Abstract
Background
People experiencing homelessness (PEH) have disproportionately high rates of early-onset geriatric conditions and decades-early mortality. These phenomena are well documented but not well understood, particularly for younger adult PEH (e.g., 18–39 vs. 40+). This study compared functional limitation prevalences in a sample of adults experiencing homelessness vs. the general adult population in Los Angeles County (LAC).
Methods
Data came from the Periodic Assessment of Trajectories of Housing, Homelessness, and Health Study (PATHS), a longitudinal study of adults experiencing homelessness in LAC; 2019–2023 American Community Survey data defined the adult LAC population (weighted N = 7, 790,124). The PATHS analytic sample comprised participants who responded to functional limitation questions, identified as female or male, and were not missing age and race/ethnicity (N = 633). In this cross-sectional analysis, overall, age-specific, and age- and sex-standardized functional limitation item prevalences were calculated and compared in PATHS vs. LAC.
Results
PATHS participants ages 25–59, men, and non-Hispanic Black participants were overrepresented vs. LAC. Overall and age-specific functional limitations were higher in PATHS vs. LAC (55% vs. 13% reporting any functional limitation, respectively), and age-specific prevalences did not uniformly increase in PATHS. The age- and sex-standardized any functional limitation prevalence was 59%, 6.2 times higher in PATHS vs. LAC.
Discussion
This work adds to literature documenting a substantial functional limitation burden among unhoused populations. Overall and item-specific prevalences were elevated but showed no consistent trend by age in PATHS, and age and sex distribution differences did not explain overall differences between PATHS and LAC.
Introduction
Homelessness in the United States (US) is a public health crisis, with over 771,000 people experiencing homelessness (PEH) and over 274,000 people experiencing unsheltered homelessness as of 2024.1 Evidence shows that PEH have disproportionately high rates of chronic diseases, early-onset geriatric conditions, and decades-early mortality compared to housed populations.2–4 This phenomenon of “accelerated aging” among PEH has been well documented, with aging trajectories often 10–15 years faster than in the general population.5,6 Understanding the specifics of accelerated aging among PEH, which might reflect the extremes of human weathering (i.e., the process by which marginalized populations experience structural adversities, leading to premature aging and mortality7,8), is important to help address morbidity and mortality among PEH.
Previous work has consistently found elevated prevalences of accelerated aging markers among PEH.2,9–13 In a recent meta-analysis, PEH aged 40 years and older averaged 36–50% higher prevalence of limitations to instrumental and basic activities of living (i.e., I/ADLs), over 3 times higher fall risk, and over 2.5 times higher frailty rates compared to housed counterparts.4 Much of the existing work has sampled participants from homeless shelters and/or service-providing facilities, or focused on older age groups. Additional work examining accelerated aging markers in samples of PEH that are representative of the broader unsheltered population, including those not accessing services, is particularly important for understanding and quantifying the burden of accelerated aging in this population. Additionally, researchers believe that premature aging may begin in early adulthood for PEH (e.g., ages ≥ 18 years),9 so it is relevant to assess outcomes in younger age groups.
This study aimed to characterize functional limitations, a marker of disability and strong predictor of future mortality,14 by age in a representative sample of PEH ages 18 years and older in Los Angeles (LA) County using data from the Periodic Assessment of Trajectories of Housing, Homelessness, and Health Study (PATHS). Hypotheses included (1) functional limitation onset would begin at much younger ages, resulting in a high and flatter age pattern in PATHS compared to the expected age trend of exponentially increasing prevalences at older ages in the general LA County adult population, and (2) age-specific functional limitation prevalences would be higher in PATHS vs. the LA County adult population.
Methods
Study samples
PATHS, established in December 2021, is a longitudinal cohort study that aims to understand the health and housing trajectories among unhoused LA County residents through web-based monthly and biannual surveys.15 Participants were recruited in-person through the 2022, 2023, and 2024 LA Homeless Count Demographic Surveys.16 PATHS eligibility criteria included self-reported living in a homeless shelter or unsheltered setting for at least one night in the previous month, age ≥ 18 years old, residence in LA County, and access to a text-enabled mobile phone at enrollment, where the link to prescreen surveys were sent. Complete study protocols have been previously published.17 Study protocols and procedures were approved by the University of California, Los Angeles IRB.
