Thriving to Survive: A Case Study of Local Collaborative Efforts to Decrease Infant Mortality

Despite decades of decline in overall infant mortality rates (IMR), across the United Sates, IMR among Black infants remains 2 to 3 times higher than among white infants.¹ The state of Ohio continues to lag most of the nation and has the fourth highest Black IMRs (12.8 per 1000) among the 27 states with data disaggregated by race.² In Franklin County, the most populous county in Ohio, non-Hispanic Black infants account for 31% of the approximate 18 000 births that occur annually but 54% of infant deaths (see Table 1).³ Healthy People 2030 set the goal of reducing the US IMR to 5.0 deaths per 1000 live births. This goal was achieved for non-Hispanic white infants in Franklin County in 2019 (IMR = 4.3), 11 years in advance of the goal date, yet in 2021, Franklin County Black infant mortality was still 2.9 times higher than this goal.⁴ Over the past decade, the increased disparity ratio has largely been driven by the greater improvement in the white IMR compared with their Black counterparts (see Table 2). Between 2011 and 2021, white IMR in Franklin County improved by 37% compared with a 15% improvement in Black IMR, a trend that is mirrored at the national level. At this pace, it is estimated that it will take at least 49 years, or until 2073, for health equity in local IMR to be reached.

The cause of the racial inequity in IMR is complex. There is no single cause that, if addressed, would alone improve outcomes and reduce disparities. Although genetics can contribute to racial inequities in infant health outcomes, the degree of the effect is estimated to be small,^5–7 confirming that race is a social rather than a biological construct. Instead, differences in social status, born of historical and contemporary structural inequities within our society impacting Black and other racially minoritized communities, are the strongest drivers of IM disparities.^8,9 

The multifactorial underpinnings of infant mortality and infant mortality disparities cannot be effectively addressed by any single sector or system. Cross-system advocacy and partnership hold the greatest promise for achieving meaningful improvement. Here, we describe the collaborative efforts undertaken to reduce IMR inequities in Franklin County, Ohio over the past decade.

As one of the largest pediatric hospitals in the nation, Nationwide Children’s Hospital (NCH) is working with community partners to improve the health and well-being of all children in the region, shifting the institution’s focus from primarily individual medical care to encompass active engagement in population health. This shift was operationalized within the framework of the Pediatric Vital Signs initiative, which targeted 8 county-level child health and well-being metrics, including IMR, for measurable improvement.¹⁰ For infant mortality, the aim of the Pediatric Vital Signs initiative is to reduce the overall IMR from 7.1 to 5.9 deaths per 1000 live births and to simultaneously decrease the racial disparities by 50%. Although efforts to improve IMR in the county predate the Pediatric Vital Signs initiative, this initiative helped strengthen local effects by creating a framework calling for shared accountability across community partners, including health care organizations, to improve these metrics.

A key driver diagram approach was employed to develop a combination of approaches that yield universal improvement through broad community-wide campaigns and direct targeting of interventions for Black birthing people and the communities in which they live. Targeted approaches that address the unique barriers experienced by Black birthing people were overweighted in recognition of the fact that the IMR gap will only be reduced if Black infant mortality decreases at a faster rate than that in populations where goals have already been achieved. This 2-pronged approach has been referred to as targeted universalism.¹¹ 

Given Franklin County’s high overall IMR and long-standing racial inequities, collaborative efforts were developed to facilitate improvement. In 2014, the City of Columbus convened academic, medical, and social service partners throughout the city to form the Greater Columbus Infant Mortality Task Force. The Task Force engaged in a review of local data and best practices from comparable cities to develop a set of recommendations for improving IMR and reducing disparities. A public-private partnership, CelebrateOne, now serves as the backbone of a countywide collective-impact process¹² focused on decreasing infant mortality and improving maternal and child health outcomes.¹³ During this time, Franklin County’s 4 largest health systems (NCH, The Ohio State University Wexner Medical Center, OhioHealth, and Mount Carmel Health System) also acted, forming the Ohio Better Birth Outcomes (OBBO) collaborative.¹⁴ Other OBBO partners included the Columbus Public Health department and 3 federally qualified health centers. OBBO has served as the lead entity in the countywide collaborative structure responsible for health care system–based interventions across the prenatal and perinatal periods. Together, CelebrateOne and OBBO collaborate with more than 60 partner agencies to implement the recommendations from the infant mortality task force and develop new strategies to reduce the IMR and racial inequities in birth outcomes.

