Understanding service navigation pathways and service experiences among homeless populations

Abstract

Previous homelessness research examined common pathways into homelessness, yet not much is known about how people navigate through services while experiencing homelessness. This study explored the service pathways of homeless individuals in the U.S. context, which show their connection with multiple organizations and their lived experiences of using services over time. We conducted 12 semi-structured in-depth interviews to grasp the history of service pathways, including the number of organizations, time gaps between services, and referral patterns. We also conducted participant observation shadowing with a subset of the study participants to understand how they interact with caseworkers. The length of service pathways varied, from less than five years to more than two decades. On average, participants went through at least three and up to eight organizations. Regarding service experiences, systemic- and individual-level themes were drawn for negative or positive experiences, such as strict organizational policies and the caseworker’s demeaning attitudes (negative), or supportive organizational culture and strong employee competencies (positive). The findings of this study provide deeper insights into homeless populations’ service trajectories and their experiences throughout the service-navigating process.

Keywords

Homelessness homeless pathways service pathways social services participant observation thematic analysis

Introduction

The United Nations High Commissioner of Human Rights Office acknowledges adequate housing as a human right, yet homelessness remains an issue in most prosperous nations. In fact, homelessness is an increasingly major social problem due to the affordable housing crisis, increasing rents, and the soaring risks of eviction during the current pandemic (Center on Budget and Policy Priorities, 2021; Rogers and Power, 2020). In the United States, on any given night, over 600,000 people are homeless according to the national point-in-time (PIT) estimate (National Alliance to End Homeless [NAEH], 2021). However, the actual population who experience homelessness may be larger due to the undercount of people who seek shelter during the winter months, when the PIT count is conducted, as well as the number of unaccompanied youth (Boone, 2019).

The status of homelessness is difficult to track because it is more comprehensive than the absence of a roof over one’s head. The U.S. Department of Housing and Urban Development (HUD) defines the “literally homeless” as those with no habitable place to live, including unsheltered and sheltered populations (HUD Exchange, 2022). Homelessness involves not only those sleeping on sidewalks, under bridges, in parks or abandoned buildings, or in cars but also people living in emergency accommodations such as homeless shelters or hotel/motel with vouchers (HUD Exchange, 2022; NAEH, 2021).

Research has examined homelessness at various stages, for example, the trajectories into homelessness, during homelessness, and out of homelessness. The experience of homelessness may be brief or the mark of a long journey without adequate housing and a threat to the well-being of homeless persons. The concept of pathways has been useful in understanding the individual experiences of homelessness, especially from a longitudinal perspective. This study focuses on the pathways during homelessness, particularly related to service navigation and utilization.

Understanding pathways in homelessness research

Previous research mostly applied the concept of pathways in the phase into or out of homelessness. People’s paths into homelessness are complex, interlinked, and often the result of behavioral and biographical factors. Poverty and low access to housing are typically the leading causes of homelessness. However, many other factors lead people into homelessness, including mental illness, substance abuse, domestic violence, disabilities, racial inequalities, and incarceration history (Couloute, 2018; Johnson et al., 2008; National Law Center on Homelessness & Poverty, 2015; Sample and Ferguson, 2020). On the other end, in the paths of exiting homelessness, housing and psychological and social factors were found to be important (Johnson et al., 2008).

The causes above highlight how one possibly enters or exits homelessness, yet little is known about how people navigate services and what contributes to remaining homeless. As developed by Clapham (2002), we adopted the term “pathways” to indicate the time and steps taken to navigate and access needed services. Clapham (2002, 2005) primarily discussed the movement of individuals and their housing routes taken over time. However, Clapham (2003) specifically used this model to address homelessness and housing services and suggested further research on the complexities of homelessness and prevention.

The concept of pathways is also found in literature considering mental and physical health. Service pathways in those areas were first coined in the 1980s in the United Kingdom as “Integrated Care Pathways,” which are also referred to as “clinical pathways,” “critical pathways,” “care plans,” “care paths,” and “care maps” (Center For Policy on Ageing, 2014: 266). Similarly, in the United States, care pathways were conceptualized to assist people with physical or mental health problems in navigating services. For example, pathways were examined to identify children’s points of entry into mental health services (e.g., educational, medical, juvenile justice, or child welfare systems) (Farmer et al., 2003; Powell et al., 2007).

