OBJECTIVE

Social risk screening is recommended by the American Academy of Pediatrics, but this practice is underutilized in NICUs. To address this gap in social care, we aimed to increase rates of: (1) systematic social risk screening and (2) connection with community resources, each to ≥50% over a 14-month period.

METHODS

We conducted a quality improvement initiative from November 2020 to January 2022. We adapted a screening tool and used Plan-Do-Study-Act cycles to integrate screening and referral to resources into clinical workflow. Primary outcome measures included the percentage of (1) families screened and (2) connection with resources. We examined screening by maternal race/ethnicity and primary language. Process measures were (1) time from admission to screening and (2) percentage of referrals provided to families reporting unmet needs and requesting assistance. We used statistical process control to assess change over time and χ2 tests to compare screening by race/ethnicity and language.

RESULTS

The rates of systematic screening increased from 0% to 49%. Among 103 families screened, 84% had ≥1, and 64% had ≥2 unmet needs, with a total of 221 needs reported. Education, employment, transportation, and food were the most common needs. Screening rates did not vary by race/ethnicity or language. Among families requesting assistance, 98% received referrals. The iterative improvement of a written resource guide and community partnerships led to increased rates of connection with resources from 21% to 52%.

CONCLUSION

Leveraging existing staff, our social risk screening and referral intervention built the capacity to address the high burden of unmet needs among NICU families.

In the United States, 44% of children <3 years of age live in poor or near poor households,1  which places them at risk for worse health and developmental outcomes than their more privileged peers.2  Families with young children who live in poverty are more likely to have unmet basic needs such as food or housing insecurity.1  To offset the detrimental effects of poverty on child health, the American Academy of Pediatrics recommends universal social risk screening and referral to community resources in pediatric clinical care.3,4  Although the uptake of social risk screening has increased rapidly in outpatient pediatric settings,5  it remains largely underutilized in NICUs,6  representing a missed opportunity to address social adversity at the earliest stage of life.7 

A randomized trial in Boston revealed that a social risk screening and referral intervention in pediatric outpatient settings increased families’ receipt of resources.8  No such evidence exists for inpatient pediatric settings. The authors of a recent study examined the prevalence of social risks identified through inpatient screening but did not report on the outcome of referrals.9  How social risk screening and referral interventions perform in NICUs and whether implementation of this evidence-based practice improves not only the identification of unmet needs but also a connection with community resources among families of hospitalized infants, is unknown.

Systematic social risk screening and referral is particularly relevant in the NICU setting because families of preterm infants are disproportionally low-income,10  and the burden of unmet needs is higher in the preterm compared with the term population.11  Unmet needs are exacerbated during the prolonged NICU hospitalization owing to the costs associated with frequent meals and transportation, childcare for other children at home, and forgone income from lost time at work.12,13  Before this quality improvement (QI) initiative, social risk screening had been broadly implemented across outpatient settings in our health system,14  but this practice had not been integrated into inpatient settings, including the NICU. A 2017 to 2019 chart review indicated that, in our NICU and another safety-net NICU in Massachusetts, without a systematic approach, unmet needs were assessed and documented infrequently and variably (0.2% for utilities, 2.8% for transportation difficulties, 6.9% for food insecurity, and 38% for housing instability).15  Although the vast majority of NICUs employ social workers, staffing models and the scope of their role are variable across NICUs and span a large array of tasks, including assessments of family structure, caregiver mental health, domestic violence, substance use, and other complex social issues (e.g. custody), in conjunction with assessments of unmet basic needs. In safety net settings like ours, the bandwidth of social workers and the resources at their disposal are often limited.16 

To fill this gap in social care, we launched a QI initiative to implement social risk screening and referral in our NICU in November 2020. To understand both the uptake of screening and the effectiveness of referrals, we established the following SMART (specific, measurable, achievable, realistic, timely) aims: (1) increase systematic social risk screening of eligible families (length of stay ≥1 week), and (2) increase the rates of families’ connections with resources, each from baseline to ≥50% over 14 months. In addition, we aimed to achieve no disparity in the implementation of screening by race/ethnicity and primary language.

