1. INTRODUCTION

The United States (U.S.) Department of Housing and Urban Development (HUD) defines homelessness in four categories that include: 1) literally homeless; 2) imminent risk of homelessness; 3) homeless under other Federal statutes; and 4) fleeing and attempting to flee domestic violence (Housing and Urban Development, 2012).

A 2021 Point-in-Time (PIT) Count reports that 580,466 people were experiencing homelessness in America during January 2020; 70% were adults; 30% were families with children; 6% were unaccompanied youth under age 25; 19% were chronically homeless; and 6% were veterans (National Alliance to End Homelessness, 2021). In Texas, about 27,229 people experienced homelessness during the year 2020; 1,912 were families, 1,948 were veterans, 1,408 were unaccompanied youth aged 18-24, and 4,033 were experiencing chronic homelessness (United States Interagency Council on Homelessness, 2021). Homelessness is a global phenomenon that has insufficient quantifiable data (Busch-Geertsema et al., 2016) but that has significant and many times long-lasting health impacts.

The frequent hardships experienced by people who are unhoused are greater than those experienced by the general population (Oppenheimer et al., 2018). Common lived adversities for unhoused individuals include social disconnect, substance use, hospitalization, economic vulnerability, stigma, unmet basic human needs and amenities, poverty, sexual violence and other forms of crime, and limited access to healthcare services (Baxter et al., 2019). These negative experiences are shaped by social determinants of health, which Artiga & Hinton (2018) define as “the conditions in which people are born, grow, live, work and age that shape health” and “include factors like socioeconomic status, education, neighborhood and physical environment, employment, social support networks, as well as access to health care”. Homelessness, with its associated inequalities and physical and psychosocial risks, has been classified as a social determinant of health itself and a major contributing factor for poor health (Stafford & Wood, 2017) across mental, physical, emotional, psychosocial, and sexual health domains (Nichols & Mays, 2021).

The negative impacts of homelessness on a person’s health can place them at greater risk of developing chronic conditions (Nanjo et al., 2020), which in turn may predispose them to various cancers (Tu et al., 2018). Screening, early diagnosis, and appropriate treatments are key components to reducing cancer mortality (World Health Organization, 2021) and can lead to a reduction in costs, disabilities, and death from other chronic conditions (Fragala et al., 2019). However, people who are unhoused are at greater risk of having untreated conditions due to their forced prioritization of meeting immediate human needs over accessing care or finding themselves unable to access those systems due to complex structures, language, requirements, stigma, discrimination, and negative experiences with health and human services professionals; being ultimately forced to overutilize emergency departments as their primary source of care (Omerov et al., 2020). Lack of access to care is one of several social determinants that can dictate every step of cancer care including prevention and intervention efforts (Healthy People 2030, 2021). Cancer is currently the second leading cause of death amongst people who are unhoused (Biedrzycki, 2018). In the U.S., bronchus and lung cancer are the most common cancers found in unhoused individuals, causing more than a third of cancer-related deaths, which suggests a need for more cessation efforts on tobacco use (Baggett et al., 2015). Men experiencing homelessness also show high incidents of prostate, colon, and rectum cancer while women experience more breast and cervix uteri cancers (Lawrie et al., 2020). People experiencing homelessness do not have the privilege of managing cancer treatment in a safe environment with family and friends, exacerbating stressors that impact their quality of life (QoL) (Biedrzycki, 2018).

QoL is a multifaceted concept that the World Health Organization (WHO) defines as “an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” (World Health Organization, 2012). However, because of the subjective nature of QoL, there is still no universal agreement on the definition of this concept (Moons et al., 2006) and although objective variables, such as health, income, and employment can be used to measure one’s QoL, they do not provide enough insight to the subjective experiences of unhoused individuals and how those unique circumstances impact their QoL (Hubley et al., 2014). Subjective measurements on QoL can include opportunities for individual, familial, and community growth, interpersonal and intrapersonal relationships, and having basic human needs met (Brown & Schippers, 2018). Despite the lack of consensus on QoL, Moons et al., (2006) found that satisfaction with life was the most applicable approach to accurately defining QoL and defines it as “the degree to which a person positively evaluates the overall quality of his/her life”. An unhoused person’s satisfaction with life can influence the perception of their health and impact how they cope with and overcome the hardships of homelessness (Munoz et al., 2017).