Regarding homelessness, Continuum of Care (CoC) was established in the United States in 1995 to streamline the services and assistance funding process. Yet, the CoCs only monitor the entrance of people who are homeless into Coordinated Entry Assessment, but not their pathways throughout services (Fowler et al., 2019). Also, previous literature mainly focused on pathways into homelessness or the re-entry process (Hong et al., 2018). For example, Burns and Sussman (2019) observed 15 older adults using qualitative shadowing methodology to better understand their pathways entering homelessness, but not the pathways within service systems. Addressing this gap, a recent study suggested the need to understand the interaction between homeless service providers and people who are homeless during the service pathways (O’Shaughnessy and Greenwood, 2021). The researchers applied Clapham’s pathway framework and identified the need for adult service users to have a sense of autonomy and dignity (O’Shaughnessy and Greenwood, 2021).

A shortage of research in service pathways limits our understanding of the challenges and needs of people experiencing homelessness and how they use services, which is critical to improving service planning (Mayberry, 2016). These insights can benefit social workers to map out the service pathways to be more efficient and effective for homeless populations. The current study explores homeless individuals’ long-term service navigation processes and their service experiences. We adopted qualitative methods to provide retrospective insights into how people move through services over time.

Methods

Design

We used mixed methods to understand the service pathways of people who are homeless in Albany, New York. We reached three notable homeless-serving organizations in the city starting from August 2019 to recruit study participants. With recruited participants, semi-structured in-depth interviews were first conducted to gain insights into how people are referred from one service to another. We considered all services that address housing, healthcare, and any other needs. Then, using the interview data, we drew each participant’s pathway map to show how the person has used services through various organizations over time. Next, we shadowed participants in a setting where they met with a service provider and received direct social service. After the shadowing opportunity, in-depth interviews were conducted again to debrief the observed experience. Collected interview data were analyzed using thematic analysis.

Use of shadowing

Shadowing is “a research technique which involves a researcher closely following a member of an organization over an extended period of time” (McDonald, 2005: 456). In a qualitative research domain, shadowing has been used as a “perspective-gathering mechanism” in a wide range of fields, including education, sociology, anthropology, ethnography, and phenomenology (Czarniawska, 2007; Quinlan, 2008: 1483). Shadowing generates holistic insights into the dynamics of relationships and behaviors based on the firsthand observation of participants (McDonald, 2005; Watts, 2010). We chose to use shadowing because we wanted to gain a deeper understanding of homeless individuals’ service use experiences, the process of which is largely unknown to research. Due to the feasibility of the project, we carried out one-time observations of the interactions between the study participants and their service providers. Each shadowing session lasted from one to two hours.

Data collection

Local context

Albany is the New York State capital city, which hosts approximately 95,000 residents within the city limit and serves the broader surrounding capital district with close to one million people. With its central location, Albany is known to provide more robust services and benefits to people who are homeless with a working alliance among their homeless coalition compared to neighboring counties (Burt et al., 2010). The PIT count for homeless populations was 92,091 (or 47 per 10,000) in New York State and 750 (or 24.4 per 10,000) in Albany County in January 2019 (NAEH, 2021). Both numbers are higher than the U.S. average, which was 17 out of 10,000 during the same PIT count of January 2019.

Recruitment

The key inclusion criteria for our study were homeless individuals who (a) could share their experiences of navigating services and (b) potentially allow us to observe their service experiences in person. Through convenient sampling, participants were recruited at three local homeless-serving agencies (i.e., one family shelter, one permanent supportive housing, and one multi-service center). During the recruitment process, we explained the research to the gatekeepers of the organizations (e.g., directors) first and obtained their permission to reach out to their service recipients. We offered participants local grocery store gift cards as an incentive and as an expression of gratitude: $15 for the first interview and an additional $35 for the shadowing session and the post-shadowing interview. The type and the amount of incentives were determined after a careful discussion on their potential influence on the motivation of participants and the quality of data. Considering that homeless individuals are financially vulnerable, the research team wanted to honor their time with an amount comparable to the minimum wage. The lead researcher’s previous experiences interviewing homeless populations in the local setting and her consultation with the Institutional Review Board also guided the final decision.

Participation in and withdrawal from the study were completely voluntary. The research logistics and potential harms were informed to participants before every step of the research process. We started each interview by going through a consent form, including explaining the purpose of the study, what data will be collected, and how the data will be kept and managed. Also, we clearly informed the interviewees that their participation or responses would not impact their services at the organizations. The authors’ Institutional Review Board approved the research procedure before implementing the study.

