sgt103 / adv

Antidepressants Variability Ratio Bayesian Meta-Analysis

Antidepressants Variability Ratio Bayesian Meta-Analysis

This repo contains the code to reproduce the results of the paper "On the treatment effect heterogeneity of antidepressants in major depression. A Bayesian meta-analysis", Volkmann et al..

Installation - using Docker (recommended)

In order to rerun this analysis the following two programs need to be installed:

• Docker
• make. For installation on MacOS or Linux we recommend to first install brew and then run brew install make.

Running the analysis

Open a terminal window, go to the root folder of this repository, and execute the following commands:

$ make build  # this creates the Docker image for the analysis  

$ make run  # this creates a Docker container that hosts an instance of Jupyter lab

Paste the following URL into a browser:

http://127.0.0.1:8765/lab

Click on the notebook main.ipdb and run it.

Conda users

Execute the following commands in your terminal

$ conda create --name ad_meta python=3.7 -y
$ source activate ad_meta
$ python setup.py install

Now you can import all the functions used for the analysis.

Citations

The data set in the folder /data is taken from Cipriani et al_GRISELDA_Lancet 2018_Open dataset under the Creative Commons Attribution 4.0 International Licence.