Resilience in the System: COVID-19 and Immigrant- and Refugee-Serving Health and Human Service Providers

Abstract

Introduction:

Immigrant and refugee families in the U.S. have been particularly hard hit by the COVID-19 pandemic. Health and human service providers who serve these communities have been essential in supporting them during this crisis, yet have also had to adapt the way they provide services. The current study aims to describe the challenges these service providers have faced and the adaptations they have made.

Methods:

Our research team conducted semistructured interviews with 19 service providers at 10 organizations identified as serving one or more immigrant and/or refugee communities in the state of Minnesota. We analyzed the interviews for themes and used normalization process theory (May & Finch, 2009) to understand how service providers have shown resilience and where gaps in capacity emerged.

Results:

Mechanisms of adaptation to the COVID-19 crisis included staff taking on larger workloads, utilizing existing service frameworks in new ways, shifting their services remotely and/or substantively, and utilizing the trust they had built with communities and individuals over time. Challenges that had not been fully overcome included insufficient funding for community need and restrictions on methods of interaction.

Discussion:

Key implications include allocating funding for immigrant and refugee families, developing and evaluating new service formats in collaboration with clients, providing direct support for staff in times of crisis, and using practice-based evidence to speed implementation science research.

COVIDE-19 has had a particularly devastating impact on the lives of immigrants and refugees, who are more likely to contract and die of COVID-19 (Tai et al., 2021), to lose income or housing (Gonzalez et al., 2020), to face reduced health care access, and to have increased mental health concerns such as PTSD (Mattar & Piwowarczyk, 2020; Rothman et al., 2020). These relate to preexisting disparities in social determinants of health (e.g., access to safe housing, well-paying jobs) coupled with negative social climate elements such as xenophobia and systemic racism (Garcini et al., 2020).

Health and human service providers who serve immigrants and refugees across the country are working to mitigate the impacts of COVID-19 on these populations. Although research has assessed impacts on individuals and families, attention to how COVID-19 has affected the larger health and human service systems that support them is minimal. The demand for their resources has grown as immigrant and refugee families turn to trusted providers for support. Therefore, the goal of this article is to use an implementation science lens to assess the rapid shifts in the health and human service provider system that occurred from March to August 2020. We use a resilience lens and normalization process theory (May & Finch, 2009) to address how the health and human service provider system in Minnesota adapted and expanded to meet clients’ needs, and to identify ongoing gaps. Finally, we provide recommendations for private and public organizations to support this critical infrastructure.

Health and Human Service Providers as Resilient Systems

Masten (2018) defined resilience as “the capacity of a dynamic system to adapt successfully to challenges that threaten the function, survival, or development of the system” and indicated this definition can apply to both individuals and larger systems (e.g., organizations, communities). A few recent studies have demonstrated some of the challenges that service providers faced when adapting to COVID-19, such as the transition to telehealth/remote services (Mattar & Piwowarczyk, 2020; Silva et al., 2020) and the immense need of communities who were struggling economically pre-COVID (Bernstein et al., 2020). When we think about service providers as resilient systems, we can see how the sequelae of COVID-19 may threaten the ability of the system to function in its typical manner and how the system may adapt. Questions remain about what mechanisms among service providers increase system resilience and successful adaptation. Our study examined the following questions: (1) What challenges did service providers face in their response to COVID-19, and what were the mechanisms of adaptation? (2) Were there challenges where service providers were not able to adapt? What blocked the identified adaptation mechanisms, and how might these be addressed?

Normalization Process Theory Application

Normalization process theory (NPT), an implementation science theory that examines how processes become normalized into systems, emphasizes that collective action by the individuals in the system is key to practice change (May & Finch, 2009). The mechanisms of change include: coherence (defined and organized practice based on a set of ideas about meaning), cognitive participation (engagement of the people involved), collective action (work that organizes and enacts a practice), and reflexive monitoring (formal and informal continuous evaluation; May & Finch, 2009). NPT is one of few process theories currently available in implementation science and appeared to most clearly reflect the topic under study as it describes mechanisms rather than phases (Damschroder, 2020). NPT presents the four mechanisms described above as the way in which the work of implementation is completed. Our examination of service providers’ resilience in the face of COVID-19 changes can be framed by these four NPT mechanisms that articulate the change process in a cohesive manner.

