CLINICAL SCIENCES

Racial and Ethnic Disparities

in Adherence to Glaucoma Follow-up Visits
in a County Hospital Population
Yohko Murakami, MD, MA; Bradford W. Lee, MD, MSc; Martin Duncan, BA; Andrew Kao, MD;
Jehn-Yu Huang, MD; Kuldev Singh, MD, MPH; Shan C. Lin, MD

Objectives: To identify predictors of inconsistent attendance at glaucoma follow-up visits in a county hospital
population.
Methods: Prospective recruitment from August 1, 2008,
through January 31, 2009, of 152 individuals with glaucoma, with 1-to-1 matching of patients (those with inconsistent follow-up) and controls (those with consistent follow-up). Data were collected via oral questionnaire.
Survey results were correlated with attendance at follow-up examinations, using the t test, 2 test, and multivariate stepwise logistic regression analysis to calculate the odds ratios (ORs) and 95% confidence intervals.
Results: After adjusting for covariates in the logistic regression analysis, factors independently associated with
inconsistent follow-up included black race (adjusted OR,
7.16; 95% confidence interval, 1.64-31.24), Latino eth-

Author Affiliations:
Departments of Ophthalmology,
Stanford University, Stanford,
California (Drs Murakami, Lee,
and Singh and Mr Duncan),
University of California, San
Francisco (Drs Kao and Lin),
and National Taiwan University
Hospital, Taipei, Taiwan
(Dr Huang).

nicity (adjusted OR, 4.77; 1.12-20.29), unfamiliarity with

necessary treatment duration (adjusted OR, 3.54; 1.269.94), lack of knowledge of the permanency of glaucomainduced vision loss (adjusted OR, 3.09; 1.18-8.04), and
perception that it is not important to attend all follow-up visits (adjusted OR, 3.54; 1.26-9.94).
Conclusions: Demographic factors, including race and
ethnicity, may directly or indirectly affect adherence to
recommended glaucoma follow-up visits. Lack of information regarding irreversible vision loss from glaucoma, need for lifelong treatment, and lack of visual symptoms may be significant barriers to follow-up in this
population. Targeted glaucoma education by physicians may improve follow-up, thereby decreasing the morbidity associated with glaucomatous disease.

Arch Ophthalmol. 2011;129(7):872-878

LAUCOMA IS THE SECOND

leading cause of blindness after cataract, accounting for 12.3% of the

blindness worldwide, and
is the leading cause of irreversible vision
loss.1 Lowering intraocular pressure (IOP)
significantly reduces disease progression
and development of adverse outcomes,
such as vision loss and blindness.2-4 Several clinical studies5-7 have demonstrated
that the use of IOP-lowering medications
decreases the likelihood of measurable disease progression, and it is postulated that
patients adherent to therapy will have more
favorable outcomes than those who are
nonadherent.
Regular patient follow-up is critical for
physicians to assess the course of glaucomatous disease and appropriately adjust
therapy as needed.8,9 The American Academy of Ophthalmology suggests at least 2
follow-up visits per year in patients with
primary open-angle glaucoma in the Preferred Practice Pattern guidelines.10 Inconsistent adherence to recommended fol-

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

872

low-up hinders the ability of the physician

to track disease progression, presumably increasing the likelihood of adverse disease
outcomes.11 Kosoko et al11 reported that patients at a glaucoma clinic in inner-city Baltimore, Maryland, with inconsistent follow-up were more likely to have mild
disease than those who demonstrated consistent follow-up. Those with inconsistent
follow-up also were more likely to express
displeasure with long wait times in clinics.
Although those with inconsistent follow-up were less likely to have been prescribed glaucoma medications than those
with consistent follow-up, it is unclear
whether this was due to more severe disease in case than control individuals. A questionnaire-based study12 at the Aravind Eye
Care System in Coimbatore, Tamil Nadu,
India, showed that little or no formal education, not using glaucoma medications, and
the perception that glaucoma follow-up is
less important when patients are adherent
to their prescribed glaucoma medication
regimen were all independent predictors of
inconsistent glaucoma follow-up.