The COVID-19 pandemic highlighted the way people who are unhoused have a greater risk of infection and mortality (Richard, 2021). The added distress to social services, the economy, and housing opportunities have exacerbated existing issues and risks for people who are unhoused (Fish, 2021) and underscored the significance of housing as a social determinant of health (Perri et al., 2020), which can be addressed through advocacy and policy (Dawes & Williams, 2020).

2. METHODS

The study was conducted in El Paso (Texas) located on the U.S.-Mexico border, with a population of 865,657 (U.S. Census Bureau, 2019), home to thousands who have migrated or are in the process from places like Mexico, Central, and South America, as well as other parts of the United States and world (World Population Review, 2021). The city’s median household income in 2019 was $46,871, per capita income was $21,683, and 18.8% of the study location population lives in poverty (U.S. Census Bureau, 2019).

The Point in Time (PIT) Count of January 2020 estimates that 843 individuals in the city are experiencing homelessness (El Paso Coalition for the Homeless, 2020). The annual count accounts for all sheltered individuals (unsheltered count is bi-annually) in a geographic area (Housing and Urban Development, 2019). Because the PIT Count has limitations, which result in an undercount (National Law Center on Homelessness & Poverty, 2017), it is likely that there are more than 843 individuals experiencing homelessness in the study location.

A local non-profit organization with more than twenty-six years serving individuals experiencing homelessness managed five of the living facilities where participants resided. Another non-profit managed a hotel serving as an overflow shelter and co-managed with the aforementioned non-profit one of the shelters where other participants lived. This shelter, referred to as a welcome center, was created in response to the COVID-19 pandemic. In mid-April of 2020, the City and County came together to have a one-stop facility for receiving individuals experiencing homelessness, where they were screened and tested for COVID-19, isolated, quarantined, and provided shelter, food, and assistance moving on to other housing arrangements.

Research design

To investigate perceived health, health risks, chronic disease, cancer knowledge, access to care, social functioning, and quality of life domains, a research team used an exploratory and cross-sectional design, chosen because it has been implemented with adults experiencing homelessness.

Sampling

The following criteria were used for enrollment of participants in the in-depth interviews: women and men between ages 18 and 75, residing in a shelter at the time of the study, speaking English or Spanish, with residency in study location. The sample consisted of n=75 adult men and women that met the inclusion criteria; this represented approximately 30% of adults experiencing homeless and residing in shelters. The study used a convenience sample (Babbie, 1990). One interview was excluded as audio was inaudible, leaving a sample of .=74 participants.

Eligible participants lived in the shelters, transitional living centers, and overflow shelter. Recruitment was done through outreach by agency managers, social networks, word of mouth, and bilingual flyers distributed through agency properties. Snowball sampling was also used beyond the first recruitment sample size phase (Browne, 2007).

Informed consent

Written informed consent was administered to each participant. During this process, the research team (Author/PI and Research Assistants) explained the study aims, risks, interview tool, and answered questions. Participants consented to having the interviews audio recorded, analyzed, and written up for dissemination. No participant names or identifiers were recorded, and consenting participants received a $20 gift card. The study protocol received approval by the University Institutional Review Board.

Data Collection and Analysis

The study employed a Participatory Action Research approach with the organization managers, several of them with lived experience of homelessness. Researchers and community partners collaborated on all research phases from development of the proposal and conceptualization of the study to the data collection, analysis, and translation of findings. An agency social worker helped formulate the original concept and co-author this article; team members translated to Spanish the study questions; and the team reviewed the tool and made adaptations.

