Evidence Table 24Economic Evidence

Study detailsPatientsInterventionsOutcome measuresEffect sizeComments
Kymes et al., 200680
USA

Economic analysis:
Cost Utility analysis

Study design
Decision analysis*

Time horizon:
Life-time

Discount rates:
Costs: 3%
Effects: 3%
Patient group: patients between 40 and 80 with OHT (IOP between 24mm Hg and 32mm Hg in one eye and between 21mm Hg and 32mm Hg in the other eye, and normal VF and optic disk in both eyes)

All patients *
N: 1636 N with glaucoma: 0
M/F: 705/931
Mean IOP at baseline (SD): 24.9 (2.7)
Ethnic origin: Asian 14, African American 408, Hispanic 59, White 1137, Other 18
Drop outs: 228
Intervention 1:
Treat no one

Intervention 2:
Treat if IOP≥24 mm Hg and annual risk of developing POAG ≥5%

Intervention 3:
Treat if IOP≥24 mm Hg and annual risk of developing POAG≥2%

Intervention 4:
Treat everyone with IOP≥24 mm Hg
Mean QALYs gained per patient (determined by progression and development of cataract)intervention 1: 13.537
intervention 2: 13.559
Intervention 3: 13.588
Intervention 4: 13.587
p value: NR
Funding:
National Eye Institute; National Institutes of Health; Merck Research Laboratories; Pfizer, Inc; Research to prevent Blindness, Inc.

Limitations:
Treatment was a mixture

Notes:
* Based on the Ocular Hypertension Treatment Study
Mean total life-time cost per patient
2006 US$, cost of medication, cataract surgery, cost associated with POAG progression, cost of blindness. Societal perspective
Intervention 1: $4,006 (£ 2,476)
Intervention 2: $4,086 (£ 2,525)
Intervention 3: $5,305 (£ 3,278)
Intervention 4: $11,245 (£ 6,949)
p value: NR
Cost-effectiveness
Cost per QALY gained
Int 2 vs Int 1: $3,670 (£2,268)
Int 3 vs Int 2: $42,430 (£ 26,222)
Int 4 vs Int 3: Int 4 is dominated
Sensitivity analysis
One-way SA

Probabilistic sensitivity analysis (Monte Carlo simulation)
Sensitive factors were: incidence of POAG without treatment (if less than 1.5%, Int 2 more cost-effective), proportion of people with OHT to be treated, reduction in risk because of medical treatment (if <30% Int 2 more cost-effective), annual probability of progression of a POAG stage, cost of one medication, increased annual risk of cataract surgery, utility loss in stage 1 POAG.

At the £20,000/QALY threshold, both Int 1 and Int 3 have a 30% probability of being the most cost-effective, while Int 2 has a 40% probability.
Stewart et al., 2008144
USA

Economic analysis:
Cost Utility

Study design
Decision model based on the Ocular Hypertension Treatment Study and Early Manifest Glaucoma Trial.

Time horizon:
5 years

Discount rates:
Costs: 3%
Effects: 0%
Patient group: patients with ocular hypertension from the Ocular Hypertension Treatment Study.Intervention 1:
No treatment

Intervention 2:
1 medication for the first 2 years.
In the last 3 years:
-

1.4 medications in non-progressing patients

-

2 medications in 75% of patients that progressed

-

3 medications in 15% of patients that progressed

Medications could be Prostaglandin Analogues, Beta-Blockers or Brimonidine.
QALYsIntervention 1: 4.45
Intervention 2: 4.48
p value: NR
Funding:
NR (one of the authors was employed by Pfizer).

