A General Gaussian Approximation Approach for Value of Information Analysis This release accompanies the publication of the article: Jalal H, Alarid-Escudero F. A Gaussian Approximation Approach for Value of Information Analysis. Med Decis Making. 2018;38(2):174-188. The three files provided below are used to apply our Gaussian approximation (GA) for computing the Expected Values of Sample Information (EVSI). The first file (EVSI_GA_Box1.R) details each of the steps that are summarized in Box 1 of the main text of the article. The second file (EVSI GA Appendix.R) details each of the steps that are summarized in Box 1 of the main text and also provides the R code for computing EVSI for different types of research studies using the GA for a single parameter, multiple parameters, and balanced and unbalanced designs. The second file (GA functions.R) provides the predict.ga function that calculates the conditional loss by computing the preposterior for each of the basis functions of the GAM model. This code can also be downloaded from https://github.com/feralaes/VOI-Gaussian-Approximation. The version used of the package mgcv was 1.8-17.