Boston Medical Center (BMC) is the largest safety net hospital in New England, with ∼2800 births annually. BMC serves an urban community; ∼80% of patients have public insurance, and ∼50% and ∼25% identify as Black and Hispanic, respectively. Nearly 30% of patients have limited English proficiency. The NICU at BMC is a 21-bed, level III unit with ∼300 admissions annually. There is 1 full-time social worker who covers the NICU, labor and delivery, and postpartum. Social work consultation is automatic after admission to the NICU. The goal is for the social worker to meet with each family within 7 days of admission (goal attained 72% of the time) and as needed for the remainder of the infant hospitalization. The social worker documents her meeting summaries in the mother and infant’s electronic health records (EHR).

We developed a multidisciplinary QI team, including NICU leadership (nursing director and physician QI lead), social service researchers, neonatologists, nurses, a social worker, a medical student, and a family partner. This team developed our SMART aims and key driver diagram (Fig 1) composed of 4 drivers: (1) lack of systematic social risk screening using a standardized tool, (2) lack of standardized processes to connect families with resources to address unmet needs identified through screening, (3) lack of a unit-wide approach to integrating social risk screening and resource referrals into a routine workflow, and (4) variable levels of staff’s knowledge and skills conducting social risk screening and providing referrals.

FIGURE 1

Key driver diagram including SMART aim, primary drivers, and change concepts. LOS, length of stay.

FIGURE 1

Key driver diagram including SMART aim, primary drivers, and change concepts. LOS, length of stay.

Close modal

During the preimplementation period, we conducted interventions related to the first 2 drivers, including the selection of a social risk screening tool and the development of a 1-page resource guide for unmet needs. Our screening tool was adapted from an existing tool, “THRIVE,” which was already being used across outpatient settings and integrated into our health system’s EHR (Supplemental Figs 6 and 7).14 

To develop the resource guide, we followed guidance from the American Academy of Pediatrics child poverty Website,17  used publicly available community resource aggregators,18,19  and engaged social workers and family navigators who had previously participated in the development of resource guides for the outpatient pediatric clinic. This process entailed weekly meetings over 2 months. For each domain of screening, the 1-page guide listed information on local community resources and services (Supplemental Fig 8). The resource guide was translated into 6 languages. The workflow for social risk screening and the provision of referrals is presented in Fig 2.

FIGURE 2

Workflow for social risk screening and provision of referrals.

a Nurses were selected as screeners because they spend the most one-on-one time with families at the bedside and often engage in conversations about unmet needs. Nurses also did not “turn over” like pediatric residents.

b Infants discharged in <1 week would receive social risk screening during their outpatient follow-up visit shortly after discharge.

c The adapted THRIVE tool included questions to assess for homelessness, housing insecurity, food insecurity, transportation difficulties, utilities, caregiving, unemployment, and educational aspirations.

d SmartPhrases are short phrases used to insert templates into a note in the EHR. This was later replaced by readily available resource guides (printed).

FIGURE 2

Workflow for social risk screening and provision of referrals.

a Nurses were selected as screeners because they spend the most one-on-one time with families at the bedside and often engage in conversations about unmet needs. Nurses also did not “turn over” like pediatric residents.

b Infants discharged in <1 week would receive social risk screening during their outpatient follow-up visit shortly after discharge.

c The adapted THRIVE tool included questions to assess for homelessness, housing insecurity, food insecurity, transportation difficulties, utilities, caregiving, unemployment, and educational aspirations.

d SmartPhrases are short phrases used to insert templates into a note in the EHR. This was later replaced by readily available resource guides (printed).

Close modal

We used the Institute for Healthcare Improvement Model for Improvement as a framework for this project.20  The third and fourth drivers guided our plan-do-study-act (PDSA) cycles focused on improvements in both systematic screening and connection with resources (Table 1).

