People Experiencing Homelessness in Miami: Insurance Enrollment Rates, Hypertension Trends

ABSTRACT

Objectives: People experiencing homelessness (PEH) have significant health needs and face significant obstacles when accessing health care. Hypertension can be a rough predictor of health care access; increased levels may reflect barriers to care. It is unknown whether problems stem from a lack of health insurance, as PEH enrollment data are scant, particularly in non–Medicaid expansion states. To begin to address existing literature gaps, we surveyed insurance status and collected blood pressure (BP) measurements of PEH presenting to a free clinic in Miami-Dade County, Florida.

Study Design: We present trends from retrospective cohort data that include insurance status information and BP measurements obtained from 200 PEH receiving care at a free clinic.

Methods: Recruitment encounters took place over 1 year. Participants were approached about the option of using deidentified data for research purposes and informed consent was obtained prior to being included in the study. Deidentified data were collected and stored in a Health Insurance Portability and Accountability Act–compliant REDCap then downloaded and analyzed in R 4.3.2.

Results: A majority (61%) of PEH were insured. Mean (SD) systolic BP of uninsured PEH was 141.0 (23) mm Hg vs 135.4 (22) mm Hg for insured PEH. Independent 2-tailed t test found the mean systolic BP of insured PEH to be significantly lower than that of uninsured PEH (P = .00035). After stratifying the cohort into 10-year age groups (eg, 30-39 years, 40-49 years, etc), we discovered that uninsured participants aged 50 to 59 years had significantly increased systolic and diastolic BP.

Conclusions: Insured PEH had lower systolic BP, despite an older mean age. After age stratification, a difference remained for those aged 50 to 59, representing the largest sample and oldest age group without Medicare. Still, the mean BP of insured PEH was elevated within range of hypertension, indicating that there are other barriers limiting the ability of PEH to access health care and that improvement is unlikely to be achieved solely by increasing insurance enrollment. Future studies should investigate additional factors, including why insured PEH choose free clinic care.

The American Journal of Accountable Care. 2024;12(2):33-40. https://doi.org/10.37765/ajac.2024.89570

People experiencing homelessness (PEH) are a vulnerable population simultaneously experiencing weighty health care needs and large barriers to receiving health care.^1,2 PEH have an increased disease burden compared with the general population, including higher rates of cardiovascular disease–associated mortality rates.³ Specifically, research has shown that PEH have increased rates of hypertension, a chronic condition that is asymptomatic at onset but can lead to cardiovascular and cerebrovascular disease if left untreated.^3,4 Hypertension rates can be used as a rough predictor of health care access because primary care is often necessary to monitor and manage asymptomatic hypertension.^3,4 The increased prevalence of hypertension among PEH suggests reduced levels of health care access, and that overcoming health access barriers may improve cardiovascular health outcomes among PEH.

Although many news outlets and politicians report insurance as the main roadblock to PEH obtaining necessary medical care,^5-8 data describing current rates of health insurance enrollment in PEH populations are scant.^8,9 Therefore, it is unclear whether health care issues such as hypertension stem from a lack of insurance among PEH. Studies that have been conducted on youth¹⁰ and military veterans¹¹ experiencing homelessness have shown that insurance access did improve health outcomes of the participants. These studies also found that as health insurance enrollment rates increased among participants, so did primary care physician visits, whereas emergency department visit and return rates decreased. We also know that health access differs from state to state, with 22.8% of marketplace enrollees in non–Medicaid expansion states reporting problems paying family medical bills compared with 14.5% in expansion states.¹² Herein, we begin to fill some of the existing information gaps by reporting insurance status information for PEH presenting to a free clinic in Miami-Dade County in Florida, a non–Medicaid expansion state. We investigated the extent of insurance coverage and compared blood pressure (BP) measurements, which we used as a proxy for general health status, between PEH with and without health insurance coverage. In doing so, we aimed to examine the role that health insurance coverage plays for PEH accessing primary care and explore subsequent health implications.

METHODS

Study Site

Our research was conducted in partnership with a locally based community 501(c) nonprofit organization registered with the Florida Department of Health that aims to provide free medical care and case management to PEH living in Miami-Dade County. The organization works to coordinate housing and health care for patients struggling to access both. The clinic is located in the geographical area surrounding Jackson Memorial Hospital, known as the Miami Medical District.