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

Table 1. Criteria for Evaluating Adherence to Recommended Follow-up Visits a

Glaucoma Disease Severity b
Recommendation
General follow-up recommendations c
Control individuals with consistent
follow-up
Case individuals with inconsistent
follow-up

Mild

Moderate

Severe

Every 5-6 months, approximately

2 visits per year
2 Visits, with 5- to 6-month
maximum intervals between visits
1 Visit or extended interval
between visits

Every 4-5 months, approximately

3 visits per year
3 Visits, with 4- to 5-month
maximum intervals between visits
2 Visits or extended interval
between visits

Every 3-4 months, approximately

4 visits per year
3 Visits, with 3- to 4-month
maximum intervals between visits
2 Visits or extended intervals
between visits

a These were general guidelines for the study cohort, although each patient was reviewed individually based on recommended and actual follow-up patterns
when assigning follow-up status.
b Mild indicates that characteristics of optic nerve abnormalities are consistent with glaucoma but the visual field is normal; moderate, visual field abnormalities exist
in 1 hemifield and are not within central 5 of fixation; and severe, visual field abnormalities exist in both hemifields or visual field loss is within central 5 of fixation.
c Glaucoma disease severity was evaluated on the American Academy of Ophthalmology Preferred Practice Patterns guidelines for primary open-angle glaucoma and
suspicion of primary open-angle glaucoma.

Disparities in the prevalence of glaucoma by race and

ethnicity have been well documented in several large
population-based studies.13-17 Previous studies15,18-20 have
found that blacks and Latinos are at greater risk for glaucoma than whites. Black race is a known risk factor for
glaucoma, evidenced by increased disease prevalence, earlier disease development, and increased progression to
blindness. Latinos also experience an exponentially greater
incidence of disease development with age compared with
whites.13,16,17 Several such studies were conducted at academic medical centers serving patient populations predominantly of 1 minority racial/ethnic group who were
then compared with whites. No large prevalence surveys, however, have directly compared glaucoma prevalence in blacks and Latinos in 1 city or county population, to our knowledge.
We conducted a case-control study of individuals with
definitive or suspected glaucoma to assess differences between those with inconsistent follow-up (ie, cases) and
those with consistent follow-up (ie, controls) at San Francisco General Hospital (SFGH), a county hospital serving a patient population that is diverse with regard to race,
ethnicity, and language. Data were examined for independent predictors of inconsistent glaucoma follow-up,
and patient-reported barriers to such follow-up were
assessed.
METHODS
We recruited 152 individuals diagnosed as having glaucoma
or suspected of having glaucoma examined at the SFGH Glaucoma Clinic in San Francisco, California, from August 1, 2008,
through January 31, 2009, with 1-to-1 matching of cases (those
with inconsistent follow-up) and controls (those with consistent follow-up). This clinic receives referrals from within SFGH
and its satellite community centers throughout San Francisco.
The clinic is located in an inner-city county hospital that provides comprehensive health care services as the safety net system throughout the San Francisco area for patients covered by
Medi-Cal government-sponsored health insurance and those
who are homeless, uninsured, and/or immigrants.
Table 1 summarizes criteria for classifying adherence to recommended follow-up patterns at SFGH based on disease severity. Classification according to these guidelines was based on data
gathered from medical records. Eligibility criteria included hav-

ing a medical record documenting the dates of all glaucoma follow-up visits scheduled and attended in the past 12 months, age
of 40 years or older, and 1 of the following diagnoses made at
SFGH: primary open-angle glaucoma, primary angle-closure glaucoma, exfoliative glaucoma, pigmentary glaucoma, low-tension
glaucoma, or ocular hypertension. These selection criteria were
designed to exclude patients who would not require ongoing follow-up at the SFGH Glaucoma Clinic, such as those seeking a
second opinion or for urgent care. The chief of the SFGH Glaucoma Service evaluated disease severity (ie, mild, moderate, or
severe) according to the American Academy of Ophthalmology
Preferred Practice Patterns guidelines for primary open-angle glaucoma based on visual field testing, cup-to-disc ratios, and applanation tonometry.10
All study participants were interviewed in their preferred
language (ie, English, Spanish, Mandarin, Cantonese, Vietnamese, or Tagalog) by a trained member of the multilingual
research team. Data were collected by oral questionnaire regarding patient demographics, perceived barriers to followup, and reasons for nonattendance. Patients self-reported their
racial/ethnic classifications. In the case of multiethnic individuals, patients were asked with which classification they most
identified. Oral informed consent was obtained from all patients before the interview. Questions were standardized across
languages. The questions were based on a questionnaire used
at the Aravind Eye Hospital Glaucoma Clinic in Coimbatore,
Tamil Nadu, India, and adapted to the present study population.12 The questionnaire was validated in a pilot study with a
randomly selected cohort of 14 patients who met the aforementioned eligibility criteria. Questions were adapted to the
SFGH patient population, response coding classifications were
calibrated, and protocol feasibility was tested. The participants in the pilot study were not included among the 152 cases
and controls who comprised the study group. On the basis of
pilot study results, it was determined that 150 patients would
be adequate to identify predictors of inconsistent adherence with
an odds ratio (OR) of 2 or greater with a power of 80% and an
of .05. Human subjects approval for this study was obtained
from the institutional review boards of the following organizations: SFGH; University of California, San Francisco; and the
Stanford University School of Medicine. The study followed the
tenets of the Treaty of Helsinki.
Statistical analysis was conducted using IBM SPSS Statistics statistical software, version 18.0 (SPSS Inc, Chicago, Illinois). Proportions were compared using the t and 2 tests, and
adjusted ORs and 95% confidence intervals (CIs) were calculated using a stepwise multivariate logistic regression model.
Variables with P.20 were initially included in the multivar-