TABLE 1

Plan-Do-Study-Act Cycles

PDSA CyclesaDateLessons Learned
#1 Training of 6 champion nurses to perform universal screening, provide 1-page resource guide to families, and document in EHR November 2020 Nurse champions were highly engaged but unable to reach screening targets during times of high census and acuity. 
  Nurse champions had technical difficulties with HER documentation. 
#2 Training of residents to perform screening/referral February 2021 Training of residents was difficult to maintain because of frequent turnover (biweekly to monthly), so this plan was abandoned. 
Availability of step-by-step instructions for EHR documentation in nurses’ station   
#3 Training of all beside nurses to perform screening/referral May 2021 Increased awareness of families’ social needs during bedside rounds allowed clinical team to address barriers to caregiver presence in the NICU (ie, transportation difficulties). 
Revision of resource guide based on family feedback  Bedside nurses had variable levels of comfort engaging in conversations about unmet social needs. 
  Nurses reported higher level of confidence/comfort screening families with whom they had an existing relationship. 
#4 Webinars on community resources by agency representatives June 2021 Increased shared knowledge among NICU staff of available community resources 
Partnership with community organization #1  Families received direct assistance from community partner on enrollment in specific services/resources. 
#5 Created “THRIVE station” with resource guides readily available (printed) and “resource cheat sheet” August 2021 Launching a new resource bolstered staff’s buy-in and increased uptake of the intervention. 
Launched Parking and Uber Health Program  High family satisfaction with transportation program 
#6 Partnership with community organization #2 October 2021 Launching a new resource bolstered staff’s buy-in and increased uptake of the intervention. 
PDSA CyclesaDateLessons Learned
#1 Training of 6 champion nurses to perform universal screening, provide 1-page resource guide to families, and document in EHR November 2020 Nurse champions were highly engaged but unable to reach screening targets during times of high census and acuity. 
  Nurse champions had technical difficulties with HER documentation. 
#2 Training of residents to perform screening/referral February 2021 Training of residents was difficult to maintain because of frequent turnover (biweekly to monthly), so this plan was abandoned. 
Availability of step-by-step instructions for EHR documentation in nurses’ station   
#3 Training of all beside nurses to perform screening/referral May 2021 Increased awareness of families’ social needs during bedside rounds allowed clinical team to address barriers to caregiver presence in the NICU (ie, transportation difficulties). 
Revision of resource guide based on family feedback  Bedside nurses had variable levels of comfort engaging in conversations about unmet social needs. 
  Nurses reported higher level of confidence/comfort screening families with whom they had an existing relationship. 
#4 Webinars on community resources by agency representatives June 2021 Increased shared knowledge among NICU staff of available community resources 
Partnership with community organization #1  Families received direct assistance from community partner on enrollment in specific services/resources. 
#5 Created “THRIVE station” with resource guides readily available (printed) and “resource cheat sheet” August 2021 Launching a new resource bolstered staff’s buy-in and increased uptake of the intervention. 
Launched Parking and Uber Health Program  High family satisfaction with transportation program 
#6 Partnership with community organization #2 October 2021 Launching a new resource bolstered staff’s buy-in and increased uptake of the intervention. 
a

Throughout our PDSA cycles, we incorporated activities related to staff education and motivation (eg, on-site trainings, family testimonials), and transparent data sharing (eg, monthly meetings to share data, dashboard panel in NICU) to promote awareness and acceptability of the intervention.

For PDSA 1 (November 2020), we trained 6 champion nurses to conduct social risk screening using a verbal prompt (Supplemental Fig 9). Using feedback from nurses and families, we selected verbal screening in the form of a structured conversation because this format was perceived as more appropriate for caregivers of hospitalized newborns to allow caregivers space to share concerns and ask questions. Nurses documented screening responses and provision of resource guide in the EHR. In PDSA 2 (February 2021) and PDSA 3 (May 2021), we focused on increasing screeners’ capacity and improving the quality of resources. We trained residents and, subsequently, all bedside nurses to perform screening, leveraging the original group of nurse champions as internal facilitators to provide ongoing problem-solving support. In parallel, we revised the resource guide on the basis of feedback from families about resources or services that were less helpful (eg, long wait times, lack of availability of language services). To update the resource guide, we contacted various local organizations and gathered additional input from social workers. The revised guide was presented to NICU staff in webinars attended by agency representatives. Over time, we built relationships with community-based organizations, which resulted in our first partnership in PDSA 4 (June 2021). This meant that families in our NICU reached out to a specific person at the community organization who provided guidance on how to enroll in various services for housing, food, utilities, and childcare.

In PDSA 5 (August 2021), we further streamlined the screening and referral processes by creating a “THRIVE station” in the NICU with accessible printed copies of the resource guide and developed a 1-page “resources cheat sheet” with information that nurses could use in their discussions with families about resources. During this time, our team also secured institutional philanthropic funding from the Department of Pediatrics21  to launch a “Parking and Uber Health Program” to provide improved transportation resources, including free parking or rides for families throughout the NICU hospitalization. In October 2021 (PDSA cycle 6), we introduced a second partnership with a community-based organization that specialized in programs for employment and education. This organization provided a multilingual link for an intake form that allowed families to receive a list of available programs, with the option to book an appointment with agency representatives to discuss the next steps. Both community partners reported monthly on the outcome of the referrals received by their organizations.

We tracked data on all infants admitted to the NICU with a length of stay ≥1 week. For multiple gestations, we only included the firstborn infant.