Limitations:
-

other relevant outcomes were omitted (e.g. blindness)

-

limited applicability (US cost data)

Mean cost per patient
2007 US $, Cost of visits, medications, and tests (central corneal thickness, gonioscopy, IOP, optic disc imaging, refraction, automated visual field).
Intervention 1: $ 2,467 (£ 1,525)
Intervention 2: $ 5,001 (£ 3,091)
p value: NR
Cost-effectiveness
incremental cost per QALY gained
Intervention 2 vs Intervention 1 $84,467 (£ 52,200)
Sensitivity analysis
One-way SA (risk of progression is changed according to risk factors)

DSA (costs are changed by + or −10%)
Intervention 2 is cost-effective in one of the following situations:
-

vertical cup to disc ratio plus 0.7 or more

-

corneal thickness plus 80μm

No change in results
Bernard 200310
France

Economic analysis:
cost-effectiveness

Study design
Decision analysis*

Time horizon:
2 years and 3 years

Discount rates:
Costs: 3%
Effects: 0%
Patient group: patients newly diagnosed with open angle glaucoma or ocular hypertension (IOP>21 mmHg and no optic nerve damage).Intervention 1:
First-line treatment with a beta-blocker followed by usual care for patients who switch therapy.

Intervention 2:
First-line treatment with latanoprost 0.005% followed by usual care for patients who switch therapy.
Proportion of patients remaining on first-line treatment
(after 1 year; after 2 years)
Int 1: 46%; 29%
Int 2: 82%; 73%
p value: NR
Funding:
Pharmacia Corporation, Peapack, USA

Limitations:
Clinical outcomes were not compared to other studies.
Limited time horizon.

Additional outcomes:
Proportion of patients undergoing surgery (7% for Int 1and 3% for Int 2 over 3 years)

Notes:
* Model inputs were taken from chart reviews.
** Calculated by NCC-AC from incremental cost per IOP-controlled day gained.
Mean time spent on the initial therapy (months)Int 1: 13.4
Int 2: 20.5
p value: <0.0001
Mean number of therapies used over 2 years (CI)Int 1: 2.08 (± 0.94)
Int 2: 1.38 (± 0.74)
p value: <0.0001
Mean IOP-controlled days (days)
(over 2 years; over 3 years)
Int 1: 653; 973
Int 2: 703; 1047
p value: <0.0001
Mean cost per patient
(over 2 years; over 3 years)
(2002 Euro Cost of management, treatment, surgery)
Int 1: € 539 (£ 366); € 817 (£ 556)
Int 2: € 580 (£ 394); € 844 (£ 574)
p value: <0.0001
Cost-effectiveness
Incremental cost per IOP-controlled year gained per patient** (over 2 years; over 3 years)
Int 2 vs Int 1: € 299 (£204); € 131 (£ 88)
Sensitivity analysis
One-way sensitivity analysis
The results were sensitive to time to therapy failure, bottle duration, assessment visit schedule for patients who switched treatments, surgical rates, and cost of surgery.
Day 200431
USA

Economic analysis:
Cost consequences

Study design
Retrospective cohort study

Duration of follow-up:
6 months

Discount rates:
Costs: NA
Effects: NA
Patient group: adult patients with COAG or OHT in at least one eye whose records were stored in large glaucoma practices in the USA.

All patients
N: 1182 (1 eye randomly chosen evaluated)
N with glaucoma: 922
M/F: 510/672
Drop outs: 0

Group 1
N: 487 N with glaucoma: 361
Age (mean±SD): 64.4±14.3
M/F: 219/268
Ethnic origin: Caucasian 325, African-American 82, Asian 6, Hispanic 12, Other and Unknown 62

Group 2
N: 490 N with glaucoma: 401
Age (mean±SD): 67±13.9
M/F: 207/283
Ethnic origin: Caucasian 303, African-American 109, Asian 1, Hispanic 8, other and unknown 69

Group 3
N: 205 N with glaucoma: 160
Age (mean±SD): 68.9±12.8
M/F: 84/121
Ethnic origin: Caucasian 114, African-American 30, Asian 1, Hispanic 5, Other and Unknown 55
Group 1:
Beta-blockers monotherapy as first or second line (71% with Timolol).

Group 2:
Latanoprost monotherapy as first or second line.

Group 3:
Bimatoprost monotherapy as first or second line.
Risk ratio to discontinue therapy compared to group2Group 1: 1.15 (95% CI: 1.03, 1.27)
Group 2: 1
Group 3: 1.08 (95% CI: 1.01, 1.16)
p value: 0.02
Funding:
Pfizer, Inc.

