Physical and Behavioral Health Characteristics of Aging Homeless Women in the United States: An Integrative Review

Abstract

Background:

The average age of the homeless population is and will continue to rise. Although women comprise a significant and growing percentage of this vulnerable population, their age- and sex-specific health characteristics are poorly understood.

Materials and Methods:

This integrative review appraises published research addressing the physical and behavioral health characteristics of aging homeless women (≥50 years) in the United States (2000–2019). The authors searched six electronic databases to identify eligible studies. Studies were screened for methodological quality by using the Johns Hopkins Nursing Evidence-Based Practice model. The review is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Results:

Ten primary studies met the review eligibility criteria. All were level III (non-experimental); nine appraised as “good” quality (level B), and one as “lower” quality (level C). Aging homeless women demonstrate elevated rates of physical health conditions, related to suboptimal nutrition, lower than expected preventive health screening uptake, and geriatric concerns. Disproportionate rates of mental health conditions are compounded by substance use and interpersonal trauma. Familial and social dynamics and socioeconomic disadvantage contribute to social health concerns. Spiritual health is a critically important yet underexplored protective factor.

Conclusions:

Studies are limited, though collective findings suggest that aging homeless women endure a disproportionate physical, behavioral, and social health burden compared with aging non-homeless women and aging homeless men. Implications for research on early aging, preventative health strategies, and homelessness among women, and clinical practice in the context of geriatric and women's health are described.

Introduction

The average age of the homeless population is increasing and will continue to rise.^1

–8 Over the past three decades, the median age of a homeless person in the United States has risen from the late 20s to about 50 years of age.^9,10 Older adults in the United States are at higher risk for homelessness than at any point in recent history, and by 2050, homelessness is projected to double nationally among people aged 65 and older.¹¹ Aging people are especially vulnerable to the negative health consequences of homelessness.¹² Due to early physiological aging,^13,14 homeless people as young as 50 years of age experience health conditions at rates on par with non-homeless older people chronologically 20 years older.¹⁴ It is projected that the aging of the homeless population will result in an estimated $5 billion annually in health care and shelter costs.⁷

Although increasing research has focused on the aging homeless populations, representation of women in most studies ranges from 4%¹² to <25%.^15,16 This is problematic, as more women are becoming homeless for the first time in mid- to late life.^6,17
–19 During midlife and older age,²⁰ women are often socially and economically vulnerable, relating to relationship breakdown,¹⁷ abusive relationships, illness, and/or employment issues.²¹ However, little is known about the unique age- and sex-specific health characteristics and needs of aging homeless women. By synthesizing studies focusing on physical and behavioral health characteristics of aging homeless women, unique health and health care needs can be identified, and population-specific research and clinical priorities established.

This integrative review evaluated the question: What are the physical and behavioral health characteristics of homeless women aged 50 and older in the United States? Findings from this review identify the state and quality of the science on this topic, implications of existing research findings, and areas requiring further research.

Materials and Methods

Article eligibility criteria

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed.²² Eligible studies included those focused on actively homeless cisgender females with a mean/median sample age ≥50 years; were conducted in the United States; were published in English in the year 2000 or later; and assessed physical and/or behavioral health characteristics. Diverse methodological approaches were eligible. Studies including non-homeless females or homeless males were eligible when used as a comparator population, including separate analyses of aging homeless women's health. Ineligible studies included those sampling at-risk for or formerly homeless, incarcerated or otherwise institutionalized populations; and/or explored topics not directly pertaining to physical or behavioral health characteristics (e.g., health care utilization). Non-empirical and non-peer reviewed studies (e.g., commentaries, protocols) were ineligible.

Conceptual definitions

For the literature search, physical health refers to the presence or absence of physiologic conditions, injury, or disabilities of the body that are separate from mental health conditions. Behavioral health focuses on emotional and psychological well-being, including mental health and substance use disorders.²³ The U.S. Department of Housing and Urban Development (HUD) definition of homelessness was used: “individuals who lack resources and support networks to obtain permanent housing.”²⁴

Information sources and search strategy

Table 1 reports results of a search strategy that was created and executed by a medical librarian (LLP) in six databases: (1) Medline (via Ovid); (2) Embase; (3) Ovid Nursing Database; (4) Cochrane Library (via Ovid); (5) Cumulative Index to Nursing and Allied Health Literature (CINAHL) (via EBSCOhost); and (6) Web of Science. The search used controlled vocabulary and free text synonyms for the concepts: homeless, women, women's health; subject headings were used to exclude research focused on pediatric populations or populations outside the United States.

Table 1.

Medline (Ovid) Search Strategy

Search number	Search query
1	Homeless^* or roofless or houseless or shelter or shelters or “street involved” or “street connected” or ((marginally or precarious or insecure or inadequate) and housing)).ti,ab.
2	exp Homeless People/
3	1 or 2
4	(Women^* or woman^* or female^*).ti,ab.
5	exp “Women's Health”/or exp “Women's Health Services”/or exp Women/
6	4 or 5
7	3 and 6
8	Limit 12 to (clinical trial protocol or comment or congress or editorial or letter or meta analysis or news or newspaper article or “review” or “systematic review”)
9	(“literature review” or “systematic review” or “scoping review” or “integrative review”).ti.
10	(exp europe/or exp oceania/or exp asia/or exp africa/or south america/or canada/or “Central America”/or “Caribbean Region”/or “Latin America”/) not exp “united states”/
11	(adolescent/or exp child/) not adult/
12	7 not (8 or 9 or 10 or 11)
13	Limit 12 to year = “2000 to current”

Search Command Key.

= denotes wildcard character.

/ = denotes Medical Subject Heading.

Exp = denotes exploded Medical Subject Heading.

ti,ab. = denotes keyword search in title and abstract.

Search outcome

The database searches yielded 6184 records from each database's inception date through December 17, 2019; citation-index searching²⁵ yielded one additional record (n = 6185). After removing duplicates, 3177 records remained; articles were collated and uploaded into Covidence^®.²⁶

Article review

Figure 1 reports the PRISMA flow diagram, which details the review process. A comprehensive data search (Table 1) accounted for the substantial number of records (n = 3177) initially included in title and abstract screening. Two reviewers with expertise in women's health and vulnerable populations (K.A.D., S.E.L.) independently screened each article's title and abstract against eligibility criteria and found that 3089 records did not meet eligibility criteria. Of the 88 remaining articles eligible for full-text screening, 78 did not meet eligibility criteria for reasons detailed in Figure 1.

FIG. 1.

PRISMA flow diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Assessment of evidence quality

All articles that met eligibility criteria were independently appraised by two reviewers (K.A.D., S.E.L.) using the Johns Hopkins Nursing Evidence-Based Practice (JHNEBP) Research Evidence Appraisal Tool.²⁷ First, the level of evidence (I, II, III) was determined. Level I indicates a randomized controlled trial or experimental study, level II a quasi-experimental study, and level III a non-experimental study. Next, the quality of each article was determined (A, B, C): level A indicates high quality, level B good quality, and level C lower quality evidence.²⁸

Data extraction and synthesis

Key study and sample characteristics were then extracted from each study in preparation for synthesis (Tables 2 and 3).

