“We don’t want to screen for the sake of screening:” A qualitative evaluation of a social needs screening and referral intervention in the NICU

Abstract

Background:

Low uptake of social determinants of health (SDH) screening and referral interventions within neonatal intensive care units (NICUs) is partly due to limited understanding of the best procedures to integrate this practice into routine clinical workflows.

Purpose:

To examine the feasibility and acceptability of a SDH screening and referral intervention in the NICU from the perspective of neonatal nurses; and to identify factors affecting implementation outcomes.

Methods:

We conducted 25 semi-structured interviews with NICU nurses. We used the Promoting Action on Research Implementation in Health Services (PARiHS) framework to guide interview questions and codebook development for directed content analysis. Themes were mapped onto the three PARiHS domains of context, evidence, and facilitation.

Findings:

Analysis yielded eight themes.

Context:

Nurses felt that stressors experienced by NICU families are magnified in a safety-net environment. Nurses shared varying viewpoints of the roles and responsibilities for social care in the NICU, and feared that scarcity of community resources would make it difficult to address families’ needs.

Evidence:

The intervention was perceived to increase identification of adverse SDH and provision of resources; and to potentially jumpstart better caregiver and infant health trajectories.

Facilitation:

Procedures that improved acceptability included dynamic training and champion support, regular feedback on intervention outcomes, and strategies to reduce stigma and bias.

Conclusion:

We identified contextual factors, concrete messaging, and training procedures that may inform implementation of SDH screening and referral in NICU settings.

INTRODUCTION

Preterm birth occurs disproportionally among families living in poverty or near poverty;¹ and families of preterm infants are more likely to experience adverse social determinants of health (SDH), such as food or housing insecurity, compared with families of infants born at term.^1,2 The detrimental effects of SDH on children’s health and development are amplified for preterm infants requiring intensive care due to their already significant risk for chronic morbidities and developmental disabilities.² SDH screening and resource referral interventions have proven to be effective in increasing caregiver’s receipt of community resources to address adverse SDH.^3,4 The American Academy of Pediatrics and payors, such as Medicare and Medicaid, promote universal SDH screening and provision of resource referrals during pediatric clinical encounters.^5,6

Despite existing recommendations, a recent national study demonstrated that only 23% of neonatal intensive care units (NICU) have standardized SDH screening and referral processes, even though the extended neonatal hospitalization of preterm infants provides longitudinal opportunities to identify and address adverse SDH.⁷ This evidence-to-practice gap illustrates the complexity of changing clinical practice in NICU settings to integrate SDH screening and referral into routine NICU care. Because nurses spend the most one-on-one time with families in the NICU, many interventions in this setting involve nursing staff as front-line implementers. However, nurses’ perspectives on all stages of the implementation process are often overlooked. This knowledge is essential to guide implementation efforts to effectively address adverse SDH in NICU settings with limited social work resources, where nurses may be primarily involved in processes to address unmet needs of families.

In order to address this research gap, we conducted a qualitative study of nurses’ perspectives as front-line implementers following the implementation of a SDH screening and referral intervention in a safety-net NICU. We aimed to understand nurses’ viewpoints on the feasibility and acceptability of integrating this intervention into their routine practice, and to understand the factors affecting implementation to inform future dissemination.

METHODS

Setting

We implemented the SDH screening and resource referral intervention from November 2020 to January 2022 in the NICU at Boston Medical Center (BMC). BMC is the largest safety-net hospital in New England, with approximately 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 one full-time social worker assigned to the NICU, who also covers labor and delivery and postpartum. While a social work consultation is placed for all NICU admissions, there is variability in the assessment of adverse SDH, even when a social work consultation occurs.⁸ Given the limited social work staffing and resources in our setting, nurses served as front-line implementers for this intervention because they interact the most with families and become aware of adverse SDH on a daily basis.

