Impute Standard Deviations for Changes from Baseline
The function imputes standard deviations for changes from baseline adopting the approach describe in the Cochrane handbook, Section 16.1.3.2.
imputeSD(SD1, SD2, SDchange, corr)
SD1: numeric vector, baseline SD.SD2: numeric vector, follow-up SD.SDchange: numeric vector, SD for changes from baseline.corr: optional numeric vector of correlations; see details below.The function imputes standard deviations for changes from baseline adopting the approach describe in the Cochrane handbook (2019), Section 6.5.2.8.
Missing SD1 are replaced by correspondig values of SD2 and vice versa.
Correlations for complete data (rows) are computed. Alternatively, correlations can be provided via argument corr. This option may particularly be useful, if no complete data is available.
Minimum, mean and maximum correlation (over rows) are computed.
Missing values of SDchange are computed by the formula provided in the handbook. The minimum, mean and maximum correlation are used leading to maximal, mean and minimal SD values that may be used for imputation as well as a sensitivity analysis.
data.frame with possibly imputed SD1 and SD2 values as well as the given SDchange values are returen. Moreover, the computed correlations as well as possible values for the imputation of SDchange are returned.
Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.handbook.cochrane.org.
Matthias Kohl Matthias.Kohl@stamats.de
SD1 <- c(0.149, 0.022, 0.036, 0.085, 0.125, NA, 0.139, 0.124, 0.038) SD2 <- c(NA, 0.039, 0.038, 0.087, 0.125, NA, 0.135, 0.126, 0.038) SDchange <- c(NA, NA, NA, 0.026, 0.058, NA, NA, NA, NA) imputeSD(SD1, SD2, SDchange) SDchange2 <- rep(NA, 9) imputeSD(SD1, SD2, SDchange2, corr = c(0.85, 0.9, 0.95))