Table 2.

Study Characteristics

Author(s), year^ref.	Location and setting	Primary health domain(s)	Purpose	Design and approach	Key findings relevant to physical and behavioral health	Evidence level/quality rating
Asgary et al., 2014³⁷	Northwest U.S. New York (Brooklyn and Manhattan), New York; two major shelter-based clinics	Physical health	To evaluate and compare rates and predictors of mammograms in homeless and low-income domicile patients.	Secondary data analysis; retrospective medical record review (September 2010 to December 2012)	Homeless women were less likely to be insured—and more likely to have Medicaid if insured—than low-income domicile women. Overall, participants had a low rate of up-to-date mammogram, considerably lower than national average. “Unknown” was reported as a mammogram result for more than half of homeless women.	IIIB
Davis-Berman, 2011¹⁸	Midwest U.S. (medium-sized city); overnight homeless shelter	Behavioral health	To ask women to tell their stories, talk about their lives in the shelter, and speak about their needs, hopes, and dreams.	Qualitative study; participant interviews, narrative analysis	Aging homeless women expressed: (1) the system is too confusing; (2) chaos in the shelter system; (3) women as nurturers; (4) fractured families. Half of the women reported a history of substance abuse, and half of the women separately reported a history of mental health disorders. Family estrangement was a central theme in the stories of the aging homeless women.	IIIB
Gonyea and Melekis, 2017¹⁷	Northeast U.S., city Nonprofit older homeless organization	Behavioral health	To investigate the intersectionality of gender, age, and homelessness among six urban homeless older women.	Qualitative sub-study; interviews, narrative analysis	Aging homeless women experience stigmatization within the context of multiple marginalized identities, experiencing complex power dynamics related to the links between sexism, ageism, and homelessness. Aging homeless women often adopt society's extant framing of the cause of aging women's homelessness as being largely about irresponsibility, incompetence, substance abuse, and mental illness, rather than the broader social, economic, and political context. Aging homeless women experience two primary pathways to homelessness: (1) a woman who was housed becomes homeless after age 50 (often previously precariously housed); (2) a chronically homeless woman ages.	IIIB
Johnson and McCool, 2003³⁹	Large urban area (region not reported); community crisis center located adjacent to a free community medical center	Physical health	To identify the eating patterns, food sources, and nutritional problems among a limited population of older homeless women located in a large urban area.	Cross-sectional study; interviews, anthropometric measurements	The nutritional intake reported by aging homeless women was sub-optimal. Approximately one-third of aging homeless women were overweight or obese (BMI ≥27). The majority of women were found to have low BMI and mid-arm muscle mass area measurements, indicating low body fat stores and potential muscle wasting.	IIIB
Kenny and Yoder, 2019⁴⁰	Southwest U.S., large metropolitan area, city in the Mountain West; homeless shelter	Behavioral health	To explore homeless women veterans' perceived reasons for homelessness, how they got there, and what they understood they needed to attain and sustain independence.	Qualitative study; interpretive interviews, case study analysis	Several themes across the five participants arose, including age, family upheaval, mental health diagnoses, substance abuse, trauma, and need for information and networking. All women consistently reported abuse as children, and coming from dysfunctional families. Almost all women experienced some type of military sexual trauma, and most women discussed personal struggles with substance abuse.	IIIC
Kisor and Kendal-Wilson, 2002¹⁹	Southwest U.S., Virginia Beach, Richmond and surrounding areas, Virginia; 13 sites, including adult protective services and selected agencies for homeless people	Behavioral health	To develop a profile that reframes stereotypes and allows for fruitful analyses of the plights and need for services of older homeless women.	Secondary data analysis; retrospective chart review (1996–1999)	Mental health problems, low income, family disputes, and abuse or neglect by family members are the primary factors in homelessness. Women tended to be closer to middle age than elderly and experienced one or two compounding situational crisis events. There was an overrepresentation of African American women; Caucasian women were more likely to rent a room, and African American women were more likely to be doubled-up with family members/friends or have lived in an institutional setting before homelessness.	IIIB
Salem and Ma-Pham, 2015³⁸	Western U.S., Central City East (Skid Row expanse), Los Angeles, California; large drop-in day center	Physical health; Behavioral health	To understand perspectives about acute and chronic health needs and challenges, preventive measures, and barriers and facilitators to health care, as well as proposed solutions for intervention among pre-frail and frail homeless women.	Qualitative study; focus groups, demographic questionnaires (May 2013 to June 2013)	Most of the sample walked independently (80%) and did not use an assistive device (65%). More than one third (35%) fell in the past 30 days; 70% fell in the past year. Women discussed health care needs and challenges; perspectives on sexual decision making; employment difficulties; and existing support systems development of future program planning.	IIIB
Washington, 2010³⁵	Western U.S., Los Angeles County, California; VA-affiliated and non-VA affiliated homeless service organization contacts. Outreach sites included four shelters, one drop-in center for homeless people, six transitional housing programs and residential substance abuse rehabilitation programs, the county jail, several soup lines, and direct outreach to people on the streets in Skid Row.	Physical health; behavioral health	To determine risk factors for homelessness in non-institutionalized women veterans, and to contrast their health and health care use with those of a matched sample of housed women veterans.	Matched case-control study; face-to-face interviews, and surveys (December 2005 to January 2006)	Characteristics associated with homelessness included sexual assault during military service, being unemployed, being disabled, have worse overall health, and screening positive for an anxiety disorder or PTSD. Characteristics protective against homelessness include being a college graduate or being married. Programs designed to ameliorate risk factors may prevent a deterioration of women's living circumstances.	IIIB
Washington, 2005³⁴	Midwest U.S., large metropolitan area; 13 institutional settings serving urban homeless populations, including residential homeless shelters (n = 6, 45.0%), warming centers (n = 3, 23.0%), a health clinic for homeless people (n = 1, 8.0%), a community mental health center (n = 1, 8.0%), a food kitchen (n = 1, 8.0%), and a hospital (n = 1, 8.0%).	Physical health; behavioral health	To examine demographic characteristics and self-reported health issues of older African American homeless women to provide a context to understand the uniqueness of this group in society.	Cross-sectional study; semi-structured interviews to administer surveys	Most older homeless African American women were unmarried, homeless more than 3 months, reported more than two diagnoses, and self-rated their health as fair or good. Medical diagnoses increased with time homeless, and rates of some chronic diseases were higher than domiciled African American elders 60 years of age and older. Nearly half (48.5%) of women who were homeless more than 12 months reported emotional/mental illness.	IIIB
Winetrobe et al., 2017³⁶	Western U.S., Los Angeles County, California; 26 housing/social service provider agencies with clients preparing to move into PSH.	Physical health; behavioral health	To understand gender-based health and key social network and support differences among chronically homeless adults transitioning into PSH.	Sub-study within longitudinal study; interviews (baseline, 3, 6, 12 months), survey administration (August 2014 to October 2015)	Compared with men entering PSH, homeless women were younger, less likely to have achieved at least a high school education, and had lower incomes. Women had more chronic physical health conditions, were more likely to have any chronic mental health condition, and had more chronic mental health conditions than men. Women had not only more relatives in their social networks and more relatives who provided support, but also more relatives with whom they had conflict; women were also less likely to have caseworkers or physical and mental health care providers in their networks.	IIIB

BMI, body mass index; PSH, permanent supportive housing; PTSD, post-traumatic stress disorder; VA, veterans affairs.