Intervention

The intervention had two main components: standardized screening for adverse SDH and provision of resource referrals to families who desired assistance. Prior to implementation, we adapted an existing SDH screening tool⁹ and developed a one-page resource handout listing available resources and services in the hospital and local community. During implementation, we used plan-do-study-act cycles to guide and support the integration of the intervention into routine clinical practice. Adaptations to the intervention were co-created with nurses. These adaptations involved changing the delivery model, with bedside neonatal nurses identifying eligible families (length of hospital stay ≥ 7 days), conducting verbal SDH screening using a standardized approach, documenting screening results in the EHR, and providing the resource handout to families with ≥ 1 adverse SDH identified who desired assistance. The time to complete the intervention was approximately 10 minutes. While the social worker did not conduct the screening, she reviewed screening results in the EHR and during weekly social work rounds with the clinical team, and followed-up with families regarding referrals provided. Full details of the development and implementation of the intervention are reported elsewhere.¹⁰

Qualitative Process Evaluation and Conceptual Framework

To conduct the process evaluation of the SDH screening and referral intervention in the NICU, we used qualitative research methods¹¹ grounded on the PARiHS framework¹² to understand the factors affecting the implementation. The PARiHS framework proposes that implementation success is the function of three interrelated elements: evidence, context, and facilitation. Evidence refers to providers’ perceptions of the strength of the evidence, and their experience with the intervention being implemented. Context comprises aspects of the culture, leadership, and internal and external environment where the intervention takes place. Facilitation refers to how the evidence-based intervention is integrated into clinical practice.¹²

Interview Guide

The interview guide included open-ended questions regarding nurses’ experiences with the intervention, how it compared with the previous standards of social care, nurses’ perceptions of their role and responsibility addressing SDH, and perceptions of the fit and value of the intervention for families of hospitalized newborns. We also explored intervention feasibility and logistical challenges to integrate SDH screening and referral into routine workflows, and contextual factors inherent to the NICU setting and how these influenced implementation processes. We probed for strategies that nurses used to accommodate the specific needs of caregivers, as well as any trade-offs, opportunity costs, and workflow adjustments required to incorporate SDH screening and resource referral into routine care.

Sampling and Data Collection

We identified a purposive sample¹³ of nurses with maximum variation in terms of role (nurse leaders, nurse champions and bedside nurses), and years in neonatal nursing practice. Nurses were recruited in person by trained research staff from October 2021 through December 2021. Verbal consent was obtained from all participants. Two non-clinical research assistants trained in qualitative interviewing conducted semi-structured interviews in person. Interviews lasted ~20–25 minutes. Data collection continued until the research team agreed that thematic saturation was met such that new insights were unlikely to be obtained through additional interviews.¹⁴

Data Analysis

Interviews were audio-recorded, professionally transcribed verbatim and de-identified for analysis. We conducted directed content analysis¹⁵ of transcripts to develop themes describing nurses’ perspectives of the feasibility and acceptability of implementing a SDH screening and referral intervention in the NICU. To conduct analysis, we initially created a codebook with a priori PARiHS constructs. Three coders (EGC, VT, MM) independently coded all transcripts. Coders met regularly to review coding consistency, resolve disagreements in coding decisions, and iteratively refine the codebook, identifying additional codes that emerged from the transcripts.

After all coding was completed, four coders reviewed the data and used a systematic iterative approach to identify preliminary themes. The larger research team discussed preliminary themes to reach consensus on final deductive themes, and mapped themes onto each PARiHS domain of context, evidence, and facilitation. Coding and analysis were conducted in Dedoose software. Data reliability was addressed through investigator triangulation,¹⁶ and member checking,¹⁷ whereby themes were presented back to a group of nurses to comment on perceived accuracy and engage in data interpretation.

Ethical Considerations

This study was approved by the Boston Medical Center and Boston University Medical Campus Institutional Review Board.

RESULTS

Thematic saturation was achieved after conducting 25 interviews with nurses, representing 42% of all NICU nursing staff at BMC. Participant nurses were involved in various aspects of implementation including NICU leaders (n=2), nurse champions (n=4), and bedside nurses who served as front-line implementers (n=19). Sampled participants’ demographic characteristics were representative of all nursing staff in the NICU at BMC (Table 1). Thirty-six percent and 56% of nurses in the sample reported having worked as a neonatal nurse for ≤5 and ≥10 years, respectively. Directed content analysis yielded eight themes across the three PARiHS domains of context, evidence, facilitation (Table 2).

