What I’m working on right now

Steve Simon


There are several research projects where I am actively looking for collaborators. I thought I’d outline these topics briefly here.


I am proposing the use of control charts to monitor adverse events. I have an interesting “twist” on these control charts because I am monitoring time gaps (also called waiting times) rather than monthly or quarterly rates. These time gaps have several advantages


In some sense

The key to accrual is to estimate the date gap or waiting time between successive patients and model this distribution of waiting times. I am working with a colleague at Kansas University Medical Center to use a Bayesian model to predict the ultimate fate of the accrual process. I also am investigating how control charts


While the need for risk adjustment (or related approaches like risk stratification) are indeed important

A more rational approach to this is to compare each individual or group to an overall average. This allows you to identify who is performing significantly better than the overall average and who is performing significantly worse than the overall average. It does not allow you to examine how each individual or group compares to the best

This approach of looking at deviations from an overall average rather than looking at all pairwise comparisons is called Analysis of Means (ANOM) and it is an underutilized and underappreciated methodology. I have some interesting ideas about using randomization tests to produce decision limits in ANOM.

Here’s an example of an ANOM chart

A common theme for all three of these areas is that they lend themselves well to simple graphical summaries. I have always enjoyed the more visual applications of statistics (like regression analysis) and I believe that a single well constructed graph can provide a wealth of information.