As of May 2024 (the end of 2024 recruitment), PATHS recruited 2,079 eligible participants, among whom 1,798 completed a baseline survey. In August 2023, questions on self-reported functional limitations were added to the baseline and 6-month follow-up surveys. This analysis comprised participants of the 2024 baseline survey (n = 404; collected between December 2023 – May 2024) or at least one 6-month follow-up survey (n = 274; collected between August 2023 – March 2025), for a total of 678 possible participants. The analytic sample was restricted to participants with male or female self-reported sex/gender (excluding “I do not identify as either male or female” and “prefer not to answer” responses; n = 657) and no missing demographic data (n = 633). For participants who responded to functional limitation questions in more than one survey (n = 392), their earliest response was selected. The supplementary analysis additionally required no missing covariate data (see Supplementary Analysis, n = 520). Data for the LA County adult population over the same approximate period came from the pooled 2019–2023 individual-level American Community Survey (ACS) data set (weighted n = 7,790,124).18 Missing responses in ACS are addressed using hot-deck imputation.19
Measures
Self-reported functional limitations, the primary outcomes of interest, were all measured as binary variables (yes/no) and were assessed with the same questions in PATHS and ACS:20
Hearing difficulty: “Are you deaf, or do you have serious difficulty hearing?”
Vision difficulty: “Are you blind, or do you have serious difficulty seeing, even when wearing glasses?”
Cognitive difficulty: “Because of a physical, mental, or emotional condition, do you have serious difficulty concentrating, remembering, or making decisions?”
Ambulatory difficulty: “Do you have serious difficulty walking or climbing stairs?”
Self-care difficulty: “Do you have difficulty dressing or bathing?”
Independent living difficulty: “Because of a physical, mental or emotional condition, do you have difficulty doing errands alone such as visiting a doctor's office or shopping?”
For the primary operationalization of each functional limitation item in PATHS, “prefer not to answer” responses were coded as “no.” This decision was informed by the assumption that “prefer not to answer” responses were non-random, and that this categorization was more conservative for the main analysis (see Statistical Analysis section). “Any functional limitation” (yes/no) was defined as endorsement of any vs. none of the 6 functional limitation items above.
Self-reported sociodemographic characteristics harmonized between PATHS and the ACS sample included categorical age at survey (in years; 18–24, 25–39, 40–49, 50–59, and 60+), sex/gender (male, female), and race and ethnicity (any Hispanic/Latino, Asian American or Pacific Islander (not Hispanic (NH)), Black/African American (NH), Multiracial (NH) or another race, and White (NH)).
Additional sociodemographic and health characteristics reported to describe the PATHS sample included veteran status (yes/no), LGB+ identity (lesbian, gay, bisexual, or another identity vs. heterosexual or straight), Supplemental Security Income/Social Security Disability Insurance status (SSI/SSDI; currently receive, received before but not currently, never received but applied, never received or applied, or I don’t know), number of benefits received (excluding SSI/SSDI; 0, 1, or 2 or more), whether participants ever experienced trauma (yes/no), difficulty finding a job or remaining employed due to a physical, mental, or emotional problem (yes/no), and prevention from working due to a physical, mental, or emotional problem (yes/no). Note that all “prefer not to answer” responses for these and the covariates described in the following paragraph were set to missing.
Self-reported covariates collected to describe the PATHS sample and that were included in multivariate logistic regression models (see Supplementary Analysis) included: housing status at prescreen survey (doubled-up, sheltered, or unsheltered), homelessness history at baseline (<1 year, 1–5 years, 10–19 years, or ≥20 years), ever incarcerated (yes/no), and frequency of sleeping on or next to a freeway at baseline (almost all the time, some of the time, or occasionally/never), any substance use disorder (yes/no), number of mental (continuous, range 0–8) and physical health (continuous, range 0–16) conditions, health insurance status at baseline (insured vs. not insured), and ever received substance use treatment (yes/no). Health condition and government benefit response options are provided in Appendix Table 1.
Statistical analysis
Demographic and functional limitation item descriptive statistics were calculated to compare PATHS vs. the ACS sample weighted up to the LA County adult population using Census survey weights.21 Additionally, sociodemographic, housing, health, and government benefit characteristics stratified by any vs. no functional limitation in the PATHS sample were reported.
Functional limitation burden in PATHS vs. the LA County adult population was assessed by comparing overall and age-specific functional limitation prevalences with 95% confidence intervals (CIs) in PATHS, and survey-weighted prevalences and 95% CIs in the LA County adult population. Additionally, age- and sex-standardized prevalence of any functional limitation and standardized functional limitation ratios (SFLRs) were calculated in PATHS with the LA County adult population as the standard.
