A Community-Centered Approach to Strengthening Perinatal Care Connections

Abstract

Purpose

Racial disparities in maternal health outcomes are a public health crisis in the U.S. Adequate connection to pregnancy-related resources is a strategy for improving maternal outcomes (Trost et al., in Pregnancy-related deaths: Data from maternal mortality review committees in 36 States, 2017–2019. Centers for disease control and prevention. https://www.cdc.gov/reproductivehealth/maternal-mortality/erase-mm/data-mmrc.html, 2022), yet patients receive little support navigating complex systems. We tested the feasibility of a transition-of-care program that identifies individuals in early pregnancy who visit the emergency department (ED) and facilitates connections to needed healthcare and supportive resources.

Description

This pilot used a regional Health Information Exchange (HIE) to identify people from EDs across four counties in South Jersey with evidence of a current or recent pregnancy and limited connection to care. Eligible patients were assigned to a partner site who contacted them to offer scheduling support for pregnancy-related care and connection to supportive resources. The pilot initially focused on prenatal care but expanded to include other supports based on patient needs.

Assessment

Of the 2073 eligible patients, 896 were contacted, and 379 accepted one or more types of support. Support was accepted across racial, ethnic, age, and insurance groups.

Conclusion

This pilot illustrated that a perinatal transition of care program from the ED to appropriate pregnancy-related services and resources is feasible. The ED provides a unique opportunity to identify and engage people early in pregnancy who might face barriers to accessing timely care. The model reduced reliance on self-navigation and addressed common access challenges.

Significance

Timely engagement in pregnancy-related care is essential for reducing maternal morbidity and mortality and improving birth outcomes. ER-based interventions offer a promising strategy for closing care gaps, especially for vulnerable populations. This pilot adds new evidence on system-level approaches that promote equity in maternal health.

Introduction

The rising maternal mortality rate and persistent racial disparities in maternal health outcomes are urgent public health issues in the United States (Hoyert, 2022). Early and adequate prenatal care is critical for preventing and treating pregnancy-related complications (Trost et al., 2022), such as gestational diabetes and hypertension (Carter et al., 2017, Cleveland Clinic, 2022). Left undiagnosed and untreated, comorbidities can contribute to poor health outcomes for the mother and child. Most medical and social interventions become possible when patients initiate care. However, first trimester prenatal care rates are disproportionately lower for individuals of color in N.J., contributing to racial disparities during pregnancy and beyond (New Jersey Department of Health, 2022).

Patients with established prenatal care are less likely to have pregnancy-related emergency department (ED) visits (Roman et al., 2021). The ED serves as an entry to healthcare, particularly those facing barriers to care. Comprehensive reviews have demonstrated that the ED can be an effective setting for connecting patients to community resources (Walter et al., 2021), with facilitated appointment scheduling increasing follow-up rates by up to 40% (Zorc et al., 2003) and structured interventions successfully linking patients to care (Scott et al., 2024). Examples of successfully implemented ED-based connection models include those for substance use disorders (Bernstein & D’Onofrio, 2013), contraception (Hoehn et al., 2019), and asthma management (Zorc et al., 2003). These interventions may be important for addressing documented disparities in care among patients with multiple pregnancy-related ED visits, which disproportionately affect patients based on race, insurance type, age, and social needs (Ho et al., 2022; Malik et al., 2017; Miller et al., 2019). Connection to an obstetrician/gynecologist could be important for ED patients who experience early pregnancy loss, as one study found these patients were less likely to have established prenatal care (Benson et al., 2023).

While the ED presents an opportunity to connect patients to preventative care, the feasibility of activating a workflow to connect pregnant ED patients to routine pregnancy-related services has not been studied (Malik et al., 2017). This pilot was designed to engage individuals early in pregnancy who presented to the ED with limited evidence of connection to pregnancy care. The pilot responded to the conundrum that most pregnancy interventions only begin once an individual initiates care independently, therefore relying on the patient to navigate a fragmented system with limited support. The pilot tested the feasibility of using real-time data to identify and offer telephonic outreach to pregnant people following an ED visit, and to gain a better understanding of this population.

Methods

The pilot utilized the Camden Coalition Health Information Exchange (HIE), a data platform that collects medical records from multiple local health systems, to identify people with pregnancy-related needs from 11 regional EDs across three health systems in four counties in South Jersey. In compliance with N.J. law, patients are informed of their right to opt out of contributing their data to the HIE when accessing care at participating sites. Patients who do not opt out consent to the use cases of the HIE including quality improvement. Further consent was not required as this was a quality improvement study, not subject to Institutional Review Board oversight.