Methods

Participants

We conducted semistructured interviews with 19 nonprofit health and human service providers from 10 agencies who primarily serve immigrants and refugees. We used a stratified sampling method with the goal of equivalent numbers of urban and rural providers, with multiple providers serving three of the main immigrant and refugee groups in the state of Minnesota. Our final sample included 8 primarily Somali-serving providers, 6 primarily Latinx-serving providers, 3 primarily Karen-serving providers, and 2 providers who served multiple immigrant groups. Ten providers were in urban areas, and 9 providers were in rural areas. Providers had been working with immigrant communities between 7 months and 29 years (M = 6.32 years, SD = 8.66).

Procedure

We contacted health and human service agencies who served immigrants and/or refugees, by phone and/or e-mail with an invitation for up to three staff to participate in an interview. Participants received a $100 gift card in appreciation for their time. Two research team members conducted interviews in June–August 2020. Questions in the interview asked agency staff to summarize their knowledge about the experiences of the families they serve following the onset of COVID-19 in six major areas including jobs, housing, food, health and health care access, family relationships, and coping strategies. The University of Minnesota IRB determined the study to be exempt from review.

Analysis

Interviews were professionally transcribed and then analyzed using inductive content analysis (Elo & Kyngäs, 2008). The two authors who led the interviews were the primary coders. They first conducted open coding, starting with jointly coding half an interview and assigning as many codes as necessary to describe the content. They alternated separate and joint coding for five interviews, meeting to discuss initial codes, focusing on passages assigned to the same categories to verify that they held the same meanings. After these five interviews, the team of researchers then grouped the lists of codes into higher- and lower-order groups, creating a categorization matrix. A categorization matrix is a system of categories and subcategories; in our work, the broadest categories addressed family relationships, housing, employment, outside supports (from agencies and other sources), barriers to support, coping strategies, and agency changes. Example quotes from transcripts were included for each category and subcategory. We retained a code only if it appeared in at least two distinct interviews. After establishing the categorization matrix, the rest of the interviews were coded by the two main coders and three additional coders. Two team members, who did not participate in the interview process, independently coded several manuscripts (1–3 unique manuscripts for each) using the matrix, to establish data trustworthiness. The research team met after all interviews were coded to review the matrix, discuss questions that had arisen, and review the results. Main and independent coders had approximately 80% consistency, with most of the remaining 20% of codes explained with adjacent codes or subcodes. Based on team discussion, the categorization matrix was finalized.

For the current article, the first author applied a deductive content analysis to the transcript quotes that related to agency supports and changes, using the NPT framework to map relevant existing codes onto each of the mechanisms of normalization, including coherence, cognitive participation, collective action, and reflexive monitoring. This summary was then presented to the full research team for discussion and validation. The full team also reviewed the final version linking the codes, themes, and theoretical processes.

Results

Agencies and interviewed staff rapidly adapted to the new challenges they faced in the COVID-19 crisis, exhibiting resilience in four primary strategies: continuing engagement through larger caseloads, shifting services, building on community trust, and utilizing existing service frameworks. Within these strategies, systems demonstrated mechanisms of coherence, cognitive participation, collective action, and reflexive monitoring. Barriers to collective action, including insufficient funding and restrictions on interaction methods, impeded providers’ ability to meet the needs of their clients. A summary of system resilience factors and NPT mechanisms is presented in Table 1. T1 Each strategy and the NPT mechanisms within it is described below.

Table 1.