Supplementary analysis:
Multivariable logistic regression models were estimated to compare crude and adjusted prevalences of any functional limitation for each age at survey category (18–24, 25–39, 40–49, 50–59, and 60+) in the PATHS analytic sample with complete covariate information (n = 520). Covariate blocks were identified based on prior work examining correlates of accelerated aging markers among PEH5 and available measures in PATHS. Model 0 included age at survey only. Model 1 included age at survey and adjusted for sex/gender and race and ethnicity. Model 2 additionally adjusted for housing status at prescreen, homelessness history at baseline status, ever incarcerated, and frequency sleeping on or next to a freeway. Model 3 additionally adjusted for any substance use disorder, number of mental health conditions, and number of physical health conditions. Model 4 additionally adjusted for health insurance status and ever received substance use treatment. Finally, Model 0 was estimated in ACS using a weighted logistic regression model for comparison. Unadjusted and multivariate-adjusted age-specific functional limitation prevalences and 95% CIs were estimated using the emmeans R package22 for a reference person with a common set of covariate values, including housing and health values deemed “lower risk” (e.g., unhoused <1 year; Appendix Table 2).
Several sensitivity analyses were conducted. First, the main analysis was rerun with all “prefer not to answer” functional limitation responses in PATHS redefined as endorsement of a limitation. Second, sociodemographic, housing, and health characteristics, as well as functional limitation age trajectories were compared in the PATHS analytic sample (n = 633) vs. the sample with complete covariate information (n = 520) to evaluate the potential influence of covariate missingness on supplemental findings. Third, select demographic, housing, and health characteristics were compared in the analytic sample vs. the 2022–2024 LA Homeless Count Demographic Survey to evaluate the representativeness of the PATHS sample. All analyses were conducted in R version 4.3.2.
Results
Compared to the 2019–2023 LA County adult population, the PATHS analytic sample included a greater share of participants aged 25–59 and a lower share of the youngest and oldest age groups (Table 1). A higher proportion of PATHS participants identified as male and Black/African American (NH), and a lower proportion identified as Hispanic/Latino or Asian American or Pacific Islander (NH). Overall, 55% of PATHS participants self-reported any functional limitation vs. 13% of the LA County adult population (Table 1). This trend persisted across functional limitation items, with differences ranging from 9% vs. 3% for hearing difficulty to 38% vs. 5% for cognitive difficulty in PATHS and LA County, respectively.
Table 1.
Sociodemographic characteristics and functional limitations in the PATHS analytic sample and LA County adult population.
| PATHSa N = 633 | LA County adult populationb,c N = 7,790,124 | |
|---|---|---|
|
| ||
| Age at survey, % | ||
| 18–24 | 9 | 11 |
| 25–39 | 39 | 29 |
| 40–49 | 25 | 17 |
| 50–59 | 18 | 17 |
| 60+ | 9 | 26 |
| Female sex/gender, % | 47 | 51 |
| Race and ethnicity, % | ||
| Any Hispanic/Latino | 38 | 45 |
| Asian American or Pacific Islander (NHd) | 3 | 16 |
| Black/African American (NHd) | 26 | 8 |
| Multiracial (NHd) or another race | 5 | 3 |
| White (NHd) | 28 | 27 |
| Hearing difficulty, % | 9 | 3 |
| Vision difficulty, % | 11 | 3 |
| Cognitive difficulty, % | 38 | 5 |
| Ambulatory difficulty, % | 21 | 7 |
| Self-care difficulty, % | 13 | 4 |
| Independent living difficulty, % | 26 | 6 |
| Any functional limitation, % | 55 | 13 |
| Number of endorsed functional limitations (0–6; median (IQR)) | 1 (0 – 2) | 0 (0 – 0) |
PATHS data collection period for this analysis: August 2023-March 2025.
LA County adult population estimates derived from the 2019–2023 5-year American Community Survey.
Weighted percentages presented.
NH = not Hispanic
In the PATHS analytic sample, 87% of participants were unsheltered at their prescreen survey, 56% reported ever being incarcerated, and 82% reported experiencing homelessness ≥1 year at baseline (Table 2). Participants with any vs. no functional limitation had a higher median number of self-reported physical and mental health conditions. Additionally, a higher proportion of participants with any vs. no functional limitations reported a substance use disorder, ever receiving substance use treatment, receiving or applying for SSI/SSDI, and ever experiencing trauma.