Table 3.

Study Sample Characteristics

Author(s), year^ref.	Study population	Sample size	Age^a	Race/ethnicity	Duration of homelessness
Asgary et al., 2014³⁷	Homeless and low-income domicile women	N = 100 (44 [44%] homeless women; 56 [56%] low-income domicile women)	Range: 50–74 years (58.7 ± 6.4)	White (8, 18.1%); Black (19, 43.1%); Hispanic (8, 18.1%); Asian (3, 6.8%); Unspecified (6, 13.6%).	Not reported
Davis-Berman, 2011¹⁸	Currently homeless older women; residing in an overnight shelter	N = 10	Range: 52–70 years (58)	Caucasian (7, 70%); African American (3, 30%).	Most participants were recently homeless: eight at shelter for <6 months; two at the shelter for 9 months.
Gonyea and Melekis, 2017¹⁷	Older homeless women	N = 6	Range: 50–64	Caucasian (5, 83.3%); mixed race (1, 16.67%).	The women's shelter stays ranged from 5 to 20 months.
Johnson and McCool, 2003³⁹	Older homeless women residing in a large urban area	N = 52	Range: 50–80 years (56)	Majority Caucasian (# not reported); 2 African American; 2 Hispanic.	Not reported
Kenny and Yoder, 2019⁴⁰	Homeless women veterans	N = 5	≥50 years (55 ± 7)	Unknown/not documented	Not reported
Kisor and Kendal-Wilson, 2002¹⁹	Older homeless women	N = 223	Range: 50–92 (59.25)	African American (94, 42.2%); Caucasian (110, 49.3%); Asian/Pacific Islander (3, 1.3%); American Indian (6, 2.7%); Latina (3, 1.3%); Unknown (7, 3.1%)	Length of service: 0–3 months (216, 96.9%); 4–6 months (6, 2.7%); 7–9 months (1, 0.4%).
Salem and Ma-Pham, 2015³⁸	Pre-frail and frail middle-aged and older homeless women	N = 20	Range: 43–62 years (53.4)	African American (14, 70%); Caucasian/White (3, 15%); mixed race/ethnicity (3, 15%).	Lifetime years homeless: <1 year (4, 20%); 1–3 years (7, 35%); 4–6 (4, 20%); 7–9 years (3, 15%); 10–13 years (1, 5%); 17–19 years (1, 5%).
Washington, 2010³⁵	Women veterans (homeless women veterans; housed women veterans)	N = 198 (33 [16.67%] homeless female veterans; 165 [83.3%] housed female veterans)	49.7 ± 6.9	White (36.4%); Black (54.5%); Hispanic (6.1%); Other (3.0%)	Lifetime years homeless, median (IQR): 2.1 (3.5)
Washington, 2005³⁴	Older African American homeless women	N = 100	Range: 50–74 (52.55 ± 3.57)	African American (100, 100%).	1–3 months homeless (n = 36, 36%); 3–12 months homeless (n = 31, 31%); >12 months homeless (n = 33, 33%).
Winetrobe et al., 2017³⁶	Homeless women and men entering PSH	N = 418 (117 [28%] homeless women; 301 [72%] homeless men)	52.85 ± 7.27	Black (52.99, 62%); White (23.08, 27%); Hispanic/multiracial/other race (23.93, 28%).	Mean lifetime duration of literal homelessness, women = 4.32 ± 5.14, men = 6.70 ± 7.36.

Range, mean age ± SD reported, as available.

IQR, interquartile range; SD, standard deviation.

Results

Study characteristics and quality rating

Eligible studies were published between 2002 and 2019, representing all five major U.S. regions: West, Southwest, Southeast, Midwest, and Northeast (Table 2). Sample sizes ranged from 5 to 223 homeless women ≥50 years old, with the majority of participants being African American/Black and Caucasian/White (Table 3). Six were quantitative studies, and four were qualitative studies. All 10 studies were categorized as level III (non-experimental), with 9 (90%) appraised as “good” (B) quality, and 1 (10%) of “lower” (C) quality (Table 2).

Synthesis of key study findings

After careful review of the 10 articles, an inductive thematic synthesis approach²⁹ was used to generate descriptive themes to categorize the articles. Initial themes included: (1) physical health and (2) behavioral health. After further discussion, the authors identified two additional themes that emerged from the articles: (3) social health, which refers to one's relationships with other individuals, one's community, and/or society, which are influenced by norms, policies, and systems,^30,31 and (4) spiritual health, which refers to one's sense of connection or relationships with an ultimate being or superior existence, with oneself, others, and/or nature.^32,33

Physical health

Findings pertaining to the theme of physical health of aging homeless women were categorized into two domains: (1) general health, (2) chronic physical conditions, and four sub-domains: (a) risk factors for chronic physical conditions; (b) suboptimal nutrition; (c) cancer and cancer screening; (d) geriatric conditions.

General health

General physical health was a critical feature of overall health that was emphasized across articles. General physical health was described as comparatively low among aging homeless women, who experienced lower self-reported physical health scores in comparison with other populations, including non-homeless aging women.^34,35 Aging homeless women were found to be less healthy than aging homeless men, indicated by a greater number of medical diagnoses.³⁶ Having “poor” or “fair” self-rated health was a significant predictor of homelessness among aging women veterans,³⁵ and nearly 70% of aging homeless African American women self-rated their health as “poor” or “fair,”³⁴ reflective of overall poorly self-rated general health.

Chronic physical conditions

Chronic physical conditions of various types were prevalent among aging homeless women. Specifically, across the seven studies that addressed physical health issues, 75%³⁷ to 100%¹⁸ of aging homeless women had at least one documented chronic physical condition diagnosis. Higher rates of chronic physical condition diagnoses among aging homeless women than non-homeless aging people were noted,³⁴ and aging homeless women were more likely to have chronic physical conditions than aging homeless men.³⁶ Only one study detailed chronic physical condition prevalence by condition type, reported by aging homeless African American women, including: hypertension (60.8%); arthritis (50.5%); respiratory illness (44.3%); “ulcers” (32%); “heart trouble” (28.9%); anemia (28.9%); diabetes (12.4%); liver issues (10.3%); and HIV/AIDS (3.1%). In this study, the number of chronic conditions appeared to increase with length of time homelessness.³⁴ Aging homeless women endorsed the presence of physical pain as another chronic condition.³⁸ Although various types of chronic physical conditions were documented as prevalent across studies, details regarding self-perceived management and impact of these conditions were lacking.

