Keywords
homelessness - ophthalmology - healthcare disparities - student-run clinics - ophthalmology
education - medical education
Homelessness is one of society's greatest challenges facing nearly 1 million in the
United States each day[1] and its prevalence has increased in communities across the United States.[2] The effects of homelessness pose significant consequences at both the societal and
individual level. At the societal level, homelessness costs the United States economy
about $2 trillion annually.[3] At the individual level, strong associations exist between homelessness and a multitude
of comorbidities such as heart disease, diabetes, hypertension (HTN), depression,
and substance use disorders.[1]
[4]
[5]
[6]
[7] The homeless population faces higher barriers in accessing primary and preventive
care services and as a result has higher rates of acute care utilization and hospitalizations.[3]
[8] Despite homelessness being a widespread problem, there is limited information about
the eye health of this population[9]
[10] and the available resources for proper treatment. The current literature identifies
a higher prevalence of eye disease among subsets of homeless populations located abroad,[11]
[12]
[13] but multiyear comprehensive assessments of ophthalmologic health in this population
within the United States are deficient.[14] This study will be the first to investigate the ocular health of the homeless living
in an urban setting in the United States.
With a rapidly growing homeless population and an anticipated increased need for ophthalmologic
services, the University of California San Francisco Homeless Shelter Eye Clinic (UCSF-HSEC)
was launched on August 23, 2017 at the Multi-Services South Homeless Shelter, and
subsequently the Navigation Center in San Francisco.
The main aim of this study was to examine the prevalence of eye disease, demographics,
and prevalence of comorbidities such as HTN, type 2 diabetes (DM2), and human immunodeficiency
virus/acquired immunodeficiency syndrome (HIV/AIDS) of patients who utilized the UCSF-HSEC.
A secondary aim was to elucidate the role that student-run clinics can play in connecting
the homeless population to healthcare and community services.
Methods
A retrospective cohort study was performed examining patients who received care at
the UCSF-HSEC. All patients seen at the monthly UCSF-HSEC from August 2017 to December
2019 were included. This study was approved by the Institutional Review Board at the
University of California San Francisco. This study complied with the Health Insurance
Portability and Accountability Act and followed the tenets of the Declaration of Helsinki.
At each monthly clinic, UCSF medical student volunteers canvassed all floors of the
homeless shelter to offer comprehensive eye exams to the residents. All exams were
conducted a dedicated clinic room in the homeless shelter. Patient rooms were partitioned
off with portable room dividers and curtains for privacy. For the consenting participants,
medical students obtained demographic information such as age, gender, race, ethnicity,
birthplace, and education level along with a focused medical and ocular history. With
regard to medical history, patients were only asked if they had a history of HTN,
DM, or HIV/AIDS. A comprehensive ocular exam was then performed primarily by a senior
resident with faculty attending oversight that comprised of visual acuity (near and
far with a Snellen chart), extraocular movements, confrontational visual field, pupillary
responses, intraocular pressure (Ton-Pen), autorefraction (Retinomax), and fundoscopy
with an indirect ophthalmoscope. UCSF ophthalmology residents and attendings allowed
medical students to practice their ocular examination skills. Students were provided
with the opportunity to make the initial assessment and residents and faculty ophthalmologist
worked with medical students to revise and refine their assessments and plans to assign
The International Classification of Diseases, 9th Revision (ICD-9) and ICD-10. If
indicated, a patient received health coaching by a medical student on their diagnosis
along with smoking cessation counseling.
Visual impairment was defined as visual acuity of less than or equal to 20/40 but
better than 20/200 in the better seeing eye, while legal blindness was defined as
visual acuity of less than or equal to 20/200 in the better seeing eye.[15] The visual acuity recorded was the corrected visual acuity when the patient had
eyeglasses or contacts, or the uncorrected visual acuity when the patient did not
have correction.
Patients were diagnosed as glaucoma suspects if the patient had at least one of the
following findings: intraocular pressure was greater than 21 mm Hg, evidence of optic
nerve involvement, or a visual field abnormality.[16] A patient was diagnosed with glaucoma in our records if they endorsed receiving
a prior diagnosis of glaucoma by an eye provider that was verified with information
in their medical record at Zuckerberg San Francisco General Hospital (ZSFG).
Patients in need of follow-up care were given prearranged appointments at ZSFG and
community partner, Project Homeless Connect who provided patients with free eyeglasses
if indicated.
Data obtained from direct patient interviews, physical exam findings, and ICD-9/10
codes were all recorded and stored in a secure database.
Patient demographic characteristics and common comorbidities such as HTN and diabetes
were assessed. The prevalence of common diagnoses was stratified by race and ethnicity.
Results
Patient Demographics
A total of 123 patients were evaluated at the UCSF-HSEC. The average age of the patients
was 51 years with a range between 25 and 82 years. Of the 123 patients, 74% identified
as male and 21% identified as female. The gender identity for the remaining 5% was
unknown or the patients preferred not to disclose.
All of the patients stated that they were homeless. The average year of homelessness
was 7 years with a range of 0 to 60 years.