Study Population

We surveyed the insurance status of a sample of 200 PEH presenting at a free clinic. Patient information, including demographic details and insurance status, was obtained by physicians and medical students via clinical encounters conducted from February 2022 to February 2023. On initial contact with each patient, and at every subsequent follow-up visit, consent was obtained both for medical care and clinical management as well as for research purposes. Patients were informed that refusal to share their information for research did not preclude them from receiving medical care. During clinical encounters with patients who consented to research participation, study-relevant information was collected and transferred to a Health Insurance Portability and Accountability Act–compliant REDCap then deidentified prior to downloading for research analysis. The clinic prides itself on building rapport with its patients and has very few instances of research participant refusal. For those who did refuse, we do not know the reason for their refusal as they were not asked to complete a refusal questionnaire out of respect for patient boundaries and bandwidth.

Insurance Information

The study site is a free medical clinic that does not accept patient insurance; however, insurance information is routinely documented to aid in patient navigation should a patient need a referral. Participants who reported having insurance were asked to provide a card or other proof of insurance (eg, personal and group numbers). A research coordinator attended the clinic to recruit participants and collect study-relevant information. The sample participants were those who presented to the clinic on the days when a research coordinator was present to recruit and obtain informed consent. The study coordinator then worked with the clinic staff to download deidentified insurance information so it could be transferred to and analyzed by the research team.

Hypertension

Systolic and diastolic BP was measured by trained staff using an Omron HEM-907XL BP monitor. This study adhered to the 2017 American College of Cardiology/American Heart Association BP measurement protocol and guidelines for hypertension classification.¹³ As such, staff instructed participants to rest in an upright seated position with both feet flat on the floor during measurement, and although participants may have had elevated BP—defined as at least 130 mm Hg systolic and at least 80 mm Hg diastolic—suggestive of hypertension, participants were classified as having “measured hypertension” rather than “diagnosed hypertension.” We took this approach because diagnosed hypertension criteria require 2 separate elevated BP recordings from 2 different visits. Here, we only had access to a 1-time BP measurement.

Statistical Analysis

Data were analyzed using R 4.3.2 (R Foundation for Statistical Computing). We stratified patient demographic factors (eg, age, race), as well as systolic and diastolic BP measurements, by insurance status. For demographic variables, we calculated relative risk (RR), 95% CI, and P value. To maintain consistency, we used the same categories for race and ethnicity as are in our clinic survey tool, in which patients can select Hispanic, non-Hispanic African American, non-Hispanic White, or other/not recorded. Ordinary least squares regression models were used to characterize the associations of age, race/ethnicity, gender, and insurance status with systolic and diastolic BP readings.

The models were as follows:

In these models, β0 is the intercept term; β1, β2, β3, β4, and β5 are coefficients associated with the respective predictor variables (race, sex, language, age, insurance); and 𝜖 is the error term. The reported coefficients of the models represent the combined estimates of the outcome (BP readings) after adjustment for all β variables specified above.

Ethics Approval

This observational study was completed as part of ongoing research by University of Miami faculty and students comprising the clinic staff and research team to understand the impact of housing on health care. Research protocols for participant recruitment and informed consent comply with regulations outlined by the Florida Department of Health¹⁴ and were approved by the Human Services Office of Human Research Protections as well as University of Miami Institutional Review Board (#20230265).

RESULTS

The results below were based on our sample of 200 deidentified PEH. Demographic details can be found in Table 1. The mean age of the sample was 54.4 years, with the largest number of patients aged 50 to 59 years. Seventy-two percent of the patients surveyed self-identified as male and 28% self-identified as female. No patients surveyed identified as nonbinary or other. Twenty-nine percent of patients reported being of Hispanic ethnicity, with 20% of patients overall preferring to be interviewed in Spanish. Forty-three percent of patients were African American and 12% were non-Hispanic White. Of the PEH surveyed, 61% had some form of insurance; 39% were uninsured. Of the 61% of PEH with insurance, the most common insurance types were Medicaid (28%), Medicare (11%), Jackson-11 cards (6%), Simply Insurance (6%), Florida Blue (5%), Molina (5%), and Sunshine Health (4%).