The survey was available in English and Spanish, it consisted of closed and open-ended questions, was developed by the study team using QuestionPro. The instrument included the Perceived Health Status Scale (Ware, 1976), which measures personal health, and the Satisfaction with Life Scale (Diener et al., 1985), which gauges QoL. Both scales have favorable psychometric properties, have been used to study QoL in marginalized populations, and are in the public domain. The Social Functioning SF-12 Health Survey, developed from a 36-question health survey and validated for homeless populations (Gordon et al., 2019; Larson, 2009), which measures QoL through the lens of physical and mental health, was also used. The instrument was field tested with 4 managers to establish content validity and improve content.

The instrument consisted of general and socio demographic characteristics, experiences with systems of care, knowledge and attitudes associated with cancer and other chronic disease risk, access to preventative services during COVID-19, and recommendations to improve care. The PI (author) trained interviewers (research assistants) virtually on ethics, safety, precautions, interview techniques, data collection, and documentation. COVID-19 guidelines were followed for safety.

Three interviewers (Author/PI and Research Assistants) conducted interviews from February 1, 2021, until June 1, 2021. Interview length ranged from 15-55 minutes, longer interviews resulting when participants needed explanation of questions and/or had hearing or vision issues. Interviews were conducted in private or semi-private venues, depending on availability of space. The instrument was to be administered face-to-face using a laptop. At times, internet unavailability necessitated the use of paper surveys with handwritten responses. When this was the case, interviewers input responses into QuestionPro. With the exceptions of the open-ended questions, there were no transcriptions. No adverse events occurred during the interview process, and all participants completed the interviews.

The analysis involved mixed methods integrating standard quantitative (Creswell, 2009; Morgan, 2014) and qualitative protocol (Green & Thorogood, 2014) and included descriptive statistics. For the qualitative analysis, we used deductive coding and thematic analysis, afterwards the data was translated to/from Spanish to English. We assured analytic validity through an iterative process of reading, coding, and discussing among the research team, where disagreements were addressed and resolved.

3. FINDINGS

Participant Demographics

Participant demographics reflect self-reporting. For these demographics, see Table 1.

Experiences of Homelessness

Participants came from seven shelters. Twenty-six (35.13%) were at the welcome center, 18 (24.32%) at the emergency shelters, 10 (13.51%) at the overflow hotel, 8 (10.81%) at a living center for elders, 7 (9.45%) at a living center for elder women, and 5 (6.75%) at the center for veterans. Their length of stay varied from less than a week for 10 (13.51%), to 1-4 weeks for 11 (14.86%), to 1-3 months for 15 (20.27%), to 3-5 months for 7 (9.45%), to 6-12 months for 19 (25.67%), to 1-5 years for 10 or (13.51%), to more than 5 years for 2 (2.70%). Participants reported a range in the number of times they had experienced homelessness, 36.48% (.=27) stating this was the first time, 20.27% (.=15) stating it was their second time, 12.16% (.=9) stating it was their third time, and 31.08% (.=23) indicating they had experienced it 4 or more times. They were more positive in the description of their living conditions, with a combined 62.16% describing them as good (.=23 (31.08%)), very good (.=9 (12.16%)), or excellent (.=14 (18.91%)). For those describing the living conditions as fair (.=11 (14.86%)) or poor (.=16 (21.62%)), the main reasons had to do with food (.=9 (12.16%)) and personnel (.=5 (6.75%)).

Social Functioning

We administered the SF-12 Health Survey and used 11 of the 12 items due to an error: one question was omitted. Thus, we are reporting percentages, mean, and standard deviation on the items we asked. The survey refers to home, and we substituted with shelter to make it more relevant.