Table 1.

Characteristics of nurse participants n=25

Participant characteristics	N (%)
Role
Nurse leader	2 (8)
Nurse champion	4 (16)
Bedside nurse	19 (76)
Race/ethnicity
Hispanic	1 (4)
Non-Hispanic White	16 (64)
Non-Hispanic Black/African American	2 (8)
Non-Hispanic Asian	0 (0)
Other	3 (12)
Prefer not to answer	3 (12)
Years as NICU nurse
1–5	9 (36)
6–9	2 (8)
10–20	6 (24)
>20	8 (32)

Table 2.

Themes and supporting quotations

Theme description	Example quote
Context
Numerous stressors experienced by NICU families are magnified in a safety-net environment	Quote 1A: “It’s definitely beneficial, especially because I feel like our population has a lot of needs that not necessarily other hospitals do. Many of our families have low incomes and many needs…I think everyone sees that our families need more than we’ve been able to provide for them, and this is a path to get us there.”(participant 12) Quote 1B: “NICU families have so many obstacles to come up against between getting here, daycare, other kids at home, work, etc. Honestly, I don’t know how they do it. And then trying to deal with the trauma of having a baby in the NICU. These families are so resilient, they deserve that we do better by them, that we do anything we can do to help them. I mean, it’s just so much going on. I mean, you visit somebody in the hospital…After a week you’re like ‘I can’t take this anymore.’ And here, they’re visiting their babies for months and months. If we can use this opportunity to identify needs and give them resources, I think it’s a great step forward.” (participant 20)
Nurses had varying viewpoints of the roles and responsibilities for social care in the NICU	Quote 2A: “Some nurses may feel like ‘why do I have to do this on top of everything else I have to do?’ Or ‘this isn’t really our job.’ That’s why the NICU has social workers you know. Even when they are positive about the program. It’s just like who should really institute things and whose responsibility is it to do this.”(participant 13) Quote 2B: “I mean, it’s part of our job. I mean, it’s – I think it’s important. I think it’s important for the parents. I don’t think it’s that difficult or too much to ask I guess …of us to do. I feel like it kind of flows into just conversation, teaching, just basic sharing with the parents that we already do. Like if we’re in and out of the room all day, and if the parent is staying the whole day then you kind of build more of a relationship with them and you talk about these things. Not to say that doctors don’t, it’s just I think the way our workflow works versus theirs.”(participant 16) Quote 2C: “I would say sometimes it’s hard for social work to do everything. For example, we get a lot of patients with so many social needs, so it’s hard for social work to provide absolutely everything for every patient. So, it just kind of gives us the opportunity to help them [families] more. It kind of gives the nurses an insight into the specifics of that a family might need. It gives us almost permission to ask some of those questions that sometimes have been reserved for social work.”(participant 4)
Scarcity of community resources makes it difficult to address families’ needs	Quote 3A: “I know from experience that community agencies are overwhelmed. There’s too many people in need of help, and few resources to help them.”(participant 25) Quote 3B: “I am worried that community agencies won’t be able to help our families as much as we hope they will. There is so much bureaucracy in some of these agencies. And sometimes they are just not able to meet the needs of families, housing is a big one.”(participant 9)
Evidence
The intervention enabled greater identification of unmet social needs and increased provision of resources	Quote 4A: “These are things that have been going on here for many, many years. It’s nice to identify these things and have other supports to give. I think families often have questions about resources but they not always ask. With THRIVE [the intervention], it’s just more available to them. We have an easy tool to provide them with when they ask for resources. Or if they’re uncomfortable asking it’s a way to get them to disclose the help they need –without them having to ask first.”(participant 18) Quote 4B: “We have one parent that sticks out in my mind. The mother was Spanish speaking and before she was screened, we didn’t know there were any issues even though she had already met with the social worker. Well, come to find out the mother was staying here every single night because the dad lost his job and they didn’t have money for food. So she was staying here for food. After the THRIVE screen was done by one of our nurses who speaks Spanish, we found all of this out and they got her connected to the pantry that same day. That mom returned from the pantry with big smiles on her face, and just felt such a relief.”(participant 8) Quote 4C: “I actually attended one of the meetings where an agency was presenting and learned about what they can offer, which is terrific. I’m not sure that we even were aware at how many community assists and supports there were available to our parents.”(participant 20) Quote 4D: “I appreciate the immediacy of it, especially for the Uber Health to be able to access rides, and discounted parking and also foodbank, that’s huge. I can’t believe I didn’t know about the BMC foodbank before, that we can provide access to that. It just eliminates a lot of the what ifs, especially for families who may be new to the area or new to the country.”(participant 11)
