Categorizing the Stage of Glaucoma From
Pre-Diagnosis to End-Stage Disease
RICHARD P. MILLS, MD, MPH, DONALD L. BUDENZ, MD, PAUL P. LEE, MD, JD,
ROBERT J. NOECKER, MD, MBA, JOHN G. WALT, MBA, LISA R. SIEGARTEL, MPH,
STACY J. EVANS, MD, AND JOHN J. DOYLE, DRPH
● PURPOSE: To provide a reliable, comprehensive staging
system to assess glaucoma stage in the absence of an
universally accepted glaucoma staging system (GSS) on
the basis of visual field results.
● DESIGN: Literature review and GSS adaptation.
● METHODS: After a review of published GSSs was
conducted, the Bascom Palmer (Hodapp-Anderson-Par-
rish) GSS was selected as an appropriate platform for a
retrospective GSS on the basis of visual fields. The
system was modified by a panel of glaucoma specialists,
and additional modifications were made after pilot testing
to cover the full range of disease progression, from
preglaucoma diagnosis to complete blindness; the ordered
stages reflect the typical progression of glaucoma.
● RESULTS: The GSS is comprised of six ordered stages
and is on the basis of the Humphrey visual field. The
completed GSS was validated by reviewing patient charts
from 12 US glaucoma centers.
● CONCLUSIONS: The GSS allows accurate staging of
100% of glaucoma on the basis of visual fields and other
data, enabling evaluation of disease progression and re-
source utilization at various glaucoma stages. Additionally,
treatment costs may be assigned to determine cost-effective-
ness of treatment. Research utilizing the GSS has found
that cost of care increases with increasing disease severity.
The GSS may be used as the basis for creating treatment
guidelines, which have the potential to delay glaucoma
progression and lower treatment costs. (Am J Ophthal-
mol 2006;141:24 –30. © 2006 by Elsevier Inc. All rights
reserved.)
See accompanying Editorial on page 147.
Accepted for publication Jul 16, 2005.
From the University of Washington, Seattle, Washington, (R.P.M.);
Bascom Palmer Eye Institute, Miami, Florida (D.L.B.); Duke University,
Durham, North Carolina (P.P.L.); University of Pittsburgh, Pittsburgh,
Pennsylvania (R.J.N.); Allergan, Inc, Irvine, California (J.G.W.); and
Analytica International, New York, New York (L.R.S., S.J.E., J.J.D.).
This study was supported by a research grant from Allergan, Inc.,
Irvine, California.
This project was previously presented as a poster at the Association for
Research in Vision and Ophthalmology (ARVO) May 5–10, 2001.
Inquiries to John Walt, MBA, Allergan, 2525 Dupont Drive,
Irvine, CA 92623-9534; fax: 714-246-4241; e-mail: walt_john@
allergan.com
24 © 2006 BY ELSEVIER INC. ALL
I
N THE UNITED STATES, GLAUCOMA IS THE SECOND
leading cause of blindness in the general population,
and the leading cause of blindness in black patients.1
Although only half of the individuals who have the disease
are aware of their condition, glaucoma affects approxi-
mately 2.5 million people, including three percent of those
age 55 years and older.1 Annual US healthcare costs for
glaucoma total an estimated $2.5 billion, including $1.9
billion in direct costs and $0.6 billion in indirect costs.2
Additionally, costs for the treatment of a newly diagnosed
case of open-angle glaucoma have been estimated at $1055
per year.3 For the approximately 120,000 patients who
have become blind as a result of glaucoma, costs for
benefits, healthcare, and reduced tax revenues total $1.5
billion per year.4
Primary open-angle glaucoma (POAG), accounting for
more than 90% of US cases of glaucoma, is a chronic,
progressive disease characterized by optic disk cupping and
visual field loss.5 Although this form of glaucoma is
commonly associated with elevated intraocular pressure
(IOP), more than two-thirds of patients with IOP exceed-
ing 21 mm Hg do not have glaucoma.1 As 15% of patients
with glaucoma have a normal IOP of 21 mm Hg or less on
a consistent basis, there are factors other than IOP that
likely contribute to disease development.1,6 – 8
A glaucoma staging system (GSS) provides a way of
measuring the progress of glaucoma in patients who
have the disease. With clearly defined stages of disease
progression, it becomes possible to observe disease
progression, and thus gauge the effectiveness of treat-
ment at each stage. The definition of each disease stage
needs to be adequately precise to allow patients at
different stages of disease and from different glaucoma
treatment centers to be meaningfully compared. Fur-
ther, such a system must allow for precise categorization
without ambiguity, and must be easily usable and
provide consistent (reliable) results.