One of the earliest actions was to identify communities with the greatest opportunity for improvement. In 2014, the Task Force worked with academic collaborators to identify neighborhoods where IMR was the highest. Using data from 2007 to 2011, a hot spot analysis identified 8 neighborhoods that would become target areas for infant mortality prevention.¹⁵ When compared with the county, these 8 neighborhoods had high rates of social inequities, including higher proportions of the population who lived below the federal poverty level, received Supplemental Nutrition Assistance Program benefits, or were housing-cost burdened (ie, spending more than 30% of their income on housing costs alone).¹⁵ Furthermore, these neighborhoods were predominately Black and account for approximately 40% of the county’s infant mortality but only 20% of the county’s population.

Consistent with the national trends, the leading clinical causes of infant deaths in Franklin County are prematurity, congenital anomaly, and sleep-related deaths, accounting for 70% of all infant deaths.¹⁶ As a result, much of the effort to reduce IMR focused on¹ eliminating preventable sleep-related deaths,² reducing the number of premature births, and³ connecting birthing people with the needed resources to temporarily mitigate the social inequities associated with compromised birth outcomes (transportation, housing and food insecurities, improved access to care, etc). In the next sections we highlight some of the collective impact strategies throughout Franklin County that target each of these areas.

Sleep-related infant deaths accounted for 14% of all infant deaths in Franklin County between 2019 and 2022.¹⁷ Over the past decade, there were, on average, slightly more than 20 sleep-related infant deaths per year, with 28 sleep-related infant deaths in 2022.¹⁷ As one of the leading causes of infant deaths in Franklin County, several strategies were implemented to reduce the number of sleep-related deaths that occur each year. One strategy was an educational campaign to promote the ABCs of safe sleep to ensure babies sleep alone, on their backs, and in an empty crib. The safe sleep messages were delivered to the community via print, radio, and television advertisements and by community health workers (CHWs), sororities and fraternities. Building on social capital within communities, CelebrateOne also aimed to train 1000 Safe Sleep Ambassadors each year. In 2022, 1334 sleep ambassadors were trained.⁴ After completing a training led by sleep experts at Columbus Public Health, these ambassadors were empowered to engage in conversations with birthing people in their community about the importance of safe sleep practices for infants. Additionally, 2121 cribs were distributed to families in 2022 to address economic and/or environmental barriers to safe sleep. This exceeded the annual goal to distribute 1700 that year.⁴ 

In the health care setting, OBBO is currently developing training modules to educate all NCH staff on the most updated safe sleep guidelines from the American Academy of Pediatrics. This training aims to equip providers with the tools needed to promote safe sleep during encounters with families during well-child and sick visits in the primary care clinics. To create safe sleep environments where they do not exist, pack ‘n plays are provided to families, when needed. Additionally, OBBO has partnered with local delivery hospitals to provide safe sleep education videos that can be used in clinics or shared directly to families. As these interventions are implemented, process measures are tracked and compared to county-level outcome data to refine the approaches taken to reduce sleep-related infant deaths. Outcome data on the impact of this OBBO initiative as well as the sleep ambassadors and crib distribution programs on infant sleep-related infant death are forthcoming.

