Skip to main content

Table 2 Key model inputs

From: Cost-effectiveness of fluocinolone acetonide implant (ILUVIEN®) in UK patients with chronic diabetic macular oedema considered insufficiently responsive to available therapies

 

Base case

Low value

High value

Source

Patients characteristics

 Starting age – phakic

63.80

56.24

71.36

ICE-UK

 Starting age – pseudophakic

68.50

66.90

70.10

ICE-UK

 Proportion of males

61.4%

54.0%

68.5%

ICE-UK

 Percentage of patients with DMO affecting both eyes at baseline

76.6%

70.0%

82.6%

ICE-UK

 Percentage of FE with a cataract at baseline

27.2%

18.1%

37.3%

ICE-UK

 Percentage of pseudophakic FE at baseline

52.6%

45.1%

60.1%

ICE-UK

 Baseline distribution of BCVA levels

ICE-UK

Probabilitiesb

 Probability of developing DMO in the fellow eye, per cyclea

5.4%

2.4%

9.7%

ICE-UK

 Probability of developing a cataract in the FE

6.8%

4.7%

9.2%

FAME data, assumed to be the same as for laser photocoagulation [17]

 Probability of retreatment with FAc implant

35.2%

24.5%

48.9%

Based on FAME data on ≥15 letter improvement in BCVA [17]

 Probability of retreatment with dexamethasone

66.7%

48.2%

82.8%

Mastropasqua et al., 2015 [37]

 Probability of receiving study treatment in the FE

30.4%

22.2%

39.1%

ICE-UK

 Probability of cataract surgery – change of intercept used within the logit model

Logit model on FAME data depending on the treatment phase and BCVA level

Mortality

 Relative mortality risk (diabetes)

1.95

1.64

2.33

Preis et al. – 2009 [38]

 Relative mortality risk (DMO)

1.23

1.16

1.31

Christ et al. – 2008 [39], Ranibizumab Technology Appraisal Guidance 2013 Section 13.4, page 13. [40]

Transition probabilities

 Transition Matrices between different BCVA levels for FAc implant and laser photocoagulationc

Multinomial logistic model applied to patient-level data from the ICE-UK study for pseudophakic patients and the full study population.

 Odds -ratio of BCVA improvement for dexamethasone vs. FAc implantb

0.90

0.60

1.43

Calibration method, based on the NMA results for mean change in BCVA letter score from baseline to 24 months for pseudophakic population with chronic DMO

 Odds-ratio of BCVA improvement for anti-VEGFs vs. laser photocoagulationb

2.36

2.15

2.60

 Probability of worsening to lower BCVA level during the end-of-life phaseb

3.5%

2.3%

5.0%

NICE ERG report for aflibercept [25]

Usual care treatment mix

 Laser photocoagulation

28.3%

20.3%

38.2%

ICE-UK, low and high value defined for laser, for the rest of treatments calculated proportionally to the base case scenario

 Ranibizumab 0.5 mg/injection

62.6%

69.6%

53.9%

 Bevacizumab 1.25 mg/injection

9.1%

10.1%

7.8%

 Aflibercept 2 mg/injection

0%

0%

0%

Number of drug administrations per treatment regimen

 FAc 0.2 μg/day implant [per 3 years]

1

FAc 0.2 μg/day implant SPC [9]

 Laser photocoagulation [per year]

0.98

0.74

1.23

NICE ERG report of aflibercept [25]

 Usual Care [per year]

1.68

1.31

2.05

ICE-UK

 Dexamethasone [per 6 months]

1

Dexamethasone SPC, conservative assumption [8]

Drug costs

 FAc 0.2 μg/day implant

£5500.00

Alimera Science

 Ranibizumab

£551.00

£330.60

DM + D

 Bevacizumab

£242.66

£145.60

BNF (January 2017)

 Aflibercept

£816.00

£489.60

BNF (January 2017)

 Dexamethasone

£870.00

£522.00

BNF (January 2017)

Health state utilities

 Utilities depended on the BCVA levels

According to the study by Czoski-Murray et al. 2009 [24], worse BCVA level was related to lower health state utility. Both eyes were considered when determining utility depending on visual acuity: values varied from 0.8559 for both eyes with the highest possible visual acuity, to 0.2634 for both blind eyes. Baseline utility in the model was estimated to be 0.63 and 0.66 in the pseudophakic and phakic population, respectively.

Utility decrements

 Retinal detachment repair

0.130

0.079

0.181

Evidence Review Group report for aflibercept submission (p. 82) [25]

 Vitreous haemorrhage

0.020

0.012

0.028

 Disutility for anxiety associated with injections applied during treatment with anti-VEGFs

0.071

0.043

0.099

UK population norms EQ-5D [41]

 Percentage of patients with anxiety while on treatment with anti-VEGFs

0.173

0.133

0.218

Senra et.al. 2017 [26]

  1. aFor patients with DMO affecting the study eye only
  2. bProbabilities fixed over time in the model
  3. cProbabilities differentiated between efficacy and maintenance phase, i.e. 0 to 3 months or after 3 months
  4. Anti-VEGF Anti-Vascular Endothelial Growth Factor, BCVA Best-Corrected Visual Acuity, BNF British National Formulary, DM + D Dictionary of Medicines and Devices, DMO Diabetic Macular Oedema, ERG Evidence Review Group, FAc Fluocinolone Acetonide, FE Fellow Eye, ICE-UK ILUVIEN Clinical Evaluation-UK, NICE National Institute for Health and Care Excellence, NMA Network Meta-Analysis, SPC Summary of Product Characteristics