We first conducted in-depth interviews to learn about the participant’s service pathways. At the end of the interview, we asked participants if they had any upcoming appointments with service providers within a month. Those who met the criteria continued with the shadowing session, during which we conducted observation and a follow-up interview to inquire about their service experiences. Before the shadowing session, we informed the caseworker and the site manager about the research and asked for their verbal permission to observe the session.

Our research team particularly sought contexts where we could observe detailed service procedures and interactions between the participant and a caseworker or a service coordinator; thus, simple service activities such as getting a prescription were not selected for shadowing. Therefore, not everyone who participated in the first interview led us to shadowing opportunities. Attrition also occurred when we did not hear back from study participants since their first interview. As a result, 12 people participated in the first interview and four people in the shadowing session between September 2019 and January 2020.

Data

Quantitative data were collected regarding the demographic characteristics of the participants and their experiences in service navigation and use. Demographic data included age, sex, race/ethnicity, size of household, income level, educational attainment, and employment status. For service experiences, we gathered detailed information about every organization that participants have worked with in the past. For example, we asked, “Have you worked with any human service organizations in the past?” “Among the organizations you mentioned above, which ones are you currently working with?” “Which one was the first organization you had contact with?” “What year was it when you were connected to that organization?” “What services do/did you use from that organization?” and “Have they referred you to any other organizations?” Based on the responses, we created service pathway maps by identifying the first organization the person contacted and drawing arrows to the following ones referred to by the previous organization. When the referral source was not an organization but family or friends, such information was specified on the map. Pathway maps visually showed how many organizations each participant had engaged in their service navigation history and how those systems were connected.

Qualitative data were collected regarding the participants’ experiences of navigating and receiving needed services. For example, during the initial interview, we asked, “How did you start working with this organization?” and “Have you had any difficulties or barriers with scheduling or getting an appointment with this organization?” These questions allowed us to gain the participants’ sources of information and referrals and any challenges receiving services and establishing a relationship with service providers (e.g., caseworker). After shadowing observations, we asked questions more specific to the observed session and the service agency. These questions included “What do you think about the services you received today?” “What went well or did not go well in your opinion?” and “Is there anything else you’d like to add to your overall interactions with this organization?” While observing the meetings, researchers also kept a shadowing log which included notes on the content and the impression of the interactions between the client and the caseworker. During the post-shadowing interview, the notes were shared with the participant to confirm that the researchers’ understanding of the context was accurate.

Data analysis

Pathway maps showed a unique service trajectory for each participant. We analyzed each map to gather descriptive information regarding when and how the person was connected to service organizations (e.g., the year that a participant engaged with a service organization for the first time, the total number of organizations identified in the pathway map, and times taken between each service experience). The quantifiable information helped the research team identify how long participants have navigated services and what common issues they have dealt with since becoming homeless.

Interview data were analyzed using thematic analysis. All voice files for the pre-and post-shadowing interviews were transcribed. The first and third authors read the transcripts separately and coded phrases or sentences relevant to participants’ subjective experiences, opinions, perceptions, and feelings regarding navigating and receiving services. Then the authors collectively discussed the codes and identified similarities and discrepancies. We dropped the insignificant or irrelevant codes to the research questions and merged the ones named differently but pointing to the same theme. Then each researcher reviewed the codes again to draw themes independently. Lastly, the authors finalized the major themes and discussed how to structure them in a way that is conducive to understanding the data. Shadowing logs were not included in the analysis as they did not have a direct response from the participants.

Results

Descriptive statistics

Participant characteristics

Table 1 shows descriptive statistics of the participant demographics. All 12 participants were adults between 19 and 62 (average=43.8) years old residing at either a temporary shelter or permanent supportive housing. The male and female ratio was 5:5. In terms of race or ethnicity, seven African Americans, four Whites, and one biracial person participated. According to the HUD income limit, all participants were low income, with a majority reporting less than $10,000 income per year and unemployment status. Five out of 12 participants attained an associate degree or some college education, three achieved a high school degree or General Educational Diploma (GED), and four did not complete high school. Those who were recruited at the family shelter and the multi-service center had at least one child living with them (ID = 1, 6, 7, 8, 9, 10, 12), whereas the others recruited at the permanent supportive housing had no other family or dependent staying with them (ID = 2, 3, 4, 5, 11).

Table 1.

Descriptive characteristics of study participants.