The purpose of this article is to report the develop-
ment of this GSS from an existing GSS. We aimed to
create a staging system that can be used to stage patients
retrospectively by chart review without physician inter-
RIGHTS RESERVED. 0002-9394/06/$32.00
doi:10.1016/j.ajo.2005.07.044
pretation of a patient’s clinical presentation by adapting
an existing GSS. By assigning cost of treatment, and
reviewing its effectiveness at each stage, the relative
cost-effectiveness of glaucoma treatment at various
stages of disease progression can also be quantified and
compared using this GSS.
METHODS
● PRELIMINARY RESEARCH: A literature review was
conducted to evaluate previously developed GSSs. Ex-
amined GSSs included the staging systems used in the
Advanced Glaucoma Intervention Study (AGIS)9 –11
and the Collaborative Initial Glaucoma Treatment
Study(CIGTS),10 –12 Esterman binocular scale,13 and the
Bascom Palmer system.11,14 –16 These existing GSSs were
considered as potential “seed systems” for modification.
Consideration was also given to the American Academy
of Ophthalmology GSS, but this was not used because it
was limited to only three stages (no field loss, moderate
field loss, severe field loss).
● AGIS: AGIS used the central 24 to 2 total deviation
printout of the Humphrey Field Analyzer to determine a
patient’s visual field score.10 This study scored three
visual field sectors: the nasal, superior, and inferior
areas. Where visual field threshold values were de-
pressed from the normal values, as measured in decibels
(dB) by five to nine dB, this was considered abnormal,
depending on the field location. To qualify as abnormal,
a depression had to be greater in the superior field than
in the inferior, and a depression had to be greater in the
periphery than centrally. Scores were assigned on the
basis of the dB depressions found in the various areas. A
score of zero indicated no visual field loss, while 20 was
the maximum possible score. One stage of field loss
progression was arbitrarily defined as an increase in
score of four.
● CIGTS: The CIGTS staging system used a scoring
methodology similar to that used by the AGIS.10,12,17
Glaucoma staging in this study was also on the basis of
the central 24 to 2 program of the Humphrey Field
Analyzer, but the categories of probability values on the
total deviation probability plot rather than dB devia-
tions as in AGIS were used. Scores were scaled from 0 to
20. A patient was considered to have progressed when
the CIGTS score increased by three or more.
● ESTERMAN BINOCULAR SCALE: Mills and Drance
used an Esterman visual function score derived from an
automated binocular visual field test on the CooperVi-
sion Diagnostics Dicon AP2000 perimeter (CooperVi-
sion, Fairport, New York).13 The scoring system
included both central and peripheral visual field and was
VOL. 141, NO. 1 GLAUCOMA STAGING SYSTEMS
weighted according to the functional importance of
different areas of the visual field. The American Medical
Association accepted the Esterman scoring system in
1984 as a standard for rating vision capabilities.18
Patients with advanced glaucomatous loss also re-
sponded to a 15-item quality of life questionnaire to
assess problems in activities of daily living they encoun-
tered as a result of their visual field loss. The authors,
however, did not attempt to use Esterman test scores to
assign standardized stages of disease progression.13
● BASCOM PALMER (HODAPP-ANDERSON-PARRISH) GSS:
The Bascom Palmer staging system was chosen as the
foundation for development of a new GSS (Table
1).15,16,19 Like the CIGTS and AGIS systems, this
system allows for stage assignments on the basis of HVF
testing, making it easily applicable to a multicenter
retrospective chart review. This staging system assigns
glaucoma patients to different stages of disease progres-
sion on the basis of a combination of mean defect (MD)
score and one of the following: pattern deviation prob-
ability plot score (indicating deviation from a normal-
ized visual field pattern), dB plot (stages 2 to 4) or, for
stage 1, either corrected pattern standard deviation/
pattern standard deviation (CPSD/PSD) or glaucoma
hemifield test results. Despite the utility of this system
on the basis of its use of visual field loss as an indication
of progression, pilot testing found that this GSS fails to
capture the full range of stages in glaucoma progression,
from patients with early and minimal visual field defects
to patients who are blind from end-stage disease.