Limitations:
No differentiation between treatments used as a first- or second- choice.
The short follow-up does not allow including the costs associated with disease progression (e.g. surgery).
Mean IOP at baseline not reported.

Additional outcomes:
Main reasons for changing or adding to current medication before 6 months of therapy were IOP not controlled and adverse events.
Patient visits were fewer with latanoprost (p=0.01).
The number of ocular adverse events was fewer with beta-blockers.
IOP at the last visit before the therapy is changed (mmHg±SD)Group 1: 17.9±3.7
Group 2: 17.3±3.9
Group 3: 18.0±3.6
p value: <0.0001
Mean cost per patient per 6months of therapy 2004 US$ Direct costs only: cost of drugs (average wholesale price) + visits and procedures resulting from adverse events as well. Cost of drug based on both eyes receiving treatment and assuming perfect complianceGroup 1: $ 119 (76+43) (£ 74)
Group 2: $ 154 (116+38) (£ 95)
Group 3: $ 164 (124+40) (£ 101)
p value: <0.0001 (drugs), p=0.07 (visits and procedures)
Cost-effectivenessNR
Sensitivity analysisNR
Goldberg 200648
USA

Economic analysis:
cost-effectiveness

Study design
Decision analysis based on RCT*

Time-horizon:
1 year

Discount rates:
Costs: NA
Effects: NA
Patient group: patients with POAG or OHT (IOP 22–34 mmHg) in at least one eye.

All patients*
N: 715
M/F: 307/408
Drop outs: 86
Ethnic origin: 583 non-black, 132 black

Group 1
N: 241
Age (mean): 61
M/F: 101/140
Drop outs: 27
Mean IOP at baseline: NR
Ethnic origin: 195 non-black, 46 black

Group 2
N: 474
Age (mean): 61.7
M/F: 206/268
Drop outs: 59
Mean IOP at baseline: NR
Ethnic origin: 388 non-black, 86 black
Group 1:
Timolol twice daily morning and evening as first-line.

Group 2:
One drop of Bimatoprost 0.03% once-daily in the evening as first-line.
Percentage of patients achieving target pressure (17mmHg) after 12 months.Group 1: 37%, 27%, 16%, 9%, 5%
Group 2: 58%, 47%, 31%, 21%, 12%
p value: <0.05
Funding:
Allergan, Inc.

Limitations:
The study assumes success is achieved after dual therapy and patients are perfectly compliant.
The study does not consider surgical treatment, adverse events or endpoints other than IOP (e.g. blindness).
Limited time horizon.

Notes:
*Higginbotham 200262. Data from another RTC excluded because it has a 3-month follow-up.
** calculated by NCC-AC according to costs and algorithm reported in the study.
Mean annual cost per patient**
2003 US$, (cost of initial and adjunctive medication based on average wholesale prices + cost of visits, if target pressure 17mmHg)
Group 1: $828 (£ 517), $ 896 (£ 559), $964 (£ 601), $1032 (£ 644), $1063 (£663).
Group 2: $1043 (£ 651), $1066 (£665), $1112 (£694), $1151 (£ 718), $1183 (£ 738).
p value: NR
Cost-effectiveness** Incremental cost per additional treatment successGroup 2 vs Group 1: $1024 (£ 639)
Sensitivity analysis
one-way sensitivity analysis
ICER was $850 (£ 530), $987 (£ 616), $992 (£ 619), $1714 (£ 1069) if target pressure was respectively16mmHg, 15mmHg, 14mmHg, 13mmHg.
Results were sensitive to the average wholesale prices (if branded Timolol was used, bimatoprost would become at least 30% more cost effective at target IOP 17), to changes in treatment success rates, to the adjunctive agent chosen (if brimonidine, bimatoprost would be dominant).
Halpern 200254
USA

Economic analysis:
Cost consequences

Study design
Decision analysis based on a RCT (Netland 2001)

Duration of follow-up:
1 year

Discount rates:
Costs: NA
Effects: NA
Patient group: black patients with POAG or OHT.