Main Outcomes

The primary screening outcome was the percentage of eligible families that were screened for 7 social risk factors using the THRIVE tool within 2 weeks of admission based on documentation in the EHR. The timing of screening aimed to maximize time for the receipt of resources. Screening was not restricted to this timeframe. The primary referral outcome was the percentage of unmet needs for which families connected with resources within 3 months of referral. Connection with resources was defined as enrollment in a new hospital or community program or service.8 

Process Measures

The screening process measure was the time from admission to screening. The referral process measure was the percentage of families with ≥1 unmet need identified through screening that were provided a resource guide if they desired assistance.

Balancing Measure

The balancing measure was the percentage of families that declined screening, as documented in the EHR. We used this measure as a proxy for the acceptability and appropriateness of social risk screening in the NICU context.

Other Characteristics

Infant characteristics abstracted from the EHR included gestational age, sex, birth weight, and length of hospital stay. We also collected maternal race/ethnicity and primary language to track outcomes by these metrics, with the goal of identifying and addressing disparities in the implementation process.22  We categorized maternal race/ethnicity as non-Hispanic Asian, non-Hispanic Black, non-Hispanic white, Hispanic (any race), and other. Primary language was categorized as English versus non-English.

Data about the screening and provision of referrals were obtained from the EHR. Data on the outcome of referrals were obtained through 3 main sources: follow-up calls with families 2 to 4 months after NICU discharge, reports from community partners, and Uber Health use logs.

Descriptive statistics were calculated for maternal and infant demographics. The outcome and process measures were analyzed by using statistical process control charts.23  For rates of screening, we used the total number of families screened over the total number of eligible families with infants hospitalized that month. For rates of connection with resources, we used the total number of unmet needs for which families enrolled in new resources within 3 months of referral over the total number of unmet needs identified among all families in a given month. We used probability-based rules to identify changes in the data that have <5% probability of occurring by chance (a “shift” of 6 or more consecutive points increasing or decreasing).24  When a shift occurred, the average of the shifted points represents the new mean. χ2 tests were used to compare screening by maternal race/ethnicity and language. Analyses were conducted by using QI macros for Microsoft Excel and Stata/SE 17.0.

This QI project was deemed exempt by the Boston Medical Center/Boston University Medical Campus Institutional Review Board.

A total of 212 infants were eligible for screening from November 2020 to January 2022 (Table 2). The median length of hospital stay was 31.0 (range 7–109) days. Non-Hispanic Black and Hispanic mothers of any race represented 41.5% and 29.3% of the population, respectively. Forty-five percent of mothers had a non-English primary language.

TABLE 2

Characteristics of Mother–Infant Dyads Among Infants Hospitalized ≥1 Week in the Boston Medical Center NICU from November 2020 to January 2022 (n = 212)

Infant Characteristicsn (%)
Gestational age at birth  
 <28 wk 28 (13.2) 
 28–31 wk 37 (17.4) 
 32–37 wk 128 (60.4) 
 ≥38 wk 19 (9.0) 
Birth wt  
 ≤1500 g 58 (27.4) 
 1501–2500 g 97 (45.7) 
 ≥2500 g 57 (26.9) 
Female sex 95 (44.8) 
Length of stay, d, median (range) 31 (7–109) 
Maternal characteristics  
Race/ethnicity  
 Non-Hispanic Black/African American 88 (41.5) 
 Non-Hispanic White 39 (18.4) 
 Non-Hispanic Asian 2 (0.9) 
 Hispanic or Latino, any race 62 (29.3) 
 Othera 9 (4.3) 
 Unknown 12 (5.7) 
Primary language  
 English 110 (51.9) 
 Spanish 47 (22.2) 
 Haitian Creole 29 (13.7) 
 Portuguese 7 (3.3) 
 Other 14 (6.6) 
 Unknown 5 (2.4) 
Infant Characteristicsn (%)
Gestational age at birth  
 <28 wk 28 (13.2) 
 28–31 wk 37 (17.4) 
 32–37 wk 128 (60.4) 
 ≥38 wk 19 (9.0) 
Birth wt  
 ≤1500 g 58 (27.4) 
 1501–2500 g 97 (45.7) 
 ≥2500 g 57 (26.9) 
Female sex 95 (44.8) 
Length of stay, d, median (range) 31 (7–109) 
Maternal characteristics  
Race/ethnicity  
 Non-Hispanic Black/African American 88 (41.5) 
 Non-Hispanic White 39 (18.4) 
 Non-Hispanic Asian 2 (0.9) 
 Hispanic or Latino, any race 62 (29.3) 
 Othera 9 (4.3) 
 Unknown 12 (5.7) 
Primary language  
 English 110 (51.9) 
 Spanish 47 (22.2) 
 Haitian Creole 29 (13.7) 
 Portuguese 7 (3.3) 
 Other 14 (6.6) 
 Unknown 5 (2.4) 
a

Arab n = 1, Middle Eastern n = 2, Indian n = 2, multiracial n = 4.