Table 2.
Any functional limitation-stratified sociodemographic, housing, and health characteristics among PATHS participants in the analytic sample (N = 633).
| No functional limitation N = 283 | Any functional limitation N = 350 | Overall N = 633 | |
|---|---|---|---|
|
| |||
| Age at survey (years), n (%) | |||
| 18–24 | 30 (11) | 26 (7) | 56 (9) |
| 25–39 | 126 (45) | 122 (35) | 248 (39) |
| 40–49 | 64 (23) | 97 (28) | 161 (25) |
| 50–59 | 43 (15) | 71 (20) | 114 (18) |
| 60+ | 20 (7) | 34 (10) | 54 (9) |
| Female sex/gender, n (%) | 125 (44) | 170 (49) | 295 (47) |
| Race and ethnicity, n (%) | |||
| Any Hispanic/Latino | 105 (37) | 137 (39) | 242 (38) |
| Asian American or Pacific Islander (NHa) | 12 (4) | 6 (2) | 18 (2.8) |
| Black/African American (NHa) | 79 (28) | 83 (24) | 162 (26) |
| Multiracial (NHa) or another race | 15 (5) | 17 (5) | 32 (5) |
| White (NHa) | 72 (25) | 107 (31) | 179 (28) |
| Hearing difficulty, n (%) | - | 54 (15) | 54 (9) |
| Vision difficulty, n (%) | - | 67 (19) | 67 (11) |
| Cognitive difficulty, n (%) | - | 239 (68) | 239 (38) |
| Ambulatory difficulty, n (%) | - | 135 (39) | 135 (21) |
| Self-care difficulty, n (%) | - | 80 (23) | 80 (13) |
| Independent living difficulty, n (%) | - | 167 (48) | 167 (26) |
| Number of endorsed difficulties (0–6), median (IQR) | - | 2 (1–3) | 1 (0–2) |
| Homelessness history at baseline, n (%) | |||
| Less than 1 year | 66 (23) | 49 (14) | 115 (18) |
| 1–4 years | 122 (43) | 148 (43) | 270 (43) |
| 5–9 years | 49 (17) | 94 (27) | 143 (23) |
| 10–19 years | 29 (10) | 39 (11) | 68 (11) |
| 20 years or more | 16 (6) | 18 (5) | 34 (5) |
| Missing (n) | 1 | 2 | 3 |
| Ever incarcerated, n (%) | 133 (49) | 205 (61) | 338 (56) |
| Missing (n) | 11 | 15 | 26 |
| Frequency sleeping on or next to a freeway at baseline, n (%) | |||
| Almost all the time | 42 (16) | 92 (27) | 134 (22) |
| Some of the time | 43 (16) | 62 (18) | 105 (17) |
| Occasionally/Never | 185 (69) | 186 (55) | 371 (61) |
| Missing (n) | 13 | 10 | 23 |
| Housing status prescreen survey, n (%) | |||
| Doubled-up/Housed | 10 (4) | 8 (2) | 18 (3) |
| Sheltered | 25 (9) | 41 (12) | 66 (10) |
| Unsheltered | 248 (88) | 301 (86) | 549 (87) |
| Number of self-reported physical health conditions, median (IQR) | 0 (0–1) | 1 (0–2) | 1 (0–1) |
| Missing (n) | 15 | 15 | 30 |
| Number of self-reported mental health conditions, median (IQR) | 0 (0–1) | 2 (1–3) | 1 (0–2) |
| Missing (n) | 19 | 20 | 39 |
| Any self-reported substance use disorder | 53 (20) | 124 (38) | 177 (30) |
| Missing (n) | 19 | 20 | 39 |
| Self-reported health insurance, n (%) | |||
| Insured | 208 (74) | 272 (78) | 480 (76) |
| Not insured | 53 (19) | 39 (11) | 92 (15) |
| I don’t know | 19 (7) | 37 (11) | 56 (9) |
| Missing (n) | 3 | 2 | 5 |
| Ever received SUD/AUD treatment, n (%) | 54 (19) | 123 (36) | 177 (29) |
| Missing (n) | 6 | 7 | 13 |
| SSDI/SSI receipt status, n (%) | |||
| Currently receive | 18 (7) | 59 (17) | 77 (12) |
| Received before but not currently | 10 (4) | 18 (5) | 28 (5) |
| Never received but applied | 25 (9) | 87 (25) | 112 (18) |
| Never received or applied | 218 (79) | 169 (49) | 387 (63) |
| I don’t know | 6 (2) | 9 (3) | 15 (2) |
| Missing (n) | 6 | 8 | 14 |
| Number of government benefits received (excluding SSI/SSDI), n (%) | |||
| 0 | 69 (25) | 78 (23) | 147 (24) |
| 1 | 128 (47) | 153 (44) | 281 (45) |
| 2 or more | 78 (28) | 114 (33) | 192 (31) |
| Missing (n) | 8 | 5 | 13 |
| Reported ever experiencing trauma, n (%) | 101 (45) | 215 (78) | 316 (64) |
| Missing (n) | 61 | 76 | 137 |
| Veteran status, n (%) | 6 (2) | 12 (4) | 18 (3) |
| Missing (n) | 2 | 6 | 8 |