All patients
N: 132
M/F: 56/76
Drop outs: 19

Group 1
N: 40
Age (mean): 62.3
M/F: 15/25
Drop outs: 7
Mean IOP at baseline: 25.8

Group 2
N: 43
Age (mean): 58.6
M/F: 18/25
Drop outs: 3
Mean IOP at baseline: 26.2

Group 3
N: 49
Age (mean): 62.6
M/F: 23/26
Drop outs: 9
Mean IOP at baseline: 25.3
Group 1:
Timolol 0.5%, one drop at 8 AM and at 8 PM as first-line.

Group 2:
Latanoprost 0.005% One drop at 8 PM plus placebo at 8 AM as first-line.

Group 3:
Travoprost 0.004% One drop at 8 PM plus placebo at 8 AM as first-line.
Mean IOP during the 1-year follow-up (mm Hg±SD)Group 1: 20.5±3.4
Group 2: 18.7±2.4
Group 3: 17.3±2.5
p value: <0.05 (group 1 and 2 vs 3)
Funding:
Alcon Research, Ltd.

Limitations:
It is not clearly stated if the costs of medication have been included.
It is not clear when the IOP at follow-up was measured.
Limited follow-up.

Notes:
*Calculated by averaging various algorithms that link IOP with visual field defect
** Inpatient costs: increased VFDS × mean number of hospitalisation per year due to severe visual field defect × average length of stay × cost per day as reimbursed by Medicare.
Outpatient costs: Medicare 2000 reimbursement values × increased VFDS
Mean increase in visual field progression rates*Group 2 vs Group 3: 19%
Group 1 vs Group 3: 27.5%
p value: Sig
Mean increase in annual cost per patient 2000 US$, inpatient and outpatient costs, based on the likelihood of increased Visual Field Defect Score (VFDS)**Group 2 vs Group 3: $170 (£ 108)
Group 1 vs Group 3: $ 247 (£ 156)
p value: NR
Cost-effectivenessNR
Sensitivity analysisNR
Rouland 2003125
France

Economic analysis:
Cost-effectiveness

Study design
decision analysis based on retrospective cohort study

Duration of follow up:
one year

Discount rates:
Costs: NA
Effects: NA
Patient group:
second-line adult patients with COAG or OHT (IOP>21 mmHg and no optic nerve damage) in at least one eye for whom treatment was changed or stopped, presenting in 37 centres in France.

All patients
N: 283 (549 eyes)*
N eyes with glaucoma: 425
Age (mean): 65±1.5
M/F: 155/128
Mean IOP at baseline: 20.0±4.3

Group 1
N: 209 eyes
Mean IOP at baseline: 19.5±3.9

Group 2
N: 90 eyes
Mean IOP at baseline: 19.3±4.7

Group 3
N: 39 eyes
Mean IOP at baseline: 20.9±3.7
Group 1:
Beta-blocker as a second-line treatment

Group 2:
Latanoprost as a second line treatment

Group 3:
Unfixed combination of Latanoprost+Timolol as a second line treatment
Mean IOP reduction per treated eye (mmHg)Group 1: 2.1
Group 2: 3.0
Group 3: 5.3
p value: 0.02 (group 1 vs group 2 only)
Funding:
Pharmacia corporation, Peapack, NJ, USA

Limitations:
Short follow-up
Clinical outcomes were not compared to other studies and RCTs.

Additional outcomes:
average number of days remaining on the same treatment (longer for Group 2 and 3)

Notes:
* other groups treated with CAI and other combinations not reported here as a CEA was not performed
** calculated by NCC-AC from data reported in the study
*** calculated by NCC-AC (different figures reported by authors)
Proportion of eyes remaining on the same second-line treatment after 1 yearGroup 1: 69%
Group 2: 84%
Group 3: 80%
p value: 0.0068 (group 1 vs group 2 only)
Mean annual cost per patient** (2001, Euros
direct costs: visits, medical procedures, drugs, surgery including trabeculectomy, trabeculoplasty, combined cataract- trabeculectomy, iridotomy, and 10% of cataract surgery) estimated from National Sources.
Group 1: € 179 (£ 124)
Group 2: € 273 (£ 189)
Group 3: € 329 (£ 228)
p value: <0.0001 (group 1 vs group 2 only)
Cost-effectiveness*** additional cost per 1 mmHg of control gained after 1 year of treatmentGroup 2 vs Group 1: £ 72
Group 3 vs Group 1: £33
Group 3 vs Group 2: £24
Sensitivity analysisNR
Rouland 2005126
France