Prematurity and related conditions account for nearly a third of infant deaths in Franklin County.⁴ Infants born at less than 32 weeks’ gestation, classified as very preterm, account for most prematurity-related infant deaths. With focused attention, over the past decade, the reduction of very preterm births has slowly fallen. In 2011, the overall very-preterm birth rate was 2.2% compared with 1.9% in 2023, a 13.6% decrease.^3,4 In that period, Black very-preterm births decreased by 21%, from 3.4% to 2.7%,⁴ whereas the white very-preterm birth rate decreased by 23%, from 1.7% to 1.3%.⁴ 

One clinical intervention used to decrease prematurity rates was administration of progesterone to patients identified as good candidates for the treatment within the OBBO clinics. This treatment was thought to reduce the probability of preterm labor in birthing people with a history of preterm labor or with a shortened cervix.¹⁸ OBBO trained providers on how to screen patients for progesterone treatment fit, using guidelines promulgated by the American College of Obstetricians and Gynecologists (ACOG). As a result, utilization increased substantially. However, in April 2023, the Food and Drug Administration (FDA) withdrew approval for the progesterone brand Makena and its generics as an approved drug treatment to reduce preterm birth, citing a trial that failed to show the drug’s ability to reduce the risk of preterm birth or improve infant health.¹⁹ As a result, there is no FDA-approved clinical treatment option for physicians; ACOG has called for more research on effective interventions to address preterm birth.

The loss of progesterone as a medical preventive measure reduced the already limited number of treatment options that OBBO could recommend, but medical care is not the only, or even the primary, driver of prematurity. The associations between preterm birth and social determinants of health including housing, poverty, maternal education, food insecurity, neighborhood exposures to crime, and racism are well documented.⁶ To mitigate these exposures, collaborative countywide efforts focused on delivering individual and community-level social needs interventions. One such intervention was Healthy Beginnings at Home (HBAH) designed to reduce the risk of poor birth outcomes associated with the experience of housing instability or homelessness during pregnancy.²⁰ Currently, housing policies do not prioritize receipt of public and/or rent-assisted housing for pregnant birthing people at high risk of infant mortality. To determine whether such policies could make a difference, partners in Columbus, including CelebrateOne, CareSource—a Medicaid Managed Care Organization—the Columbus Metropolitan Housing Authority, the Homeless Families Foundation—a holistic, strength-based organization that offers homeless prevention and rehousing education, and stabilization services—and Nationwide Children’s developed the HBAH intervention to assess whether providing rental assistance with housing stabilization services to pregnant birthing people experiencing insecure housing could improve birth outcomes.²¹ 

In addition to developing interventions that directly address social determinants of health such as housing, Franklin County collaborators have implemented strategies to connect birthing people to existing services and resources when person needs exist. One model for this work is maternal and child home visiting. Through the Center for Family Safety and Healing, an NCH subsidiary focused on family welfare, Nationwide Children’s provides 2 evidence-based models of home visiting: Healthy Families America (HFA) and Nurse-Family Partnership (NFP). Facilitated by trained home visitors, HFA provides home visits to pregnant people and parents of newborns until the child’s third birthday. HFA focused on giving information on prenatal care, child development, and positive parenting practices and is connected to community resources. NFP is also a nationally recognized home visiting program that pairs registered nurses with Medicaid-eligible high-risk pregnant people to provide comprehensive pregnancy supports, education, and referrals for identified social needs to improve maternal and child health outcomes and parenting skills. Traditionally, NFP is limited to people during their first pregnancy; however, in 2022, Ohio Medicaid expanded eligibility to include multiparous birthing people.

Another strategy implemented to connect birthing people with needed resources is connecting high-risk birthing people to CHWs who can assist with the often-daunting process of accessing assistance. The Pathways Community HUB is an example of one of the several programs using this strategy in Franklin County.²² This model relies on CHWs to engage with at-risk pregnant people to connect them with evidence-based and best practice interventions available within the community.