| LGB+ identity, n (%) | 36 (13) | 64 (19) | 100 (16) |
| Missing (n) | 13 | 8 | 21 |
| Difficulty finding a job or remaining employed due to physical, mental, or emotional problem, n (%) | |||
| Yes | 38 (16) | 202 (69) | 240 (45) |
| No | 184 (75) | 69 (23) | 253 (47) |
| Prefer not to answer | 23 (9) | 23 (8) | 46 (9) |
| Missing (n) | 38 | 56 | 94 |
| Prevented from working due to physical, mental, or emotional problem, n (%) | |||
| Yes | 40 (16) | 176 (60) | 216 (40) |
| No | 185 (76) | 86 (29) | 271 (50) |
| Prefer not to answer | 20 (8) | 32 (11) | 52 (10) |
| Missing (n) | 38 | 56 | 94 |
NH = not Hispanic
Age-specific overall and individual functional limitation prevalence point estimates were consistently higher in PATHS vs. the LA County adult population, including at ages 18–24 and 25–39 years, with the largest differences observed for cognitive difficulty (Figure 1, Appendix Table 3). Prevalences in PATHS generally did not increase by age, whereas expected age gradients were present in the LA County adult population. Additionally, with the exception of ambulatory difficulty, there was some evidence in PATHS that overall and individual functional limitation prevalence point estimates at ages 50–59 years were less than or equal to those at 60+ years, although estimates were imprecise. After age and sex-standardization, the overall functional limitation prevalence was slightly higher than the crude prevalence in PATHS (59% vs. 55%; Appendix Table 4), and the SFLR suggested that any functional limitation prevalence in PATHS was 6.2 times higher vs. LA County after accounting for age and sex.
Figure 1.
Functional limitation prevalence trajectories and 95% confidence intervals by age in the PATHS analytic samplea and LA County adult population.b
aPATHS data collection period for this analysis: August 2023-March 2025.
bLA County adult population estimates derived from the 2019–2023 5-year American Community Survey.
In the supplementary analysis, multivariate-adjusted age-specific any functional limitation prevalence point estimates were lower with each successive covariate set except between Models 0 and 1 (Appendix Figure 1, Appendix Tables 5–6). Adjusted age-specific prevalence point estimates from Model 4 were closest to those observed in the LA County adult population, but non-monotonic age trends persisted across all models, and estimates were imprecise.
Sensitivity analyses recategorizing “prefer not to answer” responses in PATHS and comparing the PATHS analytic vs. supplementary analysis samples yielded similar results to the main analysis (Appendix Figures 2–4, Appendix Table 7). Participants in the PATHS analytic sample skewed younger, a higher proportion identified as White (NH), and higher proportions reported health conditions compared to the 2022–2024 LA Homeless Count Demographic Survey samples (Appendix Table 8).
Discussion
In this study, overall and item-specific functional limitation prevalences were elevated in PATHS, a representative sample of PEH in LA County, relative to the LA County adult population during the same approximate period. These trends persisted across nearly all age categories for each functional limitation item, with the largest difference observed for cognitive difficulty. The age- and sex-standardized functional limitation prevalence was similar to the overall crude prevalence in PATHS and over 6 times higher than the overall prevalence in the adult LA County population. In supplemental analyses, demographic-, housing-, health-, and health care access-adjusted age-specific functional limitation point estimates were closer to crude estimates among LA County adults but imprecise.