The level of perception that participants are not limited by their general health was high (59.4% or 44 participants), as was reporting that physical health and emotional problems had not caused them to accomplish less. Although pain was a factor, 35 (48%) said pain did not interfere at all, and another 9 (12.3%) stated it only interfered a bit. Emotionally, except for feeling sad and blue—51 (69%) dealt with these at least some of the time—they felt calm, peaceful and had energy. Finally, participants feel positively about their health, 23 (31%) describing it as good, 6 (8.1%) as very good, and 11 (14.8%) as excellent. The mean score was 6.671975 and standard deviation 9.7962.

Quality of Life

We learned about life satisfaction using the SWLS, which explores the following: (1) In most ways, my life is ideal; (2) the conditions of my life are excellent; (3) I am satisfied with my life; (4) so far, I have gotten the important things I want in life; (5) if I could live my life over, I would change almost nothing. Responses revealed that 51 (68.91%) have some level of satisfaction with their life, with 22 (29.72%) slightly satisfied, 18 (24.32%) satisfied, and 11 (14.86%) extremely satisfied. Conversely, 20 participants (27.03%) indicated a range of dissatisfaction with their life, with 9 (12.16%) slightly dissatisfied, 5 (6.75%) dissatisfied, and 6 (8.10%) extremely dissatisfied.

Access to Physical and Mental Health Care

We learned that 57 (77%) participants had some access to care in the past year, with 47 (63.51%) in the past three months. When asked about when they had last accessed a health care provider, 16 (21.62%) did not answer. Mental health conditions were prominent, with almost one-third (24 or 32.43%) indicating they had one or more, the primary conditions being depression, anxiety, PTSD, bi-polar disorder, and schizophrenia. A little over one-third (27 or 36.49%) were receiving services and an equal number were not. Though asked why they were not receiving services, most participants did not respond. Of the answers provided, not having insurance, not experiencing symptoms, and negative experiences with mental health professionals were cited.

Chronic Disease and Cancer Risk Factors

While 45 participants (60%) understood what cancer is—an abnormal growth of cells—29 (39.1%) did not. Thirty-one (41.8%) participants revealed a family history of cancer. Family cancers named were leukemia, stomach, breast, lung, pancreatic, lymphoma, prostate, ovarian, colon, kidney, thyroid, shoulder, heart, and brain. Thirty (40.5%) had been screened for cancer, 42 (56.7%) had not. The cancers screened for most frequently were breast, colon, cervical and lung. Eight (10.8%) of the 74 stated they had been diagnosed with some form of cancer.

Fifty-five participants (74.3%) reported at least one health condition, the most frequent ones being high blood pressure (29.7%, .=22), mental health diagnosis (25.6%, .=19), visual impairment (22.9%, .=17), depression (22.9%, .=17), arthritis (18.9%, .=14), high cholesterol (18.9%, .=14), and chronic pain (17.5%, .=13). Although 38 (51.3%) participants indicated they were receiving services for these conditions, 26 (35.1%) reported they were not. The shelter clinic (Federally Qualified Health Center) was the medical provider for 22 (29.7%) participants; private doctors 3 (4%); private and public hospitals 4 (5.4%); and other source of care 1 (1.3%). They reported they were not at all satisfied with substance abuse treatment (.=56, (75.6%)); dental care, physical and occupational therapy (.=55 (74.3%)); mental health, transportation, prevention (.=45 (60.8%)); medical (.=42 (56.7%)); and nutritional support (n=41 (55.4%)). However, 15 (21%) reported services as excellent in transportation, mental health, nutritional support, preventive, and medical care. Thirty-eight (51%) participants have a history of tobacco use, while 36 (49%) do not, and 37 (50%) reported having a history of drinking alcohol.

Participants perceived themselves at low cancer risk in areas of unprotected sex 61 (82.4%), multiple sexual partners 59 (79.7%), drug use 59 (79.7%), workplace risk 58 (78.3%), obesity 54 (72.9%), and family cancer history 49 (66.2%); for moderate risk in unhealthy eating 20 (27%), sun exposure 19 (25.6%), and pesticides 16 (21.6%); and high risk for tobacco use 19 (25.6%).