Systematic social risk screening did not exist before the project implementation. The rates of screening reached 49% during the implementation period (Fig 3). Among 103 families screened, 84.4% reported ≥1 and 64.1% reported ≥2 (range 2–6) unmet needs. Overall, 221 unmet needs were reported, and families desired assistance for 190 (86%). Education, transportation, employment, and food were the most prevalent reported needs (Fig 4). The median time to screening was 10 days (range 5–16), and 98% of families who reported ≥1 unmet need and requested assistance received referrals. Rates of connection with resources increased from a mean of 21% to 52% over time. There were no significant differences in screening rates by maternal race/ethnicity (P = .92) or primary language (P = .84; Fig 5). Zero families declined screening.

FIGURE 3

Statistical process control charts of outcome measures over time among eligible infants (length of stay ≥1 week). (A) Percentage of families screened among all eligible and (B) percentage of unmet needs addressed by community resources among all needs identified.

PDSA 1: Training champions in screening/referral. PDSA 2: Support EHR documentation. PDSA 3: Training bedside nurses, resource guide revision. PDSA 4: Community partnership #1. PDSA 5: Reduce referral time burden, Uber Health. PDSA 6: Community partnership #2.

FIGURE 3

Statistical process control charts of outcome measures over time among eligible infants (length of stay ≥1 week). (A) Percentage of families screened among all eligible and (B) percentage of unmet needs addressed by community resources among all needs identified.

PDSA 1: Training champions in screening/referral. PDSA 2: Support EHR documentation. PDSA 3: Training bedside nurses, resource guide revision. PDSA 4: Community partnership #1. PDSA 5: Reduce referral time burden, Uber Health. PDSA 6: Community partnership #2.

Close modal
FIGURE 4

Percentage of unmet needs identified through standardized screening and percentage of requests for resource information.

FIGURE 4

Percentage of unmet needs identified through standardized screening and percentage of requests for resource information.

Close modal
FIGURE 5

Stratification of social risk screening by race/ethnicity and primary language.

Note: 2 non-Hispanic Asian families who were screened are not shown.

FIGURE 5

Stratification of social risk screening by race/ethnicity and primary language.

Note: 2 non-Hispanic Asian families who were screened are not shown.

Close modal

Using a QI approach, we implemented a systematic social risk screening and referral intervention within a safety net NICU in Massachusetts. In this acute inpatient care setting serving primarily low-income families, integrating this practice into routine clinical workflow was feasible and acceptable. We observed no disparities in screening by maternal race/ethnicity or primary language. Systematic screening revealed a substantial burden of unmet needs among families of hospitalized infants. The iterative improvement of a written resource guide and community partnerships increased our capacity to address unmet needs.

Neonatal hospitalizations, which may last weeks to months, provide numerous opportunities for identifying unmet needs and connecting families to resources. Thus, instituting systematic social risk screening and referral during NICU hospitalizations represents a unique opportunity to address the unmet needs of families. We found a high prevalence of both short- and longer-term needs among families in our NICU. Short-term needs related to the NICU hospitalization itself, such as food, parking, and transportation, are financially burdensome and stressful for families of hospitalized children, particularly for low-income families.12,13  Addressing these short-term needs contributes to a greater frequency of family presence in the NICU, thereby increasing opportunities for families to participate in their infants’ care and engage in key health-promoting behaviors, such as breastfeeding and skin-to-skin care.25  Systematic screening in our unit also revealed a substantial burden of longer-term needs, which are essential for families’ economic mobility (eg, workforce development, educational advancement). Although the processes to address these higher-order needs are often complex, early involvement of social services, as well as cross-sector partnerships yielded positive referral outcomes. Overall, our rates of connection with resources were higher than those reported in previous prospective studies conducted in outpatient pediatric settings (range of 39% to 47%),8,26  partly due to the mutiple opportunities while inpatient to follow-up on referrals and assist families in troubleshooting barriers to connect with resources.