Finally, because individual-level, community-level, and health resources are strongly influenced by federal, state, and local policy, the Pediatric Infant Mortality Vital Signs initiative has adopted policy advocacy and government partnerships as core components of its approach. The beginning of our work coincided with the earliest phases of the Affordable Care Act (ACA). ACA Medicaid expansion increased insurance coverage and access to care for millions of individuals and families across the United States. The expansion was also associated with improvements in infant mortality. States who chose to expand Medicaid to include childless adults had greater declines in IMR between 2014 and 2016 compared with non-Medicaid expansion states, with the greatest decline in Black infants.²³ As one of the early adopters of Medicaid expansion in 2013, Ohio did not see such strong results. Overall and race-specific IMR slightly rose in Ohio during the same study period, suggesting that for us, insurance access was possibly necessary but not sufficient to affect change. However, Ohio Medicaid, with substantial input from provider partners like OBBO and community partners like CelebrateOne, continues to innovate, developing and implementing additional improvements to positively impact birth outcomes. Advocacy has supported policy solutions, including extension of Medicaid postpartum coverage to 12 months, expanded funding for access to nurse home visiting, and leveraging of federal funding from the Covid Aid, Relief and Economic Security and the American Rescue Plan Acts to support state, county, and local agencies in implementing innovative strategies to address social determinants including housing, food security, transportation, and broadband access. Examples of these strategies included funding development of affordable housing units, supporting housing stabilization programs, and expanding local food pantry systems.

The Pediatric Vital Signs program selected its 8 population health metrics because they were important, not because they were easy. As noted in the introduction, IMR is stubbornly high, and gaps are not closing at the rates that were targeted. Rather than being defeated by slow progress, the IM vital signs team celebrates the intermediate wins and uses the learnings from each activity to further tailor our approach. For example, in the HBAH pilot study, 100 families who were experiencing housing instability or homelessness and made less than 30% of the area median income were recruited through community partners and randomly assigned to receive the housing intervention (n = 50) or usual care (n = 50). Families were enrolled in the pilot between 2018 and 2019 and those randomly assigned to the HBAH intervention received 15 months of rental assistance and housing stabilization services. Both intervention and control groups received usual care, which included referrals to social services, access to prenatal care, and job coaching services offered through community partners. On average, it took 62 days to secure housing for the families enrolled in the study, highlighting the affordable housing crisis many families face. Common barriers to housing included having a poor or no credit score (96%), history of criminal justice involvement (48%), no income (46%), and/or outstanding electric bill (60%). Results were promising. Seventy-eight percent of the 51 live births in the intervention were full-term and healthy birth weight compared with 55% among the 44 live births control group. The average length of stay for newborns admitted to the neonatal intensive care unit (NICU) was 8 days for those in the intervention group compared with 29 days in the control group, driving large differences in average paid per claim for infants at the time of birth until their initial release from the hospital ($4175 vs $21 521, or an average saving of $17 346 for newborns in the NICU). HBAH also improved housing stability, with two-thirds of the participants having good prospects for maintaining their housing as they exited the study.²¹ 

The early results of HBAH illustrate the potential impact of interventions that provide resources directly to those in need. Since the initial pilot, HBAH has expanded, with the goal of providing housing services to 300 birthing people across 4 Ohio counties. The financial investment required to achieve this goal is great; however, the health care savings could offset the costs. Using 2021 Fair Market Rent for a 2-bedroom unit, the cost to provide a full subsidy for rent to a pregnant person for 15-month is approximately $15 620, nearly $2000 less than the savings documented in the pilot study. This pilot study was funded by the Ohio Housing Finance Agency and several other public and private organizations. Initiatives like HBAH require advocacy and/or lobbying to federal, state, and local agencies, philanthropic organizations, and even managed care organizations to fund the interventions. This lobbying can be headed by the lead organizations who plan to implement the intervention but should include a collaborative team of partners throughout the county including but not limited to health care organizations, public health departments, housing authorities, and nonprofits who focus on housing issues in the community. As researchers continue to build evidence of these interventions’ effectiveness, the results can demonstrate the impact of large investments in population health.