Summary of System Factors and NPT Mechanisms in Minnesota During COVID-19 Pandemic

NPT mechanism	Able to meet needs through ...	Gaps in capacity due to ...
Coherence	Larger caseloads/workloads Utilizing existing service frameworks
Cognitive participation	Larger caseloads/workloads Utilizing existing service frameworks Shift in services
Collective action	Trust Utilizing existing service frameworks	Insufficient funding
Immediate factors:		Restrictions on interaction methods
Relational integration Interactional workability
Reflexive monitoring	Utilizing existing service frameworks  Shift in servicesTrust

How Service Providers Continued to Meet Clients’ Needs

The key to resilient service organizations was their staff. In the first categories of larger caseloads and utilizing service frameworks, the mechanisms of coherence and cognitive participation are prominent, while collective action and reflexive monitoring play a role in agencies’ trust in the community, utilization of existing service frameworks, and shifts in services. Each of these NPT processes is described below.

Continued Engagement through Larger Caseloads

Providers reported that their caseloads increased rapidly at the beginning of the pandemic, especially around requests for advocacy and financial assistance as client layoffs occurred. These caseloads required intense, ongoing cognitive participation from staff in order to provide the needed services, which is exemplified by one provider stating, “What we’ve been noticing a lot is the need to be able to pivot in a manner that meets the need that emerges among the working population and understanding their particular needs and how we can assist.” Another provider indicated her thought process when referring families for programs: “I try to think of programs where the main thing they’re going to be asked for is a name and a date of birth, or they’re not going to be asked for an ID because then there’s a greater chance that the patients will go there.” Providers remained committed to providing service to increased caseloads due to a coherent goal of service to their community. One provider who serves immigrants and refugees from multiple ethnic backgrounds indicated a threefold increase in calls. Another provider for the Latinx community said, “Advocates are overloaded. . . . I’ll be like, okay, I got the crisis line and I’m busy the entire eight hours. So the volume of calls, the number of participants, really went up.” One staff member indicated that the agency had increased their fundraising activities: “Our director is still working on finding ways we can find more funding for more families.” Another provider indicated that when they were able to open a low-barrier program for rent assistance, they were inundated with calls: “It was like we ended with like 2,000 voicemails in our voicemail . . . because they know we’re going to help them and it’s the only option they have because the other programs require so many things” (emphasis added). The providers were clearly motivated to help their communities however they could.

Shift in Services

Providers primarily shifted from working in person to working from home, or occasionally working in the office, which was closed to the public. Many providers shifted their job responsibilities as well as their format, from providing in-person therapeutic services and support to addressing more basic needs. As one provider stated, “We acknowledged that we cannot run our regular programs without ensuring that our families had access to basic needs such as food, diapers, or financial support for their grants and bills.” This indicates staff listened to their clients and used reflexive monitoring, or continuous evaluation, to hear their clients’ changing needs to shift to basic services. Based on this reflexive monitoring, sometimes staff just provided connection with clients who felt isolated, particularly during the state-wide stay-at-home order in April 2020. One provider related a common response from her Latinx clients: “Mainly they’ve been saying thank you for calling, that it helps a lot to be heard.”

In other cases, service providers helped families navigate the new way that other systems were operating. One provider who already helped parents navigate a difficult educational system helped them deal with the sudden transition to remote learning: “So a lot of it has been local advocacy work, making sure that they call the teachers, they call the schools and be like, ‘Hey, can we get access, can you check in with my child more often? We’re seeing that my child is falling behind, why haven’t you contacted me?’ It’s definitely been navigating the educational system, which is already an issue, but it’s more.” This provider indicated her continued cognitive participation in adapting to new ways of advocating for her clients, a clear example from NPT, which indicates that providers must have “powers of intervention and agency” and that they are“ able and prepared” to use them (May & Finch, 2009, p. 543).

Trust

One of the most important ways that providers were able to continue meeting clients’ needs was through leveraging trusting relationships from the many years they had been providers: “Most of the clients they trust our work and the job that we’re doing. So they keep contacting us or they keep coming back to us with any systems for help they’re needing.” Another provider indicated that she was working with many families she had never met:

Some of [the families] I don’t even know in person, not even in video. I know them by text or WhatsApp. . . . They’re not even participants of the program. They don’t come to [program], neither as students—it was just former students. Even from 25, 27 years ago. They came back and was like, we need information about this. You know where to find this. I have a friend that is going through this. From finding food to finding shelter to finding an attorney for domestic abuse. . . . Everything came up and people refer others to us. Really I’m trying to keep up. Sometimes it’s not too easy.