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

873

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

Table 2. Demographic Characteristics of the Study Populations a

Characteristic
Age, mean (SD), y
40-49
50-59
60-69
70-79
80
Sex
Female
Male
Race/ethnicity
White
Black
Latino
Asian/Pacific Islander
English proficiency
Yes
No
Primary language
English
Spanish
Mandarin/Cantonese
Tagalog
Other
Educational level b
High
Medium
Low
Disease severity
Mild or moderate
Severe
Outside service use
Seen only at SFGH for glaucoma-related care
Seen outside SFGH
Employment status
Not working, retired, unemployed, or laid off
Employed
Health insurance coverage type/status
Private
Government coverage (ie, Medicare, Medi-Cal,
or San Francisco Health Plan)
No insurance
Glaucoma care history
No interventional treatment
Laser surgery treatment
Missing

No. of Case
Individuals
(n = 76)

No. of Control
Individuals
(n = 76)

Unadjusted OR for Inconsistent

Follow-up (95% CI)
(Reference Group, Controls)

P Value

64.3 (9.5)
4 (5.3)
19 (25.0)
36 (47.4)
12 (15.8)
5 (6.6)

64.8 (8.8)
4 (5.3)
15 (19.7)
33 (43.4)
21 (27.6)
3 (3.9)

NA
1 [Reference]
1.27 (0.27-5.92)
1.09 (0.25-4.72)
0.57 (0.12-2.71)
1.67 (0.23-12.2)

NA
NA
.76
.91
.57
.62

44 (57.9)
32 (42.1)

52 (68.4)
24 (31.6)

1 [Reference]
1.58 (0.81-3.06)

NA
.18

6 (7.9)
21 (27.6)
31 (40.8)
18 (23.7)

13 (17.1)
13 (17.1)
26 (34.2)
24 (31.6)

1 [Reference]
3.50 (1.07-11.5)
2.58 (0.86-7.75)
1.63 (0.52-5.10)

NA
.04
.09
.41

39 (51.3)
37 (48.7)

41 (53.9)
35 (46.1)

1 [Reference]
1.11 (0.59-2.10)

NA
.75

29 (38.2)
29 (38.2)
10 (13.2)
5 (6.6)
3 (3.9)

24 (31.6)
25 (32.9)
14 (18.4)
8 (10.5)
5 (6.6)

1 [Reference]
0.96 (0.45-2.05)
0.59 (0.22-1.57)
0.52 (0.15-1.80)
0.50 (0.11-2.29)

NA
.92
.29
.30
.37

25 (32.9)
18 (23.7)
33 (43.4)

23 (30.3)
22 (28.9)
31 (40.8)

1 [Reference]
0.87 (0.41-1.82)
0.92 (0.39-2.16)

NA
.70
.85

42 (55.3)
34 (44.7)

52 (68.4)
24 (31.6)

1 [Reference]
1.57 (0.81-3.03)

NA
.18

67 (88.2)
9 (11.8)

72 (94.7)
4 (5.3)

1 [Reference]
2.42 (0.71-8.22)

NA
.16

55 (72.4)
21 (27.6)

56 (73.7)
20 (26.3)

1 [Reference]
0.94 (0.46-1.92)

NA
.86

5 (6.6)
69 (90.8)

2 (2.6)
74 (97.4)

1 [Reference]
0.37 (0.07-1.99)

NA
.25

6.46 108 (0-)

NA

1 [Reference]
0.58 (0.25-1.34)
NA

NA
.20
NA

2 (2.6)
58 (76.3)
11 (14.5)
7 (9.2)

0
55 (72.4)
18 (23.7)
3 (3.9)

Abbreviations: CI, confidence interval; NA, not applicable; OR, odds ratio; SFGH, San Francisco General Hospital.
a Percentages may not total 100 because of rounding.
b High indicates any undergraduate university or community college coursework and beyond; medium, any secondary school coursework through secondary
school completion or general educational development certification; and low, no formal education through completion of primary school.

iate regression model with age and sex then successively eliminated based on higher P values. All variables except age and
sex had P.05 in the final regression model.
RESULTS