We then stratified insurance rates by demographic characteristics to determine whether doing so would present any differences in insurance rates (Table 2). PEH who preferred to have their history taken in Spanish were significantly more likely to be uninsured (P < .05). All the patients who preferred their history be taken in Spanish also identified as Hispanic, which made further investigation on the effects of ethnicity vs preferred language impossible here (90% of the uninsured Hispanic patients listed Spanish as their preferred language, so the sample size was almost identical). PEH reporting non-Hispanic White ethnicity who stated their preferred language was English were less likely to be uninsured (ie, they were more likely to be enrolled in some form of insurance). PEH aged 20 to 29 years were significantly more likely to be uninsured (RR, 5.47; 95% CI, 1.2-25.7; P = .031); however, the sample size for this population was very small.

Detailed results from our linear regression analysis with ordinary least squares assessment modeling systolic and diastolic BP outcomes of the cohort can be found in Table 3. For systolic BP, the model demonstrated a significant association between mean BP and participant gender (β = 18.53; P = .012), indicating that male participants had decreased systolic BP compared with female participants. The other predictors, including race/ethnicity, language preference, age, and insurance status, did not reach significance. Overall, the model fit for the systolic BP outcomes, with an R² of 0.29 with residual SE of 17.59, indicated the factors assessed had limited impact on the variability of systolic BP in the cohort. The diastolic BP model similarly found none of the evaluated predictors to have a significant impact on cohort mean diastolic BP, except gender (β = –8.70; P = .0348), with male participants showing decreased mean diastolic BP. The overall model demonstrated a fit with an R² of 0.31 with a residual SE of 10.03.

DISCUSSION

Our study presents several factors that may impact insurance status in a cohort of PEH. Spanish language speakers were at a significantly increased risk of being uninsured, and English language fluency was one of the largest preventive factors against being uninsured. Because 90% of Hispanic PEH included in the study preferred to have their appointments in Spanish, it is difficult to determine whether increased risk is a product of lack of sociocultural outreach or simply a language barrier. It could be due to lack of Spanish-language outreach and registration services compared with English-language resources or perhaps because of the undocumented immigrant population in South Florida,¹⁵ although this study did not examine immigration status. More research should be done on why the Hispanic and Spanish-speaking PEH population is at higher risk of being uninsured compared with the English-speaking and non-Hispanic PEH population because this can inform better outreach efforts.

The study found that among PEH, those with insurance did have better controlled systolic BP than PEH without insurance. The mean measured BP among insured participants was significantly lower than the uninsured participants’ measured mean BP, even though the mean age of insured participants was older, which increases risk of elevated BP. When stratifying the data by age, there were significant differences in both systolic and diastolic BP between insured and uninsured participants aged 50 to 59 years. Age is widely known to be associated with hypertension, and 54.5% of the general population in the US receives a diagnosis of hypertension in their 50s.⁴ We found significant differences in systolic and diastolic BP based on insurance status among participants in the age group when most people are diagnosed with hypertension, which is also the age group immediately before the participants would qualify for Medicare-funded health services.^4,16 The lower levels of systolic BP among insured PEH compared with uninsured PEH may indicate that the insured group indeed has better access to BP control medications or BP monitoring, which can be extrapolated to possible reductions in barriers to chronic disease management and care.

That being said, the insured PEH group had an overall mean systolic BP of 135.4 mm Hg, which is still elevated and within the range of hypertension,¹³ meaning that having insurance did not completely ameliorate the risk of hypertension. We can speculate about other barriers to accessing traditional health care among PEH participants given the fact that our analysis with multivariable linear regression with adjustment and ordinary least squares assessment did not demonstrate a significant difference in the systolic and diastolic BP of PEH with insurance furthers our belief that insurance alone does not impact access to primary care for BP management for PEH. The most surprising finding of the study was that a large percentage of the PEH surveyed did have health insurance coverage. Our results indicate that the majority of the patients (61%) who chose to receive their care at a free clinic indeed had health insurance. Based on current perceptions and literature, an operating assumption of public health officials and policy makers has been that not having insurance poses one of the most significant barriers to health care access for unhoused populations.^5-8 Our findings challenge these beliefs.