● DRAFT GSS DEVELOPMENT: Following review of the
literature on existing GSSs, an expert panel was assem-
bled, consisting of four glaucoma specialists. The use of
expert panels to create guidelines has been used in a
number of fields of medical inquiry,20 –24 including
ophthalmology.25 The panels use shared medical litera-
ture and initial algorithms to form drafts of final
algorithms, typically for treatment, formulating a stan-
dard of care.22,26
Among existing GSSs considered, the Bascom Palmer
GSS was deemed most appropriate by the expert panel
as it allowed for structured severity stage assignment
based primarily on Humphrey visual field parameters in
a relatively simplified manner. The AGIS and CIGTS
scoring systems involve more complex calculations and
were thus felt to be more susceptible to errors in scoring.
Additionally, those scoring systems have not been used
to any significant degree in clinical practice. In contrast,
the Bascom Palmer GSS has been used clinically for
routine patient care. While there has been no compar-
ison of the Bascom Palmer GSS to the other systems,
Katz and associates did find that the AGIS and CIGTS
scoring systems diverged considerably in their evalua-
tion of visual field progression.27 The panel initially
25
 26

TABLE 1. Glaucoma Staging System Based on Humphrey Visual Field (Stages 0 Through 5)

Stage Humphrey MD Score Probability Plot/Pattern Deviation dB Plot (Stages 2–4) or CPSD/PSD (Stage 1) dB Plot (Stages 2–4) or Hemifield Test (Stage 1)
AMERICAN JOURNAL

Stage 0 – Ocular ⬎0.00 Does not meet any criteria for Stage 1.
hypertension/earliest
glaucoma
Stage 1 – Early glaucoma ⫺0.01 to ⫺5.00 Points below 5%: ⬎3 CPSD/PSD significant at P ⬍ .05 Glaucoma hemifield test “outside normal limits”
(P ⬍ .05) continguous AND ⬎1 of the
points below 1%
Stage 2 – Moderate ⫺5.01 to ⫺12.00 Points below 5%: 19–36 AND Point(s) within the central 5° with sensitivity Point(s) with sensitivity ⬍15 dB within 5° of
OF

glaucoma Points below 1%: 12–18 of ⬍15 dB: ⬎1 AND point(s) within the fixation: Only 1 hemifield (1 or 2)
AND OR OR
OPHTHALMOLOGY

central 5° with sensitivity of ⬍0 dB:

None (0)
Stage 3 – Advanced ⫺12.01 to ⫺20.00 Points below 5%: 37–55 AND Point(s) within the central 5° with sensitivity Point(s) with sensitivity ⬍15 dB within 5° of
glaucoma Points below 1%: 19–36 of ⬍0 dB: 1 only fixation: Both hemifields, at least 1 in each
Stage 4 – Severe ⫺20.01 or worse Points below 5%: 56–74 AND Point(s) within the central 5° with sensitivity Point(s) with sensitivity ⬍15 dB within 5° of
glaucoma Points below 1%: 37–74 of ⬍0 dB: 2–4 fixation: Both hemifields, 2 in each (ALL)
Stage 5 – End-stage No HVF in “worst HVF not possible attributable to central scotoma in “worst eye” OR “worst eye” acuity of 20/200 or worse attributable to glaucoma.
glaucoma/blind eye” “Best eye” may fall into any of above stages.

CPSD/PSD ⫽ Corrected pattern standard deviation/pattern standard deviation; dB ⫽ decibel; HVF ⫽ humphrey visual field; MD ⫽ mean deviation.
JANUARY 2006

TABLE 2. Glaucoma Staging System Stage Definitions
and Severity Criteria (Stages 0 Through 5)
Stage 0: No or minimal defect/ocular hypertension
Does not meet any criteria for stage 1.
