VIS-SIG Blog

Wonderful Wednesdays November 2021

CGI-I data
Wonderful Wednesdays

Understanding clinically meaningful differences is important to get a feeling for how relevant changes are on e.g. a patient reported outcome. These are often done by understanding how changes from baseline relate to the overall impression of the physician measured by the CGI-I.

Wonderful Wednesdays October 2021

Competing risk
Wonderful Wednesdays

The goal was to 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.

PSI Scientific Meeting

PSI Scientific Meeting: Generating Insights through Modern Applications of Data Visualisation. Find more information and sign up here: https://psiweb.org/events/event-item/2021/09/17/default-calendar/psi-scientific-meeting-generating-insights-through-modern-applications-of-data-visualisation

Wonderful Wednesdays September 2021

CGM
Continuous variable
Wonderful Wednesdays

The goal of the challenge is to produce explanatory graphics - Visually demonstrate that there is a dose response in glycemic control for Rx (the higher the dose the lower the variability) and that the Rx med and Rx high demonstrate better glycemic control that SOC and these attributes are sustained.

Wonderful Wednesdays August 2021

Data journalism
Wonderful wednesdays
COVID-19

For this edition of the Wonderful Wednesdays, the audience was asked to send examples of visualizations representing coronavirus data. The webinar was a showcase og the different types of visualizations, such as area graphs, forest plots, streamgraphs, or network analysis.

Wonderful Wednesdays July 2021

Patient Reported Outcomes
Wonderful Wednesdays

Patient Reported Outcomes (PROs) are often used within clinical trials. Prior to use as an endpoint, it is important to establish measurement properties of the instrument in the population of interest. This is conducted on blinded data, with no knowledge of the treatment allocation of each patient. Instead, anchor measures are used and compared to the score that is being evaluated. These anchor measures are usually simple questions that are designed to link conceptually to the overall target of evaluation or can be previously validated measures that assess similar concepts.

Wonderful Wednesdays June 2021

Vasculitis
Rare disease
Wonderful Wednesdays

Vasculitis is an inflammation of the blood vessels, in many cases caused by the body’s immune system attacking healthy blood vessels, causing them to become swollen and narrow. The symptoms associated with vasculitis are varied, and include asthma, allergic rhinitis (cold-like symptoms), fever, joint pain, tiredness, loss of appetite and weight loss. In serious cases, if left untreated, severe vasculitis can lead to organ failure and death. Vasculitis is treated primarily with oral corticosteroid (OCS) treatment. However long-term usage of OCS therapy at high doses is associated with a number of side-effects which worsen over time, and steroid-dependency is often a concern, so steroid doses need to be reduced if possible. However, some patients have a propensity to relapse, in which the patient experiences an acute flare-up of symptoms, usually requiring immediate treatment with high dose OCS treatment. Patients are considered to be in remission if the symptom score is zero concurrently with the OCS dose being ≤ 7.5 mg/day.

Wonderful Wednesdays May 2021

Psoriasis
Sustained response
Wonderful Wednesdays

Psoriasis is a chronic, inflammatory skin disease. For such diseases, it is of interest to not only compare treatments in terms of patients achieving a response, but also in terms of patients sustaining this response once it is achieved. This is the focus of this month’s challenge: produce a visualization to compare treatments with regards to patients sustaining a response. The synthetic data contains PASI (Psoriasis Area and Severity Index) scores for three treatment groups: two receive an active treatment in different doses; the third receives a comparator. The three arms are balanced, with 300 patients in each. The data contains PASI scores at Baseline and eight follow-up visits, spread over the course of a year. PASI scores can range from 0 to 72, with higher scores indicating a worse condition state. The criteriafor response can be chosen, but typical thresholds include a 75%, 90% or 100% reduction in PASI scores relative to Baseline (PASI75/PASI90/PASI100). More than one threshold could also be considered in the same visualization.

Wonderful Wednesdays April 2021

Mobile app data
Wonderful Wednesdays

COPD is a disease that effects adults typically aged 40 years or older. It is a chronic condition that effects the lungs of a patient. The intervention in this study was an self-management app that aimed to improve inhaler use and exercise capacity in individuals with COPD. Users of the intervention could access different videos to educate themselves around different topics to better understand COPD and what steps help improve their condition, such as exercise or smoking cessation. Users could also report other detail such as their medication use, their CAT score and even a daily symptom score on how they are feeling that day. All this information could be shared with their doctor who can then monitor and manage patients to ensure they get the right treatment for them.

Wonderful Wednesdays March 2021

Missing values
Wonderful Wednesdays

The current example data set is focusing on missing data. Missing data are present in almost any (clinical) data set. Ways to visualize different patterns of missing values is the topic of this webinar / blog entry.

Wonderful Wednesdays February 2021

DLQI
Wonderful Wednesdays

The DLQI is the most frequently used method for evaluating quality of life for patients with different skin conditions. There are 10 questions, covering the following topics: symptoms, embarrassment, shopping and home care, clothes, social and leisure, sport, work or study, close relationships, sex, treatment. Each question refers to the impact of the skin disease over the previous week and is scored from 0 to 3, giving a possible score range from 0 (meaning no impact of skin disease on quality of life) to 30 (meaning maximum impact on quality of life). This month's challenge was to use an effective visualisation to show the effect of treatment on DLQI scores across time points, possibly also reflecting the DLQI's multidimensional nature.