The intervention has the potential to jumpstart better caregiver and infant health and well-being trajectories	Quote 5A: “The baby has to go home to a safe place with food and heat and there’s a lot of stress for the families, and I think that if this program can identify needs earlier on, it can probably help a lot of these issues, as simple as transportation to come see their baby, I think it makes the bonding and just kind of being a parent more in the focus, and it makes the stress less on the parents, which in the end helps the baby.”(participant 4) Quote 5B: “The earlier that we start it, the earlier that we can actually set them [families] up with resources, the better for the baby to go home. So, to have everything set up or at least started before baby is discharged is important. So, it’s kind of like a jumpstart.”(participant 2) Quote 5C: “I like it because we can begin to address social determinants of health, and get people better care, and get families out of the cycle of poverty. You need a job to get good food and healthy food for your family. THRIVE [the intervention] touches upon things that like we wouldn’t normally. Like if the baby is going to be here for a really long time, say one of the parents is interested in resources for education like one family I had, or someone in the house is looking for work or anything like that, that can help them get on their feet and have better opportunities.”(participant 10)
Facilitation
Dynamic training and champion support increased nurses’ comfort with the intervention	Quote 6A: “We went over what the program was. We went over the screening tool, the questions that we were going to ask. We created some scripts that I thought were helpful. For example, how to introduce yourself, how to explain the THRIVE program [the intervention], and to make it so that it was actually something that was helpful and useful versus intimidating. I really liked the role play for example with a pretend parent who was suspicious of why we were asking these questions. So we did it only after feeling comfortable with it. And then after that, after learning all the questions, then we also learned about the various resources that were available and how to connect the families with those resources.” (participant 11) Quote 6B: “There were a couple of videos we watched, and then [redacted name] came around and she did it with us individually, and we would role play for a little bit with each other, and then we practiced with each other before doing it with patients. And now there’s always somebody here that if you don’t know how to do something there’s support.”(participant 15)
Regular feedback to nurses on intervention outcomes improved buy-in	Quote 7A: “I really liked it when they shared with us the data on what happened with the referrals. I loved to learn that many families had used the resources or had enrolled in the programs we offered them. We don’t want to screen for the sake of screening. One of the mothers shared her experience with the English program and she was so excited about it, it was great to hear. I’d love to see the data on the feedback coming in from families that I know they are getting, that’ll kind of tell us what we need to do, how to steer the ship.”(participant 6) Quote 7B: “We did change some resources as we found out that some of the resources, they were not as helpful for our parents or some other resources were more advanced. So, we learned and changed things, we fine-tuned our resources as we found other places that offered more or better services. I really liked that.”(participant 3)
Procedures to reduce bias and stigma increased acceptability of the intervention	Quote 8A: “It provides, I hope, a little more intimacy and trust with the parents once they know we’re on their side, and this isn’t about outing them, or letting authorities know that they’re here. We have to acknowledge their lived experiences and the many reasons they have distrust the system. I mean, I always want parents to feel comfortable with us, and the way we present THRIVE [the intervention] is a great way to provide them with, again, a sense of trust, and some community options that they may not be aware of.”(participant 7) Quote 8B: “You want to develop a relationship with the patient before you sit down and kind of ask those questions, or for them to trust you with their honest responses. So, it is definitely best for either like a primary nurse or somebody who has taken care of the baby at least a few times to ask the questions. I personally believe that matters to families.”(participant 2) Quote 8C: “Considering everybody is asked it, it kind of sets a precedent of, okay, they’re not just asking me because they think I need it, versus you let them [families] know upfront it is just done for everyone and why we do it. I think that’s important. It’s a nice way to get information without being perceived as biased.”(participant 19)

Context

Numerous stressors experienced by NICU families are magnified in a safety-net environment

Nurses described the numerous stressors experienced by caregivers of hospitalized newborns in the NICU (e.g. frequent transportation, costs of meals and parking), and that they were highly aware that these stressors were especially burdensome for low-income families in a safety-net setting. In this context, acceptability of the intervention was influenced by the perception that the intervention could potentially alleviate the high burden of unmet needs among NICU families, and by their personal commitment to improve care for vulnerable populations (Table 2, quotes 1A and 1B).