Stage 1: Early defect
MD ⱖ ⫺6.00 dB and at least one of the following:
A On pattern deviation plot, there exists a cluster of 3
or more points in an expected location of the
visual field depressed below the 5% level, at
least 1 of which is depressed below the 1% level
B Corrected pattern standard deviation/pattern
standard deviation significant at P ⬍ .05
C Glaucoma hemifield test “Outside Normal Limits”
Stage 2: Moderate defect
MD of ⫺6.01 to ⫺12.00 dB and at least one of the following:
A On pattern deviation plot, greater than or equal to
25% but fewer than 50% of points depressed
below the 5% level, and greater than or equal to
15% but fewer than 25% of points depressed
below 1% level
B At least 1 point within central 5° with sensitivity of
⬍15 dB but, no point within central 5° with
sensitivity of ⬍0 dB
C Only 1 hemifield containing a point with sensitivity
⬍15 dB within 5° of fixation
Stage 3: Advanced defect
MD of ⫺12.01 dB to ⫺20.00 dB and at least one of the
following:
A On pattern deviation plot, greater than or equal to
50% but fewer than 75% of points depressed
below the 5% level and greater than or equal to
25% but fewer than 50% of points depressed
below 1% level
B Any point within central 5° with sensitivity of ⬍0 dB
C Both hemifields containing a point(s) with sensitivity
⬍15 dB within 5° of fixation
Stage 4: Severe defect
MD of ⫺20.00 dB and at least one of the following:
A On pattern deviation plot, greater than or equal to
75% of points depressed below the 5% level
and greater than or equal to 50% of points
depressed below 1% level
B At least 50% of points within central 5° with
sensitivity of ⬍0 dB
C Both hemifields containing greater than 50% of
points with sensitivity ⬍15 dB within 5 degrees
of fixation
Stage 5: End-stage disease
Unable to perform Humphrey visual fields in “worst eye”
attributable to central scotoma or “worst eye” visual acuity
of 20/200 or worse attributable to primary open-angle
glaucoma. “Best eye” may be any stage.
Staging definitions are based on Humphrey visual field printouts.
dB ⫽ Decibels; MD ⫽ mean deviation.
VOL. 141, NO. 1 GLAUCOMA STAGING SYSTEMS
modified the Bascom Palmer GSS to reflect the full
range of severity in the development of glaucoma, from
minimal and early visual field loss to blindness from
end-stage disease. Visual acuity test results were added
for determining end-stage categorization.
Secondary modifications were then made to ensure
that the choices of visual field parameter, and their
respective threshold values within each stage, were
consistent with typical visual field progression patterns.
In particular, visual acuity was found to be an important
addition in assigning end-stage categorization. To facil-
itate ease of GSS use, staging tables were created and
were customized for various types of HVF, including 10
to 2, 24 to 2, and 30 to 2.
● SIX POINT SYSTEM: The final GSS is composed of six
stages: stage 0 (normal visual field), stage I (early), stage
II (moderate), stage III (advanced), stage IV (severe),
and stage V (end-stage). Staging criteria are based
mainly on the HVF, with MD as the primary measure.
There are three subdomains for adjustments, depending
on CPSD/PSD and hemifield test results for stages 0 and
I, the numeric (dB) plot for stages II through IV, and the
pattern deviation plot for stages I through IV. Stage V
classifications are on the basis of poor visual acuity and
inability to perform visual field testing as a result of
severe loss of field.
● GSS VALIDATION: For validation purposes, the GSS
was pretested with the review of 30 patient charts at
Duke University, Durham, North Carolina. The GSS
was used as a system for retrospectively assigning stage
to all Humphrey visual fields within the 30 charts.
Ambiguities that arose in stage assignment during the
pretest were resolved through appropriate modifications
to the GSS parameters and decision rules. The GSS was
then finalized and further evaluated for being able to
uniquely and unambiguously stage a new group of 68
glaucoma charts at six centers in the United States.
Patients were staged on the basis of the worst-eye visual
field score. The instrument was able to classify all
glaucoma patients from OHT to end-stage disease.
RESULTS
GSS DECISION RULES WERE CREATED TO DETERMINE
which parameters should be used to assign patients to
stage 0 through stage V. MD was decided on as the
primary measure for assigning stages 0 through IV, with
three additional criteria for stage adjustments depending
on the CPSD/PSD and hemifield test for stages 0
through I, dB plot for stages II through IV, and pattern
deviation plot for stages I through IV (Table 2).