The top services participants received in the past 60 days were food 58 (78.3%), transportation 29 (39.1%), social services 23 (31%), healthcare 22 (29.7%), and mental health services 21 (28.3%).

Hope and Gratitude

Hope and gratitude emerged as the salient themes when asked what participants’ wishes were. We heard them say that despite being unhoused, they have hope, specifically, hope to be with family again, to return to health, and to have a home to live in. A female participant in her early 20s stated,

“To get my daughter back and be with my mom again.”

An elder male participant answered,

“To have my own apartment, to be good friends with my wife, for me to be healthy.”

And a female in her late 50s responded,

“Wow, this is something to think about. Well, to be closer to my family to my brothers and sisters and my children. I’d like to have a house where they all could stay.”

The gratitude participants expressed was toward the community of El Paso (Texas) for the help they have received. A female in her early 60s acknowledged,

“Out of all the cities I have lived, El Paso cares the most about homeless people.”

Another female in her late 60’s pointed out,

“I’m very grateful to all the people, as much to the people who are part of groups as to those who are in their homes, because there are many people involved in all of this, and I thank them.”

Recommendations

We asked participants about recommendations related to health and mental health. They responded that certain things or services would make their lives better and that they wish society had a different understanding and perspective on individuals experiencing homelessness. A participant in his early 40s cited health insurance as important:

“I think it is very important to have medical insurance. It is not fair not to have insurance, so you die. We are all human beings.”

Another participant (late 40’s) identified a free community fair Health, Opportunity, Prevention and Education (HOPE) which canceled because of COVID-19, as something valuable to her:

“We used to do the HOPE fair, and it was really good. It’s bad that we can’t do it anymore because of COVID.”

Several participants identified having someone with whom to talk:

We all need someone to talk to, especially [a] . . . trained counselor. To go to counseling for mental health, it is just good to be able to speak to someone. It makes a homeless person feel that they’re not out there all alone. (Female, late 50s)

Another participant identified the desire to have help with communication among each other:

Workshops for everyone on how to communicate… often there is little communication. In fact, even more so for the people in shelters, because when people are in this situation, they feel separated from everyone else and need to be included. (Male, mid 20s)

The participants had messages for the public. One participant in her late 50s indicated the need for the public to have a greater understanding of homelessness and its causes:

Mental healthcare… is inadequate. I see there is a lot of depression and schizophrenia, and there is not much support and understanding of what causes homelessness. If they get adequate prescriptions for the right drugs, they get the right emotional support and psychological things, then maybe you can make a dent into the mental healthcare aspect of homelessness.

Several participants do not like how they are perceived by the public; they wish people would acknowledge their humanity:

I like people to understand that we are human beings. Choices have brought us to face homelessness . . . Some people do not have a choice, and so I just like for people to know that we are people, human beings. We are viable. I wish they could put themselves in our shoes and understand that we have family, that we’re not bad people. You know . . . it’s a stigma that goes along with it. We are sons, daughters, brothers, sisters, mothers, and fathers. . . we are responsible people. The majority are fighting and trying very hard. A lot of things can get in our way, and I believe one is the stigma that we have. (Male, early 50s)

Another participant stated:

I would like you to know that we come from all different walks of life, and we’re human beings. It doesn’t matter who lives at 1600 Pennsylvania Avenue or in the alley. We’re all human and we need health care, basic needs, food, and housing. (Female, mid 50s)

4. DISCUSSION

“The Grand Challenge of Ending Homelessness” working paper (January 2015), calls for ending homelessness, asserting that achieving this is possible and that an interdisciplinary, multi-system approach is needed where social workers at micro and macro venues are integral to bringing about this change. Social work’s values of service, social justice, and the dignity and worth of the person (National Association of Social Workers, 2021, preamble) speak directly to the challenge and call on the profession to engage earnestly and work vigorously to do so.

