Wonderful-Wednesdays

Repository to hold the data and materials for the Wonderful Wednesday webinar series https://www.psiweb.org/sigs-special-interest-groups/visualisation/welcome-to-wonderful-wednesdays

View the Project on GitHub VIS-SIG/Wonderful-Wednesdays

Competing risk example data set

Study Design

Observational, multicenter cohort study, in adult patients with PCR-confirmed infection with SARS-CoV-2, who were admitted to one of five collaborating hospitals in the region of Gelderland and North-Brabant (NL) for at least 24 hours between March and May 2020

Recovery is considered a competing risk for mortality and is taken into account as an extra outcome

Cumulative incidence curves were presented side by side for different characteristics

Goal

Find a way to visualize the impact of different risk factors (comorbidities and symptoms at admission to hospital) on the risk of death considering the competing risk of recovery.

Data set

Source paper: Nijman et al, 2021

The data can be found here.

The data set contains the following variables:

Variable Description Values
FTIME Follow-up time in days (hospital admission) Integers
STATUS Patient status at end of follow-up 0 (censored), 1 (died), 2 (recovered)
COMORB_xxx Comorbidities No / Yes
SYMPT_xxx Symptoms reported at admission 0 (No) / 1 (Yes) / Duration in days
MEWS Modified early warning score Integers
other Age, Sex, BMI, Lab values, … As appropriate