Economic analysis:
cost-effectiveness

Study design:
decision analysis based on cohort study

Duration of follow-up:
2 years

Discount rates:
Costs: NR
Effects: NR
Patient group:
second-line adult patients with COAG or OHT (IOP>21 mmHg and no optic nerve damage) in at least one eye presenting in 37 centres in France.

All patients (eyes)
N: 498 (672 eyes)
N eyes with glaucoma: 511
Age (mean±SD):
64.8±12.9
M/F: 159/187
Drop outs: 152
Mean IOP at baseline ±SD: 20.1±4.1

Group 1
N eyes: 248 eyes
Mean IOP: 19.7

Group 2
N eyes: 112 eyes
Mean IOP: 19.9

Group 3
N eyes: 39 eyes
Mean IOP: 20.5
Group 1:
Beta-blocker as a second-line treatment

Group 2:
Latanoprost as a second line treatment

Group 3:
Unfixed combination of Latanoprost+Timolol as a second line treatment
Frequency of episodes of adverse eventsGroup 1: 116
Group 2: 21
Group 3: 3
p value: NR
Funding:
Pfizer

Limitations:
Short follow-up.
Clinical outcomes were not compared to other studies and RCTs.

Additional outcomes:
average number of days remaining on the same treatment (longer for Group 2)

Notes:
* other groups include combinations, not reported here
** calculated by NCC-AC from data reported in the study
*** calculated by NCC-AC
Relative risk of adverse events vs group 1 (95% CI)Group 1: 1.00 (0.996–1.004)
Group 2: 0.40 (0.16–0.64)
Group 3: NR
p value: NR
Proportion of eyes remaining on the same second-line treatment after 2 yearsGroup 1: 41%
Group 2: 62%
Group 3: 44%
p value: NR
Mean IOP reduction after 2 years per treated eye (mm Hg)Group 1: 2.6
Group 2: 3.3
Group 3: 4.4
p value: NR
Mean 2-year cost per eye**
(2003, Euros, direct costs: visits, medical procedures, drugs, surgery, 10% of cataract surgery)
Group 1: € 388 (£ 260)
Group 2: € 556 (£ 373)
Group 3: € 731 (£ 490)
p value: NR
Cost-effectiveness***
additional cost per 1 mmHg of control gained after 2 years of treatment
Group 2 vs Group 1: £162
Group 3 vs Group 1: £128
Group 3 vs Group 2: £106
Sensitivity analysisNR
Le Pen et al., 200582
France

Economic analysis:
Cost-utility

Study design
Decision analysis based on a Markov model

Time horizon:
5 years

Discount rates:
Costs: 5%
Effects: NR
Patient group:
patients with advanced POAG in five European countries.
Intervention 1:
Timolol 0.5% twice daily as first-line.

Intervention 2:
Latanoprost 0.005% once daily as first-line.

Intervention 3:
Travoprost 0.004% once daily as first-line.
Mean daily IOP over all visit days (mmHg)*Int 3 - Int 1: −1.3
Int 3- Int 2: −1.0
p value: <0.0001
Funding:
Alcon Laboratories Inc, USA.

Limitations:
Complicated third and fourth line strategies after disease progression were not considered. Limited time horizon. Clinical outcomes were not derived from a systematic search.
Calculations of QALYs and ICUR were dubious.

Additional outcomes:
Same outcomes reported for other countries (Austria, France, Germany, and the Netherlands). The results were consistent across countries.