Although our results offer valuable insights by providing a more nuanced perspective on insurance enrollment trends of PEH, it is crucial to discuss the limitations inherent in a small observational study such as ours. First, it is important to recognize that there are countless variables impacting PEH insurance status and the process by which PEH may access health care, complicating the ability to effectively determine the causality of any one factor on any of the outcomes reported. For example, one study surveyed 134 PEH presenting to an emergency department in California (a state with Medicaid expansion) and found that 26% had never heard of the Affordable Care Act and 70% of the PEH who were uninsured were not aware that they could qualify for Medicaid.¹⁷ The study also found that PEH surveyed were significantly less likely to have access to any form of communication (eg, phone, computer, email address) compared with housed participants, thereby inhibiting study follow-up and insurance enrollment efforts.¹⁸ Second, behavioral economic theorists have conducted studies on the impact of cognitive load on poverty, which may play a role in limiting PEH from completing the forms necessary for insurance enrollment and possibly raising their mean BP rates due to chronic stress.¹⁸ These illustrate only but a few of the complex sociosituational factors that we did not assess in our study. There are also limitations to our study design stemming from the nature of the collected data, which were taken in a busy clinic setting from patients who often have few documented health records and whose self-reported history can sometimes be unreliable. Unfortunately, we do not know the reasons for research participation refusal because there was no refusal questionnaire included in the study design in order to not place additional burdens on a PEH community.

Our final limitation is also a strength of the study. Indeed, it is one of the study incentives: We cannot assess the generalizability of our findings, as there is very little information published about insurance trends in PEH populations. We cannot assess whether the results are representative of the PEH population in Florida, or even in Miami-Dade County, because this study only begins to fill a critical data shortage for this underresearched and vulnerable community.^19-21 In filling the notable gap in existing literature about health insurance trends and barriers among PEH, we hope to be able to inform public health research, advocacy, and policy at the city, county, and state levels seeking to address health care barriers and needs of PEH. In addition, our findings contribute to establishing the baseline data necessary for research comparing BP trends of PEH vs housed populations. That being said, a clear limitation impacting the generalizability of findings is that the study sample includes only PEH living in the Miami Medical District within Miami-Dade County. The patients who regularly interact with this free clinic are thus patients who live in a region that is heavily populated by outreach clinics, social services, and shelters, which may self-select for patients who are already connected to other services compared with other PEH less integrated into community resources. Future studies should continue to report PEH insurance enrollment rates and hypertension trends in Miami-Dade County, in Florida, and in different geographic regions of the US to better understand the role of insurance status in PEH health outcomes and health care access.

Although the PEH surveyed in our sample represent only a small percentage of the PEH population at large, results from similar studies conducted in different regions further our confidence in the validity of our findings. These studies have found similar, or even higher, rates of insurance enrollment among PEH. A study on insurance rates of older adult PEH at a shelter in Delaware found that 90% of their cohort had some form of insurance.²² Another study of PEH who were recently admitted to a supportive housing program in Oregon found that 59% of the individuals had Medicaid insurance on arrival.²³ These studies, conducted with PEH living in Medicaid expansion states, further indicate that, for PEH, becoming insured represents only part of the problem and that the barriers to accessing health care for PEH populations are likely multifactorial. Another study reporting outcomes from the same clinic found only 15.9% of their recruited participants had ever been diagnosed with clinical hypertension or previously prescribed BP control medication.²⁴ Comparative analysis of the PEH cohort vs the general population, represented by data collected in the CDC’s National Health and Nutrition Examination Survey, determined that the PEH were at increased risk for undiagnosed hypertension (RR, 4.4880; 95% CI, 4.055-5.873; P < .001) and untreated hypertension (RR, 1.66; 95% CI, 1.470-1.905; P < .0001).²⁴ The markedly elevated RRs were reported from the same clinic as our study and we found that 61% of patients at the clinic are insure.Together, these outcomes further suggest that health care access barriers do still exist for PEH with insurance.