Eligible patients included those with potential or confirmed pregnancy, miscarriage or nonviable pregnancy, and limited evidence of connection to care. The inclusion of patients without a confirmed viable pregnancy was intentional, recognizing the clinical uncertainty during early gestation, as well as an acute lack of resources and traditional care pathways for patients with early pregnancy loss. The ED visit did not need to be for a pregnancy-related cause. Eligible patients were identified through a custom HIE report that identified people with recent ED visits and evidence of a current/recent pregnancy. The report used admit-discharge-transfer notifications, ICD-10 codes, urine and blood human chorionic gonadotropin (hCG) lab results, and Perinatal Risk Assessments – a screen required for N.J. Medicaid providers to bill for the delivery bundle, commonly completed following the first prenatal visit, the absence of which indicated potential eligibility. One hospital system, representing an additional two ERs, did not participate in the HIE so the report was duplicated within their EMR. Eligible patients were manually entered into the HIE to ensure data continuity.

The HIE report updated daily. Any patient whose HIE record showed an ED admit date in the past seven days and one of the pregnancy-related data elements within the past four months populated into the report. Camden Coalition staff reviewed the report and the patients’ HIE record to assess eligibility. This eligibility review occurred within a few days of the ED visit.

Eligible patients were assigned to one of six partners: two hospitals, three federally qualified health centers (FQHC), and one community-based organization (CBO) that specialized in resource navigation for the perinatal population.

For the purposes of care continuity and privacy, patients were primarily assigned to the site with which they had an existing treating relationship based on evidence in the HIE (e.g., where they receive primary or previous OB care or the hospital system where they had visited the ED). When patients had relationships with multiple participating partners, other factors were considered (e.g., uninsured patients were referred to FQHCs who can see patients without insurance; individuals desiring termination were referred to a site that offers abortion services). Patients without established care at a participating health system received outreach from the CBO, which signed a subcontractor agreement allowing them to perform outreach on behalf of the Camden Coalition, a N.J. Regional Health Hub authorized to plan and execute population-level and patient-level health interventions in South Jersey.

The Camden Coalition recruited partners through a webinar. This webinar described the need to identify an outreach champion. Outreach champions – typically nurses, social workers, and outreach staff – were responsible for the telephonic post-ED outreach.

Outreach champions accessed the HIE to receive their assigned patients and documented call outcomes in a standardized form. Each site was provided with a $10,000 flexible patient cost fund, provided through philanthropic dollars, to help address barriers identified during outreach. Staff time was provided in-kind.

Outreach champions received workflow-specific training on effective outreach techniques (Fig. 1) and proper HIE utilization. Although the overall process was standardized, workflows varied across sites due to institutional policies, organizational structure, and scheduling practices.

Fig. 1.

Process for identifying and conducting outreach to patients

The Camden Coalition’s program team, including a program manager with a nursing background and a program assistant, met monthly with each site to review data and provide coaching focused on refining workflows, addressing barriers, and continuous quality improvement.

This paper describes the population seeking pregnancy-related care in the ED, assesses the feasibility of a telephonic transition-of-care outreach program, and describes the demographics of patients who accepted support.

While timely initiation and retention in prenatal care were important outcomes, these measures were not applicable to our entire population, which included individuals whose pregnancies ended in non-live births. Limitations in outpatient data available in the HIE prevented us from tracking visit attendance. Because this pilot relied on existing staff capacity, we deliberately minimized additional data collection requirements. Future analyses using administrative data sources such as birth certificate and claims data may allow for deeper insights into utilization and clinical outcomes.

The HIE report extracted demographic data such as age, race, and insurance status. Maternal age at the time of ED visit was calculated using patients’ birthdate. Gestational trimester and pregnancy status were identified from diagnoses and provider notes, then coded by program staff into discreet data fields into a customized HIE form by the program team during the eligibility review. Outreach champions documented call attempts and call outcomes (e.g., support type accepted) in the same HIE form.

Results

Patient Population

2073 patients were identified as eligible and assigned for outreach (Table 1). Across sites, 82.2% of patients received an outreach attempt. Of these, just over half (52.6%) were successfully contacted, equating to approximately 43.2% of the eligible population. Of those reached, 42.3% accepted support, corresponding to 18.3% of all eligible individuals (Fig. 1).

Table 1.

