Northwell Health in 2022 and met study inclusion criteria (9.7%), we assumed a 72% or higher reduction in risk of SMM based on previous studies^31,47 for participants in the MOMs-HT. A sample size of 293 participants per arm would achieve a power of 94% to detect a reduction of at least 7% in the SMM rate (9.7% in the MOMs-LT arm versus 2.7% in the MOMs-HT arm), based on a two-sided z-test with type I error of 0.05. We plan on recruiting up to 337 participants per arm (total of 674) to allow for up to 15% attrition.

Time to preeclampsia and initiation of low-dose aspirin (from randomization) will be estimated using Kaplan-Meier method with observations censored at the time of delivery. Median times-to-event will be reported with 95% CIs and the difference between curves will be assessed with the log-rank test. Cox regression models will be further employed to test the associations between study arms and time-to-event outcomes in univariable analysis and also adjusted for the same covariates used in the primary analysis. Associations will be quantified using hazard ratios (HRs) and 95% CIs.

Perceived social support.

Changes from randomization to 1-month and randomization to 1-year postpartum in the total scores of emotional support, informational support, and tangible support (assessed separately) will be summarized using descriptive statistics (mean ± SD and/or median ± interquartile range) for each study arm. The effect of MOMs-HT compared to MOMs-LT (binary indicator) on score changes for each of these domains will be assessed using linear regression analyses both in univariable and multivariable (adjusting for the same covariates included in the primary analysis) models. If the normality assumption is violated, we will apply a log-transformation of the outcome. Unadjusted and adjusted regression coefficient estimates, and their corresponding 95% CIs will be reported.

Physical activity.

Participants’ trajectories and longitudinal changes in minutes of physical activity (from randomization to labor and delivery and 1-year postpartum) using Fitbit data will be summarized monthly and reported numerically and graphically. Changes in minutes of physical activity over time between the two study arms will be assessed using linear mixed models. To account for the repeated observations nested within an individual, the models will include individual-level random intercepts. The main fixed effects will include study arm, time, and similar covariates included in the primary analysis. To determine whether changes in physical activity are affected by the intervention arm, we will also examine the interaction term between time and intervention arm. If the normality assumption is not satisfied, we will apply a log-transformation of the outcome.

To examine the association between patterns of physical activity and maternal health outcomes, we will apply latent class growth model (LCGM) analysis. Latent class growth models (LCGM) will be used to determine potential subgroup classifications as a function of different trajectories of change in physical activity.⁴⁸ Then, we will employ log-binomial regression models, with study arm as a binary predictor of the physical activity classes, to test if the classes are predictors of: 1) preeclampsia rates (enrollment to labor/delivery) and 2) SMM-related hospital admissions (enrollment to 1-year postpartum), adjusting for covariates mentioned above.

Qualitative analysis.

All interviews and focus groups will be transcribed and coded using NVivo. A qualitative researcher on the study team will develop a preliminary codebook based on our implementation outcomes and the Three Delays Model. The codebook will be iteratively revised during group meetings to incorporate new codes, and merge and eliminate exiting codes. Qualitative, survey, and field note data will be combined and analyzed using convergent mixed methods analysis.⁴⁹

CONCLUSIONS

There is a maternal health crisis in the US that disproportionately affects Black birthing people. SMM and risk factors for SMM may be preventable or at least managed if detected early; timely, appropriate care is provided; birthing people are supported in engaging in self-management behaviors; and health-related social needs are addressed. The MOMs Chat & Care Study aims to understand how to feasibly implement an effective and sustainable integrated care approach to address SMM disparities. In 2022, at Northwell Health, approximately 142 (9.7%) Black birthing people experienced SMM. If the MOMs-HT intervention arm successfully reduces incidence of SMM by 7%, that would be 103 fewer Black birthing people experiencing SMM and instead spending quality time with their newborn. The longitudinal design of this study, including multi-source data collection throughout the birthing continuum, will provide the opportunity to examine mechanisms such as the role of early identification and treatment of preeclampsia, social support, and physical activity in reducing the risk for SMM. Furthermore, the partnership among Northwell Health researchers, clinicians and the Center for Maternal Health as well as maternal health justice community-based organizations will help foster widespread implementation of sustainable solutions to address maternal health and healthcare inequities.