The strategies used to integrate our intervention into the NICU setting have implications at multiple levels. At the NICU level, standardizing the assessment of social risks may help units understand the burden of unmet needs of the population they serve. NICU teams may use this information to better address systematic barriers to family engagement, allocate resources, and design interventions. At the hospital level, QI initiatives like this can begin the process of building sustainable cross-sector partnerships to effectively address the most prevalent unmet needs. Successful partnerships are a win-win for both hospitals and community organizations and have the potential to scale up within other clinical settings. At the state level, perinatal quality collaborations may be leveraged to disseminate standardized social risk screening and referral across multiple NICUs, which can accelerate the spread of this potentially better practice for greater public health impact.27 

We have many “lessons learned” from this project regarding facilitators and barriers to the implementation of social risk screening and referral in the NICU setting. Facilitators that bolstered the uptake of the intervention and staff buy-in included (1) leveraging the EHR and streamlining procedures to reduce the time burden on clinically busy staff, (2) having an engaged group of nurse champions, and (3) frequently soliciting and incorporating feedback from nurses and families. Mounting evidence on the influence of social risks on health has led to greater recognition that clinical teams have a responsibility to practice social as well as technical medicine.27  Consequently, social risk screening and referral interventions have involved physicians, nurses, and residents, not just social workers.28  In our setting, the social worker did not serve as the primary screener because of limited bandwidth. Thus, it was critical to ensure the clarity of roles and to establish procedures whereby screening results and referrals were communicated to social workers. We view the intervention as an augmentation, not a replacement, of the crucial role of social workers in the NICU.

Contextual factors that may have hindered the uptake of screening included that most of this project occurred during the coronavirus disease 2019 pandemic, which imposed unique challenges due to visitation restrictions for coronavirus disease 2019-exposed families, as well as issues related to nursing understaffing that limited opportunities for screening. Additional barriers were that some staff lacked confidence and/or comfort in conducting social risk screening. This finding aligns with previous reports of inpatient providers’ perceptions around discussions of social needs.29,30  We overcame this challenge by (1) conducting training using interactive role-play, (2) using standardized language to introduce the purpose of the screener using a family-centered approach, and (3) having champions as internal facilitators. To address staff’s concerns that providing referrals to resources was ineffective we (1) shared data on referral outcomes and highlighted success stories through family testimonials, (2) conducted informational webinars led by community organizations, and (3) replaced resources that were reported as unhelpful by families.

The intervention was acceptable to families, as revealed by the 0 screening refusal rate. High caregiver acceptability of social risk screening interventions as part of pediatric well-child visits has been previously reported.31  However, less is known about the perspectives of families of hospitalized infants. We informally asked families in a follow-up phone call after discharge how they perceived the screening while their infant was in the NICU. Families indicated that highlighting the universal nature of the screening (“we offer this to all families”) was important because it offered a less-stigmatizing way to engage them in conversations about unmet needs. Families largely trusted nurses and believed that the purpose of the screening was “helping families to help the babies.” Families also appreciated being asked in subsequent encounters during the hospitalization whether the resources provided were helpful or not, which prompted efforts to overcome barriers. Although families valued having a realistic picture of the multiple steps necessary to enroll in certain resources (eg, childcare), they reported having limited bandwidth to carry out these requirements, and that they would have liked additional hands-on support (eg, assistance filling out applications).

The strengths of this study include the use of an equity-focused quality improvement framework.22,32  To achieve equity, our intervention was designed to preferentially address the needs of vulnerable social groups, and our team set equity goals from the project’s inception and ensured equity of implementation by tracking screening by race/ethnicity and primary language. In addition, we made considerable efforts to track the outcome of referrals, which guided iterative improvements of the resource guide. Limitations include our lack of baseline information on the connection with community resources. However, we observed an increase in the rates of connection over time. Because the families served in our NICU lived in a relatively small geographic area, we tailored our resource guide accordingly. NICUs that serve large geographic areas may need to develop multiple resource guides.

This QI initiative reveals the feasibility and acceptability of implementing a systematic social risk screening and referral intervention to address unmet needs in a NICU setting with limited social work support. Universal screening coupled with the systematic use of a resource guide and community partnerships increased families’ connections with resources. Careful attention to integrating processes into routine workflows, and utilizing existing staff and resources, will ensure sustainability. Further studies are needed to establish best practices for inpatient social risk screening; we offer implementation strategies that can be adopted or adapted by other teams.

