top of page
Statistical Methods for Contemporary Clinical Trials
Statistical Packages
aedot
Rachel Phillips & Suzie Cro
Author:
Stata package to create a dot plot for visually comparing adverse events in two-arm clinical trials
More Detail
This creates a dot plot for adverse event data from a two-arm clinical trial, as proposed by Amit, Heiberger, and Lane (2008). The dot plot produces two graphs plotted adjacent to each other. The first plot on the left of the graph displays the incidence of each event by treatment arm giving a visual summary of absolute differences. The second plot on the right of the graph displays either the relative risk with corresponding 95% confidence interval or the risk difference with corresponding 95% confidence interval.
aevolcano
Rachel Phillips & Suzie Cro
Author:
Stata package to create a volcano plot for visually comparing the adverse events for two-arm clinical trials.
More Detail
This creates a volcano plot for adverse event data from a two-arm clinical trial, as proposed by Zink, Wolfinger, and Mann (2013). The volcano plot is a means of displaying the incidence of multiple adverse events simultaneously. The volcano plot can help to identify potential differences in the adverse event profile between treatment arms.
aefdr
Rachel Phillips & Suzie Cro
Author:
Stata package for performing a false discovery rate (FDR) p-value adjustment for adverse event data grouped within body- systems.
More Detail
This performs a false discovery rate (FDR) p-value adjustment for adverse event data where events are nested within bodysystems from a two-arm clinical trial as proposed by Mehrotra and Adewale (Stat. in Med., 2012). The FDR procedure is a two-step approach that utilises the structure of adverse event data to adjust p-values to reduce the false discovery rate.
mimix
Suzie Cro
Author:
Stata package for imputing missing outcomes for longitudinal data in trials.
More Detail
This multiply imputes missing numerical outcomes for a longitudinal trial with protocol deviation under distinct reference group (typically treatment arm) assumptions, following the general algorithm of Carpenter, Roger and Kenward (Journal of Biopharmaceutical Statistics, 2013). Imputation options include randomised arm Missing-At-Random, Jump to Reference, Last Mean Carried Forward, Copy Increments in Reference and Copy Reference.
bottom of page