An additional important barrier to consider is the relative cost of health care, even with insurance, for individuals living at the poverty line.²⁵ Although insurance makes medical care more affordable, it does not necessarily make health care affordable and certainly does not make medical care free. For the patients who frequent the free clinic study site, even seemingly low co-pay costs can be prohibitive. They may be faced with the choice of using their insurance and $20 to get medical care and not eating that day vs using that $20 to buy their necessities. Cost-effective strategies to improve the health care outcomes of PEH could entail addressing other barriers to individuals seeking and receiving health care treatment besides health insurance. Previous studies have shown barriers to transportation to be a factor in PEH missing medical appointments and inability to receive regular medical care.^26,27 Communication barriers, in the form of language preference, were shown in our study to be a significant barrier to obtaining insurance, and they likely would continue to be a barrier to care even after insurance is obtained. This barrier is not only for PEH.^28,29 Results of the 2022 American Community Survey found that 18% of Hispanic adults were uninsured compared with 7% of non-Hispanic White adults and 9% of the general population.³⁰ In addition, many PEH do not have easy access to a phone or internet, which makes it difficult to schedule medical appointments or keep track of where and when they are.¹⁷

It is also interesting that even after insurance has been obtained, many patients choose to receive their care at a free clinic as opposed to a more traditional care setting. This could be due to the community ties the clinic has established with the local population. Findings also may be reflective of trust. There are 2 trust relationships here to talk about. One is the lack of trust that exists both from PEH toward the traditional hospital system and often from clinicians of a traditional hospital system toward the needs of PEH.³¹ There is indeed a prevailing trust barrier that exists between PEH and the medical system. The other important aspect of the conversation is the gain of trust that community clinics often develop with PEH communities.³² A qualitative study on the health needs and preferences of PEH found that free community clinics have increased impact because they can function as centers of the PEH community.³³ In the study, the PEH preferred to go there because they felt less judged and believed their needs were addressed in a more sensitive manner.³³ Future research directions should include investigating specifically why the insured patients prefer to get their care from the free clinic and what aspects of their clinic could be implemented in a traditional health care setting, as this may reduce barriers to entry. This could be done through simple surveying. This survey could also include questions on patient satisfaction with their insurance and whether they would switch their coverage if given the option.

CONCLUSIONS

PEH are a high-risk population with many unmet health care needs.^1,2 Access to health care for this community is often debated by policy makers, their reason being that insured PEH have improved health care access, resulting in better BP control, compared with uninsured PEH.^34,35 Although this perspective was partially reflected in the BP trends found in the PEH cohort, overall, the prevailing assumption that insurance is the main barrier to traditional health care access cannot be supported by our data. Merely being insured in Florida, and particularly with Florida Medicaid, was not sufficient for our patients to seek health care in a traditional clinical setting. This presents a refutation of the implicit positive connection between having insurance and seeking traditional health care. In other words, if public health and policy officials aim to have the minimum level of health care needs met for unhoused populations, expanding insurance coverage simply by providing more PEH with the insurance types that currently exist will not be sufficient in reaching that goal. More research is needed to gain a better understanding of barriers to care because this, in turn, can help inform health care reform that will better address the needs of this high-risk, high-need population.

Acknowledgments

The authors would like to thank the nonprofit Dade County Street Response, Miami Street Medicine for its patient-centered research and care and continued advocacy for this underserved population. They would also like to thank the patients of Dade County Street Response, Miami Street Medicine for trusting the authors to research and advocate for the improvement of their health.

Author Affiliations: Department of Internal Medicine, University of Miami Miller School of Medicine (AM, OM, AH, SP, JL), Miami, FL; Department of Emergency Medicine, Jackson Memorial Hospital Systems (AM, OM, JV, MS, SP), Miami, FL; Department of Philosophy, University of Oklahoma (VV), Norman, OK; Saint Anselm College Center for Ethics and Society (VV), Manchester, NH; Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai (GM), New York, NY.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information:Concept and design (AM, OM, VV, GM, JL); acquisition of data (AM, OM, AH, JL); analysis and interpretation of data (AM, OM, VV, GM , JV, MS, SP); drafting of the manuscript (AM, OM, VV, GM, JL); critical revision of the manuscript for important intellectual content (AM, OM, VV, AH, GM, JV, MS, SP, JL); statistical analysis (OM, SP); provision of study materials or patients (AH, JL); administrative, technical, or logistic support (AM, OM, AH, JL); and supervision (AH, GM, JV, MS, JL).

Send Correspondence to: Joshua Laban, MD, University of Miami Miller School of Medicine, 1600 NW 10th Ave #1140, Miami, FL 33136. Email: jlaban@med.miami.edu.