Interventions employing CHWs and home visitors have also shown promise. Preliminary outcome data for the NFP and Healthy Families America home visiting programs show that between 2019 and 2022, less than 1% (5 out of 527 births) of birthing people who enrolled prenatally had a very preterm birth (less than 32 weeks) compared with 8% among a statistically matched sample of birthing people who did not receive home visiting. However, these differences were not statistically significant. These early findings in local home visiting exceed effects in other evaluations, however; given our local findings, OBBO’s goal is to expand community capacity to serve 5000 families per year and to connect 50% of birthing people seen in OBBO-affiliated prenatal care clinics to prenatal care. Preliminary data on social-needs support from CHWs has demonstrated that engagement in the Pathways HUB model significantly reduced the chances of neonatal admission among high-risk pregnancy deliveries and generates a 236% return on investment.²⁴ The program has now grown to 10 hubs throughout the state funded by Ohio Commission on Minority Health and by Medicaid Managed Care Plans.

The role of the NCH Pediatric Vital Signs initiative for infant mortality has been to provide data support, groups facilitation, and operational support, within a comprehensive community partnership. Pediatric Vital Signs goal achievement, or at a minimum, movement toward that goal, will be a testament to the importance of broad-based collaboration and advocacy. There is no single funding mechanism used to support the Pediatric Vital Signs initiative. Therefore, measuring the total costs is complex. This initiative is based on a collaborative framework that creates shared accountability and financial investment among community partners. NCH hired 2 full-time-equivalent staff members to support the initiative. Additional costs associated with the strategies discussed here are funded by the varies community partners who lead each strategy.

The infant mortality vital signs approach to advocacy and partnership has intentionally used every success and every frustration as a learning opportunity. Lessons learned include:

1. Patience: The seemingly intractable disparities in infant mortality have been established over centuries of differential treatment of Black Americans. These gaps will not close quickly. Managing the expectations of partners is necessary to maintain engagement and energy.

2. Urgency: Although patience is necessary, it cannot outweigh the urgency of each young life lost.

3. Hope: We can make a difference in small, targeted ways. Hope supports the belief that if we keep accumulating small wins, they will one day appear much bigger.

4. Investment: Success will require the large-scale investment of time, energy, and money from a variety of sources.

5. Communality: No institution or sector can do this work in a vacuum. Partnership, advocacy, and even activism are necessary tools if substantial change is to happen. Support from an organization like a hospital or university have been essential to maintaining progress and political support.

Despite substantial efforts across Franklin County, we have yet to close the Black-white gap in infant mortality. In the 3 years preceding the COVID-19 pandemic, IMR in both racial groups declined. Unfortunately, the pandemic brought a new wave of challenges and impacts on social determinants that resulted in increases in both the IMR and the Black-white disparity ratio (see Figure 1). The collaborative efforts taking place in Franklin County, led by CelebrateOne and OBBO, in alignment with the Pediatric Vital Signs Initiative are necessary to make lasting shifts in the outcomes. Between 2021 and 2023, there was a slight decrease in IMR among both white and Black infants in Franklin County; however, the racial disparity has increased as we have seen a sharper decline in IMR among white infants. The solutions needed to address the social inequities that drive the racial differences in IMR require partners from multiple sectors to work together to provide families with the resources they need. By calling for a collaborative approach and shared accountability, the Pediatric Vital Signs initiative is accelerating local efforts to ensure that IMR goals can be met in the future.

ACA

Affordable Care Act

ACOG

American College of Obstetricians and Gynecologists

FDA

Food and Drug Administration

IMR

infant mortality rate

OBBO

Ohio Better Birth Outcomes

NCH

Nationwide Children’s Hospital

NICU

neonatal intensive care unit

HBAH

Healthy Beginnings at Home

CHW

community health worker

HFA

Healthy Families America

NFP

Nurse-Family Partnership