Clinical Global Impression Data Challenge

Background

This month’s challenge will be to Clinical Global Impression (CGI) data.

The clinical global impression – severity scale (CGI-S) is a 7-point scale that requires the clinician to rate the severity of the patient’s illness at the time of assessment, relative to the clinician’s past experience with patients who have the same diagnosis.

The Dataset

The CGI-S is an ordered categorical variable. However, CGI is often analysed also as a continuous variable. The visualisation may approach it both ways. It is especially interesting to understand how many patients achieve a score of =1, <=2, and <=3.

The data is provided in a summary format rather than indivdual patient data. The aim is to focus on presenting the summary data rather than focusing on how to analyse the data.

Two data files are provided:

• The original WWW example data set with a description in xslx (excel) format. CGI_S_3_groups_WWW_example.xlsx
• A simplified csv version of the data to be more amenable for computation. CGI_S_3_groups_csv.csv

The Challenge

Please provide data visualisations, which show these and also provide comparisons between the different groups (e.g. based on response differences or odds ratios for the different response categories).

You may also consider using Rank-based approaches like Wilcoxon tests or it’s extensions