REFERENCES

1. Baggett TP, O’Connell JJ, Singer DE, Rigotti NA. The unmet health care needs of homeless adults: a national study. Am J Public Health. 2010;100(7):1326-1333. doi:10.2105/ajph.2009.180109

2. Schanzer B, Dominguez B, Shrout PE, Caton CLM. Homelessness, health status, and health care use. Am J Public Health. 2007;97(3):464-469. doi:10.2105/AJPH.2005.076190

3. Morrison DS. Homelessness as an independent risk factor for mortality: results from a retrospective cohort study. Int J Epidemiol. 2009;38(3):877-883. doi:10.1093/ije/dyp160

4. Ostchega Y, Fryar CD, Nwankwo T, Nguyen DT. Hypertension prevalence among adults aged 18 and over: United States, 2017–2018. National Center for Health Statistics data brief 364. CDC. April 2020. Accessed December 15, 2023. https://www.cdc.gov/nchs/products/databriefs/db364.htm

5. Hayden N, Bharath D. For the homeless, medical insurance often falls by the wayside. CapRadio. September 25, 2019. Accessed April 6, 2023. https://www.capradio.org/articles/2019/09/25/for-the-homeless-medical-insurance-often-falls-by-the-wayside/

6. Jain SH. Homelessness is a healthcare issue. Why don’t we treat it as one? Forbes. November 9, 2022. Accessed April 6, 2023. https://www.forbes.com/sites/sachinjain/2021/04/17/homelessness-is-a-healthcare-issue-why-dont-we-treat-it-as-one/?sh=5695cb5077e3

7. Venkatesh AK, Chou SC, Li SX, et al. Association between insurance status and access to hospital care in emergency department disposition. JAMA Intern Med. 2019;179(5):686-693. doi:10.1001/jamainternmed.2019.0037

8. Liu M, Hwang SW. Health care for homeless people. Nat Rev Dis Primers. 2021;7(1):5. doi:10.1038/s41572-020-00241-2

9. Cha AE, Cohen RA. Reasons for being uninsured among adults aged 18-64 in the United States, 2019. National Center for Health Statistics data brief 382. CDC. September 2020. Accessed March 27, 2023. https://www.cdc.gov/nchs/products/databriefs/db382.htm

10. Terry M, Bedı G, Patel N. Healthcare needs of homeless youth in the United States. J Pediatr Sci. 2010;2(1). https://bit.ly/3VsZjlC

11. Tsai J, Doran KM, Rosenheck RA. When health insurance is not a factor: national comparison of homeless and nonhomeless US veterans who use Veterans Affairs emergency departments. Am J Public Health. 2013;103(suppl 2):S225-S231. doi:10.2105/AJPH.2013.301307

12. Zewde N, Wimer C. Antipoverty impact of Medicaid growing with state expansions over time. Health Aff (Millwood). 2019;38(1):132-138. doi:10.1377/hlthaff.2018.05155

13. Jacobs AK, Kushner FG, Ettinger SM, et al. ACCF/AHA clinical practice guideline methodology summit report: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2013;127(2):268-310. doi:10.1161/CIR.0b013e31827e8e5f

14. Department of Health. Institutional oversight of assurance. Florida Health. Accessed May 19, 2024. https://www.floridahealth.gov/provider-and-partner-resources/research/irb/PoliciesCombination.pdf

15. Beck TL, Le TK, Henry-Okafor Q, Shah MK. Medical care for undocumented immigrants: national and international issues. Physician Assist Clin. 2019;4(1):33-45. doi:10.1016/j.cpha.2018.08.002

16. Who’s eligible for Medicare? HHS. Updated December 8, 2022. Accessed December 20, 2023. https://www.hhs.gov/answers/medicare-and-medicaid/who-is-eligible-for-medicare/index.html

17. Fryling LR, Mazanec P, Rodriguez RM. Barriers to homeless persons acquiring health insurance through the Affordable Care Act. J Emerg Med. 2015;49(5):755-762.e2. doi:10.1016/j.jemermed.2015.06.005

18. Bertrand M, Mullainathan S, Shafir E. A behavioral-
economics view of poverty. Am Econ Rev. 2004;94(2):419-423. doi:10.1257/0002828041302019