The present study’s overall functional limitation prevalence ranged from 55% before standardization to 59% after age- and sex-standardization to the 2019–2023 LA County adult population. Existing studies have used a variety of measures to estimate functional limitation/impairment prevalence among PEH. One study in San Diego estimated 48% of sheltered adults experiencing homelessness met the criteria for functional impairment using the University of California, San Diego, Performance-Based Skills Assessment-Brief scale, which assesses communication and financial skills using role-play activities.23,24 A few studies that defined functional limitation using I/ADLs estimated prevalences between 30–39% for any vs. no ADL impairment and between 49–59% for any vs. no IADL impairment among PEH aged ≥50 years.5,25,26 Any functional limitation prevalence was slightly higher among participants ages ≥50 years in this study at 63%. These differences may be driven by differences in measurement and sample characteristics across studies. Most notably, 87% of participants in the PATHS sample reported living in an unsheltered setting at baseline. It is possible that the higher estimates in this study reflect the unique daily physical and emotional challenges faced by a largely unsheltered population.
Functional limitation prevalences among PATHS respondents did not exponentially or uniformly increase with age, reflecting a pattern of extreme early onset of age-related impairments. To the authors’ knowledge, this is the first study to examine age-specific functional limitation prevalences among unhoused adults aged ≥18 years. Elevated functional limitation prevalences at early ages may reflect some combination of adverse physical and mental effects of homelessness, adverse experiences prior to experiencing homelessness, and disabilities that are not on the aging continuum (e.g. developmental delays, traumatic injuries). Functional limitation patterns observed at older ages likely reflect some combination of selective survival and housing placement among unsheltered individuals. The most recent LA County homeless mortality report estimated an annual mortality rate among unhoused adults ages ≥55 years much higher than that among the general LA County adult population 55+.27,28 Separately, housing prioritization tools such as the Vulnerability Index – Service Prioritization Decision Assistance Tool consider older age (≥60 years) and functional impairment as criteria for placement.29 Many older adults with functional impairment may therefore have recently accessed housing, precluding them from study inclusion. Together, these phenomena suggest that older PEH may appear “healthier” because those with a high functional limitation burden were more likely to have died or been placed into housing, making them unobservable in this study.
Despite the somewhat unique age trajectories of functional limitation prevalence observed in the PATHS sample, most but not all of the age-specific excess burden between PATHS and LA County was explained by controlling for adverse exposures, some of which may both precede and result from homelessness. Baseline housing status explained some of the variation, but this measure is likely a very crude representation of respondents’ housing status trajectories. Apart from the age ≥60 group, adjustment for self-reported health conditions and substance use resulted in lower age-specific functional limitation prevalence estimates but did not completely explain differences between PATHS and the LA County adult population. This finding could reflect a number of phenomena, including underreporting health conditions and crude, cross-sectional health condition measurement in PATHS. Future work leveraging longitudinal measures and biomarkers are needed.
Strengths of this study included evaluation of functional limitation items in a representative sample of PEH in a large, urban area. Additionally, this study leveraged harmonized functional limitation and demographic measures from PATHS and the ACS, allowing for a direct comparison of age-specific functional limitation prevalence estimates.
Limitations
This study also had limitations, including differences in data collection years for PATHS and the ACS. Measures of functional limitations in this study differ from those in other studies, which may limit comparisons. Further, this analysis was limited to individuals identifying as male or female sex/gender due to data availability in the ACS. Thus, findings from this study may not generalize to gender-diverse populations. Additionally, unhoused LA County residents without access to a text-enabled mobile phone could not participate in PATHS. This requirement may limit generalizability to unhoused residents with greater resources and physical and mental functioning necessary to complete web-based surveys. Although select sociodemographic and health characteristics were similar between PATHS and the Demographic Survey samples, there could still be differences in unmeasured characteristics, leading to underestimation of functional limitation prevalences in PATHS. Relatedly, conclusions from this study are limited to LA County.
Conclusions
This study contributes important evidence on the extraordinarily high prevalence of functional limitations, a marker of accelerated aging, in a representative sample of adults experiencing homelessness in a large urban area. Overall functional limitation prevalence was over 6 times higher than in the general population of LA County, with onset at much younger ages. These disadvantages were partially explained but not eliminated by controlling for adverse exposures. Directions for future work include augmentation and validation of self-reported functional limitation measures through provider-based assessments, electronic health records linkage, and biomarker collection. Additionally, longitudinal research is needed to examine drivers of functional limitations trajectories among PEH.