Youth also continued engaging with agency staff, as one Somali provider indicated: “The youth have been—I actually gave them my e-mail and some of them have my personal phone number that I’m close to . . . and they keep calling me.” Staff at these agencies had built a high level of community trust over years and decades, a unique embodiment of collective action. Through their openness and ability to support their clients, they continually demonstrated to their community that they were worthy of their referrals and reputation.

Staff in medical clinics also became a vector for accurate information and trusted testing for COVID-19. Some providers discussed with clients how to stay safe at work, and inform and distribute information and explain to families the guidelines of CDC. Through their cultural connections, one group of staff at an urban medical clinic were able to engage the community in a public housing complex in a COVID testing clinic:

We met some of the elderly and the wise member of that building. And we shared the plan and what we doing. A lot of people came, actually, to be tested. I mean, like [another participant] said, but the stigma was still there. And that was one thing, really, we needed to work on—I mean, in the future, it will be one thing that we are trying to work on just to help and build—this is one of the thing, really, that I do believe. It’s the relation between the patient and the clinic. And how they see us, we are delivering the service. If they feel that they are comfortable with us, the service that we are delivering, that’s it.

In this way, providers demonstrated reflexive monitoring: They reflected on their own practice and service delivery in light of the perspectives and experiences of clients in order to consider how to change their COVID-19 testing practice to better serve their clients. This comment also demonstrates one reason that community representation is important in their service providers and messaging; It increases trust and communication.

Utilizing Existing Service Frameworks

Service providers needed to get creative to overcome the challenges of increased client needs and the public health necessities of social distancing. A coherent goal of continuing to meet client needs and continued cognitive participation generated new ways to provide the needed services by leveraging existing partnerships with other local organizations through collective action. For example, one urban provider indicated, “A lot of the resources that were accessible to some of our parents, like our mothers had access to diapers, had access to the car seats, access to someone to help them to file for insurance, medical insurance. And our elders, we started partnering up with Metro Mobility to help with deliveries, right? So that establishment that was already there helped a lot.” These partnerships and the multiple providers engaged in developing and sustaining them required collective action and ongoing cognitive participation to continue to be effective.

Another way service providers used existing frameworks was to provide virtual gatherings for previously organized programming. “My program responded in hosting virtual gatherings . . . just for the focus of making sure that the youth still saw each other’s faces, that we’re still checking in with each other, and that they had some kind of routine.” Through the change in format, families continued to engage and reach out to the providers. When one provider for Latinx families moved their youth programming to virtual delivery, they said, “Their parents want them to register because they’re like, They’re just at home. Their sleeping schedule is awful. Take them, fix that. So yeah, they’ve shown a lot of gratitude and also understanding through this process because we have taken them along the way to hear their feedback, to hear their suggestions.” This provider and others at the agency were engaged in reflexive monitoring, and the clients appreciated it.

How Service Providers Struggled to Meet Clients’ Needs

Insufficient funding and limitations of technology as a substitute for human interaction were typical reasons why providers struggled to meet clients’ needs.

Insufficient Funding

Many providers identified city, county, state, and private financial assistance programs initiated to support the community during the pandemic. Unfortunately, these programs quickly ran out of funds. One provider discussed the shortcomings of a state program: “It had a three-day deadline or four-day deadline. And then on the second day it closed down because so many people applied, right? So that has been also very challenging because there hasn’t been a lot of investment in our immigrant families from our state.” Another provider said, “We’ve referred 61 families to a private organization that had some funding to help with rent. They all qualified, but then there’s no more funding.”

Many agencies led donation drives and their directors applied for grants or otherwise sought private funding (demonstrating collective action), but there was a consistent need for more. One provider stated, “Two months ago we raised some money and we were able to help close to 2,000 families with $500 toward rent.” Providers consistently reported that the pandemic would continue for months, but financial resources were inadequate for the long term. One Latinx provider stated,

So we’ve been able fortunately to help a lot of our families with their rent and their bills. And that helped our families to stabilizing a little bit of their financial circumstances, but not to a great extent because we’re able to cover a month, maybe two, maybe some bills, but they still don’t have a job. So unfortunately, it hasn’t been a lot of sustainable change for support to our families because they have not found a job and there still isn’t enough applications.