Risk factors for chronic physical conditions

Increased risk for chronic physical conditions among aging homeless women was evident. This heighted the risk related primarily to elevated cholesterol and overweight/obesity, which were the primary documented risk factors for chronic physical conditions among aging homeless women highlighted in the included studies. One study found that 44% of 52 aging homeless women had elevated non-fasting total cholesterol levels, and 33% had elevated triglyceride levels.³⁹ In this same study, nearly one-third of the sample (31%) was overweight (15.4%) or obese (15.4%), suggesting increased risk for cardiovascular disease and diabetes.³⁹ Although many were overweight/obese, the vast majority conversely had a very low body mass index, with mid-arm muscle mass area measurements indicating low body fat stores and potential muscle wasting.³⁹ Although the presence of elevated cholesterol and overweight/obesity was documented, there was little emphasis on assessment, management, and impact of other forms or features of chronic physical condition risk in the included studies.

Suboptimal nutrition

For aging homeless women sampled in the included studies, nutritional status was non-optimized, though there was an expressed desire for education and healthier dietary opportunities. Among these women, average daily caloric intake fell below federal daily recommendations (mean = 1460 calories daily), and dietary intake was lower than recommended levels for nearly all nutrients. Data obtained from 24 hours dietary recall ranged from 220 to 3992 calories, and the average reported duration of going without food was 1.5 days.³⁹ Further, women conveyed concerns about the quality and nutritional value of shelter meals.³⁹ Barriers to obtaining nutritious foods were noted, and aging homeless women expressed their desire for education regarding healthy eating on a budget,³⁸ demonstrating an opportunity for optimizing low-cost nutritional options among aging homeless women.

Cancer and cancer screening

The impact of cancer and the value of cancer screening for aging homeless women was evident, though information across articles regarding cancer diagnoses and screening rates was limited. One study reported cancer diagnosis rates, revealing that 9% of 100 aging homeless African American women self-reported a cancer diagnosis.³⁴ In a retrospective evaluation of breast cancer screening rates and related outcomes among 26 aging homeless women with documented mammogram results, more than half of the women's mammogram results were unknown (n = 14), 10 were normal, 1 was abnormal, and 1 signified a breast cancer diagnosis.³⁷ The rate of aging homeless women who were up-to-date with breast cancer screening was lower in comparison to the national average (59% homeless vs. 72.2% non-homeless). Aging homeless women explicitly expressed a need for cancer screening,³⁸ suggesting opportunities for structured clinical approaches to improve cancer detection in this high-risk population.

Geriatric conditions

Geriatric conditions were prevalent among aging homeless women. In the included articles, exploration of geriatric conditions focused primarily on functional status and falls, with expressed need for greater evaluation of other age-associated conditions. In one sample of aging homeless women, although 80% of 20 participants reported ability to walk independently and 65% did not use an assistive walking device, 70% reported having fallen within the past year, and 35% reported having fallen within the past 30 days.³⁸ Data regarding frailty scores, bone mineral density, or fracture incidence specific to aging homeless women were not explicitly reported in any study included in this review. Although no studies reported rates or impact of memory impairment, aging homeless women expressed a need for memory testing.³⁸ Functional status and falls have been demonstrated as crucial geriatric concerns among aging homeless women, whereas evaluation of other age-associated conditions that are pertinent to the healthy aging of this population (e.g., cognitive health) requires further attention.

Behavioral health

Findings pertaining to the theme of behavioral health of aging homeless women were categorized into two domains: (1) mental health conditions, (2) substance use; and one sub-domain: (a) trauma experiences (adverse childhood events, adult abuse, sexual coercion).

Mental health conditions

Mental health issues were described in 8 of the 10 included studies,^{17
–19,34

–37,40} highlighting the importance of this domain in the overall health of aging homeless women. One study reported the prevalence of mental illness history in more than half (56.8%) of the study sample of aging homeless women.³⁷ In another study, nearly one-third of the study sample of 223 aging homeless women attributed mental health difficulties as a major cause of homelessness.¹⁹ When compared with aging homeless men, older homeless women were two and one-half times more likely to have any chronic mental health condition, and they were diagnosed with more mental health conditions.³⁶ Moreover, aging homeless women veterans were more likely to screen positive for an anxiety disorder and/or post-traumatic stress disorder, and had lower self-reported mental health scores than non-homeless women veterans, as measured by mental health items on the Short Form-12 (SF-12) survey.³⁵ Across studies, the pervasive impact of various forms of mental health conditions was apparent and highlighted as a critical feature in the comprehensive understanding of the overall health of aging homeless women.

Substance use

Substance use—in its various forms—was a noteworthy topic of focus across the majority of articles, underscoring the impact of this domain on the health of aging homeless women. Specifically, to differing degrees, 8 of the 10 studies described issues related to various types of substance use.^{17
–19,35,37

–40} In studies that reported alcohol use separate from other substances, rates were relatively low, ranging from 2.2% of 44 participants³⁷ to 18% of 52 participants.³⁹ Aging homeless women veterans were more likely than their non-homeless veteran counterparts to use tobacco³⁵; the only study to report a rate showed that just more than one-third of 44 participants had a tobacco use history.³⁷ In one study, alcohol and marijuana were the primary substances used by aging homeless women, and substance use was influenced by familial substance use, triggered by stressful situations, and used as a coping mechanism to escape from life situations; moreover, most participants expressed an interest in discontinuing substance use.⁴⁰ Mental health and substance use co-occurrence was more common among aging homeless women veterans than non-homeless women veterans,³⁵ with one study revealing a 30% co-occurrence rate.¹⁸ Substance use, particularly when considered in combination with mental health conditions, is a vital consideration necessary to understanding the health and experiences of many aging homeless women.

Trauma experiences

The prevalence and impact of trauma experiences were explored and described in 6 of the 10 included studies,^{17
–19,35,38,40} illustrating the insidious and lasting effects of diverse forms of trauma across the lifespan of many aging homeless women. It was clear that most women experienced some form of trauma at one or multiple points throughout their lives,¹⁸ often in the form of interpersonal violence. In numerous situations, abuse and/or neglect were the factors that women believed meant the difference between remaining precariously housed and becoming homeless. The experience of trauma—which is often repeated and prolonged—in many ways shapes and informs current and future life experiences and overall health of aging homeless women.

Adverse childhood events

For many aging homeless women, the experience of trauma began in early life, with several having endured multiple forms of childhood abuse. Many aging homeless women have survived multiple childhood traumas, including childhood sexual abuse.¹⁷ In one study, all aging homeless women veterans reported childhood abuse that ranged from verbal abuse, to physical violence, to prolonged sexual molestation, including with immediate and distant family relatives. Within this study sample, denial in the family surrounding the abuse was a common theme. For many of these women, their childhoods were unstable, having experienced the death of a parent or sibling, foster care involvement, or being moved from relative to relative throughout childhood. Several described childhood abuse or family upheaval as “normal,”⁴⁰ suggesting the ways in which the experience of childhood or early abuse may impact future health and life experiences.