19. Council on Homelessness: Annual Report. Florida Council on Homelessness; 2017. Accessed December 16, 2023. https://www.myflfamilies.com/sites/default/files/2022-11/Council%20on%20Homelessness%20Annual%20Report%202018.pdf

20. Shinn G. The Cost of Long-Term Homelessness in Central Florida: The Current Crisis and the Economic Impact of Providing Sustainable Housing Solutions. Central Florida Commission on Homelessness; 2014. Accessed December 15, 2023. https://shnny.org/uploads/Florida-Homelessness-Report-2014.pdf

21. Smith SM, McAuliffe K, Hall JM, et al. Hypertension in Florida: data from the OneFlorida Clinical Data Research Network. Prev Chron Dis. 2018:15:E27. doi:10.5888/pcd15.170332

22. Ye VM, Caplan RJ, Consiglio-Ward L, Ellison JM. Health care needs of homeless older adults: examining the needs of a senior center cohort. Dela J Public Health. 2019;5(5):74-80. doi:10.32481/djph.2019.12.021

23. Wright BJ, Vartanian KB, Li HF, Royal N, Matson JK. Formerly homeless people had lower overall health care expenditures after moving into supportive housing. Health Aff (Millwood). 2016;35(1):20-27. doi:10.1377/hlthaff.2015.0393

24. Seshadri S, Morgan O, Moore A, et al. Analysis of older adult blood pressure readings and hypertension treatment rates among the unsheltered population of Miami-Dade County. Aging Med (Milton). 2023;6(4):320-327. doi:10.1002/agm2.12272

25. McMorrow S, Kenney GM, Long SK, Gates JA. Marketplaces helped drive coverage gains in 2015; affordability problems remained. Health Aff (Millwood). 2016;35(10):1810-1815. doi:10.1377/hlthaff.2016.0941

26. Syed ST, Gerber BS, Sharp LK. Traveling towards disease: transportation barriers to health care access. J Community Health. 2013;38(5):976-993. doi:10.1007/s10900-013-9681-1

27. Luo J, Kibriya MG, Zakin P, et al. Urban spatial accessibility of primary care and hypertension control and awareness on Chicago’s South Side: a study from the COMPASS cohort. Circ Cardiovasc Qual Outcomes. 2022;15(9):e008845. doi:10.1161/CIRCOUTCOMES.121.008845

28. Tolbert J, Drake P, Damico A. Key facts about the uninsured population. KFF. December 18, 2023. Accessed May 19, 2024. https://www.kff.org/uninsured/issue-brief/key-facts-about-the-uninsured-population/

29. Rhoades JA. The uninsured in America, 2004: estimates for the U.S. civilian non-institutionalized population under age 65. Agency for Healthcare Research and Quality. June 2005. Accessed December 20, 2023. https://meps.ahrq.gov/data_files/publications/st83/stat83.pdf

30. American Community Survey tables for health insurance coverage. US Census Bureau. Accessed May 19, 2024. https://www.census.gov/data/tables/time-series/demo/health-
insurance/acs-hi.html

31. van den Berk-Clark C, McGuire J. Trust in health care providers: factors predicting trust among homeless veterans over time. J Health Care Poor Underserved. 2014;25(3):1278-1290. doi:10.1353/hpu.2014.0115

32. Wen CK, Hudak PL, Hwang SW. Homeless people’s perceptions of welcomeness and unwelcomeness in healthcare encounters. J Gen Intern Med. 2007;22(7):1011-1017. doi:10.1007/s11606-007-0183-7

33. Thorndike AL, Yetman HE, Thorndike AN, Jeffrys M, Rowe M. Unmet health needs and barriers to health care among people experiencing homelessness in San Francisco’s Mission District: a qualitative study. BMC Public Health. 2022;22(1):1071. doi:10.1186/s12889-022-13499-w

34. Wadhera RK, Choi E, Shen C, Yeh RW, Joynt Maddox KE. Trends, causes, and outcomes of hospitalizations for homeless individuals. Med Care. 2019;57(1):21-27. doi:10.1097/mlr.0000000000001015

35. DiPietro B. Five ways Medicaid expansion is helping homeless populations ten years after the ACA became law. Health Affairs Forefront. February 27, 2020. doi:10.1377/forefront.20200225.434660