Additional criteria were included on the basis of deci-
sion rules, which apply to both the GSS stage definitions
27
 TABLE 3. Glaucoma Staging System Based on Octopus Visual Field (Stages 0 through 5)
Stage Octopus MD Score
Stage 0 – Ocular hypertension/ ⱕ⫺0.8
earliest glaucoma
Stage 1 – Early glaucoma ⫺0.7 to ⫹4.4
Stage 2 – Moderate glaucoma ⫹4.5 to ⫹9.5
Stage 3 – Advanced glaucoma ⫹9.5 to ⫹15.3
Stage 4 – Severe glaucoma ⫹15.4 to ⫹23.1
Stage 5 – End-stage ⱖ ⫹23.2
glaucoma/blind
MD ⫽ Mean deviation; VF ⫽ visual field.
and the staging table. These decision rules are defined as
follows: if a patient meets the MD criteria for a
particular stage (stages I to IV) but fails to meet one of
the additional criteria for that stage, then the patient is
categorized in the preceding stage; if a patient meets the
MD criteria for a particular stage (stages I to IV), and
meets one of the additional criteria for a succeeding
stage, then the patient is categorized in the succeeding
stage; if a patient meets the MD criteria for a particular
stage (stages I to IV), and meets one or more of the
additional criteria for a preceding stage as well as one or
more of the criteria for a succeeding stage, then the
patient is categorized in the original stage on the basis of
MD criteria.
DISCUSSION
A MODIFIED VERSION OF THE BASCOM PALMER GSS WAS
developed to enable disease severity assignment of all
glaucoma patients on the basis of historical visual field
data recorded in patient charts, and to create a GSS that
can be used to retrospectively stage patients without
clinical judgment by a physician. The GSS was success-
fully applied by retrospectively assigning unambiguous
stages using HVFs from a total of 98 glaucoma charts
across seven centers in the United States. Automated
visual field testing with HVFs has been the gold stan-
dard for visual field testing for the past decade.28 –30 Our
GSS may be used to monitor the long-term progression
of disease in glaucoma patients; however, it requires
28 AMERICAN JOURNAL
Probability Plot/Pattern Deviation
The N-value in the Bebie Curve up to 50 (up to 100)
Where the Lower Confidence Limit Line Intersects
With the Patient Line
If Probability Plot in a less 51 and above (101 and above)
severe stage than
preliminary MD stage,
then qualify VF as that 41–50 (81–100)
LESS severe stage.
If Probability Plot is in a 31–40 (51–80)
more severe stage
preliminary MD stage, 21–30 (41–50)
then qualify VF at a
MORE severe stage.
If no intersection of 11–20 (21–40)
patient line and lower
confidence limit, must ⱕ10 (ⱕ20) OR Octopus VF not possible
qualify VF as one stage attributable to scotoma in “worst eye”
MORE than preliminary OR “worst eye” acuity of 20/200 or
MD score. worse attributable to glaucoma
further validation by comparison to other systems,
including those in use outside the United States, to
determine its ultimate applicability.
Despite the modifications made to the Bascom Palmer
GSS and the improved staging across all stages of
glaucoma these modifications allowed, there are several
limitations to our new GSS and its application. This
GSS is based primarily on visual field criteria and does
not consider other clinical factors that may be used to
assess progression of disease such as optic nerve head
examination, nerve fiber analyzer, or disk photographic
findings. Thus, the GSS serves as a practical tool for
assigning severity categories but is not necessarily a
complete proxy for glaucoma disease progression.11 Ad-
ditionally, because the developed GSS requires use of
HVFs, this restricts the population for which the GSS
can be used. In line with this, it is likely that the stages
do not represent equal intervals in the progression of
glaucoma; in other words, there may be more damage
occurring when a patient progresses from stage 0 to stage
I, compared with moving from stage III to stage IV. We
also do not know the functional significance of these
stages of disease in terms of vision-related functioning
and quality of life.
The need for a standardized GSS, nevertheless, re-
mains pressing. We chose to use a modification of the
Bascom Palmer system because it is one with which the
glaucoma community in the United States has at least
passing familiarity, is easy to score, and is currently used
clinically (as opposed to only in trials). With a GSS in
place, it becomes possible to do many analyses, includ-
OF OPHTHALMOLOGY JANUARY 2006
ing measuring the costs of treatment at each stage.
Direct costs for the treatment of open-angle glaucoma
include (1) patient visits to the ophthalmologist, (2)
surgery for glaucoma, (3) medication use, (4) visual field
examinations, and (5) other related services (for exam-
ple, gonioscopy, optic disk photographs, nerve fiber
thickness analysis, and IOP diurnal curve testing).
Despite the widespread applicability of this modified
system as a tool to retrospectively assign disease severity
and analyze progression retrospectively, this GSS does
not account for the association between IOP and disease
progression. Determination of patient stage may allow
modification of patient treatment regimens, but IOP and
other individual patient factors must be included in such
treatment decisions.