Adult abuse

Abuse during adulthood represented a major issue related to the trauma experiences of many aging homeless women. Within one study, a majority (60%) of the 10 aging homeless women detailed a history of physical abuse within an adult relationship.¹⁸ In addition, many aging homeless women described the threat of physical violence within the shelter.¹⁸ Further, many felt vulnerable to physical and sexual abuse and, thus, are reluctant to utilize settings such as shelters.¹⁹ As a result of real and perceived threats of violence, many aging homeless women remained hidden to service providers, which may impact health opportunities.

Sexual coercion

Sexual coercion, or sexual abuse, was another documented experience related to trauma experiences of many aging homeless women. Although discussed only minimally, sexual abuse, including sexual coercion, undoubtedly affects the lives of aging homeless women, including histories of “prostitution” endured by aging homeless women.¹⁷ One study conveyed perspectives regarding sexual practices, indicating challenges in negotiating safer sex—specifically, condom use—for fear of “losing” a partner.³⁸ Two studies detailed the prevalence and impact of military sexual trauma, both reporting a majority of the sample having experienced sexual trauma while in the military, ranging from harassment to blatant sexual exploitation.^35,40 Experiencing military sexual trauma was significantly associated with being homeless, and aging homeless women veterans were more likely than their housed counterparts to have endured military sexual trauma.³⁵ No studies provided information regarding sexually transmitted infection rates, despite the demonstrated frequency of adverse sexual experiences, which may influence the accuracy of understanding of the sexual health risk of aging homeless women.

Social health

Findings pertaining to the theme of social health of aging homeless women were categorized into two domains: (1) familial and social dynamics (2) socioeconomic disadvantage.

Aging homeless women constitute a distinctly marginalized population¹⁹ that experiences stigma at the intersection of sexism, ageism, and homelessness.¹⁷ Certain social features uniquely contribute to the increasing number of aging homeless women,¹⁹ related to familial and social dynamics and socioeconomic disadvantage.

Familial and social dynamics

Familial and social networks were of equally critical importance, operating both positively and negatively for aging homeless women. Many aging homeless women reported familial and social support, and a period before homelessness marked by prolonged reliance on external resources.⁴⁰ The role of the “nurturer” or “caregiver” was present across studies, which was found to be the central identity that helped aging homeless women to create a “valued sense of self.”¹⁷ The support of friends and relatives aided many low-income, marginally housed women in avoiding homelessness for a time.¹⁸ Once homeless, social connections with other homeless women operated as a support system. As two studies suggested, these social networks served as a source of mutual education, and information sharing regarding where to obtain meals, find shelter, and which places to avoid.^38,40

Conversely, many aging homeless women described the practice of associational distancing from other women, particularly within the shelter—separating themselves from the notion of being both “older” and “homeless.”¹⁷ Two studies showed how aging homeless women experienced the dysfunctional aspects of familial and social relationships, detailing histories of disputes¹⁹ and dysfunction within family units.⁴⁰ The majority of aging homeless women in their respective samples were single, separated, divorced, or widowed.^{17,19,34,38,40} In one study, 90% of 100 women had children.³⁴ In another study, aging homeless women described how having children made the “biggest impact” on life, despite complicated relationships with these children.¹⁷ Two separate studies shared how the women's tendency to nurture family and romantic partners caused trouble or enabled a level of relational dysfunction that ultimately led to their homelessness.^18,40 Although more aging homeless women than men had more relatives in their social networks who provided support, they also had more relatives with whom they experienced conflict.³⁶ Family fracturing, upheaval, and estrangement¹⁷ from children, grandchildren, siblings, and other family members was common. Relationship problems, or disputes, were cited as a primary reason for homelessness.¹⁸ Arguments with friends or family members, spousal abuse, family violence, and lack of social support accounted for up to 73.5% (164/223) of the major cited reasons for homelessness.¹⁹ Social and familial networks, in many instances, served as a protective factor, but in others promoted remaining in marginal housing situations, thereby compromising housing security and health for aging homeless women.

Socioeconomic disadvantage

Socioeconomic disadvantage was a key topic of consideration in several of the included studies. The limitation of socioeconomic opportunities and its intersection with aging and health played a critical role for many women in both becoming and remaining homeless. Many women were single and living on fixed incomes, with 71% living at or below the poverty level and not well connected with benefits services.¹⁹ Thirty-five percent of the 223 aging women in this study reported insufficient income as a major reason for homelessness.¹⁹ In comparison to aging homeless men, aging homeless women were less likely to have achieved a high school education, and relatedly, had lower incomes.³⁶ Further, among aging women veterans, being unemployed was one of the strongest predictors of homelessness.³⁵ These persistent gender disparities translate into heightened socioeconomic vulnerability among women.¹⁸ Although older homeless women relayed a desire to obtain employment, many experienced challenges in seeking and maintaining employment, citing physical health and age-related difficulties (e.g., balance/walking) as barriers.³⁸ Aging homeless women were often tasked with navigating complex socioeconomic circumstances, which may pervasively undermine both housing and health.

Spiritual health

Spiritual health was described as a key protective feature among aging homeless women. Findings pertaining to the topic of spiritual health were discussed in 3 of the 10 included studies, though this was not a central focus of any of the included articles. “[H]aving strength” and belief in a “higher power” was described as having critical importance to aging homeless women.³⁸ Many aging homeless women had less trust and faith in people, and more in the Lord,¹⁷ and many aging homeless women emphasized the importance of spiritual beliefs, “looking up,” and believing “He's always with you.”¹⁸ Altogether, these findings suggest the importance of spiritual beliefs as a source of strength for aging homeless women, which may be an underexplored protective factor that may facilitate resilience and support health in this population.

Discussion

The number of aging homeless women is steadily increasing. Synthesized evidence from this integrative review suggests that aging homeless women tend to fare worse than similarly aged non-homeless women. Aging homeless women experience a heavy burden of physical health conditions, complicated by elevated risk for progression associated with suboptimal nutrition, lower than expected cancer screening participation, and age-associated issues. In comparison to aging homeless men and non-homeless aging women, aging homeless women show drastically higher rates of mental health conditions, often complicated by the presence of substance use disorders. Early and often-sustained physical and emotional trauma, compounded by impaired social support, impacts the overall health of aging homeless women.

As women age in the United States, the number of low-income women increases, owing to many factors, including employment and health issues. Indeed, women are most at-risk for homelessness in their mid-50s, and there has been a dramatic increase in the number of older homeless women seeking shelter at night.¹⁷ However, research throughout the past decade has primarily included and focused on women who are younger than 50 years of age and homeless. Many of these studies have demonstrated a noteworthy host of health concerns that are present during the midlife phase, which may be foretelling of the health of the growing population of aging women who are at-risk for homelessness and older homeless women. Physical health concerns among later midlife homeless women include an extremely high seroprevalence of herpes simplex virus 2⁴¹ and trichomonas vaginalis,⁴² and high rates of bodily pain contributing to activity limitations; behavioral health concerns, including high rates of stimulant use^43,44 and engagement in condomless sex⁴⁵; mental health concerns, including high rates of psychiatric conditions; and social health concerns, including disconcerting rates of emotional, physical, and sexual violence.⁴⁶ With the passage of time and the aging of the homeless population, the samples investigated in these studies have now aged into their 50s. However, as demonstrated in this integrative review, sparse follow-up research has been conducted among aging homeless women (age 50 years and older) to evaluate and address the aforementioned concerns or explore geriatric-specific physical or behavioral issues.