Characteristics of the patients with inconsistent follow-up (cases) and consistent follow-up (controls) are
summarized in Table 2. Of the 186 patients recruited
for the study, 14 were involved in the pilot study, an additional 16 were unreachable, and 4 declined to participate, citing time constraints. A total of 152 patients com-

pleted the oral questionnaire in a 1:1 matched ratio and

were included in the final analysis.
The follow-up patterns of the 152 study patients varied across ethnic groups (Table 2). The number (percentage) of black, Latino, Asian, and white patients classified as cases was 21 (27.6%), 31 (40.8%), 18 (23.7%),
and 6 (7.9%), respectively. The proportion of cases within
each race/ethnicity was 61.8% black, 54.4% Latino, 42.9%
Asian, and 31.6% white.
Table 3 shows the univariate analysis for the potential predictors of inconsistent follow-up. A stepwise multivariate logistic regression model identified indepen-

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

874

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

Table 3. Predictors of Inconsistent Follow-up a

Predictor
Transportation
Convenient
Somewhat or very inconvenient
No. of transfers (on public transportation) necessary to travel to clinic
No transfers
1 Transfer
Travel time to clinic
0-15 min
16-30 min
31-60 min
60 min
Physical difficulty coming to clinic
Easy or not too difficult
Somewhat or very difficult
Symptoms or lack thereof at diagnosis
Asymptomatic
Symptomatic
Missing
Difficulty taking time away from work or home
Easy or not too difficult
Somewhat or very difficult
Missing
Recollection of being counseled regarding glaucoma by clinic staff
Yes
No
Glaucoma knowledge from family and friends
Yes
No
Glaucoma knowledge from pamphlets
Yes
No
Perceived importance of attending follow-up visits if adhering
to medication regimen and noticing no vision change
Very important
Not very important
Understanding regarding necessary length of glaucoma treatment
Permanent
Until symptoms resolve
Understanding of permanency of glaucoma-induced vision loss
Permanent
Reversible
Not sure
Knowledge of IOP-glaucoma relationship
High IOP increases risk
No or not sure
Knowledge of glaucoma being a disease defined by optic
nerve damage
Yes
No or not sure
Perceived difficulty in attending follow-up appointments
Easy or not too difficult
Somewhat or very difficult
Perceived importance of follow-up visits
Very important
Somewhat or not important

No. of Case
Individuals
(n = 76)

No. of Control
Individuals
(n = 76)

Unadjusted OR for Inconsistent

Follow-up (95% CI)
(Reference Group, Controls)

P Value

61 (80.3)
15 (19.7)

54 (71.1)
22 (28.9)

1 [Reference]
1.66 (0.78-3.53)

NA
.19

41 (53.9)
35 (46.1)

34 (44.7)
42 (55.3)

1 [Reference]
0.69 (0.37-1.31)

NA
.26

13 (17.1)
27 (35.5)
30 (39.5)
6 (7.9)

9 (11.8)
26 (34.2)
36 (47.4)
5 (6.6)

1 [Reference]
0.72 (0.26-1.97)
0.58 (0.22-1.54)
0.83 (0.19-3.58)

NA
.52
.27
.80

56 (73.7)
20 (26.3)

49 (64.5)
27 (35.5)

1 [Reference]
1.54 (0.77-3.09)

NA
.22

30 (39.5)
32 (42.1)
14 (18.4)

31 (40.8)
29 (38.2)
16 (21.1)

1 [Reference]
1.14 (0.56-2.32)
NA

NA
.72
NA

60 (78.9)
13 (17.1)
3 (3.9)

48 (63.2)
24 (31.6)
4 (5.3)

1 [Reference]
2.31 (1.06-5.01)
NA

NA
.03
NA

39 (51.3)
37 (48.7)

51 (67.1)
25 (32.9)

1 [Reference]
1.94 (1.00-3.73)

NA
.049

11 (14.5)
65 (85.5)

3 (3.9)
73 (96.1)

1 [Reference]
0.24 (0.07-0.91)

NA
.04

9 (11.8)
67 (88.2)

7 (9.2)
69 (90.8)

1 [Reference]
0.76 (0.27-2.14)

NA
.60

52 (68.4)
24 (31.6)

65 (85.5)
11 (14.5)

1 [Reference]
2.73 (1.22-6.08)

NA
.01

45 (59.2)
31 (40.8)

53 (69.7)
23 (30.3)

1 [Reference]
1.59 (0.81-3.10)

NA
.18

37 (48.7)
1 (1.3)
38 (50.0)

44 (57.9)
3 (3.9)
29 (38.2)

1 [Reference]
0.40 (0.04-3.97)
1.56 (0.81-3.00)

NA
.43
.18

35 (46.1)
41 (53.9)

45 (59.2)
31 (40.8)

1 [Reference]
1.70 (0.89-3.23)