Demographics and characteristics of individuals who were eligible and eventually supported through the Safer Childbirth Cities workflow

	Eligible		Supported
	N = 2073	%	N = 379	%
Race/Ethnicity
Black/African American	731	35.3	137	36.1
White/Caucasian	655	31.6	81	21.4
Hispanic	462	22.3	110	29.0
Other	98	4.7	31	8.2
Asian/Pacific Islander	62	3.0	11	2.9
Unknown	55	2.7	7	1.8
Native American	9	0.4	2	0.5
Multi racial	1	0.0	0	0.0
Age (years)
< 18	45	2.2	5	1.3
18—34	1712	82.6	331	87.3
35 and older	316	15.2	43	11.3
Insurance
Medicaid—HMO	1180	56.9	221	58.3
Commercial	442	21.3	62	16.4
Uninsured/Presumptive Eligibility	238	11.5	55	14.5
Other	92	4.4	18	4.7
Medicaid—no HMO	80	3.9	14	3.7
Out of State Medicaid	22	1.1	4	1.1
Health-System Insurance	15	0.7	4	1.1
Medicare	4	0.2	1	0.3
Trimester at Triage
1st Trimester	1584	76.4	294	77.6
2nd Trimester	186	9.0	30	7.9
3rd Trimester	29	1.4	4	1.1
Unknown Gestational Age	213	10.3	39	10.3
Not Pregnant	61	2.9	12	3.2

Compared to the eligible population, a smaller share of the supported population was identified as White (21.4% vs. 31.6%), and a larger share were of Hispanic ethnicity (29.0% vs. 22.3%), suggesting differences in engagement across groups. Most eligible and supported patients had Medicaid coverage. Patients who were uninsured or presumptively eligible for Medicaid comprised a larger share of the supported population than of the eligible population (14.5% vs. 11.5%), while patients with commercial insurance were less represented among those supported (16.4% vs. 21.3%).

There were no meaningful age differences between those eligible and those supported; the mean maternal age in both groups was 27.8 years. Among eligible patients, 74.8% had a confirmed pregnancy at ED discharge. However, 21.2% of eligible patients and 21.1% of supported patients had an uncertain pregnancy status, such as potential miscarriage/nonviable pregnancy or inconclusive lab/imaging results (Table 2). Overall, 76.4% of eligible patients were in the first trimester, and acceptance rates were consistent across trimesters (Table 1).

Table 2.

Pregnancy Status (at time of triage) of individuals who were eligible and eventually supported through the Safer Childbirth Cities workflow

	Eligible		Supported
	N = 2073	%	N = 379	%
Confirmed Pregnancy	1551	74.8	284	74.9
Potential Miscarriage	236	11.4	44	11.6
Unclear	179	8.6	32	8.4
Confirmed Miscarriage	38	1.8	8	2.1
Possible Ectopic Pregnancy	26	1.3	4	1.1
Recently Terminated	17	0.8	2	0.5
Confirmed Ectopic Pregnancy	14	0.7	3	0.8
No Pregnancy	8	0.4	2	0.5
Recently Delivered	4	0.2	0	0.0

Among the 379 individuals who accepted services, the most frequently accepted support type was scheduling assistance, followed by answered questions about care (e.g. education on pregnancy-related care) (Table 3). More than a third (38.5%) of supported patients accepted more than one type of support.

Table 3.

Types of support provided to individuals who accepted support

	N = 379
Appointment scheduling	121
Answered questions about care	85
Notification of incentive	76
Gift card	75
Referral to additional services	70
Transportation coordination	59
Grocery voucher	49
Reminder of previously scheduled appointment	20

Identified Barriers to Engagement and Appointment Access

Patients who did not accept support cited various reasons, including existing connections to care, discomfort with the call, uncertainty with plans for their pregnancy, or reporting no barriers to care. Outreach champions also reported that the brief nature of the call sometimes made it challenging to build sufficient trust and communicate the purpose of the call. With coaching, capacity-building, and quality improvement efforts, outreach champions showed improvement in addressing patient discomfort and building rapport telephonically.

Patients who accepted support frequently had unanswered questions or were seeking reassurance about navigating their care. One key tactic deployed by outreach champions was to provide patients with the nurse triage line before their appointment for clinical questions or concerns.

Barriers affecting pregnant individuals’ ability to establish and maintain consistent care were documented throughout the pilot. Numerous scheduling barriers were identified including insurance eligibility confusion; health system policies that prevent scheduling too early/late in pregnancy; OB/GYN offices requiring proof of pregnancy prior to scheduling; and long hold times. The proactive outreach and scheduling support offered helped overcome some of these barriers. Additionally, barriers like lost wages due to time away from work to attend appointments, poor access to transportation, and childcare made it difficult to prioritize pregnancy care. People who accepted scheduling support commonly accepted additional assistance, such as transportation, gift cards to cover out-of-pocket costs, and referrals to food access programs.