Indeed, results of this review reveal that chronic physical condition rates among aging homeless women are concerningly high, yet few data are available on specific physical condition diagnoses and preventive efforts tailored to the sex- and age-specific needs of this population. Although many contemporary studies focus on aging, there is a dearth of disaggregated sex-specific analyses regarding the impact of health conditions and the social determinants of health on older populations.⁴⁷ This deficit of sex-specific research may lead to service gaps for women and bias provider perceptions,⁴⁸ diagnosis, and treatment patterns, which can ultimately impact treatment outcomes, morbidity, and mortality. Greater focus is needed on the sex-specific manifestations of geriatric conditions in aging homeless women, such as frailty, cognitive impairments, and other functional impairments that progress in older age. Further, although mental health conditions are a prominent feature among aging homeless women, little is known regarding the prevalence, impact, and management of specific mental health conditions. More investigation is needed into the features and treatment of mental health conditions, such as depression and anxiety, with attention to the presence of co-occurring substance use disorders. Many of the studies documented noteworthy histories of multiple forms of trauma and its intersection with socioeconomic disadvantage; however, none represented a longitudinal analysis of the presence and/or impact of the trauma on the lived experiences of aging homeless women. Lastly, an understanding of how the social determinants of health influence physical and behavioral health outcomes of aging homeless women is imperative to leveraging protective factors and identifying effective health promotion strategies.

Implications for future research

Homelessness and housing research

Annually, the HUD point-in-time count estimates the number of homeless people living across the United States, which is believed to underestimate⁴⁹ the true magnitude of homelessness, particularly among women. Many low-income midlife and older women may be marginally housed and not self-define as “homeless.” Marginally housed people are a traditionally difficult-to-study population,⁴⁹ as it is challenging to count people who are provisionally sheltered,⁵⁰ or invisible from public spaces and typical homeless service sites. Although studies inside and outside of the United States generally find more homeless men than women,⁵¹ this may reflect gender differences in the numbers of people most visibly homeless, inaccurately accounting for those who are doubled-up, couch surfing, staying in hotels, and staying in unsafe domestic situations.⁵² It is, therefore, necessary to consider and count low-income marginally housed women at high risk for homelessness in future research.

Aging research, early aging and geriatirc conditions

Although the incidence of homelessness among aging adults is expected to continue rising, homeless policies and practices still largely focus on younger populations.⁵³ Owing to several factors, the onset of geriatric conditions² such as dementia occurs much earlier for homeless people. However, little is understood regarding the mechanisms by which this phenomenon occurs. Relatedly, preventive opportunities that could promote extended longevity and quality of life are not well understood. Future aging research should focus on historically underrepresented sub-groups of the aging homeless population, focusing on how to best delay the onset and provide better clinical management of geriatric conditions, including dementia, depression, and frailty. Further, future longitudinal analyses aging homeless women should focus on high-risk physical, behavioral, and social health issues, emphasizing the presence and impact of the trauma.

Mortality research

Though homeless people are on average three to four times more likely to die than non-homeless people,⁵⁴ no studies included in this integrative review addressed the risk for excess death among aging homeless women. In fact, although homelessness is considered an independent risk factor for death attributable to specific causes,⁵⁵ no U.S.-based studies have focused exclusively on mortality among homeless women, but rather they have focused on excess death seen in younger homeless populations. Existing research has demonstrated that leading causes of death among aging homeless women aged 65–84 years are: (1) cancer, (2) heart disease, and (3) diabetes,⁵⁶ which may be preventable and/or reversible if conditions are identified early and women are engaged in care. To equitably promote healthy aging and reduce the increased risk for mortality experienced by homeless women, preventive health opportunities should effectively target specific conditions or causes of death,⁵⁵ early in the aging trajectory. In so doing, the illness burden experienced by aging homeless women with potential disproportionate mortality risk may be addressed. Among aging homeless women, research into the rates and causes of death among homeless women is needed, with a focus on aging homeless women.

Implications for clinical practice

Women's health practice

Homeless women are at increased risk of injury and illness, though they are less likely to obtain needed health care services, including primary and preventative care services.^56

–61 As a result, pre-existing problems that were once more manageable with early intervention are likely to worsen.⁶² Despite a heavy burden of cancer risk factors,⁵⁹ preventive screening uptake for women's health cancers remains suboptimal. Given below-average cancer screening rates and above-average cancer risk factors,⁶³ homeless women are often diagnosed with breast cancer at significantly later stages than non-homeless women.⁶⁴ Implementation of low-barrier, multidisciplinary point-of-care⁶⁵ interventions should be tailored to the needs of aging homeless women.

Geriatric practice and need for trauma-informed care

Aging homeless women disproportionately experience trauma, including high rates of physical and emotional abuse, high-risk sexual behaviors, along with mental health concerns and substance use disorders. Clinical care must be sensitive to age while focusing on trauma symptom recognition and integrating knowledge about trauma into assessments and plans. Guided by the framework of a trauma-informed approach,⁶⁶ employing a shared decision-making approach,⁶⁷ and leveraging clinical decision support tools,⁶⁸ clinicians should be equipped to offer substance use screening with aging homeless populations, and follow positive screens with brief interventions and referrals to treatment.⁶⁹

Limitations of the eligible studies

Despite the important contributions of the 10 level III studies identified in this review, there were a number of significant limitations, underscoring the need for future empirical studies. None of the studies utilized an experimental design. Methods of data collection largely relied on participant self-report, with some studies utilizing retrospective chart reviews. Self-reported data are often subject to social desirability bias and recall bias.⁷⁰ Further, studies that utilized provider/staff report (n = 2)^19,37 did not provide details regarding the sources from where data were obtained, the instruments used for data extraction, or the proportion of participant self-report versus provider report. In addition, some findings were limited by small sample sizes in quantitative and qualitative studies alike, which can affect generalizability and transferability.

Conclusions

Aging homeless women endure a disproportionate physical, behavioral, and social health burden compared with aging non-homeless women and aging homeless men. The paucity of published research in this area brings to light the need for large contemporary cohort evaluations of the age- and sex-specific health risk factors and characteristics of aging homeless women. Future research, clinical initiatives, and policy efforts must work in unison to develop opportunities for healthy aging, protect against the disproportionate impact of novel health threats, and extend quality of life among this socially vulnerable population.

Footnotes

Author Disclosure Statement

No competing financial interests exist for K.A.D., L.L.P., J.F., and S.J.B.. T.P.B. receives royalties from UpToDate for authorship of a topic review on homeless health care. S.E.L. has no competing financial interests for this article. Unrelated to this work, S.E.L. is a nonpaid board member of the nonprofit organization Healing Our Community Collaborative, Inc., Boston, MA, and has received one-time speaker fees from the Association of Nurses in AIDS Care in 2017 and 2019.