What does ending homelessness have to do with the study findings? Health is a determinant of homelessness (National Healthcare for the Homeless Council, 2019), and because of this, we believe social workers’ greater understanding of the realm of health—mental, physical, and systems of care—will be indispensable in their efforts to improve health conditions and end homelessness. High self-reporting of hypertension, depression, high cholesterol, and visual impairments was consistent with the research. Also highlighted by the research (Aldridge et al., 2019; Parsell et al., 2018), many un-homed adults were not accessing systems of care. This could be due to purposeful avoidance of healthcare due to negative experiences with it (Omerov, 2019).

The low self-perception of cancer risk was concerning based on reported risk factors like alcohol and tobacco use, cancer family history, and cancer diagnosis among many. This may be an area where intervention could be successful, as studies reveal that unsheltered adults report a desire to change health risk factors (Taylor et al., 2016) and are receptive to accessible and affordable cancer screenings (Asgary et al., 2015; Wilde et al., 2013).

Some of our findings are different from what the literature reports: most of the un-homed individuals were connected to a system of care—though it should be noted, a significant number were not using these services—and the majority perceived their health to be in the range of good to excellent. The SF-12 scores were, for the most part, high; QoL scores report some level of life satisfaction; and despite adversities faced, they displayed a hopeful outlook and exhibited gratitude. These last two findings bear looking at. One might expect that most adults who are unhoused would be very dissatisfied with their lives. Yet this is not the case with this study’s participants and may suggest something worth studying further.

Clearly, there must be factors in their lives that are causing them to give a positive assessment of their lives. Flike & Aronowitz’s (2021) systemic mixed studies review highlights that housing and living conditions, age, sex, health—both mental and physical—alcohol and drug use, personal mind-set, and the health system all affect QOL in some negative or positive way. Although this study of adults who are without a home recorded the presence of these factors in their lives, its original purpose was not to examine their correlation or association. However, now seeing this surprising finding, and knowing from the research that QoL affects the health of individuals experiencing homelessness (Flike & Aronowitz, 2021), which, in turn has a relationship to homelessness (Perri et al., 2020), we see QoL as an important area to return to and study. Additionally, the hope participants displayed bears looking at more closely. It suggests an area of potential psychological strength that individuals experiencing homelessness may have that could be leveraged to improve their QoL (Munoz et al., 2016) and perhaps exit homelessness. Interestingly, there is not much research on the positive mindsets—i.e., feeling hopeful or grateful--of older adults who are un-homed. Whereas, Rew et al. (2019) point out, there is growing research on such strengths, such as having hope and gratitude, in youth who are without a home.

5. LIMITATIONS

Findings of this study have limitations. We used a convenience sample; therefore, findings are not generalizable to other populations experiencing homelessness. Possible biases may have been introduced due to lack of privacy when being interviewed. Neither the SF-12 Health Survey nor the QoL scale are culturally specific for Hispanics, who made up 69% of the sample. Survey length may have caused fatigue, discouraging elaboration on open-ended questions.

6. CONCLUSION

Findings have important implications. First, QoL and Life Satisfaction, as potential areas for social work interventions among those who are unhoused, represent areas for further research. Second, the positive mind sets displayed in the hope and gratitude of the unhoused adults are areas of strength for social workers to research and leverage in efforts to increase the QoL of unhoused adults and help them address their comprehensive needs. Third, research on why, when free services are available to treat their chronic conditions, participants are not accessing them, is needed. Not only would accessing care contribute to health, but, as research shows, it can contribute to improving QoL (Flike & Aronowitz, 2021), which in turn may exert a positive influence on health (Munoz et al., 2016). Fourth, our findings on cancer risk, chronic conditions, and risk factors point to areas where education, outreach, and screenings are needed, areas where social workers in their roles as educators, advocates, and change makers can strive to make these things happen for and with the unhoused. Fifth, additional research is needed to inform service planning and policy that will benefit unhoused individuals. Sixth, the findings of chronic conditions and health risks among the unhoused in El Paso (Texas) are not unique to the unhoused in the United States; these are findings with international equivalents (Vallesi et al., 2021) that will need attention and mitigation wherever they are found if their relationship to homelessness is to be reduced and someday eliminated. We call on social workers everywhere to be integral agents in the inter- and intra-professional, cross-sector, and even cross-border, collaboration needed to address the complex issues of health and homelessness, and ultimately bring greater health to vulnerable individuals while moving closer to eliminating homelessness.