NA
.11

10 (13.2)
66 (86.8)

11 (14.5)
65 (85.5)

1 [Reference]
1.12 (0.45-2.81)

NA
.82

57 (75.0)
19 (25.0)

50 (65.8)
26 (34.2)

1 [Reference]
0.64 (0.32-1.30)

NA
.22

70 (92.1)
6 (7.9)

64 (84.2)
12 (15.8)

1 [Reference]
2.19 (0.78-6.17)

NA
.14

Abbreviations: CI, confidence interval; IOP, intraocular pressure; NA, not applicable; OR, odds ratio.
a Percentages may not total 100 because of rounding.

dent predictors of inconsistent follow-up after adjusting

for covariates (ie, age, sex, disease severity, employment status, marital status, and health insurance coverage status). These predictors included black race, Latino ethnicity, having little or no recollection of receiving

glaucoma-related education from clinic staff, receiving

information regarding glaucoma from family members,
being unaware of the risk of irreversible blindness due
to glaucoma and the need for lifelong treatment, and perceiving follow-up visits not to be important if patients

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

875

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

Table 4. Significant Predictors of Inconsistent Adherence

to Glaucoma Follow-up Visits After Multiple Logistic
Regression Analysis

Table 5. Breakdown of Case and Control Individuals

by Race/Ethnicity and Educational Level a

Educational Level
by Race/Ethnicity b

OR (95% CI) (Reference Group,

Control Individuals)

Risk Factor
Perceived importance of
attending follow-up visits
if adhering to medication
regimen and no
noticeable vision changes
Very important
Not very important
Race/ethnicity
White
Black
Latino
Asian/Pacific Islander
Understanding of glaucoma
treatment duration
Permanent
Until symptoms
resolve
Understanding permanency
of glaucoma-induced
vision loss
Permanent vision loss
Reversible vision loss
Not sure
Recall being counseled
regarding glaucoma by
clinic staff
Yes
No
Disease severity
Mild or moderate
Severe
Employment status
Not working, retired,
unemployed, or laid off
Employed

Unadjusted for
Inconsistent
Follow-up

Adjusted for
Inconsistent
Follow-up

1 [Reference]
1 [Reference]
2.73 (1.22-6.08) 3.54 (1.26-9.94)

High
White
Black
Latino
Asian/Pacific Islander
Medium
White
Black
Latino
Asian/Pacific Islander
Low c
Black
Latino
Asian/Pacific Islander

P
Value

NA
.02

1 [Reference]
1 [Reference]
NA
3.50 (1.07-11.53) 7.16 (1.64-31.24) .009
2.58 (0.86-7.75) 4.77 (1.12-20.29) .04
1.63 (0.52-5.10) 3.27 (0.78-13.71) .10

1 [Reference]
1 [Reference]
1.59 (0.81-3.10) 3.54 (1.26-9.94)

NA
.02

1 [Reference]
1 [Reference]
0.40 (0.04-3.97) 0.31 (0.02-4.61)
1.56 (0.81-3.00) 3.09 (1.18-8.04)

NA
.40
.02

1 [Reference]
1 [Reference]
1.94 (1.00-3.73) 3.16 (1.26-7.94)

NA
.01

1 [Reference]
1 [Reference]
1.57 (0.81-3.03) 2.10 (0.92-4.78)

NA
.08

1 [Reference]

NA

1 [Reference]

0.94 (0.46-1.92) 1.28 (0.53-3.09)

No. of
Cases
(n = 76)

No. of
Controls
(N = 76)

3 (12.0)
8 (32.0)
5 (20.0)
9 (36.0)

8 (34.8)
2 (8.7)
7 (30.4)
6 (26.1)

.39
.07
.67
.49

3 (9.4)
13 (40.6)
12 (37.5)
4 (12.5)

5 (14.7)
9 (26.5)
12 (35.3)
8 (23.5)

.79
.04
.99
.69

0
14 (73.7)
5 (26.3)

2 (10.5)
7 (36.8)
10 (52.6)

.89
.04
.08

P Value

a Percentages may not total 100 because of rounding.

b High educational level defined as any undergraduate

university or
community college coursework and beyond; medium, any secondary school
coursework through secondary school completion or general educational
development certification; and low, no formal education through completion
of primary school.
c White ethnicity not presented for this variable because all such patients
in the cohort reported completion of a secondary school degree or receipt of
a general educational development certificate.

college coursework or beyond) and Asian/Pacific Islander patients of low educational background demonstrating consistent follow-up patterns (P = .07 and P = .08,
respectively).
Factors not found to be predictive of follow-up patterns included employment status, marital status, health
insurance coverage status, self-reported ability to pay for
medications, history of laser treatment for glaucoma, and
self-reported inconvenience of transportation to clinic
(Table 3 and Table 4).