Wonderful Wednesdays January 2021

Prediction
Wonderful Wednesdays

The data set is about a retrospecitve study on finding "Predictors of Residual Tumor in Breast-Conserving Therapy". There are 500 subjects included. The outcome variable is reexcision and it describes, if there has been a reexcision necessary after the (initial) surgery. In a first step, a prediction model for reexcission needed to be set up. The goal of this exercise was to visualize the results. The audience was intended to be a "non-technical" one.

Festive Reindeer Plots

Christmas
Rudolph

In this entry, we do a little festive fun. We hope you will enjoy it!

Wonderful Wednesdays December 2020

Meta-analysis
Wonderful Wednesdays

Mark Baillie guides us through the submissions for the meta-analysis data set. The purpose of the challenge was to explore how data visualisation can be deployed to find insights when faced with big data. Big data in the pharma setting typically means wide data i.e. more variables than observations. We went through a number of interesting submissions from a shiny application of a forest plot, an interactive application to explore relationships between variables, an application that provides a workflow for developing a prognostic model, through to exploratory plots using small multiple ridge and scatter plots. A lesson from the challenge is to always visualise your data to avoid surprises.

Wonderful Wednesdays November 2020

Mediation analysis
Wonderful Wednesdays

Zachary Skrivanek guides through a number of data visualisations explaining a mediated treatment effect on patient reported quality of life. In addition the problem of missing data should be handled within the graphical representation. A Lollipop plot and a bar chart were presented as well as multiplot solutions using correlation plots or a combination of scatter plots, modelling plots and distribution plots. A point of discussion was the usability of a parallel coordinates plot. Another proposal used a Bayesian model displaying the results in an impressive grable – a combination of graphic and table. The last approach used an innovative way of storytelling with data called scrollytelling.

Wonderful Wednesdays October 2020

Adverse events
Co-occurence
Wonderful Wednesdays

Abi Williams presented data visualisations on co-occurrence of adverse events. How to display events that occur at the same time in the same patient and highlight differences of treatment or gender? A variety of very different approaches was discussed. The first visualisation was a heatmap for the frequency of co-occurrences. A Lollipop plot combined the frequency with the overlap time. Then a Shiny App made interactive exploration of up to 4 co-occurring events possible. The presentation of an UpSet plot brought up a discussion on the advantages over the vinn diagram. The shiny app AdEPro was referred to as an useful exploration tool for adverse event in general. Another tool for exploring the co-occurrence in particular was the force directed network graph. The final approach was embedded in a PowerBI Dashboard.

Wonderful Wednesdays September 2020

Adverse events
Wonderful Wednesdays

How to display safety data? This month's challenge has shown there are very different ways to visualize adverse event data. Although the example data set was from a two-arm study and relatively simple, the display of type of AE, frequency, timing, severity and seriousness is not easily combined in one plot.

Wonderful Wednesdays August 2020

Event data
Wonderful Wednesdays

In this month’s webinar, we discussed an exacerbation example data set. The data is based on the RISE study for patients with moderate COPD. The primary endpoint is the number of exacerbations during a six month treatment period. Event data – but patients can (and do) have multiple exacerbations. Statistical analysis used a Negative Binomial model. The dataset also included other variables which (may) effect the exacerbation rate: % Predicted Normal of Forced Expiratory Volume in 1 second (FEV1), Exacerbations in previous year, and Region / Gender. The challenge was to produce a data visualisation incorporating the information on the number of exacerbations observed. The discussed visualisations included a straightforward and clear presentation of the number of exacerbations, a Power BI app, longitudinal plots, and a display of patient-level outcomes.

Wonderful Wednesdays July 2020

Longitudinal
Continuous data
Wonderful Wednesdays

This month's challenge was to summarise changes in haemoglobin (Hgb) concentration over time in patients with anaemia associated with Chronic Kidney Disease (CKD). An experimental medicine was compared with a control group, but in addition to demonstrating efficacy, there was a special interest in visualising intra-individual variability, as large changes in Hgb are a potential safety concern. The visualisations presented ranged from grouped line plots over time, Sankey-diagrams and line plots with quantile bands for summary views and a lasagna plot for the display of individual data. Some general hints were given on how to define suitable color palettes for graphs depending on the type of data. Presented by Steve Mallett.

Wonderful Wednesdays June 2020

Wonderful Wednesdays
Subgroups
Discrete endpoint

This month the discussion is about visualizations of the satisfaction data using cumulative distribution functions, a tree plot, staggered bar plots, a power BI dashboard and two interactive funnel plots using plotly and RShiny. These examples nicely demonstrate that different displays support different purposes. Some visualizations are telling a story about the data and others help to explore the data. Presented by Bodo Kirsch.

Wonderful Wednesday April 2020

survival data
Kaplan-Meier
Wonderful Wednesdays

In this webinar we discuss the first submissions looking at some great ways to visualise a categorical response variable over time. We talk about what the SIG members like about each plot and give some pointers on areas we think could be improved. More specifically, we talk about: the value of a title with conclusions, the use of colour, the benefits of animation to display time, ways to declutter your graph and examples of clean design, enclosure to highlight important parts of the data, the benefits of the sankey diagram, how different charts help to answer different questions, and many more aspects of great visualizations.

Wonderful Wednesdays March 2020

Wonderful Wednesdays

The VIS SIG is happy to introduce a new initiative called “Wonderful Wednesdays”. This initiative will provide the participants to not only learn theoretically about visualisation principles but also apply them to relevant examples from the field of healthcare and the development of new therapies. This webinar kicks off the webinar series by giving an introduction into principles of visualization and an overview of the process. The first dataset is also introduced.

Wonderful Wednesdays Overview

Wonderful Wednesdays
Overview

This provides a short overview of our Wonderful Wednesdays Webinars.

More articles »

VIS-SIG Blog