ID	Age	Sex	Race	Household Size (adult/child)	Income	Education	Employment	Other Issues
1	50	F	Bi-racial/Multi-racial	3 (2/1)	$20,000–$29,999	Some college/Associate degree (AD)	Employed	Disability/special education
2	39	M	White	1 (1/0)	<$10,000	High school (HS)/GED	Underemployed	HIV, incarceration
3	53	F	Black/African American	1 (1/0)	$10,000–$19,999	AD	Underemployed	HIV, mental health, substance abuse
4	55	M	Black/African American	5 (5/0)	<$10,000	Less than HS	Unemployed	HIV
5	55	M	Black/African American	1 (1/0)	<$10,000	Less than HS	Unemployed	HIV, mental health, incarceration
6	19	M	Black/African American	3 (2/1)	<$10,000	Less than HS	Unemployed
7	42	F	White	2 (1/1)	<$10,000	AD	Unemployed
8	41	M	Black/African American	2 (1/1)	<$10,000	High School	Unemployed
9	44	F	White	3 (1/2)	<$10,000	High School	Unemployed	Mental health, disability
10	28	F	Black/African American	2 (1/1)	<$10,000	Less than HS	Unemployed
11	62	M	Black/African American	1 (1/0)	<$10,000	Some college/AD	Unemployed	HIV
12	38	F	White	3 (1/2)	<$10,000	Some college/AD	Unemployed	Mental health, substance abuse, domestic violence

Service navigation pathways

The service pathways can be conceptualized in two ways: the length of time passed since the person became homeless and the number of service organizations that the participant contacted since they became homeless. The length of pathways largely depended on the first contact with a service organization. Those who started their engagement with service organizations in earlier years naturally showed longer service pathways. Two participants mentioned that their first service occurred in the 1990s, three in the 2000s, five after 2017, and two responded, “I don’t remember.” In terms of the number of service organizations, participants went through at least three and up to eight (average=5.4) organizations. The service organizations varied from government agencies to medical centers, temporary shelters, and other community-based organizations. Among various situational factors, the participants who had mental health and substance abuse issues tended to engage with a higher number of organizations. Some pathways also demonstrated how incarceration could disrupt service navigation. For example, one participant passed through seven organizations at five different times, in 2001, 2008, 2010, 2018, and 2019 while he was in and out of jail. During the period of incarceration, the participant was disconnected from services and, once released, reconnected to some of the previous services or engaged with new service agencies introduced by the re-entry coordinator.

Although there were some waiting times until the approval of service applications, the participants’ pathways to temporary services were shorter and quicker than the others requiring more long-term solutions such as employment and permanent housing. The time between services (i.e., time taken for being referred from one organization to another to utilize services at a later organization) also varied from less than one month up to six years. One notable pattern was that the referral time was relatively shorter for addressing immediate needs than addressing long-term needs, especially for those already connected to a centralized service hub such as a local government’s social service agency (i.e., Department of Social Services [DSS]). In those cases, it took less than one month with a maximum of two to connect people to benefits and services such as temporary housing, supplemental security income (SSI), and other non-cash benefits (e.g., Supplemental Nutrition Assistance Program [SNAP] or Medicaid).

Participants also showed various referral patterns. Some participants experienced linear referral patterns without time gaps or interruptions from one service organization to the next. For example, Figure 1 shows a pathway in which the participant first worked with a staff in a medical center, which then led him to temporary housing and later to permanent supportive housing, where he currently receives additional help with job placement and social support. In these cases, all participants’ needs were gradually met over time. However, in other cases (see Figure 2), service pathways did not show smooth referrals between services. The connection was particularly sparse when participants were merely seeking immediate financial assistance (e.g., SNAP and SSI). People who did not receive referrals from previous organizations often sought service-related information on their own or through their social networks (e.g., friends, family members, acquaintances).

Figure 1.

Example of a linear service pathway.

Figure 2.

Example of a non-linear service pathway.

Qualitative research findings on service experiences

As a response to the open-ended questions, “What do you think about the services you received today?” “What went well or did not go well in your opinion?” and “Is there anything else you’d like to add to your overall interactions with this organization?” participants’ experiences of service navigation broadly reflected positive or negative sentiments. Furthermore, the themes in each sentiment comprised both individual- and systemic-level factors that articulate what made their experiences positive or negative. Table 2 shows significant themes from qualitative data analysis, categorized by sentiment and the level of experience.

Table 2.

