Root Cause Analysis

Requests from hospitals for assistance in facilitating root cause analyses of sentinel events are up this year.  This appears to be related to the Joint Commission’s more intensive focus on this area.

Although the process for completing a root cause analysis remains rather opaque in most instances, the final products are remarkably similar.  They usually bear a strong resemblance to the framework promulgated by the Joint Commission or to a conceptually similar format required by regulation. 

If a particular format is not required by regulation, I think some modifications of the Joint Commission’s suggested form can make the document much more user-friendly. 

Those of you who know me know that I’m usually more inclined to use a concise narrative style rather than a series of checkboxes when trying to describe something.  So, the Joint Commission’s boxes for “Root Cause?”, “Ask ‘Why?’”, and “Take Action?” don’t do much for me.  I’d rather analyze each issue until the root causes are identified at which point they can be simply listed as a bulleted list at the end of the narrative that concisely sums up the analysis of that specific issue.  Doing this eliminates the need for the first two boxes since the group continues to “Ask Why” until they’ve reached a “Root Cause.”  A major benefit of this is that each topic can be fully developed in a single place on the report rather than scattered around a lengthy form like installments in a Victorian novel. 

When the analysis is complete, each root cause can be copied into a table where planned interventions and measures of effectiveness can be listed, along with other critical information like responsible person and due dates.  If a root cause is not going to be acted upon, the same table can be used to explain why.  The third check box is now gone.  In its place is a clear description of the interventions for each root cause or the reasons for not intervening. 

I think these modifications of the reporting template (along with a few others) make the results—and the reasons for the results—much more understandable.  Of even more significance, though, is the process that I recommend to develop an understanding of the event. 

Knowing that in the end I’ll need to correlate the results with the minimum required scope of a root cause analysis, based on the type of event, I simply forget about the need to do this throughout the bulk of the analysis.  I believe that trying to make a group think in narrowly defined categories hinders, rather than helps, analysis.  With a good facilitator, the group can conduct a thorough analysis using a few standard tools and techniques of quality improvement, such as brainstorming, nominal group technique, and a prioritization matrix.  Afterwards, each of the root causes can be associated with one or more components of the required scope such as staffing levels or supervision of staff.  With good facilitation you’ll likely find that you’ve touched on all the required processes or else decided that a particular process was not relevant. 

A solidly conducted root cause analysis can be an effective component of an organization’s quality improvement strategy.  If you’re facing one of these events and want some assistance we’re here to help.