.58

COMMENT

Abbreviations: CI, confidence interval; NA, not applicable; OR, odds ratio.

were regularly using medications and did not notice any

visual symptoms (Table 4).
Black race and Latino ethnicity were strong predictors
of inconsistent follow-up even after adjustment for age and
sex (OR, 7.16 [95% CI, 1.64-31.24]; P = .009; and 4.77
[1.12-20.29]; P = .04, respectively). A breakdown of cases
and controls by race/ethnicity and educational background (used as a proxy for socioeconomic status) is given
in Table 5. Latino patients having a low level of education (ie, no formal education through completion of primary school) and black patients with a medium level of
education (ie, any secondary school through completion
of a secondary school degree or a general educational development certificate) were more likely to have inconsistent follow-up patterns compared with their white counterparts with similar educational backgrounds (P = .04
for each). Although not statistically significant, a trend
was observed of black patients with high educational level
(ie, having any undergraduate university or community

Disease severity at the time of diagnosis and rate of disease progression are well-established risk factors for vision loss in patients with glaucoma.3,4 Recently, much attention has been given to nonadherence to prescribed
glaucoma medication regimens as an important predictor
of adverse outcomes.11,21 Remarkably little attention, however, has been given to inconsistent adherence to recommended follow-up visits as a predictive factor for vision
loss. The physician treating glaucoma has many treatment options, including medications, laser trabeculoplasty, and incisional glaucoma surgery, all of which can
lower IOP. Insufficient IOP lowering with 1 approach because of lack of efficacy of or inconsistent adherence to
glaucoma therapy generally results in advancement to the
next step in this treatment algorithm. Such assessment and
advancement of therapy can only take place, however, if
patients are seen on a timely basis for follow-up. Although high-quality confirmatory evidence is lacking, one
can hypothesize that inconsistent follow-up is a risk factor for inconsistent outcomes regarding any chronic de-

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

876

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

generative disease for which multiple potentially effective

therapeutic options exist.
This study demonstrates that race/ethnicity and educational background are important factors that affect the
follow-up patterns of patients with glaucoma in an ethnically diverse, resource-limited county hospital population. Black and Latino patients were found, on average, to show less consistent follow-up relative to whites.
The proportion of black patients with inconsistent follow-up (61.8%) was approximately double that observed among white patients (31.6%), resulting in a calculated OR of 7.16 (P = .009) even after adjustment for
covariates (ie, age, sex, disease severity, employment status, marital status, and health insurance coverage status). Latino patients had a calculated OR of 4.77 (P = .04)
compared with their white counterparts after similar covariate adjustment. These results are consistent with the
trends of lower follow-up and screening rates observed
in black and Latino patients with breast and cervical cancer compared with their white counterparts.22,23
Patients understanding of glaucoma disease mechanisms greatly influences their adherence to recommended follow-up visits. In particular, those unaware of
the chronic and insidious nature of glaucoma and the need
for permanent treatment despite regular use of prescribed medications and lack of symptoms were more
likely to have inconsistent follow-up in this study. Uncertainty regarding these issues was highly correlated with
patient perceptions regarding insufficient glaucomarelated education from clinic staff. This finding dovetails with the observation in the breast cancer literature
that misconceptions regarding breast lumps and appropriate follow-up care were more common among women
of ethnic minority backgrounds, those with lower socioeconomic status, and those with less access to health care.24
The identification of modifiable risk factors for inconsistent follow-up may allow assessment of targeted
solutions, such as increasing physician-initiated patient
education in an effort to eliminate the knowledge gap and
to decrease inconsistent adherence to follow-up. Given
our results, targeted interventions aimed at patients with
multiple predictors of inconsistent follow-up, such as minority race/ethnicity, low educational background, and
those with limited knowledge of glaucoma, have the greatest potential of increasing rates of glaucoma disease followup. Additional larger studies in the future can help refine recommendations regarding groups to target for
improvement in follow-up.
Several prior studies11,12 have reported that mild glaucomatous disease and lack of glaucoma symptoms are predictors of inconsistent adherence to follow-up. In contrast, individuals with severe disease and those with ocular
pain and/or vision changes at the time of their initial glaucoma diagnosis were more likely to show inconsistent adherence to recommended follow-up in this study. Although causality is difficult to assess in such a study, a
possible explanation for these findings is that inconsistent adherence to appointments in the past led to greater
disease severity and also predicted future follow-up behavior. This hypothesis is supported by reports from the
Institute of Medicine25-27 and the US Department of Health
and Human Services28 documenting that racial/ethnic mi-