Providers clearly indicated that their ability to engage in collective action only took them so far. Insufficient collective action was demonstrated through insufficient provision of funding for clients’ needs from the broader community, including city, county, state, and federal governments as well as private groups.

Restrictions on Interaction Methods

Many providers indicated that they struggled to engage their clients through virtual means. Some of this was logistical, for example, clients did not have reliable Internet or did not have enough minutes on their phones. For some it was an issue of literacy; as one provider said, “Some clients doesn’t know how to use technology, like texting or sending a picture through the message. It’s even harder for some clients who doesn’t know how to read and write, which we couldn’t help them—they cannot give all their information over the phone.” This gap in capacity demonstrates a relational integration issue, which is an immediate factor of collective action (May & Finch, 2009).

Many health care providers reported that some clients did not see virtual care as useful:

Some of them do have to have the meet your doctor on Zoom appointments. It’s really—it’s mumbo-jumbo to them. It’s like, “This is not going to work.” So it’s not important. They’re not counting it as important as that one-on-one visit. But the ones that have health concerns, the ones with children that need appointments, they’re doing that. Whenever possible, they’re opting out for the in-person, and the in-person appointments are more than the tele appointments.

It is possible that, given time and continued telehealth access, more people would adapt to the telehealth model, but it may be that structural barriers (e.g., lack of reliable Internet/phone service) preventing full engagement in telehealth could result in reduced engagement in health care by many immigrants and refugees.

An organization’s status in the community allowed new clients to trust them despite not meeting in person, the change in service format to virtual contributed to difficulty in building trust with others. Particularly for undocumented immigrants, this trust is needed due to rhetoric and policies that instill fear (Garcini et al., 2020; Wieling et al., 2020). One medical provider stated,

So I work a lot with patients that are undocumented . . . if I haven’t had referral prior to COVID from them, the first interaction I’m trying to have with them is sometimes over the phone, and it’s just hard to build that connection and to establish that rapport and let them know that—I mean, I have no association with reporting whether a person does not have a social security number or is working with a false name. . . . And it’s hard to build that trust over the phone.

Discussion

The present study aimed to examine the challenges health and human service providers faced in serving their clients during the COVID-19 pandemic, and how they were or were not able to overcome those challenges. Challenges faced included increases in service demand and types of services requested, and/or changes in service delivery method or limitations in allowed methods, all while continuing to manage client relationships. While COVID-19 service changes are anything but “normal,” normalization process theory mechanisms were clearly demonstrated by how quickly agency staff shifted their processes and were generally resilient in the face of significant service delivery disruption. Service providers had a coherent goal of continuing to serve their clients, which meant that they were willing to take on larger workloads and utilize existing service frameworks in creative ways. They engaged in collective action through these service frameworks and by leveraging the trust they had built with their communities over time. They cognitively participated in the change process, showing active engagement with their larger workloads, continuing to connect clients to resources in existing service frameworks, and when those existing frameworks were inadequate, shifting to different communication approaches or fundamentally changing the services provided. Then, they continued to adapt by reflexively monitoring how their new processes were or were not allowing them to meet the needs of their clients.

Yet providers experienced limitations in the remote provision of services, categorized as issues with the immediate factors of interactional workability and relational integration (May & Finch, 2009), New processes were sometimes not well-suited for the clients’ skill sets or preferred mode of interaction. Finally, an unresolvable limitation was a need for larger community collective action of accessing funding for the families they served. Agency staff were constantly scrambling to find ways to access funds for clients’ basic financial needs such as rent. Unfortunately, programs that surfaced were quickly depleted due to the overwhelming demand.

Implications for Research, Practice, and Policy

Health and human service providers made enormous changes to their service provision due to COVID-19. Some changes seemed to be working well and others were only substitutes until in-person services could safely resume.