Supplementary Material
Acknowledgements
This does not necessarily represent the views of the United States Government or the Department of Veterans Affairs.
Funding:
This work was supported by the NIH NICHD grant number 5P2CHD041022–24. This work was funded in part by the Conrad N Hilton Foundation grant #32793 “Periodic Assessment of Trajectories of Housing, Homelessness and Health Study.” BJS is supported by Veterans Affairs IK2-CX002648 and National Institute on Aging R03AG082989.
Footnotes
Declaration of Interest: None.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
CRediT Author Statement
Taylor Mobley: Conceptualization, Methodology, Software, Data Curation, Formal Analysis, Writing – Original Draft, Visualization; Jessie Chien: Conceptualization, Data curation, Writing-Review and Editing; Danielle Paulazzo: Data Curation, Writing-Review and Editing; Benjamin Seligman: Conceptualization, Writing-Review and Editing; Benjamin Henwood: Conceptualization, Writing-Review and Editing, Supervision, Funding Acquisition; Randall Kuhn: Conceptualization, Writing-Review and Editing, Supervision, Funding Acquisition
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.de Sousa T, Henry M, Abt Global. The 2024 Annual Homelessness Assessment Report (AHAR) to Congress. 2024. Accessed April 22, 2025. https://www.huduser.gov/portal/sites/default/files/pdf/2024-AHAR-Part-1.pdf
- 2.Mantell R, Hwang YIJ, Radford K, Perkovic S, Cullen P, Withall A. Accelerated aging in people experiencing homelessness: A rapid review of frailty prevalence and determinants. Front Public Health. 2023;11:1086215. doi: 10.3389/FPUBH.2023.1086215/BIBTEX [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Richards J, Kuhn R. Unsheltered Homelessness and Health: A Literature Review. AJPM Focus. 2023;2(1):100043. doi: 10.1016/J.FOCUS.2022.100043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Suh K, Beck J, Katzman W, Allen DD. Homelessness and rates of physical dysfunctions characteristic of premature geriatric syndromes: systematic review and meta-analysis. Physiother Theory Pract. 2022;38(7):858–867. doi: 10.1080/09593985.2020.1809045 [DOI] [PubMed] [Google Scholar]
- 5.Brown RT, Hemati K, Riley ED, et al. Geriatric Conditions in a Population-Based Sample of Older Homeless Adults. Gerontologist. 2017;57(4):757–766. doi: 10.1093/GERONT/GNW011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Funk AM, Neil Greene R, Dill K, Valvassori P. The Impact of Homelessness on Mortality of Individuals Living in the United States: A Systematic Review of the Literature. J Health Care Poor Underserved. 2022;33(1):457–477. doi: 10.1353/HPU.2022.0035 [DOI] [PubMed] [Google Scholar]
- 7.Brown RT, Evans JL, Valle K, Guzman D, Chen YH, Kushel MB. Factors Associated With Mortality Among Homeless Older Adults in California: The HOPE HOME Study. JAMA Intern Med. 2022;182(10):1052–1060. doi: 10.1001/JAMAINTERNMED.2022.3697 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Simons RL, Lei MK, Klopack E, Beach SRH, Gibbons FX, Philibert RA. The effects of social adversity, discrimination, and health risk behaviors on the accelerated aging of African Americans: Further support for the weathering hypothesis. Soc Sci Med. 2021;282:113169. doi: 10.1016/J.SOCSCIMED.2020.113169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cronin T, Healy D, McCarthy N, Smith SM, Travers J. Prevalence and risk factors of frailty in people experiencing homelessness: A systematic review and meta-analysis. J Frailty Aging. 2025;14(2):100029. doi: 10.1016/J.TJFA.2025.100029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in Older People. Lancet. 2013;381(9868):752. doi: 10.1016/S0140-6736(12)62167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Salem BE, Nyamathi A, Phillips LR, Mentes JC, Sarkisian C, Brecht ML. Development of a frailty framework among vulnerable populations. Advances in Nursing Science. 2014;37(1):70–81. doi: 10.1097/ANS.0000000000000013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sullivan G, Dumenci L, Burnam A, Koegel P. Validation of the brief instrumental functioning scale in a homeless population. 2001;52(8):1097–1099. doi: 10.1176/APPI.PS.52.8.1097/ASSET/IMAGES/HF21T1.JPEG [DOI] [PubMed] [Google Scholar]
