PERT Analysis by Physician

The following was submitted by one of the most highly regarded physicians at Western State Hospital:

Dear Medical Staff,


In light of the Medical Staffs concern about the increasing assaults coinciding with the start of PERT in CFS, as well as subsequent concern by The Joint Commission, the hospitals Quality Management department created a report to address this concern.  The report was generated by the Quality Management Department, headed by Mark Simon-Haines and Theresa Becker.  I have attached the report to this email.

The report provides sixteen graphs of monthly assaults.  Each graph depicts monthly assaults on the individual wards of the Center for Forensics Services (CFS) for the past three years.  I found the report rather confusing (16 different graphs) but I believe it concludes that any increase if it exists is probably a natural fluctuation.  The report also emphasizes that only a few wards saw any increases so not caused by PERT.

It wasn’t clear to me why the report avoided looking at CFS as a whole and why it didn’t look more at overall trends and fluctuations in assaults.  I have limited access to the hospitals data, but because I found the Quality Management report rather confusing, I decided to work with this limited data and create some spreadsheets to tally up assaults, injuries and seclusion/restraint hours.  Here’s what I got.

In this first year of PERT’s operation in CFS, a dramatic reduction in Seclusion and Restraint hours occurred (-85%) compared with the previous year.  This associations seems undeniable.

However, accompanying this reduction in seclusion and restraint hours was a 47% increase in assaults and 20% increase in injuries (see the table below).

screenshot 30

 This increase in assaults is actually quite disturbing when considering the 19% decrease in CFS between 2012 and 2013 (a year before PERT began) and the overall hospital downward trend in assaults since 2005.  On average, since 2005, the hospital has seen a 5% annual reduction in assaults.  The 47% increase in assaults in CFS accompanying the start of PERT does NOT appear to reflect a normal fluctuation of assault rates, when viewed against the backdrop of the hospital wide data since 2005 (see table below).

screenshot 31


© Paul Vilja 2017