Skip to main content

Table 1 The baseline values for model parameters

From: From accuracy to patient outcome and cost-effectiveness evaluations of diagnostic tests and biomarkers: an exemplary modelling study

Input parameter

Value*

Range$

Distribution

Source

Probability of early death (within 30 days)

EuroSCORE risk equation

 

[37]

Probability death is due to fatal stroke event

13.2%

7.5 – 20.2%

Beta(14,92)

[25]

Probability that a stroke event is not fatal

80.8%

75.3 – 85.8%

Beta(173,41)

[38]

Probability that stroke is caused by emboli

71.0%

63.3 – 77.8%

Beta(107,44)

[38]

Manual palpation: sensitivity

31%

22 – 40%

Beta(33,73)

[31]

Manual palpation: specificity

98%

96 – 100%

Beta(179,4)

[31]

Modified TEE: sensitivity

97%

94 – 100%

Beta(124,4)

[39]

Modified TEE: specificity

80%

74 – 85%

Beta(173,44)

[39]

Risk reduction of post-operative stroke due to adaptations when applying the modified TEE

0.46

0.196 – 0.910

LogNormal (−0.864, 0.392)

[25]†

Probability of a major adaptation of surgery given a patient with atherosclerosis

2.7%

2.3 – 3.1%

Beta(163,5888)

[40]

Risk of death (year 1–2 post-surgery): age < 61 years

1.18%

0.82 – 1.42%

Uniform

[41] ‡

Risk of death (year 1–5 post-surgery): age 61–70 years

2.33%

1.95 – 2.71%

Uniform

[41] ‡

Risk of death (year 1–5 post-surgery): age > 70 years

4.35%

3.49 – 5.24%

Uniform

[41] ‡

Utility of the ‘Post Stroke’ health state

0.439

0.186 – 0.653

Triangular

[42]^

Utility of the ‘No complications’ health state

1

-

-

Default

Cost of death (Euros)

2,736

1,368 – 5,471

Uniform

Expert opinion [43]

Diagnostic cost of modified TEE tool (Euros)

212

184 –243

Gamma

Manufacturer

Cost of fatal stroke event (Euros)

2,736

1,368 – 5,471

Uniform

#

Discount rate for costs

4.0%

-

-

Dutch guidelines

Discount rate for effects

1.5%

-

-

Dutch guidelines

  1. The baseline values that were used as input parameters for the model, along with their range and distribution (if applicable).
  2. * All costs were recalculated to 2012 euros.
  3. $ Range concerns the full range for uniform distributions and 95% confidence intervals for other distribution types.
  4. # The cost of a fatal stroke was assumed equal to the cost of death, as the costs of additional medical care delivered to stroke patients prior to death is already incorporated in the costs of surgery (Additional file 1).
  5. † This risk reduction was based on the comparison of two groups in the original article: group C (13 strokes in 690 patients) versus group A (12 strokes in 268 patients).
  6. ‡ Values were recalculated from the actuarial survival curves by age in the source material and concern annual mortality risks applied in the first five years after cardiac surgery.
  7. ^ Estimates of the utility of patients after major stroke vary widely therefore a large range was used (min = 0.11, max = 0.71, most likely value = 0.50, resulting in an expected value of 0.439).