Nurses had varying viewpoints of the roles and responsibilities for social care in the NICU

Nurses had varying viewpoints on whether or not conducting SDH screening and referral in the NICU should be part of the nurses’ role. In some cases, nurses felt that addressing SDH should be the sole responsibility of social workers (Table 2, quote 2A). In other cases, nurses reflected on the fact that they often engaged in conversations about social needs in their routine interactions with families. Therefore, the intervention was perceived to simply formalize the conversations they were already having (Table 2, quote 2B).

In our NICU context, nurses also described their perspective about the limited bandwidth of the social worker to address the high burden of unmet needs among families. Therefore, they considered the intervention to augment and support the role of social work (Table 2, quote 2C).

Scarcity of community resources makes it difficult to address families’ needs

Some nurses pointed out structural challenges that were perceived to negatively influence acceptability of the intervention. These challenges were related to perceived paucity of community resources as well as bureaucracy-related barriers that families face when seeking resources in the community (Table 2, quotes 3A and 3B).

Evidence

The intervention enabled greater identification of unmet social needs and increased provision of resources

Based on their everyday experience in the NICU, nurses recognized that social needs were very common among the families they cared for. They believed that addressing unmet needs of families was essential to caring for preterm infants. For this reason, acceptability of the intervention was influenced by the perception that it increased identification of unmet needs and enabled provision of information about resources (Table 2, quote 4A). One nurse recounted an instance where a food emergency was uncovered through screening even when the family had previously met with the social worker (Table 2, quote 4B).

Some nurses expressed that prior to the implementation of the intervention, they were unaware of many available resources in the local community and even in the hospital (Table 2, quotes 4C and 4D).

The intervention has the potential to jumpstart better caregiver and infant health and well-being trajectories

Nurses generally considered that the intervention supported a better transition from the NICU to home by alleviating stressors and short-term needs that prevented caregivers from being present in the NICU, and by setting up better home environments for infants after NICU discharge (Table 2, quotes 5A and 5B).

In addition, some nurses expressed hope about the potential of resources for longer-term needs (i.e. career coaching, educational advancement) as a means of “breaking the cycle of poverty” through opportunities for economic mobility (Table 2, 5C).

Facilitation

Dynamic training and champion support increased nurses’ comfort with the intervention

Many nurses mentioned that training strategies such as role-play and videos made them feel more prepared and increased their comfort with the intervention (Table 2, quote 6A). Other nurses highlighted the importance of continued on-site support by champions to troubleshoot problems that arose after initial training (Table 2, quote 6B).

Regular feedback to nurses on intervention outcomes improved buy-in

Nurses described procedures that boosted their buy-in with the intervention including learning about intervention outcomes through champions sharing aggregate data and family testimonials (Table 2, quote 7A). Nurses also valued that the use of outcome data prompted replacement of resources that were reported as unhelpful by families (Table 2, quote 7B).

Procedures to reduce bias and stigma increased acceptability of the intervention

Many nurses reflected on the structural inequities that often underlie families’ mistrust of the health system, and emphasized their desire to approach SDH screening and referral in a way that fostered trust with families (Table 2, quote 8A). Therefore, nurses embraced elements of the intervention that were perceived to build trust, and reduce bias and stigma such as leveraging pre-existing relationships with the family (i.e. primary nurses conducting screening), and highlighting the universal nature of screening (Table 2, quotes 8B and 8C).