We thank the nursing staff at BMC without whom this initiative would not have been possible. Special thanks to our nurse champions Judith Burke, Marilyn Valle, Ramona Silva, Patricia Edman, Marsha Adams, and Joanna Bushfield. We also want to acknowledge our community partners Julia Power, Ashley Pizarro, and Larissa Charistable from Action for Boston Community Development, and Kurt Faustin and Christian Espino from Jewish Vocational Services. We also thank our unit coordinators Janet Adames and Brenda Walsh for leading and collecting data on the parking and Uber Health Program and the BMC Department of Information Technology for their technical support. Special thanks to the BMC Department of Pediatrics leadership (Dr Robert J. Vinci) for its unwavering commitment to our initiative and funding support for our transportation program.

Drs Cordova-Ramos and Parker contributed to the conceptualization, design, and analysis of the study and drafted and edited the manuscript; Ms Jain, Ms Torrice, and Ms McGean contributed to the design of the study and data collection and edited the manuscript; Drs Drainoni, Buitron de la Vega, and Vinci and Ms Burke and Ms Stickney contributed to the design of the study and reviewed and edited the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING: This work was supported by the W.K. Kellogg Foundation (P0131665 PI Parker), the Boston Medical Center Department of Pediatrics Kids Fund, and the Evans Center for Implementation and Improvement Sciences (CIIS Department of Medicine, Boston University School of Medicine. Dr Cordova-Ramos is supported by BU-CTSI Grant Number 1KL2TR001411

CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest relevant to this article to disclose.