Major themes for service experiences.

Sentiment	Level	Themes	Code frequency
Negative	Systemic	Service coordination (e.g., communication, wait-time)	9
		Welfare policies (e.g., sanctions)	3
		Accessibility infrastructure (e.g., transportation, technology)	3
Negative	Individual	Service provider’s attitude (e.g., discriminative, demeaning)	5
Negative	Individual	Service provider’s incompetence	3
Positive	Systemic	Agency support	2
Positive	Individual	Worker-client relationship (e.g., trust, partnership)	5
Positive	Individual	Service provider’s competence	1

Negative experiences—systemic level

Most participants discussed a lack of service coordination within an organization or across agencies in different counties as a critical issue. They often had to wait for weeks to get status updates on their services in the absence of staff members or their response. Also, service appointments were sometimes miscommunicated and canceled for no reason, which caused a waste of time for service users. The following quotes demonstrate several examples of unstable service coordination and staff shortages and how they have caused inefficiencies.

Several weeks… They wouldn’t return my calls. I kept calling and calling and calling and calling and calling, and finally, I called one morning, and she answered the phone, and I was shocked. And she said all she needed to know was a single letter. She could have sent a letter way ahead of time. But she didn’t.

I had intensive care management, and they were supposed to come to my house. Every time they were saying they’ll come to my house, they never showed up. Then they were trying to blame me and say I wasn’t home. It was not true because my window was right outside the driveway, so I could see if anybody comes as I sit there and wait for them, and then no one showed up. So, it was true there were lies, and it was all kinds of crazy misinformation given, and I just got tired. Then I just said, forget it, I want to go to someplace else.

Participants also experienced strict policies that were often punitive. Most frequently addressed were sanctions for declining an affordable housing option recommended by the state agency. Many participants stated that they did not like the quality and environment of the housing that their caseworker had found for them. However, they felt almost forced to accept it because if not, they would not have anywhere to stay for at least the next few weeks until they get another recommendation. The following quotes from two different participants describe the similar struggles experienced while waiting for permanent supportive housing.

If I denied it, then they (DSS) said that I would basically be denying affordable housing, and then I wouldn't have nowhere to live because you get sanctions, so I had no choice, so I took it. I didn't want to.

She (the caseworker) said, oh, you know, you can't turn down for housing. I said, listen, one thing, you're not going to put me anywhere. ... I'm not doing that. ... I got a son... I don't want them to just put him anywhere... I didn't know I'd get a sanction for that. She was like, you're going to have sanction for that. I didn't know that, that's crazy. That's crazy. That's really ridiculous.

In addition, participants mentioned the lack of infrastructure as a barrier. For example, when service providers used slow Wi-Fi or an outdated computer, it prevented smooth procedures of service delivery. In addition to infrastructure barriers within the organizations, some participants commented on the lack of transportation options as a barrier for them to travel to visit offices and submit the necessary paperwork in time.

All the systemic-level factors described above (i.e., coordination issues, inflexible and punitive policies, and lack of infrastructure that can enhance accessibility) were closely related to elongated wait times. In the participants’ service-navigating experiences, waiting for an extended period to connect to the next service or resources, often with unexpected hurdles in between, led to negative emotions and mental stress. Participants described such a waiting period using expressions such as “very stressful,” “tired,” “my mental was dying,” and “that was killing me.” The quote below demonstrates a similar emotional toll involved with indefinite waiting.

The longer you stay here, it’s a lot of stress that you’re going through. Waiting around for things to happen, and didn’t. They hype you up to the point where you think it’s going to happen and then it doesn’t.

Negative experiences—individual level

Some negative service experiences were related to their interactions with individual service providers. Some participants shared that they experienced discriminative attitudes from staff, as shown in the examples below:

It is like, honestly, if you could treat them a little bit better in a homeless unit as opposed to treating people great when they’re in normal DSS, you know. Sometimes life just happens. And, you know, it’s like (they are treating me like) I’m a person that constantly going homeless. No, this is the first time ... It’s the first time I’ve ever been homeless.

I got attitude, and I was being real condescending attitude and real disrespectful. That’s how you get treated when you’re in a homeless unit … As soon as I get her attitude and then instantly turned me into the opposite of what I am right now… when I get talked to like that, I shut down, you know, I close everything.

Furthermore, participants shared that such demeaning experiences with caseworkers disturbed their working relationship, the overall service process, and trust in the service agencies.