Funding Information

No funding was received for this article.

References

Culhane

, Metraux

, Byrne

, Stino

, Bainbridge

. The age structure of contemporary homelessness: Evidence and implications for public policy: Age structure of homelessness. Anal Soc Issues Public Policy, 2013; 13:228–244.

Brown

, Kiely

, Bharel

, Mitchell

. Geriatric syndromes in older homeless adults. J Gen Intern Med, 2012; 27:16–22.

Brown

, Steinman

. Characteristics of emergency department visits by older versus younger homeless adults in the United States. Am J Public Health, 2013; 103:1046–1051.

Hahn

, Kushel

, Bangsberg

, Riley

, Moss

. Brief report: The aging of the homeless population: Fourteen-year trends in San Francisco. J Gen Intern Med, 2006; 21:775–778.

Kushel

. Older homeless adults: Can we do more?. J Gen Intern Med, 2012; 27:5–6.

Proehl

. Social justice, respect, and meaning-making: Keys to working with the homeless elderly population. Health Soc Work, 2007; 32:301–307.

Culhane

, Treglia

, Bryne

, et al. Emerging Crisis of Aged Homelessness: Could Housing Solutions Be Funded by Avoidance of Excess Shelter, Hospital, and Nursing Home Costs? 2019;1–21. Available at: https://www.aisp.upenn.edu/wp-content/uploads/2019/01/Emerging-Crisis-of-Aged-Homelessness-1.pdf Accessed November 27, 2020 .

National Coalition for the Homeless. Homelessness among elderly persons, 2009. Available at: https://www.nationalhomeless.org/factsheets/Elderly.pdf Accessed November 27, 2020 .

Fazel

, Geddes

, Kushel

. The health of homeless people in high-income countries: Descriptive epidemiology, health consequences, and clinical and policy recommendations. Lancet, 2014; 384:1529–1540.

10.

Lee

, Guzman

, Ponath

, Tieu

, Riley

, Kushel

. Residential patterns in older homeless adults: Results of a cluster analysis. Soc Sci Med, 2016; 153:131–140.

11.

Goldberg

, Lang

, Barrington

. How to prevent and end homelessness among older adults. Special Report Justice in Aging 2016:1–10. Available at: https://www.justiceinaging.org/wp-content/uploads/2016/04/Homelessness-Older-Adults.pdf Accessed November 27, 2020 .

12.

Brown

, Barnett

, Frahm

, Schinka

, Schonfeld

, Casey

. Health risk factors and differences in outcomes between younger and older veterans using VA transitional housing. Psychiatr Serv, 2015; 66:33–40.

13.

Gelberg

, Linn

, Mayer-Oakes

. Differences in health status between older and younger homeless adults. J Am Geriatr Soc, 1990; 38:1220–1229.

14.

Brown

, Hemati

, Riley

. Geriatric conditions in a population-based sample of older homeless adults. J. Public Health, 2017; 41:240–249.

15.

Tong

, Tieu

, Lee

, Ponath

, Guzman

, Kushel

. Factors associated with food insecurity among older homeless adults: Results from the HOPE HOME study. J Public Health, 2019; 41:240–249.

16.

Lee

, Mangurian

, Tieu

, Ponath

, Guzman

, Kushel

. Childhood adversities associated with poor adult mental health outcomes in older homeless adults: Results from the HOPE HOME study. Am J Geriatr Psychiatry, 2017; 25:107–117.

17.

Gonyea

, Melekis

. Older homeless women's identity negotiation: Agency, resistance, and the construction of a valued self. Sociol Rev, 2017; 65:67–82.

18.

Davis-Berman

. Older women in the homeless shelter: Personal perspectives and practice ideas. J Women Aging, 2011; 23:360–374.

19.

Kisor

, Kendal-Wilson

. Older homeless women: Reframing the stereotype of the bag lady. Affilia, 2002; 17:354–370.

20.

Kutza

, Keigher

. The elderly “new homeless”: An emerging population at risk. Soc Work, 1991; 36:288–293.

21.

Butler

, Weatherley

. Pathways to homelessness among middle-aged women. Women Polit, 1995; 15:1–22.

22.

Moher

, Shamseer

, Clarke

, et al.; PRISMA-P Group. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev, 2015; 4:1.

23.

Centers for Control and Prevention [CDC]. Mental health, 2018. Available at: https://www.cdc.gov/mentalhealth/index.htm Accessed November 27, 2020 .

24.

United States Department of Housing and Urban Development (HUD). HUD's definition of homelessness: Resources and guidance, 2019. Available at: https://www.hudexchange.info/news/huds-definition-of-homelessness-resources-and-guidance/ Accessed November 27, 2020 .

25.

Conn

, Isaramalai

, Rath

, Jantarakupt

, Wadhawan

, Dash

. Beyond MEDLINE for literature searches. J Nurs Scholarsh, 2003; 35:177–182.

26.

Covidence Systematic Review

Software

, Veritas Health

Innovation

. Melbourne, Australia. Available at: at www.covidence.org

27.

Dang

, Dearholt

. Johns Hopkins Nursing Evidence-Based Practice: Model and Guidelines, 3rd ed. Indianapolis, IN: Sigma Theta Tau International, 2017.

28.

Johns Hopkins Nursing Evidence-Based Practice Appendix

: Evidence Level and Quality

Guide

. Available at: https://www.hopkinsmedicine.org/evidence-based-practice/_docs/appendix_c_evidence_level_quality_guide.pdf Accessed November 27, 2020 .

29.

Thomas

, Harden

. Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Med Res Methodol, 2008; 8:45.

30.

Umberson

, Karas Montez

. Social relationships and health: A flashpoint for health policy. J Health Soc Behav, 2010; 51:S54–S66.

31.

World Health Organization (WHO). Social determinants of health. Available at: https://www.who.int/health-topics/social-determinants-of-health#tab=tab_1 Accessed November 27, 2020 .

32.

Ghaderi

, Tabatabaei

, Nedjat

, Javadi

, Larijani

. Explanatory definition of the concept of spiritual health: A qualitative study in Iran. J Med Ethics Hist Med, 2018; 11:3.

33.

Chirico

. Spiritual well-being in the 21st century: It's time to review the current WHO's health definition?. J Health Soc Sci, 2016; 1:11–16.

34.

Washington

. Identification and characteristics of older homeless African American women. Issues Ment Health Nurs, 2005; 26:117–136.

35.

Washington

, Yano

, McGuire

, Hines

, Lee

, Gelberg

. Risk factors for homelessness among women veterans. J Health Care Poor Underserved, 2010; 21:82–91.

36.

Winetrobe

, Wenzel

, Rhoades

, Henwood

, Rice

, Harris

. Differences in health and social support between homeless men and women entering permanent supportive housing. Womens Health Issues, 2017; 27:286–293.

37.

Asgary

, Garland

, Sckell

. Breast cancer screening among homeless women of New York City shelter-based clinics. Womens Health Issues, 2014; 24:529–534.