BMC

Boston Medical Center

EHR

electronic health record

PDSA

plan-do-study-act

QI

quality improvement

SMART

specific, measurable, achievable, realistic, timely

1.
Koball
H
,
Moore
A
,
Hernandez
J
.
Basic Facts About Low-Income Children Under 18 Years
.
New York, NY
:
National Center for Children in Poverty
;
2021
2.
Shonkoff
JP
,
Garner
AS
;
Committee on Psychosocial Aspects of Child and Family Health
;
Committee on Early Childhood, Adoption, and Dependent Care
;
Section on Developmental and Behavioral Pediatrics
.
The lifelong effects of early childhood adversity and toxic stress
.
Pediatrics
.
2012
;
129
(
1
):
e232
e246
3.
Council on Community Pediatrics
.
Poverty and child health in the United States
.
Pediatrics
.
2016
;
137
(
4
):
e20160339
4.
American Academy of Pediatrics
.
AAP publications reaffirmed or retired
.
Pediatrics
.
2021
;
148
(
2
):
e2021052583
5.
Sokol
R
,
Austin
A
,
Chandler
C
, et al
.
Screening children for social determinants of health: a systematic review
.
Pediatrics
.
2019
;
144
(
4
):
e20191622
6.
Cordova-Ramos
EG
,
Kerr
S
,
Heeren
T
, et al
.
National prevalence of social determinants of health screening among US neonatal care units
.
Hosp Pediatr
.
2022
;
12
(
12
):
1040
1047
7.
Parker
MG
,
Garg
A
,
McConnell
MA
.
Addressing childhood poverty in pediatric clinical settings: the neonatal intensive care unit is a missed opportunity
.
JAMA Pediatr
.
2020
;
174
(
12
):
1135
1136
8.
Garg
A
,
Toy
S
,
Tripodis
Y
, et al
.
Addressing social determinants of health at well child care visits: a cluster RCT
.
Pediatrics
.
2015
;
135
(
2
):
e296
e304
9.
Fritz
CQ
,
Thomas
J
,
Gambino
J
, et al
.
Prevalence of social risks on inpatient screening and their impact on pediatric care use
.
Hosp Pediatr
.
2020
;
10
(
10
):
859
866
10.
Martin
JA
,
Hamilton
BE
,
Osterman
MJK
,
Driscoll
AK
.
Births: final data for 2018
.
Natl Vital Stat Rep
.
2019
;
68
(
13
):
1
47
11.
Cordova-Ramos
EG
,
Brochier
A
,
Tripodis
Y
, et al
.
Beyond income: material hardship and the health and healthcare of premature children. [Published online ahead of print November 2022]
J Perinatol
12.
Parker
MG
,
de Cuba
SE
,
Rateau
LJ
, et al
.
Household unmet basic needs in the first 1000 days and preterm birth status
.
J Perinatol
.
2022
;
42
(
3
):
389
396
13.
Lewis
TP
,
Andrews
KG
,
Shenberger
E
, et al
.
Caregiving can be costly: a qualitative study of barriers and facilitators to conducting kangaroo mother care in a US tertiary hospital neonatal intensive care unit
.
BMC Pregnancy Childbirth
.
2019
;
19
(
1
):
227
14.
Buitron de la Vega
P
,
Losi
S
,
Sprague Martinez
L
, et al
.
Implementing an EHR-based screening and referral system to address social determinants of health in primary care
.
Med Care
.
2019
;
57
(
Suppl 6 Suppl 2
):
S133
S139
15.
Parker
MG
,
Garg
A
,
Brochier
A
, et al
.
Approaches to addressing social determinants of health in the NICU: a mixed methods study
.
J Perinatol
.
2021
;
41
(
8
):
1983
1991
16.
Crable
EL
,
Biancarelli
D
,
Walkey
AJ
,
Drainoni
ML
.
Barriers and facilitators to implementing priority inpatient initiatives in the safety net setting
.
Implement Sci Commun
.
2020
;
1
:
35
17.
American Academy of Pediatrics
.
Poverty and child health
.
18.
Find Help- The Social Care Network
.
Search and connect to support
.
Financial assistance, food pantries, medical care, and other free or reduced-cost help starts here. Available at: https://www.findhelp.org/. Accessed August 21, 2020
19.
Community Action Partnership
.
Find your CAA
.
Available at: https://communityactionpartnership.com/find-a-cap/. Accessed August 21, 2020
20.
Langley
G
,
Moen
R
,
Nolan
K
, et al
.
The Improvement Guide: A Practical Approach to Enhancing Organizational Performance
, 2nd ed.
San Francisco, California
:
Jossey-Bass Publishers
;
2009
21.
Boston Medical Center
.
The kids fund
.
Available at: https://www.bmc.org/pediatrics/philanthropy/kids-fund. Accessed July 20, 2022
22.
Parker
MG
,
Hwang
SS
.
Quality improvement approaches to reduce racial/ethnic disparities in the neonatal intensive care unit
.
Semin Perinatol
.
2021
;
45
(
4
):
151412
23.
Institute for Healthcare Improvement
.
QI essentials toolkit
.
24.
Montgomery
DC
.
Introduction To Statistical Quality Control
, 4th ed.
Hoboken, NJ
:
Wiley
;
2000
25.
Klawetter
S
,
Greenfield
JC
,
Speer
SR
,
Brown
K
,
Hwang
SS
.
An integrative review: maternal engagement in the neonatal intensive care unit and health outcomes for U.S.-born preterm infants and their parents
.
AIMS Public Health
.
2019
;
6
(
2
):
160
183
26.
Hassan
A
,
Scherer
EA
,
Pikcilingis
A
, et al
.
Improving social determinants of health: effectiveness of a web- based intervention
.
Am J Prev Med
.
2015
;
49
(
6
):
822
831
27.
Horbar
JD
,
Edwards
EM
,
Ogbolu
Y
.
Our responsibility to follow through for NICU infants and their families
.
Pediatrics
.
2020
;
146
(
6
):
e20200360
28.
Chung
EK
,
Siegel
BS
,
Garg
A
, et al
.
Screening for social determinants of health among children and families living in poverty: a guide for clinicians
.
Curr Probl Pediatr Adolesc Health Care
.
2016
;
46
(
5
):
135
153
29.
Schwartz
B
,
Herrmann
LE
,
Librizzi
J
, et al
.
Screening for social determinants of health in hospitalized children
.
Hosp Pediatr
.
2020
;
10
(
1
):
29
36
30.
Leary
JC
,
Rijhwani
L
,
Bettez
NM
, et al
.
Clinical stakeholder perspectives on pediatric inpatient screening for social needs
.
Acad Pediatr
.
2022
;
22
(
3
):
470
477
31.
Fleegler
EW
,
Lieu
TA
,
Wise
PH
,
Muret-Wagstaff
S
.
Families’ health- related social problems and missed referral opportunities
.
Pediatrics
.
2007
;
119
(
6
):
e1332
e1341
32.
Reichman
V
,
Brachio
SS
,
Madu
CR
, et al
.
Using rising tides to lift all boats: equity-focused quality improvement as a tool to reduce neonatal health disparities
.
Semin Fetal Neonatal Med
.
2021
;
26
(
1
):
101198

Supplementary data