Because glaucoma is often asymptomatic in its earlier
stages, this may result in delayed diagnosis followed by
increased medical vigilance in care through the later
stages of the disease. It is important, therefore, to realize
the value of early diagnosis accompanied by early
treatment, as research has found that moderate or
advanced visual field losses found in glaucoma patients
at the time of diagnosis is a strong predictor of disease
progression leading to blindness.31 Using the modified
GSS, a retrospective analysis of 151 patient charts from
12 sites was conducted in the United States.32 This
study found the estimated average annual direct cost of
treatment to range from $623 per patient with early-
stage POAG to $2511 per patient with end-stage dis-
ease.19,32 Medication costs were found to be the largest
portion of total direct healthcare costs at each stage.
The transformed Bascom Palmer GSS was further
modified for use in European countries, such as Germany
and Austria, where patients’ visual field testing ordi-
narily is conducted with Octopus perimeters. A panel of
glaucoma specialists from the United States and Ger-
many used formulas from published literature to convert
the Humphrey threshold values to Octopus values
(Table 3).32,33 This modified GSS was applied to a
retrospective chart review of 194 patients with glau-
coma among 16 sites in Austria, France, Germany, Italy,
and the United Kingdom.33,34 In this study, as in the US
study, increasing use of healthcare resource utilization
was found at each succeeding stage of glaucoma, accom-
panied by increasing treatment costs.
This staging system has already demonstrated its
usefulness in analyzing the breakdown of costs associ-
ated with different stages of disease progression in
patients with POAG. It may provide an even greater
service in providing a new standard for diagnosing, monitor-
ing, and cost-effectively treating this devastating disease. If
early treatment can considerably delay disease progression
and save costs, then this new staging system may serve as a
model to encourage early diagnosis and careful, standardized
monitoring of disease progression, accompanied by a stage-
by-stage strategy for treatment.
VOL. 141, NO. 1 GLAUCOMA STAGING SYSTEMS
REFERENCES
1. American Academy of Ophthalmology GP. Primary open-
angle glaucoma. Preferred practice pattern. San Francisco:
American Academy of Ophthalmology, 2000:1–36.
2. Glick H, Brainsky A, McDonald RC, Javitt JC. The cost of
glaucoma in the United States in 1988. Chibret Int J
Ophthalmol 1994;10:6 –12.
3. Kobelt-Nguyen G, Gerdtham UG, Alm A. Costs of treating
primary open-angle glaucoma and ocular hypertension: a
retrospective, observational two-year chart review of newly
diagnosed patients in Sweden and the United States. J
Glaucoma 1998;7:95–104.
4. Distelhorst JS, Hughes GM. Open-angle glaucoma. Am Fam
Physician 2003;67:1937–1944.
5. Riordan-Eva P, Vaughan DG. Eye. In: Tierney LM Jr,
McPhee SJ, Papadakis MA, editors. Current medical diag-
nosis and treatment. New York: Lange Medical Books/
McGraw-Hill, 2001:185–216.
6. Alward WL. Medical management of glaucoma. N Engl
J Med 1998;339:1298 –1307.
7. Infeld DA, O’Shea JG. Glaucoma: diagnosis and manage-
ment. Postgrad Med J 1998;74:709 –715.
8. Barnebey H. Screening for glaucoma. In: Fihn SD, DeWitt
DE, editors. Outpatient medicine. Philadelphia: Saunders,
1998:119 –120.
9. Advanced Glaucoma Intervention Study. 2. Visual field test
scoring and reliability. Ophthalmology 1994;101:1445–1455.
10. Katz J. Scoring systems for measuring progression of visual
field loss in clinical trials of glaucoma treatment. Ophthal-
mology 1999;106:391–395.
11. Chen PP, Park RJ. Visual field progression in patients with
initially unilateral visual field loss from chronic open-angle
glaucoma. Ophthalmology 2000;107:1688 –1692.
12. Musch DC, Lichter PR, Guire KE, Standardi CL. The
Collaborative Initial Glaucoma Treatment Study: study de-
sign, methods, and baseline characteristics of enrolled pa-
tients. Ophthalmology 1999;106:653– 662.
13. Mills RP, Drance SM. Esterman disability rating in severe
glaucoma. Ophthalmology 1986;93:371–378.
14. Hodapp E, Parrish RK, Anderson DR. Clinical decisions in
glaucoma. St. Louis: Mosby, 1993:53.