Sometimes frustration was expressed toward incompetent case managers who they believed to cause service delays, as demonstrated in the case below:

The case manager over there, he doesn’t do anything. I go to DSS; she told me your case manager never handed me your paperwork… That doesn’t make sense to me. I gave him stuff, he made copies… And then she told me we never received that from him. I’m like, really?

The quote above suggests that service delivery would fail when multiple case managers deal with miscommunication issues and potentially staff shortages and incompetency.

Positive experiences—systemic level

While participants’ positive service experiences were mainly related to the individual level, one participant stated that support within the organization created a positive service experience. The person reported that the agency had a caring culture and that staff made ongoing and holistic efforts to meet clients’ needs. The critical role of organizations in increasing clients’ service satisfaction is demonstrated in the following quote:

I joined with them (agency), and they have been on my side ever since. I am so appreciative that I had signed up with an agency that has so much care towards individuals with medical issues such as myself... They helped me through a lot. They really have. Not everyone is actually receiving these wholesome services in these organizations.

Positive experiences—individual level

The factors that promoted positive service experiences were most frequently related to a trustful relationship with their caseworker. The following quote exemplifies the significant role of the worker’s willingness to help and actively collaborate with clients to achieve their mutual goal:

He has a good idea of basically what is going on in my life and basically anything what I am attempting or trying to accomplish. And he’s all for it. So that’s the good thing about it. He’s definitely on board with it… And I’m on board with him to help me, so I feel comfortable…This is new to him because he's not used to ever approaching any situation like this… but he has accomplished a great deal. And I truly commend, I really do.

Service providers’ competence was also a recognizable factor. One participant expressed satisfaction with the service received when finding the caseworker resourceful and knowledgeable. The following quote describes how a competent service provider can help the participant feel assured about the services received:

She (the caseworker) seemed to know what she was talking about, which was helpful. Sometimes when you go to places like this, they don't really know. You can tell they're just like... pulling stuff out of wherever... She knew what would help me and what wouldn’t. Even though I didn’t need a housing… she was prepared to help with housing, but she was able to help with other resources.

These findings suggest that clients’ burden of service navigation can be lightened when service providers are adept to deal with the complexities that service pathways present. Knowing that the pathways involve multiple services over an extended period, service providers’ up-to-date knowledge and their efforts to build trust and maintain an ongoing partnership with clients would be critical for delivering sustainable benefits to clients.

Discussion

Summary

The findings from this study shed light on the trajectory of navigating services after becoming homeless. Various service pathways show that homeless people deal with multiple service providers and organizations to access needed services, which then cause them to stay homeless for an elongated time. While navigating services, participants perceived problems with organizational policies and procedures, such as insufficient resources (e.g., staff, technology), too much paperwork, long wait time, and slow operation, as barriers. Nevertheless, many participants were satisfied with the services, emphasizing that helpful attributes of service providers (e.g., knowledge, trustworthiness, and a sense of partnership) made a difference in their navigation experiences.

Despite some variances, the findings suggest that connecting homeless people to centralized resource centers such as government agencies (e.g., DSS, Social Security Administration) early on would accelerate addressing their most immediate needs, which often include temporary housing and financial assistance. However, our findings also suggest that the role of government agencies is limited and often perceived as restrictive in the process of searching for long-term solutions such as permanent housing and employment. These findings provide insights for advocating policy change and ways to shorten service navigation pathways for homeless individuals.

Limitations

We recognize several limitations of this study. Since the inquiries into service pathways mostly concern people’s past service navigation history, the response to the interviews relied on people’s memories, which generated some missing information in the participants’ pathway maps. Retrospective pathway mapping was useful for understanding how homeless populations move through services over time and what aspects of their pathways between services have in common. However, these limitations suggest that synchronous pathway mapping at the time of each service use or referral may increase the accuracy of tracking and evaluating the efficiency of service delivery and coordination.

The shadowing method implemented in this study allowed the researchers to empathize with the study participants’ experiences and their later accounts of positive and negative experiences. However, we also faced obstacles in recruitment. We originally envisioned observing many participants in action, capturing the moments of service experiences in real-time, but we learned that not everyone’s situation offers a suitable opportunity for observation. Another limitation was the number of observations. The one-time observation for each participant was often not enough to fully understand the context of the session and the participant’s perception of a service organization or a provider that had been formulated over time. Fortunately, post-observation interviews and observer’s notes allowed researchers to reaffirm some of those unanswered areas. However, the challenges we experienced suggest that having multiple shadowing opportunities could be useful to get a more comprehensive understanding of people’s service use experiences.