Notes:
* data from Netland 2001110
** Calculated from an algorithm that links IOP with VFD
*** unclear calculation
****ICUR as reported in the study= €23,828 (£ 15,989)
Time without a VFD=disease progression over 5 years (years)**Int 1: 2.812
Int 2: 3.285
Int 3: 3.417
p value: NR
Patients experiencing a new visual field defect after 5 years of treatment**(%)Int 1: 72.8%
Int 2: 59.4%
Int 3: 55.7%
p value: NR
QALYs over 5 years***Int 1: 3.6001
Int 2: 3.6164
Int 3: 3.6210
p value: NR
Mean cost per patient over 5 years in the UK
2003 Euro (€ 1.5 = £1). Cost of drugs, visits, surgery, laser, taken from national sources (UK GP Research Database and BNF)
Int 1: € 790 (£ 530)
Int 2: € 1,041 (£ 698)
Int 3: € 993 (£ 666)
p value: NR
Cost-effectiveness
ICUR = incremental cost per QALY gained (2003 €) calculated from difference in costs and QALYs as reported above****
Int 3 vs Int 1: €10,150 (£6,767)
Latanoprost is dominated by Travoprost
Sensitivity analysis
Probabilistic SA based on a Monte Carlo simulation (variables included were the cut-off value adopted for defining stability, the utility loss associated with a new VFD and the cost of a stable and progressive patient).
Probability ICUR Int 3 vs Int 1 <45,000€/QALY is 98.8%.
Cottle & Begg, 198827 Canada

Economic analysis:
CEA

Study design
cohort study

Duration of follow-up:
12 months (mean)

Discount rates:
Costs: NA
Effects: NA
Patient group:
consecutive patients with newly diagnosed, untreated POAG (IOP => 21 mmHg in at least one eye, glaucomatous visual field loss).

All patients
N: 71 (130 eyes)
N with glaucoma: 71
Age (mean ± SD): 64 (±13.1)
M/F: 34/37
Drop outs: 0
Mean IOP at baseline (all eyes): 28.7 (± 6.13)
Ethnic origin: all white

Group 1
N: 85 eyes*

Group 2
N: 20 eyes*

Group 3
N: 10 eyes*

Group 4
N: 19 eyes*

Group 5
N: 8 eyes*
Group 1:
Timolol 0.25% (Beta-blocker)

Group 2:
Timolol 0.5% (Beta-blocker)

Group 3:
Dipivefrine 0.1% (Sympathomimetic)

Group 4:
Pilocarpine 2.0%

Group 5:
Pilocarpine 1.0%
Number of eyes controlled in terms of satisfactory IOPGroup 1: 39 (46%)
Group 2: 10 (50%)
Group 3: 8 (80%)
Group 4: 7 (37%)
Group 5: 5 (62%)
p value: NR
Funding:
IMS, Inc., supplied the costs of the drugs.
The study received a Grant 6610- 1272-42 from the National Health Research and Development Program, Department of National Health and Welfare, Canada

Limitations:
Very small sample size.
Some patients were included in more than 1 group.

Notes:
* the same eye could be included in more than one group when the treatment was changed
**calculated by NCC based on monthly costs and on the assumption that treating both eyes has the same cost of treating 1 eye (bottle is discarded anyway after 1 month).
Number of severe adverse reactionsGroup 1: 9 (11%)
Group 2: 0 (0%)
Group 3: 2 (20%)
Group 4: 2 (10%)
Group 5: 1 (12%)
p value: NR
Usefulness Quotient (number of patients whose condition was controlled with no severe adverse reaction divided by the number of patients who started on the treatment)Group 1: 0.39
Group 2: 0.50
Group 3: 0.60
Group 4: 0.36
Group 5: 0.50
p value: Not Sig
Mean annual cost per eye treated**
1982 Can $, mean wholesale cost per bottle of drug, included the medication discarded during the study and the surplus remaining at the end.
Group 1: $42 (£17)
Group 2: $50 (£21)
Group 3: $29 (£12)
Group 4: $13 (£5)
Group 5: $12 (£5)
p value: NR
Mean annual cost per eye treated
if 54 BNF prices are used.
Group 1: £44
Group 2: £36
Group 3: £46
Group 4: £30
Group 5: £32
Cost-effectiveness **
incremental cost per year per additional patient controlled without side effects
Group 1 and 2 dominated by Group 3 and 5.
Group 2 vs Group 1: $73 (£30)
Cost-effectiveness**
Incremental cost per year per additional patient controlled without side effects, calculated by NCC-AC using 54 BNF prices.
Group 1 dominated by 2.
Group 3 vs Group 2: £10.
Group 1 and 2 dominated by Group 5.
Sensitivity analysisNR
Stewart et al., 2002143
USA

