Skip to main content

Table 3 Agreement between HTA bodies in the uncertainty raised about the same evidence (raised versus not raised); and when the same uncertainty was raised, agreement about how it was dealt with (addressed versus not addressed)

From: Why do health technology assessment coverage recommendations for the same drugs differ across settings? Applying a mixed methods framework to systematically compare orphan drug decisions in four European countries

Kappa scores
(95 % confidence intervals)
standard error (SE)
number of observations (n)
Level of agreement in the uncertainties raised (raised versus not raised)
Level of agreement in interpreting the same uncertainties
(addressed versus not addressed)
  NICE SMC TLV HAS
NICE 1 −0.06
(−0.235 to 0.124)
SE = 0.091
n = 117
−0.15
(−0.434 to 0.143)
SE = 0.147
n = 44
0.01
(−0.172 to 0.183)
SE = 0.090
n = 110
SMC 0.51
(0.203–0.814)
SE = 0.156
n = 29
1 −0.30
(−0.588 to 0.018)
SE = 0.145
n = 43
0.08
(−0.108 to 0.261)
SE = 0.094
n = 110
TLV 1.00
(1.00–1.00)
SE = 0.00
n = 7
0.72
(0.232–1.00)
SE = 0.249
n = 7
1 −0.07
(−0.324 to 0.180)
SE = 0.128
n = 44
HAS −0.08
(−0.227 to 0.067)
SE = 0.075
n = 24
0.18
(−0.272 to 0.630)
SE = 0.230
n = 22
−0.50
(−1.00 to 0.235)
SE = 0.375
n = 4
1
  1. Cohen’s kappa scores (κ) rank agreement levels from poor (κ = 0) to perfect agreement (κ = 1) and where minus values of κ correspond to cases when agreement was less than expected by chance
  2. NICE National Institute for Health and Care Excellence (NICE), SMC Scottish Medicines Consortium, TLV Dental and Pharmaceutical Benefits Board, HAS Haute Autorité de Santé