- 13.Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of Illness in the Aged: The Index of ADL: A Standardized Measure of Biological and Psychosocial Function. JAMA. 1963;185(12):914–919. doi: 10.1001/JAMA.1963.03060120024016 [DOI] [PubMed] [Google Scholar]
- 14.Kritchevsky SB, Forman DE, Callahan KE, et al. Pathways, Contributors, and Correlates of Functional Limitation Across Specialties: Workshop Summary. J Gerontol A Biol Sci Med Sci. 2018;74(4):534. doi: 10.1093/GERONA/GLY093 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kuhn R, Henwood B, Lawton A, Chien J. UNDER THREAT: Surveying Unhoused Angelenos in the Era of Camping Enforcement. 2022. [Google Scholar]
- 16.Henwood B, Hickey S, Kwack S, et al. 2023 Los Angeles Continuum of Care Homeless Count: Methodology Report. 2023. Accessed September 8, 2024. https://www.lahsa.org/documents?id=7699-2023-usc-homeless-count-methodology-report
- 17.Kuhn R, Chien J, Hernandez NG, et al. Periodic Assessment of Trajectories of Housing, Homelessness, and Health Study (PATHS): Protocol for a Prospective Cohort Study of People Experiencing Homelessness. JMIR Res Protoc 2025;14:e74266 https://www.researchprotocols.org/2025/1/e74266. 2025;14(1):e74266. doi: 10.2196/74266 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ruggles S, Flood S, Sobek M, et al. IPUMS USA: Version 15.0 [usa00025]. Preprint posted online 2024. doi: 10.18128/D014.V4.0 [DOI] [Google Scholar]
- 19.Census Bureau U. Chapter 15: Improving Data Quality by Reducing Nonsampling Error. 2022. Accessed March 12, 2026. https://www2.census.gov/programs-surveys/acs/methodology/design_and_methodology/2022/acs_design_methodology_ch15_2022.pdf
- 20.US Census Bureau. How Disability Data Are Collected from The American Community Survey. 2021. Accessed May 12, 2025. https://www.census.gov/topics/health/disability/guidance/data-collection-acs.html
- 21.Census Bureau U. American Community Survey and Puerto Rico Community Survey Design and Methodology (Version 3.0) Chapter 11: Weighting and Estimation. Published online 2022. [Google Scholar]
- 22.Lenth R V ’emmeans: Estimated Marginal Means, Aka Least-Squares Means. 2024. doi: 10.1080/00031305.1980.10483031 [DOI] [Google Scholar]
- 23.Mahmood Z, Vella L, Maye JE, et al. Rates of Cognitive and Functional Impairments Among Sheltered Adults Experiencing Homelessness. https://doi.org/101176/appi.ps202000065. 2021;72(3):333–337. doi: 10.1176/APPI.PS.202000065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mausbach BT, Harvey PD, Goldman SR, Jeste DV., Patterson TL. Development of a Brief Scale of Everyday Functioning in Persons with Serious Mental Illness. Schizophr Bull. 2007;33(6):1364. doi: 10.1093/SCHBUL/SBM014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Brown RT, Kiely DK, Bharel M, Mitchell SL. Geriatric syndromes in older homeless adults. J Gen Intern Med. 2012;27(1):16–22. doi: 10.1007/S11606-011-1848-9/TABLES/2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Brown RT, Kiely DK, Bharel M, Mitchell SL. Factors Associated with Geriatric Syndromes in Older Homeless Adults. J Health Care Poor Underserved. 2013;24(2):456–468. doi: 10.1353/HPU.2013.0077 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.California Vital Data Report. 2025. Accessed September 9, 2025. https://cal-vida.cdph.ca.gov/VSQWEB/ [Google Scholar]
- 28.Ferrer B, Davis M, Mahajan A, Nicholas W, Stroud L. Mortality Rates and Causes of Death Among People Experiencing Homelessness in Los Angeles County: 2017–2023 March 2025. Accessed June 2, 2025. http://publichealth.lacounty.gov/chie [Google Scholar]
- 29.Vulnerability Index-Service Prioritization Decision Assistance Tool (VI-SPDAT) Prescreen Triage Tool for Single Adults. 2014. Accessed September 7, 2025. http://www.orgcode.com/product-category/training/spdat/
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.