DISCUSSION

In this qualitative evaluation of the implementation of a SDH screening and referral intervention, we sought to understand the acceptability and feasibility of integrating this practice into routine clinical workflow from the perspective of neonatal nurses. We used the PARiHS framework to identify factors related to the context, evidence, and facilitation that influenced these implementation outcomes. We found that the intervention was perceived as highly acceptable and feasible, and potentially “a jumpstart” to address adverse SDH among low-income caregivers of hospitalized infants, thereby setting up better caregiver and infant health and well-being trajectories.

In the NICU, as in many inpatient settings, nurses are the frontline personnel that interact the most with families, compared with other types of providers. As a result, nurses frequently play a crucial role in implementing patient-facing healthcare innovations.¹⁸ Despite the crucial role of nurses in front-line care, many neonatal intervention studies do not consider the context of implementation from the perspective of nurses. Lack of consideration of nursing perspectives may explain why many interventions with proven effectiveness are not successfully implemented and sustained in real world settings. We addressed an important research gap by directly targeting nurses’ perspective regarding the feasibility and acceptability of a social intervention in the NICU setting to guide implementation efforts.

We found that acceptability of the SDH screening and referral intervention was strongly influenced by contextual factors related to the safety-net NICU setting. Previous research in non-safety-net settings has identified staff reluctance to take on implementation responsibilities that extend beyond current work roles as a barrier to the adoption of practice innovations.¹⁹ However, we found that staff’s role flexibility was less of an issue among neonatal nurses in our safety-net NICU who believed in the mission of providing social as well as clinical care to families.²⁰ Many nurses described the immense emotional and financial toll imposed on families during the prolonged NICU hospitalization. In this context, the intervention was perceived as responsive to the needs of a vulnerable population during a time of uniquely high stress. Factors associated with the high acceptability included the co-development with nursing of procedures to reduce stigma and promote trust with families. Overall, nurses believed that SDH screening and referral interventions in the NICU should incorporate procedures for facilitating provider-family relatedness, as well as address power imbalances and families’ vulnerability by ensuring universal, private screenings with clearly stated intent.

Prior literature exploring stakeholder’s perspectives around SDH screening has described staff’s discomfort performing SDH screening.^21–23 In this study, we found effective strategies that were co-created with nurses to increase acceptability of the intervention. Nurses felt that concerns of stigma or bias that families may experience by being screened for adverse SDH were eased by utilizing a standardized approach that clearly communicated the fact that screening was universal, explaining the purpose and potential benefit of screening, setting expectations, and providing an option to decline. Our findings confirm what other authors have reported: that NICU staff not only have awareness of bias and its impact on their interactions with patients, but they also have ideas to mitigate the impact of bias.²⁴

Despite the acuity and busyness of the NICU environment, nurses generally perceived that it was feasible to integrate SDH screening and referral into routine clinical workflows. For example, time constraints were described in the context of trying to ensure families were given proper time for discussions arising from screening. Prior implementation studies have reported that, in safety-net settings, providers often identify creative workarounds to navigate time and resource constraints, but this flexibility may lead to burnout.^25,26 Therefore, implementation leaders should pay close attention to the time and work demands made from front-line implementers, and develop structures to maximize supports and minimize burdens. Ultimately, it will be difficult to maintain social care interventions embedded within clinical care without organizational commitment, financial investments, innovative reimbursement models, and strong cross-sector partnerships.

There are some limitations to our study. We examined nurses’ perspectives in an urban, safety-net NICU. We may have garnered different perspectives if we had interviewed neonatal nurses working in other types of NICUs. However, NICUs implementing SDH screening and referral interventions are likely located within safety-net, urban settings. Additionally, our interviews were restricted to a relatively short window, mid-way of the implementation, when implementation efforts were ongoing. It is possible that nurses’ perspectives changed over time when educational and training efforts became less intensive. Despite these limitations, this study adds to the literature by exploring the essential perspectives of nurses as implementers of a SDH screening and referral intervention in the NICU setting.

CONCLUSIONS

From the perspective of neonatal nurses, we identified context, evidence, and facilitation factors that influenced the implementation outcomes of acceptability and feasibility of a SDH screening and referral intervention in the NICU. Themes that emerged through this qualitative work uncovered concrete messaging and training procedures that may be helpful to others looking to implement SDH screening and referral interventions in their own NICUs.