Acknowledgments

We wish to acknowledge and thank our partners, the Opportunity Center for the Homeless and Endeavors; our funder, the New Mexico State University Fred Hutch Partnership for the Advancement of Cancer Research U54; all participants; shelter and living center staff; Dr. Amy Wagler for providing data analysis and report preparation; and all other individuals who helped make this research study possible.

7. REFERENCES

Aldridge, R. W., Menezes, D., Lewer, D., Cornes, M., Evans, H., Blackburn, R. M., Byng, R., Clark, M., Denaxas, S., Fuller, J., Hewett, N., Kilmister, A., Luchenski, S., Manthorpe, J., McKee, M., Neale, J., Story, A., Tinelli, M., Whiteford, M., Hayward, A. C. (2019). Causes of death among homeless people: A population-based cross-sectional study of linked hospitalization and mortality data in England. Welcome Open Research, 4(49), 1-17. https://doi.org/10.12688/wellcomeopenres.15151.1

Artiga, S., & Hinton, E. (2018, May 10). Beyond health care: The role of social determinants in promoting health and health equity. KFF. https://www.kff.org/racial-equity-and-health-policy/issue-brief/beyond-health-care-the-role-of-social-determinants-in-promoting-health-and-health-equity/.

Asgary, R., Sckell, B., Alcabes, A. et al. (2015). Perspectives of cancer and cancer screening among homeless adults of New York City shelter-based clinics: A qualitative approach. Cancer Causes Control, 26, 1429–1438. https://doi.org/10.1007/s10552-015-0634-0

Babbie, E. R. (1990). Survey Research Methods. Wadsworth Publishing Co., Belmont.

Baggett, T. P., Chang, Y., Porneala, B. C., Bharel, M., Singer, D. E., & Rigotti, N. A. (2015). Disparities in cancer incidence, stage, and mortality at Boston Health Care for the Homeless Program. American Journal of Preventive Medicine, 49(5), 694–702. https://doi.org/10.1016/j.amepre. 2015.03.038

Baxter, A. J., Tweed, E. J., Katikireddi, S. V., & Thomson, H. (2019). Effects of housing first approaches on health and well-being of adults who are homeless or at risk of homelessness: Systematic review and meta-analysis of randomised controlled trials. Journal of Epidemiology and Community Health, 73(5), 379–387. https://doi.org/10.1136/jech-2018-210981

Biedrzycki, B. (2018). Homeless with cancer: An unrecognized problem in the United States. Clinical Journal of Oncology Nursing, 22(6), 676–679. https://doi.org/10.1188/18.cjon.676-679

Brown, R. I., & Schippers, A. (2018). The background and development of quality of life and family quality of life: Applying research, policy, and practice to individual and family living. International Journal of Child, Youth and Family Studies, 9(4), 1–11. https://doi.org/10.18357/ijcyfs94201818637

Browne, K. (2007). Snowball sampling: Using social networks to research non‐heterosexual women. International Journal of Social Research Methodology, 8(1), 47–60. https://doi.org/10.1080/1364557032000081663

Busch-Geertsema, V., Culhane, D., & Fitzpatrick, S. (2016). Developing a global framework for conceptualising and measuring homelessness. Habitat International, 55, 124-132. https://doi.org/10.1016/j.habitatint.