15. Katz J, Sommer A, Gaasterland DE, Anderson DR. Compar-
ison of analytic algorithms for detecting glaucomatous visual
field loss. Arch Ophthalmol 1991;109:1684 –1689.
16. Anderson DR. Automated static perimetry. St. Louis: Mosby,
1995.
17. Janz N, Wren PA, Group CS. Implementing quality of life
in a clinical trial: The Collaborative Initial Glaucoma
Treatment Study. In: Anderson DR, Drance SM, editors.
Encounters in glaucoma research 3: how to ascertain
progression and outcome. New York: Kugler Publications,
1996:45– 62.
18. American Medical Association. Guides to the Evaluation of
Permanent Impairment. Chicago: American Medical Asso-
ciation, 1984:141–151.
19. Quigley HA. Identification of glaucoma-related visual field
abnormality with the screening protocol of frequency dou-
bling technology. Am J Ophthalmol 1998;125:819 – 829.
29
20. Evans C. The use of consensus methods and expert panels in
pharmacoeconomic studies. Practical applications and method-
ological shortcomings. Pharmacoeconomics 1997;12:121–129.
21. Torrance GW, Blaker D, Detsky A, et al. Canadian guide-
lines for economic evaluation of pharmaceuticals. Canadian
Collaborative Workshop for Pharmacoeconomics. Phar-
macoeconomics 1996;9:535–559.
22. Normand SL, McNeil BJ, Peterson LE, Palmer RH. Eliciting
expert opinion using the Delphi technique: identifying per-
formance indicators for cardiovascular disease. Int J Qual
Health Care 1998;10:247–260.
23. Lomas J, Anderson G, Enkin M, Vayda E, Roberts R,
MacKinnon B. The role of evidence in the consensus
process. Results from a Canadian consensus exercise.
JAMA 1988;259:3001–3005.
24. Frances A, Kahn D, Carpenter D, Frances C, Docherty J. A
new method of developing expert consensus practice guide-
lines. Am J Manag Care 1998;4:1023–1029.
25. Wilkinson CP. Evidence-based medicine regarding the pre-
vention of retinal detachment. Trans Am Ophthalmol Soc
1999;97:397– 404; discussion 404 – 406.
26. Pearson SD, Margolis CZ, Davis S, Schreier LK, Sokol HN,
Gottlieb LK. Is consensus reproducible? A study of an
algorithmic guidelines development process. Med Care 1995;
33:643– 660.
27. Katz J, Congdon N, Friedman DS. Methodological variations in
estimating apparent progressive visual field loss in clinical trials of
glaucoma treatment. Arch Ophthalmol 1999;117:1137–1142.
30 AMERICAN JOURNAL
28. Ellish NJ, Higginbotham EJ. Evaluating a visual field screen-
ing test for glaucoma: how the choice of the gold standard
affects the validity of the test. Ophthalmic Epidemiol 2001;
8:297–307.
29. Burnstein Y, Ellish NJ, Magbalon M, Higginbotham EJ.
Comparison of frequency doubling perimetry with Hum-
phrey visual field analysis in a glaucoma practice. Am J
Ophthalmol 2000;129:328 –333.
30. Mills RP, Barnebey HS, Migliazzo CV, Li Y. Does saving time
using FASTPAC or suprathreshold testing reduce quality of
visual fields? Ophthalmology 1994;101:1596 –1603.
31. Oliver JE, Hattenhauer MG, Herman D, et al. Blindness and
glaucoma: a comparison of patients progressing to blindness
from glaucoma with patients maintaining vision. Am J
Ophthalmol 2002;133:764 –772.
32. Lee PP, Walt JG, Doyle JJ, et al. A multi-center, retrospec-
tive pilot study of resource utilization and costs associated
with severity of disease in glaucoma. Arch Ophthalmol 2005;
Forthcoming.
33. Zeyen T, Roche M, Brigatti L, Caprioli J. Formulas for
conversion between Octopus and Humphrey threshold val-
ues and indices. Graefes Arch Clin Exp Ophthalmol 1995;
233:627– 634.
34. Traverso CE, Walt JG, Kelley SP, et al. A multinational
retrospective study of resource utilization and costs associated
with glaucoma disease progression in Europe. Br J Ophthal-
mol 2005; Forthcoming.
OF OPHTHALMOLOGY JANUARY 2006