Skip to main content

Table 3 PSA distributions and parameters, base case values

From: A cost-effectiveness analysis model of Preventicus atrial fibrillation screening from the point of view of statutory health insurance in Germany

Original name

Type

Param 1

Param 2

Base Case

Description

General

 

Undetected AF in % detected AF

Beta

37.9

962.1

0.038

Assumption (prevention, 1/3

ratio undetected AF to detected AF)

Sensitivity Preventicus screening b

Beta

229

21

0.92

from DETECT AF [31]

Specificity Preventicus screening b

Beta

342.624

1.376

0.99

from DETECT AF [31]

Positively validated screening results (after Holter ECG) b

Beta

9.9875

1.7625

0.85

Adoption acc. to Wachter et.al [44].

Prevention

 Marcumar proportion in OAC medication

Triangular

0.05 / 0.2 / 0.5

0.29

 

 Increased stroke rate with AF without prevention

Normal

4.2

0.235

4.20

mean: Wolf et.al [40]. SD: assumption

 Reduction of stroke rate through prevention b

Normal

0.686

0.05

0.70

Assumptions, based on Hart [23] and López-López [26]

Strokes and Mortality

 Stroke rate normal population

Normal

1.00

0.14

1.00

SD according to Kolominisky-Rabas [49]

 Frequency of recurrent stroke

Beta

52.48

275.52

0.160

Hardie [48]: Perth registry

 Mortality, year 1 (SMR)a

Normal

3.7

0.3

3.7

Bromum-Hansen [46], year 0–1

 Mortality, subsequent years (SMR) a

Normal

1.92

0.13

1.92

Bronnum-Hansen [46], years 2–5

 Mortality, recurrence (SMR factor)

Triangular

1.0 / 1.5 / 2.0

1.5

Assumption according to Hardie [48]

 Stroke costs, year 1

Gamma

434.0277

0.02344

18,517

Kolominsky-Rabas [3] / variation

coefficient derived from lifetime,

2006 prices

 Stroke costs, subsequent years

Gamma

434.0277

0.07922

5479

discounted costs, 2006 prices

 Increased mortality, AF subpopulations

Normal

25&

5%

25%

Wolf, Mitchel [52]

Frequency of bleeding

 Severe bleedings, without OAC:

Normal

0.023

0.00153

0.023

Friberg [29, 50], Granger: Apixaba n[29]

 Severe bleedings, with OAC

Normal

0.039

0.0026

0.039

Mix of VKA and NOAC

 Less severe bleedings, without OAC

Normal

0.08256

0.00242

0.0823

Assumption: ratio with/without same as with severe bleeding

 Less severe bleedings, with OAC

Normal

0.14

0.0041

0.14

Krejczy [53],Grange r[29],

Cerebral haemorrhages

 Increase in the incidence of brain haemorrhage through VKA

Normal

4.00

0.22

4.00

Assumption, SD analogous to

Wolf [40]

 Reduction in cerebral haemorrhages through NOAC (compared with VKA)

Beta

24.34211

35.02890

0.41

Chatterjee [51], suppl., eFigure 3

Mortality cerebral haemorrhages

Beta

37.422

39.578

0.486

Fang [54]

QALYs

 Age decrement (per year)

Normal

−0.00029

0.0000225

−0.00029

Sullivan [55]

 with AF

Beta

33.82

7.93

0.810

Gauthier: HTA Canada [56] citng

Sullivan et.al [57, 58]

 with AF: reduction factor with VKA medication b

Uniform

0.953

1.000

0.987

Shah & Gage [59] (range), Gage et.al [60].

 with AF: reduction factor with NOAC

medication b

Uniform

0.990

0.998

0.994

Shah & Gage [59] (range), O’Brien, Gage [61]

 after cerebral haemorrhage, year 1

Beta

11.41332

17.19149

0.399

Golicki [62] (ICD I61 result)

 after stroke, year 1

Beta

378.6275

304.81567

0.554

Golicki [62] (ICD I63 result)

 after stroke: reduction in the year of the event

Normal

0.103

0.008

0.103

Gauthier: HTA Canada [56], citing

Sullivan et.al [57].

 after stroke or cerebral haemor- rhage, subsequent years b

Uniform

0

0.5

0.12

Shah & Gage [59] (range)

 Decrement for severe bleeding

Normal

−0,092

0.010

−0,092

Gauthier: HTA Canada [56], Freeman [63], Thomson [64]

 Decrement for less severe bleeding

Normal

−0.013

0.001

−0.013

Gauthier: HTA Canada [56], Freeman [63], O’Brien [61]

  1. a Values for men (75 years; if age-dependent) b Different average in EV (expected value) calculation
  2. Parameters: by distribution type (Param1, Param2): Beta distribution (alpha, beta), Gamma distribution (alpha, lambda), Normal distribution (mean, standard deviation), Uniform distribuiton (min,max), Triangular distribution (min/ likeleliest/max)