An adapted version of the National Institute of Health classification for race and
ethnicity was utilized.[16] Thirty-three percent of patients were identified as Black/African-America, 27% identified
as White, 11% preferred not to disclose, 8% identified with race/ethnicity categories
other than the provided options, 7% identified as Hispanic/Latino, 6% identified as
Asian, 6% identified with multiple races and ethnicities, 2% identified as American
Indian or Alaska Native, and 2% identified as Native Hawaiian or Pacific Islander
([Fig. 1]).
Fig. 1 Racial and ethnic composition of patients attending the free eye clinic from 2017
to 2019.
The highest education level achieved by participants breaks down as follows: 46% stated
they had a high school diploma, 15% had completed junior high school, 13% had a bachelor's
degree, 11% unknown, 9% completed vocational school, 5% had an associate's degree,
1% completed a master's degree, and 1% stated that they had not completed any formal
schooling ([Fig. 2]).
Fig. 2 Highest level of educational attainment of patients attending the free eye clinic
from 2017 to 2019.
Seventy percent of patients were unemployed, 14% did not disclose their employment
status, 8% stated they were retired, 7% were currently employed, and 1% were on disability
([Fig. 3]).
Fig. 3 The employment status of patients attending the free eye clinic from 2017 to 2019.
Medical History
Sixty-five percent of patients had HTN, 27% had DM2, and 8% had HIV/AIDS ([Fig. 4]). Additionally, out of the 98 (80%) patients who knew their family history, 43%
stated they had a family history of eye disease. The following rates of substance
use were noted: 38% endorsed tobacco use, 26% endorsed marijuana, 21% endorsed alcohol,
and 15% endorsed illicit drug use ([Fig. 5]). Fifty percent of the patients did not have a primary care physician (PCP).
Fig. 4 The prevalence of comorbidities of patients attending the free eye clinic from 2017
to 2019.
Fig. 5 The prevalence of substance use of patients attending the free eye clinic from 2017
to 2019.
Ophthalmologic Findings
Uncorrected near visual acuity (UNVA was collected in 55% of the study participants
(68/123). Sixty-eight percent (46/68) of the patients were classified as visually
impaired by having an UNVA of 20/40 or less, and 9% (6/68) were classified as blind
by having an UNVA of 20/200 or less. The average uncorrected near logarithm of the
minimum angle of resolution (logMAR) was 0.43 with a standard deviation of 0.32.
Corrected near visual acuity (CNVA) was collected in 62% (76/123) of the study participants.
Forty percent (40/76) of the patients were classified as visually impaired by having
a CNVA of 20/40 or less, and 1.3% (6/76) were classified as blind by having a CNVA
of 20/200 or less. The average corrected near logMAR was 0.21 with a standard deviation
of 0.23.
The most prevalent eye disease was refractive error at a prevalence rate of 70% with
presbyopia and myopia accounting for 77 and 42% of the refractive error diagnoses,
respectively.
A visually significant cataract was the second most prevalent pathology, which was
diagnosed in 13% of patients. Diabetic retinopathy was categorized as proliferative
and nonproliferative defined as the appreciation and absence of neovascularization
on fundoscopy respectively.[17] Diabetic retinopathy had a prevalence rate of 11%: 7% had nonproliferative diabetic
retinopathy and 4% had proliferative retinopathy. Cranial nerve and corneal lesions
were appreciated in 17 and 11% of the patients, respectively, while 8% of patients
were found to have normal ocular exams ([Table 1]).
Table 1
Prevalence of ocular disease
|
Diagnosis
|
Total
|
|
Chorioretinal
|
|
|
AMD
|
1
|
|
Chorioretinal scar
|
1
|
|
Choroidal lesion suspicious
|
1
|
|
Choroidal nevus
|
1
|
|
Diabetic macular edema
|
1
|
|
Hypertensive retinopathy
|
1
|
|
Lattice
|
1
|
|
Macular hole
|
0
|
|
Macular scar
|
1
|
|
Nonproliferative diabetic retinopathy
|
9
|
|
Proliferative diabetic retinopathy
|
4
|
|
Posterior vitreous detachment
|
4
|
|
Retinal detachment
|
1
|
|
Conjunctiva
|
|
|
Dry eye
|
10
|
|
Pingueculitis
|
1
|
|
Cornea
|
|
|
Astigmatism
|
12
|
|
Epithelial basement membrane dystrophy
|
1
|
|
Epitheliopathy
|
2
|
|
Cranial nerve
|
|
|
Afferent pupillary defect
|
8
|
|
Bilateral hemianopsia
|
1
|
|
Diplopia
|
1
|
|
Duane's syndrome
|
1
|
|
Ptosis
|
1
|
|
Refractive amblyopia
|
2
|
|
Strabismus
|
4
|
|
Visual field defects
|
13
|
|
Eyelid
|
|
|
Blepharitis
|
0
|
|
Suspicious cancer of eyelids
|
1
|
|
Glaucoma suspect/Glaucoma
|
25
|
|
Globe
|
|
|
Eye trauma
|
6
|
|
Iris coloboma
|
1
|
|
Lacrimal
|
|
|
Epiphora
|
1
|
|
Lens
|
|
|
Anisometropia
|
1
|
|
Aphakia
|
2
|
|
Cataracts
|
16
|
|
Hyperopia
|
9
|
|
Myopia
|
36
|
|
Presbyopia
|
66
|
|
Lymphoid
|
|
|
Proptosis
|
2
|
|
Refractive error
|
86
|
|
Normal ocular exam
|
10
|
Glaucoma had a prevalence rate of 20%. When stratifying the prevalence of glaucoma
by race/ethnicity, Black/African-Americans had a prevalence rate of 30%, Asians had
a prevalence rate of 29%, Hispanic/Latinos had a prevalence rate of 25%, and White/Caucasians
had a prevalence rate of 18% ([Fig. 6]).