Allocate Funding for Immigrant and Refugee Families

Providers’ ability to connect clients with basic needs support was limited by funding availability. This capacity gap was fundamentally a need for collective action to prioritize the needs of immigrant and refugee communities at various policy levels, that is, city, county, state, and national, as well as in the private philanthropic organization sector.

Assess and Design New Formats Collaboratively With Clients

Based on our research finding that some immigrant families were reluctant to use virtual appointments, agency staff and clients may need to cocreate effective service-provision alternatives that will work long-term. Researchers and organizational leaders should intentionally collect and analyze quantitative and qualitative data around access, use, and outcomes of services that changed and stayed the same to assess what worked and didn’t work during the pandemic, and make decisions about future service delivery based on those evaluations. Organizational leaders could use rapid quality improvement tools such as PDSA (Plan-Do-Study-Act) cycles to examine and improve their service provision. The sudden implementation of remote, high-tech, or virtual services as a replacement for many in-person services offers an unplanned “test” of hybrid service provision barriers and facilitators.

Value and Leverage Trust Built with Communities Over time

One of the consistent themes in this study that emerged as a resilience factor as well as a barrier to continued care, was trust. Individual providers and agencies who had solid foundations of trust with their clients were able to leverage that trust into continued support and care for clients. Providers who struggled with establishing trust, such as when meeting first-time clients on the phone, felt that their ability to provide quality care was more limited, which echoes other research on the subject (Held et al., 2020). Agencies should consciously prioritize and strategize ways to establish and sustain community trust due to its pivotal importance as a key factor to effectively serving their clients at all times, but particularly critical in times of crisis. They may find valuable the established literature on incorporating trauma-informed principles in engagement with immigrant and refugee communities (Linton et al., 2019;Miller et al., 2019).

Provide Direct Support for Staff

An important finding in this study was the need to support agency staff logistically as well as mentally/emotionally. While not asked specifically, many providers volunteered that they were experiencing stress due to rapid changes and increases in workload, their own personal experience of the pandemic, and their inability to always meet clients’ needs. Several providers indicated they received helpful support from their supervisors or colleagues. While this finding was technically outside the scope of work for the study, it is critical in the success of implementation for staff to feel supported in order to engage in the heavy work demanded from them. Specifically, staff experienced significant increases in workload. Without additional support, the increased workload along with pandemic-induced personal stress may lead to burnout and staff turnover, which could compromise the NPT mechanisms demonstrated herein, the system’s resilience, and the adapted services provided to clients (Nilsen & Bernhardsson, 2019). Support for staff has been emphasized as important in other studies of providers during COVID-19 (e.g., Horesh & Brown, 2020).

Use Practice-Based Evidence

Finally, in the wake of the pandemic, several voices have urged greater nimbleness in the implementation science field to be able to adapt to crises such as COVID-19 (Chambers, 2020; Taylor et al., 2020). The current authors also suggest that implementation science may benefit from practice-based evidence and natural experiments as foundations for these nimbler research endeavors.

Study Limitations

While this study was unique in its application of an implementation science process theory to the experiences of health and human service providers during the COVID-19 pandemic, there are some limitations of note. The study’s human subjects protocol did not allow researchers to ask personal questions of the providers’ own experiences during the pandemic, which may have led to deeper understanding of NPT mechanisms. Another limitation was that all providers were within one U.S. state and served immigrants and refugees residing there; inclusion of providers from different states and countries serving a greater range of communities may have revealed different policies, strategies, and services leading to different outcomes

Conclusion

Health and human service providers had to make significant adaptations to their service delivery during the COVID-19 pandemic. They demonstrated extraordinary resilience in continuing to meet the needs of their clients within unique constraints that surfaced from the pandemic. Policy implications include the need for increased funding for service providers and their immigrant and refugee clients whose experience increased vulnerability during times of crisis. Practice implications include the following: 1) coevaluate with clients the approaches that were put in place during the crisis, 2) provide more direct emotional and logistical support to staff, and 3) recognize and value the trust that has been built with communities. Research implications are to generate practice-based evidence to inform implementation science and service delivery science, as the application of normalization process theory in this study demonstrates that agencies and their staff are able to implement significant change when the need arises.