Implications for future research and practice

This study addresses the knowledge gap in pathways in the homeless service system and further suggests that more research is needed. Our study is aligned with previous literature suggesting that the service navigation journey can be torturous for those who experience homelessness, especially when people face institutional barriers, limited transportation options, and insufficient communication from providers (Burt et al., 2010; Mayberry, 2016). Mayberry (2016) specifically discussed “catch-22” situations as a unique challenge for homeless people, in which their needs cannot be immediately addressed unless a prerequisite is met. Such incongruity in rules or policies makes it impossible for participants to obtain the desired outcome. For example, a participant from this study could not make it to his appointments because the Medicaid transportation service was discontinued. For renewing the service, he needed to update a doctor’s note proving his physical conditions; yet the health service provider would not provide the note if he did not come to the office.

Furthermore, this study confirms that the service pathways often require financial support and various others for health and social needs. The co-occurring conditions of homelessness, such as significant mental health problems, substance use, HIV, or domestic violence, can elongate and complicate pathways. Future studies can further enlighten how such conditions affect the characteristics of pathways. Psychological factors such as stigma or a sense of belonging could be important moderating factors influencing the length of service pathways and the quality of outcomes (Johnson et al., 2008). The impact of incarceration on long-term service navigation is also an area to be further examined. Our findings suggest that incarceration disrupts service use and thus can cause repeated and inefficient pathways. These inquiries can be addressed by focusing on subgroups of homeless populations who deal with particular situations and conducting an in-depth study of their service pathways and hidden barriers to receiving services.

Methodologically, the pathway mapping and shadowing method can be scaled up with further considerations. As mentioned above, future research may consider observing service navigation experiences multiple times to get a fuller picture of the service trajectory that each participant goes through. Multiple shadowing would provide more opportunities in real-time for researchers to check in with the participants while observing them in natural interactions with service providers. Researchers would also find it beneficial to include more participants in the study to find a clearer pattern of how homeless populations receive services over time.

For practice, the insights from this research can facilitate discussions on how individual caseworkers and agencies as a system can improve the service referral process and experiences for homeless populations. Both policymakers and practitioners have acknowledged the need for simplifying service pathways for homeless populations in the field. For example, a report for the HUD recognized the need to improve the service pathways by “smoothing mechanisms,” which means the processes that reduce barriers and address problems at a street level (Burt et al., 2010: xvii). Some of the attempts to smooth mechanisms include offering transportation, one-stop intake services, feeding programs, outreach, and admission to shelters or other facilities designed for the homeless. Service navigation hubs, which refer to coalitions of service providers in one space providing accessible and connected services, can be another optimal response to reduce service pathways for people who are homeless (Colletti, 2018). These approaches can also shift the mind of social workers from simply making referrals to envisioning the pathways down the road and trying to minimize the trauma that people who are homeless could experience during the pathways. Social workers can organize ways to collect the narratives of homeless populations at a city or a regional level to reduce duplication of services in the pathways. Practitioners can also discuss the possibility of utilizing an existing data management tool (e.g., Homeless Management Information System [HMIS]) to understand people’s service pathways more in-depth. For example, HMIS can potentially be utilized to inform policy decisions by identifying duplicated service use histories, analyzing service utilization patterns, and evaluating the effectiveness of homeless programs at the client and organizational levels (HUD Exchange, 2021). More systemic evaluations and the use of current tools such as HMIS will improve policy and practice to meet the goal of ending homelessness.

Footnotes

Acknowledgements

We sincerely thank our study participants who shared their time and stories for this research. We also acknowledge Emily Herendeen, MSW, for her support for the study as a research assistant.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Foundation [grant number ECCS-1737443].

ORCID iD

Eunwoo Lee

References

Boone

(2019) Is there a better way to count the homeless? Bloomberg, 4 March. Available at: https://www.bloomberg.com/news/articles/2019-03-04/the-problem-with-hud-s-point-in-time-homeless-count (accessed 1 May 2022).

Burns

Sussman

(2019) Homeless for the first time in later life: uncovering more than one pathway. The Gerontologist 59(2): 251–259.

Burt

Carpenter

Denton

, et al. (2010) Strategies for improving homeless people's access to mainstream benefits and services. Available at: www.ssrn.com/abstract=1581462 (accessed 1 March 2021).