Fig. 6 The race and ethnicity of patients with glaucoma attending the free eye clinic from
2017 to 2019.
Discussion
The ultimate purpose of this study was to assess the patient characteristics and prevalence
of ophthalmologic diseases of patients receiving care at the institution-sponsored
student operated eye clinic during a 2-year period. The main findings of this retrospective
cohort study demonstrate that there is a higher prevalence of eye disease among the
homeless population of San Francisco as compared with the overall prevalence rates
of ophthalmologic disease nationally.[18] Additionally, the high prevalence rates of comorbidities such as HTN, DM2, and HIV/AIDS
reinforce the findings of prior studies that identified higher burdens of disease
among the homeless.[19] When analyzing the social determinants of health, participants in this study had
lower levels of educational attainment and higher unemployment rates compared with
both San Francisco and national averages.[20]
[21]
This study reiterates the presence of healthcare disparities since certain minorities
were overrepresented in this study when compared with the prevalence of these ethnic
and racial categories outlined in the records of San Francisco County and the national
census. For example, the racial and ethnic composition of San Francisco is the following:
40% White/Caucasian, 34% Asian, 15% Hispanic, and 5% Black/African-American.[22] At the national level, the percentage composition of these populations is 60% White/Caucasian,
18% Hispanic/Latino, 13% Black/African-American, 13% Asian, and 1% American Indian
and Alaska Native, and 0.2% Native Hawaiian or other Pacific Islander.[23] In this study, the percentage of Black/African-American patients is 33% that is
seven times the average for San Francisco and three times the national average. Prior
studies have demonstrated that Black/African-Americans and Asians have a higher susceptibility
of developing glaucoma[24]
[25] and the results in this study have reconfirmed these trends. This underserved population
demonstrated a higher prevalence of ocular disease in aggregate when compared with
the general population and thus highlights the major role that social determinants
of health play in patient outcomes.
Limitations were identified in the collection of VA and thus the determination of
visual impairment. UNVA was utilized for patients who never owned eyeglasses or contact
lenses. UNVA was calculated instead of best corrected visual acuity for these patients
and use of the former may have led to an overestimation of the prevalence of visual
impairment among this population but it appeared to be a more accurate proxy of their
visual acuity since they were less likely to obtain correction. To mitigate this concern,
we have formed a partnership with a local organization Project Homeless Connect that
provides free eyeglasses for our referred patients. Limitations were also centered
around the patient selection process as patients with eye concerns were most likely
to agree to receiving an eye examination. Patient selection was not randomized. Therefore,
the context of this clinic may have led to an increased likelihood of attracting patients
with preexisting eye diseases. Our study population may not be representative of the
demographic and ocular pathology trends at the national level that limit the generalizability
of these findings to general populations. However, the findings in this study are
applicable to other micropopulations that consist of predominantly homeless and underserved
populations since previous studies in this subset have discovered similar ophthalmologic
trends.[26]
[27] Lastly, as a retrospective study, it was hard to track patient outcomes over time
and it was difficult to ascertain trends during this 2-year period.
Lastly, this study emphasizes the integral role student-run clinics play in providing
healthcare to the community's most vulnerable who have a demonstrated need for ophthalmologic
care. This study provided the opportunity to provide care for a growing population
who often do not receive care within existing healthcare systems. Additionally, patients
were connected to PCPs as 50% of participants previously did not have a PCP. With
the high prevalence of tobacco, alcohol, and marijuana, there are opportunities for
health counseling and expansion of health education services that were provided by
the premedical and medical students. Patients underwent health education sessions
with hopes of understanding their illness and increased rate of adherence to medical
care. In addition to providing increased opportunities to care for the community's
most vulnerable, UCSF-HSEC provided increased exposure to the field of ophthalmology
and increased opportunities for faculty and resident teaching and mentorship of students.
Future direction should assess the rates of follow-up to our partner primary care
providers and ophthalmology providers at the local county hospital and community partner
organization Project Homeless Connect. Additionally, the evaluation of the efficacy
of our smoking cessation programs could be assessed. Further study could also investigate
the impact of the clinic on medical student specialty choice and career focus.