38.

Salem

, Ma-Pham

. Understanding health needs and perspectives of middle-aged and older women experiencing homelessness. Public Health Nurs, 2015; 32:634–644.

39.

Johnson

, McCool

. Dietary intake and nutritional status of older adult homeless women: A pilot study. J Nutr Elder, 2003; 23:1–21.

40.

Kenny

, Yoder

. A picture of the older homeless female veteran: A qualitative, case study analysis. Arch Psychiatr Nurs, 2019; 33:400–406.

41.

Kelly

, Cohen

, Grimes

, Philip

, Weiser

, Riley

. High rates of herpes simplex virus type 2 infection in homeless women: informing public health strategies. J. Womens Health, 2016; 25:840–845.

42.

Riley

, Cohen

, Dilworth

, et al. Trichomonas vaginalis infection among homeless and unstably housed adult women living in a resource-rich urban environment. Sex Transm Infect, 2016; 92:305–308.

43.

Riley

, Shumway

, Knight

, Guzman

, Cohen

, Weiser

. Risk factors for stimulant use among homeless and unstably housed adult women. Drug Alcohol Depend, 2015; 153:173–179.

44.

Riley

, Kral

, Cohen

, Dilworth

, Shumway

, Lynch

. Levamisole-contaminated cocaine use in HIV-infected and uninfected unstably housed women. J Womens Health, 2016; 25:936–943.

45.

Meacham

, Bahorik

, Shumway

, Marquez

, Riley

. Condomless sex and psychiatric comorbidity in the context of constrained survival choices: A longitudinal study among homeless and unstably housed women. AIDS Behav, 2019; 23:802–812.

46.

Riley

, Cohen

, Knight

, Decker

, Marson

, Shumway

. Recent violence in a community-based sample of homeless and unstably housed women with high levels of psychiatric comorbidity. Am J Public Health, 2014; 104:1657–1663.

47.

World Health Organization (WHO). Women, ageing and health: A framework for action: focus on gender, 2007. Accessed from: https://apps.who.int/iris/handle/10665/43810 Accessed November 27, 2020 .

48.

Weisman

, Cassard

. Health Consequences of Exclusion or Underrepresentation of Women in Clinical Studies. Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies: Volume 2: Workshop and Commissioned Papers. 1994. DOI:10.17226/2343. Available at: https://www.ncbi.nlm.nih.gov/books/NBK236583/ Accessed November 27, 2020 .

49.

Koh

, O'Connell

. Improving health care for homeless people. JAMA, 2016; 316:2586–2587.

50.

Smith

, Castañeda-Tinoco

. Improving homeless point-in-time counts: Uncovering the marginally housed. Soc Curr, 2019; 6:91–104.

51.

Toro

. Toward an international understanding of homelessness. J Soc Issues, 2007; 63:461–481.

52.

Means

, Rorie

JAL

, Mehta

. The status of homeless women in Massachusetts: Are we adequately addressing the social determinants of their health? A white paper from health care without walls, 2019. Available at: https://www.healthcarewithoutwalls.org/wp-content/uploads/HCWW-Homeless-Women-and-Social-Determinants-of-Health-White-Paper.pdf Accessed November 27, 2020 .

53.

Grenier

, Barken

, Sussman

, Rothwell

, Bourgeois-Guérin

, Lavoie

. A literature review of homelessness and aging: Suggestions for a policy and practice-relevant research agenda. Can J Aging Rev Can Vieil, 2016; 35:28–41.

54.

O'Connell

. Premature mortality in homeless populations: A review of the literature. Nashville: National Health Care for the Homeless Council, Inc., 2005:1–19.

55.

Morrison

. Homelessness as an independent risk factor for mortality: Results from a retrospective cohort study. Int J Epidemiol, 2009; 38:877–883.

56.

Baggett

, Hwang

, O'Connell

. Mortality among homeless adults in Boston: Shifts in causes of death over a 15-year period. JAMA Intern Med, 2013; 173:189–195.

57.

Asgary

, Sckell

, Alcabes

, Naderi

, Ogedegbe

. Perspectives of cancer and cancer screening among homeless adults of New York City shelter-based clinics: A qualitative approach. Cancer Causes Control, 2015; 26:1429–1438.

58.

Asgary

, Naderi

, Wisnivesky

. Opt-out patient navigation to improve breast and cervical cancer screening among homeless women. J Womens Health, 2017; 26:999–1003.

59.

Asgary

. Cancer screening in the homeless population. Lancet Oncol, 2018; 19:e344–e350.

60.

Bharel

, Casey

, Wittenberg

. Disparities in cancer screening: Acceptance of Pap smears among homeless women. J Womens Health, 2009; 18:2011–2016.

61.

Bharel

, Creaven

, Morris

, et al., eds. Health care delivery strategies: Addressing key preventive health measures in homeless health care settings. Nashville: Health Care for the Homeless Clinicians' Network, National Health Care for the Homeless Council, Inc., 2011:1–29. Available at: https://nhchc.org/wp-content/uploads/2020/02/Preventive-Health-Care-Delivery-Startegies.pdf Accessed November 27, 2020 .

62.

Cameron

, Abrahams

, Morgan

, Williamson

, Henry

. From pillar to post: Homeless women's experiences of social care. Health Soc Care Community, 2016; 24:345–352.

63.

Chau

, Chin

, Chang

, et al. Cancer risk behaviors and screening rates among homeless adults in Los Angeles County. Cancer Epidemiol Biomarkers Prev, 2002; 11:431–438.

64.

Baggett

, Chang

, Porneala

, Bharel

, Singer

, Rigotti

. Disparities in cancer incidence, stage, and mortality at Boston Health Care for the Homeless Program. Am J Prev Med, 2015; 49:694–702.

65.

Bharel

, Santiago

, Forgione

, León

, Weinreb

. Eliminating health disparities: Innovative methods to improve cervical cancer screening in a medically underserved population. Am J Public Health, 2015; 105 Suppl 3:S438–S442.

66.

Substance Abuse and Mental Health Services Administration. SAMHSA's concept of trauma and guidance for a trauma-informed approach (HHS Publication No. 14–4884). 2014. Available at: https://ncsacw.samhsa.gov/userfiles/files/SAMHSA_Trauma.pdf Accessed November 27, 2020 .

67.

Friedrichs

, Spies

, Haerter

, Buchholz

. Patient preferences and shared decision making in the treatment of substance use disorders: A systematic review of the literature. PLoS One, 2016; 11:e0145817.

68.

Burdick

, Kessler

. Development and use of a clinical decision support tool for behavioral health screening in primary care clinics. Appl Clin Inform, 2017; 8:412–429.

69.

Spinelli

, Ponath

, Tieu

, Hurstak

, Guzman

, Kushel

. Factors associated with substance use in older homeless adults: Results from the HOPE HOME study. Subst Abuse, 2017; 38:88–94.

70.

Althubaiti

. Information bias in health research: Definition, pitfalls, and adjustment methods. J Multidiscip Healthc, 2016; 9:211–217.