norities and patients of lower socioeconomic status are

more likely to have limited access to care, to have frequently missed opportunities for preventive care, and to
report insufficient patient-physician communication. These
trends are believed to contribute to the high rates of latestage breast, endometrial, cervical, and colorectal cancer
diagnoses in those patients.25-28 Whether similar patterns
are evident in patients with glaucoma, thus allowing for
greater disease progression and adverse symptoms at the
initial office visit, is an area for future investigation.
This study suggests that blacks and Latinos, 2 racial/
ethnic groups previously identified as being at high risk
for glaucomatous disease, may benefit from specialized
education and counseling to improve follow-up. Targeted physician-initiated educational interventions might
primarily focus on 3 variables: the risk of irreversible blindness from glaucoma, the long-term nature of treatment,
and the importance of attending follow-up visits despite
the ongoing use of medications and lack of visual symptoms. The effect of such interventions should be assessed, with the primary outcome being visual preservation. Glaucoma education by physicians may be
important in correcting the pervasive lack of awareness
regarding the insidious nature of disease progression,
which is predictive of inconsistent follow-up. Our work
suggests that advocating for patient-centered education
is important when caring for a multilingual patient population in a resource-limited setting. With health system
reform in full swing, pressure is growing to deliver highquality, cost-effective care. An understanding of the population-specific pattern of each practice is likely a necessary prerequisite to providing high-yield interventions
in an effort to prevent avoidable blindness.
One limitation of this study is the small number of
patients classified as white cases. Despite the aim of the
study to interview more patients in this reference category, limited numbers of white patients with inconsistent follow-up were available. Such a small sample size
in 1 group is clearly suboptimal for many reasons, including the resultant wide CIs, which limit the race/
ethnicity analysis. Future studies will aim for larger sample
sizes to address this issue.
Generalizing the current findings to other practices
must be done with caution given site-specific variability
in demographics, patient counseling, and availability of
care. However, this study demonstrates the adaptability
of our research method, originally applied at Aravind Eye
Hospital in India, to other geographic locations. We believe this survey will be a useful tool in determining predictors of and barriers associated with inconsistent patient adherence to glaucoma follow-up visits in a variety
of settings. The application of these findings in examining the influence of educational interventions on adherence to follow-up should be the focus of future studies.
Submitted for Publication: April 20, 2010; final revision received October 5, 2010; accepted October 11,
2010.
Correspondence: Shan C. Lin, MD, University of California, San Francisco, 10 Koret Way, Room K-325, San
Francisco, CA 94143-0730 (LinS@vision.ucsf.edu).
Financial Disclosure: None reported.

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

877

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015

Additional Contributions: We thank Rita Popat, PhD,

at Stanford University; Bennie Jeng, MD, and Dandan
Wang, MD, at the University of California, San Francisco; and the San Francisco General Hospital staff for
their invaluable help with this project.
REFERENCES

14.

15.

16.

1. Giangiacomo AC. The epidemiology of glaucoma. In: Grehn F, Stamper R, eds.

Glaucoma. Berlin, Germany: Springer; 2009:13-21.
2. Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch
Opthalmol. 2002;120(6):701-730.
3. Leske MC, Heijl A, Hussein M, Bengtsson B, Hyman L, Komaroff E; Early Manifest Glaucoma Trial Group. Factors for glaucoma progression and the effect of
treatment: the Early Manifest Glaucoma Trial. Arch Ophthalmol. 2003;121(1):
48-56.
4. Leske MC, Heijl A, Hyman L, Bengtsson B, Dong L, Yang Z; EMGT Group. Predictors of long-term progression in the Early Manifest Glaucoma Trial. Ophthalmology. 2007;114(11):1965-1972.
5. Epstein DL, Krug JH Jr, Hertzmark E, Remis LL, Edelstein DJ. A long-term clinical trial of timolol therapy versus no treatment in the management of glaucoma
suspects. Ophthalmology. 1989;96(10):1460-1467.
6. Kass MA. Timolol treatment prevents or delays glaucomatous visual field loss in
individuals with ocular hypertension: a five-year, randomized, double-masked,
clinical trial. Trans Am Ophthalmol Soc. 1989;87:598-618.
7. Schulzer M, Drance SM, Douglas GR. A comparison of treated and untreated glaucoma suspects. Ophthalmology. 1991;98(3):301-307.
8. DiMatteo MR. Variations in patients adherence to medical recommendations: a
quantitative review of 50 years of research. Med Care. 2004;42(3):200-209.
9. Friedman DS, Hahn SR, Quigley HA, et al. Doctor-patient communication in glaucoma care: analysis of videotaped encounters in community-based office practice.
Ophthalmology. 2009;116(12):2277-2285.
10. American Academy of Ophthalmology Glaucoma Panel. Preferred Practice Patterns: Primary Open-Angle Glaucoma, Primary Open-Angle Glaucoma Suspect,
and Primary Angel Closure. San Francisco, CA: American Academy of Ophthalmology; 2005.
11. Kosoko O, Quigley HA, Vitale S, Enger C, Kerrigan L, Tielsch JM. Risk factors for
noncompliance with glaucoma follow-up visits in a residents eye clinic.
Ophthalmology. 1998;105(11):2105-2111.
12. Lee BW, Sathyan P, John RK, Singh K, Robin AL. Predictors of and barriers associated with poor follow-up in patients with glaucoma in South India. Arch
Ophthalmol. 2008;126(10):1448-1454.
13. Quigley HA, West SK, Rodriguez J, Munoz B, Klein R, Snyder R. The prevalence