Discussion

Timely and consistent prenatal care is a fundamental protective factor for preventing adverse maternal outcomes. This pilot illustrated that addressing diverse pregnancy-related services and social supports allows individuals to prioritize their health during early pregnancy. The pilot also showed that a perinatal transition-of-care program from the ED to pregnancy-related services is feasible. The volume of eligible patients presenting to the 11 regional ERs indicates that using ED visit data is a promising strategy for improving connections to care – the annual volume was both meaningful at the population level, and manageable as a daily outreach workflow. Further, patients identified were commonly in early pregnancy, sometimes discovering they were pregnant during their ED visit. This represents a population that would otherwise not be identified unless and until they established care. By using the Health Information Exchange (HIE), we identified this population daily, allowing our partner sites to offer quick follow-up and connection to care. Individuals reached during early pregnancy were often appreciative of an ED visit follow-up call, and welcomed support scheduling an appointment, receiving transportation, or discussing other social needs.

Given our broad inclusion criteria, engagement exceeded our expectations, with approximately half of eligible patients successfully contacted. Based on comparable Camden Coalition workflows and the fact that patients were not forewarned about these calls, we had anticipated lower contact rates. We now hypothesize that patients may be more likely to answer an unknown call (with a local area code) following an ED visit, assuming it may be related to their care. Of those reached, nearly half accepted some type of support, including both appointment scheduling and social resource connection. This relatively high acceptance rate may stem from patients’ existing affiliation with the outreaching health system. Outreach staff identified these pre-existing connections as program strengths, reporting that most patients understood the call purpose and appreciated the proactive support.

Early pregnancy is often marked by uncertainty and loss, as reflected in the pregnancy status data. Outreaching people experiencing a potential loss or undesired pregnancy can be uncomfortable, even for healthcare providers. The pilot illustrated a need for more provider training to support this population. The Camden Coalition, having tested engagement tactics internally, provided technical assistance and coaching to build outreach champions’ confidence having sensitive conversations common to early pregnancy. Coaching included building a repository of best practices, reputable services, and community. Through outreach champion feedback, it became evident that pathways into care for patients experiencing pregnancy loss or considering termination are not well-established. Building capacity to support patients experiencing these situations filled a knowledge gap and cultivated new patient-centered workflows.

The present study has several limitations. First, manual chart review and data entry throughout the workflow introduced potential transcription errors and inconsistencies in patient identification. Second, variation in outreach workflows across the sites likely influenced contact and acceptance rates in ways we cannot quantify. Third, our inability to track actual resource utilization following outreach represents a significant limitation, as patients may have accessed services at non-participating locations or through alternative means. Finally, the study’s limited geographic scope of 11 regional ERs in South Jersey may restrict generalizability.

Further analysis is needed to determine whether this approach improves connection to appropriate resources and whether improved maternal outcomes are achieved. Further analysis is also needed to better understand the characteristics, including social and clinical risk factors, of the individuals who accepted support which could contribute to more efficient triage and identification workflows and resource allocation to serve those at higher risk for facing barriers to care.

Our findings demonstrate the feasibility of implementing a proactive approach to identifying and engaging pregnant patients following ED visits, with implications for clinical practice and our broader health and social care systems. EDs present an under-explored opportunity to identify individuals in early pregnancy and connect them to appropriate care and support. With existing data infrastructure in HIEs and EMRs, our study suggests that modest investments in staff training and flexible patient support funds can yield engagement with a meaningful proportion of patients, particularly those from marginalized communities. Organizations implementing similar programs should develop protocols to address the full spectrum of pregnancy needs, including pregnancy loss and termination services, as these populations have less established pathways into care. Future implementation should incorporate standardized training, robust quality assurance mechanisms, and comprehensive outcome tracking.

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: Michelle Adyniec, Erica Hartmann, Audrey Hendricks, Natasha Dravid; data collection: Audrey Hendricks, Natasha Jogleker, Erica Hartmann, Jhumna Sarkar; interpretation of results: Michelle Adyniec, Erica Hartmann, Audrey Hendricks, Natasha Jogleker, Natasha Dravid; draft manuscript preparation: Michelle Adyniec, Erica Hartmann, Audrey Hendricks, Natasha Jogleker, Jhumna Sarkar, Natasha Dravid.

Funding

This initiative was supported by Merck & Co., Inc., Kenilworth, N.J., USA, through Merck for Mothers, The Burke Foundation, The Nicholson Foundation, and the Community Health Acceleration Partnership as part of the Safer Childbirth Cities initiative.

Data Availability

Not applicable.

Material Availability

Not applicable.

Code Availability

Not applicable.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical Approval

This project was quality improvement in nature and not subject to Institutional Review Board oversight.

Consent to Participate

Participants were given the opportunity to opt out of the Health Information Exchange at their medical appointments of participating health systems and organizations. Further consent was not required as this was a quality improvement study that was part of routine clinical care, not subject to the oversight of an Institutional Review Board.

Consent for Publication

Not applicable.