Economic analysis:
cost-effectiveness

Study design
Retrospective cohort study

Duration of follow-up: up to 12 months

Discount rates:
Costs: NA
Effects: NA
Patient group: adult patients diagnosed with POAG or OHT in at least one eye previously prescribed a topical beta- blocker as monotherapy.

All patients
N: 148 (one eye from each subject)

Group 1
N: 37
Age (mean): 72.8 M/F: 16/21
Mean IOP at baseline: 20.9
Ethnic origin: 27 Caucasian, 10 Black

Group 2
N: 74
Age (mean): 75.2 M/F: 31/43
Mean IOP at baseline: 20.9
Ethnic origin: 42 Caucasian, 30 Black, 2 Hispanic

Group 3
N: 37
Age (mean): 76.4 M/F: 14/23
Mean IOP at baseline: 21.7
Ethnic origin: 24 Caucasian, 12 Black, 1 Hispanic
Group 1:
Switch from Beta- blocker to Latanoprost monotherapy

Group 2:
Beta-blocker + adjunctive therapy with Latanoprost once daily

Group 3:
Beta-blocker + adjunctive therapy with Brimonidine twice daily
Number of patients with therapeutic success (IOP decreased by 2 mm Hg or more)Group 1: 54% (20/37)
Group 2: 70% (52/74)
Group 3: 49% (18/37)
p value: 0.056
Funding:
NR

Limitations:
Short follow-up.
Retrospective study: possible selection bias.

Additional outcomes:
Treatment changes; number of visits; adverse events; difference in cost from beta-blockers to post-enrolment treatment.

Notes:
*calculated by NCC based on monthly cost
Mean IOP change from baseline to final follow-up visit (% change in IOP)Group 1: 2.8 (13.4%)
Group 2: 4.5 (21.5%)
Group 3: 4.6 (21.2%)
p value: 0.23 (on mean IOP change)
Mean annual cost per patient*
2001, US$
Average wholesale prices of medicines prescribed and reimbursement cost of visits and tests due to adverse events
Group 1: $644 (£401)
Group 2: $998 (£622)
Group 3: $1,274 (£794)
p value: 0.038 (for monthly cost)
Cost-effectiveness*
additional cost per 1 mmHg of change in IOP after 1 year of treatment
On the basis of %change in IOP Group 3 is dominated by Group 2.

Group 2 vs Group 1: $208 (£130)
Sensitivity analysisNR
Ainsworth & Jay, 19913
UK

Economic analysis:
cost analysis

Study design
RCT*

Duration of follow- up:
8 years

Discount rates:
Costs: none Effects: none
Patient group: consecutive patients of 8 ophthalmologists in 5 hospitals in Glasgow area newly diagnosed with POAG (untreated IOP of at least 26 mmHg on two occasions and field defect characteristics).

All patients
N: 104

Group 1
N: 51 (23 unilateral glaucoma)

Group 2
N: 53 (23 unilateral glaucoma)
Group 1:
Early trabeculectomy (within 4 weeks of diagnosis). Preliminary medical therapy is used if necessary to reduce the IOP to a safe level prior to surgery.

Group 2:
Conventional management: up to a maximum of three different topical or systemic drugs and late trabeculectomy if medical therapy has failed.
Mean cost per patient (unilateral** – bilateral glaucoma)
1989 GBP, cost of drugs plus 6% pharmacists’ prescription fee, outpatient visits, field tests, inpatient stay***, operation. Costs adjusted for mortality.
Group 1: £2,139 – £2,560
Group 2: £1,920 – £2,569
p value: NR
Funding:
NR

Limitations:
Population description missing.
Hospital length of stay after surgery could have decreased since time of study.