17.

18.
19.
20.

21.

22.

23.

24.

25.

26.

27.

28.

of glaucoma in a population-based study of Hispanic subjects: Proyecto VER.

Arch Ophthalmol. 2001;119(12):1819-1826.
Shimmyo M, Ross AJ, Moy A, Mostafavi R. Intraocular pressure, Goldmann applanation tension, corneal thickness, and corneal curvature in Caucasians, Asians,
Hispanics, and African Americans. Am J Ophthalmol. 2003;136(4):603-613.
Tielsch JM, Sommer A, Katz J, Royall RM, Quigley HA, Javitt J. Racial variations
in the prevalence of primary open-angle glaucoma: the Baltimore Eye Survey.
JAMA. 1991;266(3):369-374.
Kim E, Varma R. Glaucoma in Latinos/Hispanics. Curr Opin Ophthalmol. 2009;21
(2):100-105.
Varma R, Ying-Lai M, Francis BA, et al; Los Angeles Latino Eye Study Group.
Prevalence of open-angle glaucoma and ocular hypertension in Latinos: the Los
Angeles Latino Eye Study. Ophthalmology. 2004;111(8):1439-1448.
Martin MJ, Sommer A, Gold EB, Diamond EL. Race and primary open-angle
glaucoma. Am J Ophthalmol. 1985;99(4):383-387.
Sommer A, Tielsch JM, Katz J, et al. Racial differences in the cause-specific prevalence of blindness in east Baltimore. N Engl J Med. 1991;325(20):1412-1417.
Wormald RP, Basauri E, Wright LA, Evans JR. The African Caribbean Eye Survey: risk factors for glaucoma in a sample of African Caribbean people living in
London. Eye (Lond). 1994;8(pt 3):315-320.
Gwira JA, Vistamehr S, Shelsta H, et al. Factors associated with failure to follow
up after glaucoma screening: a study in an African American population.
Ophthalmology. 2006;113(8):1315-1319.
Richardson JL, Langholz B, Bernstein L, Burciaga C, Danley K, Ross RK. Stage
and delay in breast cancer diagnosis by race, socioeconomic status, age and year.
Br J Cancer. 1992;65(6):922-926.
Madison T, Schottenfeld D, James SA, Schwartz AG, Gruber SB. Endometrial cancer: socioeconomic status and racial/ethnic differences in stage at diagnosis, treatment, and survival. Am J Public Health. 2004;94(12):2104-2111.
Rauscher GH, Ferrans CE, Kaiser K, Campbell RT, Calhoun EE, Warnecke RB.
Misconceptions about breast lumps and delayed medical presentation in urban breast
cancer patients. Cancer Epidemiol Biomarkers Prev. 2010;19(3):640-647.
Institute of Medicine. The healthcare environment and its relation to disparities.
In: Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine; 2003:29-80.
Institute of Medicine. Assessing potential sources of racial and ethnic disparities in care: patient- and system-level factors. In: Smedley BD, Stith AY, Nelson
AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health
Care. Washington, DC: Institute of Medicine; 2003:125-159.
Institute of Medicine. Assessing potential sources of racial and ethnic disparities in care in the clinical encounter. In: Smedley BD, Stith AY, Nelson AR, eds.
Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine; 2003:160-179.
US Department of Health & Human Services. National Healthcare Disparities Report.
Washington, DC: US Dept of Health and Human Services; 2003.

Correction
Error in Text. In the Clinical Sciences article titled Risk
Factors for Visual Field Progression in Treated Glaucoma, by De Moraes et al, published in the May issue
of the Archives (2011;129[5]:562-568), the last sentence of the last paragraph in the right column of page
566 should have appeared as follows: This is partly corroborated by the fact that DHs are more commonly seen
in eyes with PPA and within areas where PPA is the
widest.40 This article was corrected online.

ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY 2011

878

WWW.ARCHOPHTHALMOL.COM

2011 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ on 01/08/2015