Center For Policy on Ageing (2014) The effectiveness of care pathways in health and social care. Available at: www.cpa.org.uk/information/reviews/CPA-Rapid-Review-Effectiveness-of-care-pathways.pdf (accessed 8 March 2021).

Center on Budget and Policy Priorities (2021) Why expanding housing choice vouchers is essential to ending homelessness. Available at: https://financialservices.house.gov/uploadedfiles/hhrg-117-ba00-wstate-oliviaa-20210609.pdf (accessed 1 March 2021).

Clapham

(2002) Housing pathways: a postmodern analytical framework. Housing, Theory and Society 19(2): 57–68.

Clapham

(2003) Pathways approaches to homelessness research. Journal of Community & Applied Social Psychology 13(2): 119–127.

Clapham

(2005) The Meaning of Housing: A Pathways Approach. Bristol, UK: Policy Press.

Colletti

(2018) Defining and designing navigation centers according to current California legislation. Available at: homelessstrategy.com/defining-and-designing-navigation-centers-according-to-current-california-legislation (accessed 1 March 2021).

10.

Couloute

(2018) Nowhere to go: homelessness among formerly incarcerated people. Available at: www.prisonpolicy.org/reports/housing.html (accessed 1 March 2021).

11.

Czarniawska

(2007) Shadowing: and other techniques for doing fieldwork in modern societies . Malmö: Liber.

12.

Farmer

Burns

Phillips

, et al. (2003) Pathways into and through mental health services for children and adolescents. Psychiatric Services 54(1): 60–66.

13.

Fowler

Wright

Marcal

, et al. (2019) Capability traps impeding homeless services: a community-based system dynamics evaluation. Journal of Social Service Research 45(3): 348–359.

14.

Hong

Malik

Lundquist

, et al. (2018) Applications of machine learning methods to predict readmission and length-of-stay for homeless families: the case of Win shelters in New York City. Journal of Technology in Human Services 36(1): 89–104.

15.

HUD Exchange (2021) Homeless management information system (HMIS). Available at www.hudexchange.info/programs/hmis (accessed 8 March 2021).

16.

HUD Exchange (2022) Four categories of the homelessness definition. Available at https://www.hudexchange.info/homelessness-assistance/coc-esg-virtual-binders/coc-esg-homeless-eligibility/four-categories (accessed 14 March 2022).

17.

Johnson

Hellene

Coutts

(2008) On the outside: Pathways in and Out of Homelessness. Melbourne, VIC: RMIT University.

18.

Mayberry

(2016) The hidden work of exiting homelessness: challenges of housing service use and strategies of service recipients. Journal of Community Psychology 44(3): 293–310.

19.

McDonald

(2005) Studying actions in context: a qualitative shadowing method for organizational research. Qualitative Research 5(4): 455–473.

20.

National Alliance to End Homelessness (2021) State of Homelessness: 2021 Edition. Available at: https://endhomelessness.org/homelessness-in-america/homelessness-statistics/state-of-homelessness-2021 (accessed 1 March 2021).

21.

National Law Center on Homelessness & Poverty (2015) Homelessness in America: overview of data and causes. Available at: https://homelesslaw.org/wp-content/uploads/2018/10/Homeless_Stats_Fact_Sheet.pdf (accessed 8 March 2021).

22.

O’Shaughnessy

Greenwood

(2021) Autonomy and authority: homeless service users’ empowering experiences in housing first and staircase services. Journal of Community & Applied Social Psychology 31(3): 288–304.

23.

Powell

Fixsen

Dunlap

, et al. (2007) A synthesis of knowledge relevant to pathways of service delivery for young children with or at risk of challenging behavior. Journal of Early Intervention 29(2): 81–106.

24.

Quinlan

(2008) Conspicuous invisibility: shadowing as a data collection strategy. Qualitative Inquiry 14(8): 1480–1499.

25.

Rogers

Power

(2020) Housing policy and the COVID-19 pandemic: the importance of housing research during this health emergency. International Journal of Housing Policy 20(2): 177–183.

26.

Sample

Ferguson

(2020) It shouldn’t be this hard: systemic, situational, and intrapersonal barriers to exiting homelessness among homeless young adults. Qualitative Social Work 19(4): 580–598.

27.

Watts

(2010) Ethical and practical challenges of participant observation in sensitive health research. International Journal of Social Research Methodology 14(4): 301–312.