Notes:
*From Jay198865. In Jay 1988 fewer patients.

** Cost of unilateral glaucoma includes subsequent treatment of the fellow eye if applicable.

***average length of stay=7.6 days
Cost-effectivenessNR
Sensitivity analysisWhen the length of inpatient admission is reduced to 4 days or 1 day, early trabeculectomy becomes the less costly strategy.
4 days: Group 1 £1,780
 Group 2 £ 1,875
1 day: Group 1 £ 1,130
 Group 2 £ 1,405
Henson et al., 200360
UK

Economic analysis:
cost analysis

Study design
comparative study with historical control

Duration of follow- up:
3 years

Discount rates:
Costs: NR
Effects: NA
Patient group: suspect of having glaucoma

Group 1
N: 194

Group 2
N: 93
Group 1:
Patients referred to a group of accredited optometrists working within their own practices and subsequently referred to Manchester Royal Eye Hospital if meeting referral criteria.

Group 2:
Patients referred to the GP and then to Manchester Royal Eye Hospital
3-year cost of overall scheme
2001 GBP training of optometrists, fees to optometrist, audit, minus cost savings from non-referred cases (40%) to hospital and GP
Group 2 - Group 1: 13,426
p value: NR
Funding:
Manchester Health Authority

Limitations:
Cost of false negatives was not accounted for.

Additional outcomes:
if 23 patients per month are enrolled in the scheme of group 1, the cost saving is approximately £16 per patient.
Coast 199723
UK

Economic analysis:
Cost Analysis

Study design
RCT52,140,142

Perspective:
NHS and patients

Duration of follow-up:
1 year

Discount rates:
Costs: NA
Effects: NA
Patient group:
patients with glaucoma whose IOP was satisfactorily controlled with treatment; Snellen VA of 6/18 or better in both eyes, aged 50 or above

All patients
N: 405
Drop-outs: 2

Group 1
N: 204
Drop-outs: 9

Group 2
N: 201
Drop-outs: 4
Group 1:
Monitoring by ophthalmologists with a 10-month interval

Group 2:
Monitoring by optometrists, with a 6-month interval and referral to hospital when necessary.
Cost per glaucoma visit
1994 GBP
Cost of staff, consumables, overheads.
Group 1: 50
Group 2: 29
p value: NR
Funding:
South and West Research and Development Directorate, Avon Health and the International Glaucoma Association.

Limitations:
Optometrists were volunteers, therefore the findings cannot be generalised.
Effectiveness was not estimated.
Data on patients are missing.

Additional outcomes:
46 clinics per annum could be saved from a total of 1200 clinics.

Time and costs to the patients were lower in Group 2.
Annual full cost per patient
1994 GBP
Cost of staff, training of optometrists, consumables, referrals from optometrists (19% patients), and overheads.
Group 1: 60
Group 2: 77
p value: NR
Marginal annual opportunity cost per patient
1994 GBP. Cost of staff time.
Group 1: £15
Group 2: £25
p value: NR
Cost-effectivenessNA
Sensitivity analysisWhen time spent by optometrists with patients was 60 minutes rather than 35 minutes, the annual cost per patient was £124

When rate of referrals in group 2 was 50% lower or higher than baseline annual cost per patient in group 2 was respectively £68 and £87.

When follow up interval in group 2 was similar to group 1, the annual cost per patient in group 2 was £46.

If the caseload optometrists are willing to accept is 100 patients, the marginal opportunity cost per patient becomes £45.

From: Appendix D, Evidence tables

Cover of Glaucoma
Glaucoma: Diagnosis and Management of Chronic Open Angle Glaucoma and Ocular Hypertension.
NICE Clinical Guidelines, No. 85.
National Collaborating Centre for Acute Care (UK).
Copyright © 2009, National Collaborating Centre for Acute Care.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of National Collaborating Centre for Acute Care to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.