Appendix CEAC/EVPI analysis and statistical inference σ๐๐๐ต(๐) σ๐๐๐ต(๐) The CEAC can be constructed based on (re)sampling procedures, but also directly given a normally distributed net monetary benefit (NMB) estimator ฬ ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐ต (λ) [39]: ฬ ฬ ฬ ฬ ฬ ฬ ฬ (λ) > 0} CEAC(λ) = P{๐๐๐ต (1) Equation (1) thus yield the probability that the NMB estimator ฬ ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐ต is positive. Based on the central limit theorem (CLT), the ฬ ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐ต(λ) is normally distributed [48]. Based on the standard normal ฬ ฬ ฬ ฬ ฬ ฬ ฬ (λ) – NMB(λ)) / σ๐๐๐ต(λ) ~ N(0,1) the CEAC (λ) is calculated as follows distribution with a Z score = (๐๐๐ต [48]: ฬ ฬ ฬ ฬ ฬ ฬ ฬ (λ) > 0} = P{Z > - NMB(λ) / σ๐๐๐ต(λ)} CEAC (λ) = P{๐๐๐ต (2) ฬ ฬ ฬ ฬ ฬ ฬ ฬ (λ)} = P{Z + ฬ ฬ ฬ ฬ ฬ ฬ ฬ CEAC (λ) = P{๐๐๐ต ๐๐๐ต(λ) / σ๐๐๐ต(λ) > 0} (3) where Z = standard normally distributed random variable and σ๐๐๐ต(λ) = standard deviation of the NMB estimator. A normal cumulative density function of the CEAC is then given by: ฬ ฬ ฬ ฬ ฬ ฬ ฬ (λ) / σ๐๐๐ต(λ)) CEAC (λ) = ะค(๐๐๐ต where ะค (•) = standard normal cumulative density function. An estimate of CEAC (λ) is then given by: (4) ฬ ฬ ฬ ฬ ฬ ฬ (λ)/ σ๐๐๐ต(λ)) CEAC (λ) = ะค(๐๐๐ (5) where ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐(λ) and σ๐๐๐ต(λ) are sample estimates. ฬ ฬ ฬ ฬ ฬ ฬ (λ) Because a parametric (1-α) one-sided lower-bound confidence interval (CI) has a lower bound ๐๐๐ – z1-α σ๐๐๐ต(λ), the (1-α) CI where the lower bound of the interval is not above zero is given by the inequality nmb(λ) – z1-α σ๐๐๐ต(λ) ≤ 0. By determining the cumulative density function of this term equation 6 follows: ฬ ฬ ฬ ฬ ฬ ฬ (λ) / σ๐๐๐ต(λ)) ≤ 1- α ะค(๐๐๐ (6) From equation (5) and (6) it follows that the CEAC (λ) is less or equal to the chosen level of confidence. ฬ ฬ ฬ ฬ ฬ ฬ (λ) – z1-α σ๐๐๐ต(λ) and can be used to A one-sided lower bound confidence interval is then given by ๐๐๐ test the null hypothesis H0: ฬ ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐ต(λ) ≤ 0 (i.e., control treatment should be continued) against the research hypothesis H1: ฬ ฬ ฬ ฬ ฬ ฬ ฬ ๐๐๐ต(λ) > 0 (i.e., new treatment should be implemented). Equivalently, the p value of a test being below or above the pre-specified level of significance can be used. A frequentist interpretation of a CEAC is possible because the relationship between p value and CEAC can be expressed by p(NMB (λ)) = 1 – CEAC (λ). Graphically this is reflected by the fact that the CEAC is simply the mirror image of a p-value curve [39]. CEACs in the context of frequentist analyses can be regarded as a particular case of Bayesian inference with use of non-informative priors [48]. Hence, the prior probability of H0 (i.e., for a given willingness to pay a new intervention is not more cost-effective than its comparator) being false is assumed to be 0.5. CEACs can also be applied using a non-parametric bootstrap approach [39]. The bootstrap makes no assumption about the distribution of data in the population. It works asymptotically, i.e., as the size of the original data sample increases, the bootstrap sampling distribution will tend to move towards the true sampling distribution. The bootstrap replicates of the NMB statistic nmb * b(λ), b = 1, ... , B, provide an estimate of the sampling distribution of the NMB estimator NMB (λ), defined as 1 F(a) = ๐ต ∑๐ต๐=1 ๐ผ(๐๐๐ ∗๐ (λ) ≤ a) (7) where F is a sampling distribution of the NMB estimator for any real value a and I denotes the standard indicator function. The bootstrap estimate of the CEAC can thus be calculated as: 1 ๐ต CEAC (λ) = ∑๐ต๐=1 ๐ผ{๐๐๐ ∗๐ (λ) > 0} = 1 − ๐น(0) (8) As shown for the parametric approach, there is a relationship between the CEAC and the bootstrap confidence intervals for calculating the NMB: a (1-α) one-sided lower-bound bootstrap confidence interval is given by ๐๐๐ α (λ), where ๐๐๐ α (λ) is the α-quantile of ๐น, i.e., ๐๐๐ α (λ) = F-1(α). The quantiles of ๐น are given by the [(B+1)α] and the [(B+1)(1-α)] ordered values of the bootstrap estimates ๐๐๐*b(λ), b = 1, . . . , B, respectively. The lower bound of a (1-α) confidence interval is not above zero if nmbCEAC(λ) = F-1(α) ≤ 0, or equivalently if α ≤ F(0) [39]. Because CEACs may mislead policy makers by providing insufficient information on the consequences of an incorrect decision [40,41] uncertainty is measured by the EVPI. Since perfect information would allow avoiding the chance of making a wrong decision when adopting a novel intervention, the EVPI can be defined as [8]: EVPI = λ * σ๐๐๐ต(λ) * L(D0), (9) where D0 = โNMB(λ) NT - I0โ/ σ๐๐๐ต(λ), λ = willingness to pay, σ๐๐๐ต(λ) = standard deviation of NMB, and L(D0) = unit normal loss integral for standardized distance D0, NMB(λ) NT = mean incremental net benefit of the new treatment, and I0 = point of indifference between new treatment and standard (I0 = 0).