Root Cause Analysis (RCA) is a valuable tool
for understanding and gaining insight into process
outcomes. When applied appropriately, RCA can
illuminate targets for change and may generate
testable hypotheses. The use of RCA is extremely
useful, as it lends a formal structure to efforts
to learn from past mistakes.
A key to making RCA a productive endeavor is
to make it easily accessible to those responsible
for specific processes (the Process Experts).
While unpredictable variation is best identified
using an SPC chart, the sources of this variation
will need root cause analysis for identification
that leads to actions to eliminate or minimize
these causes. The Process Experts need to be
able to see the high level picture (Score Card
and SPC trend over time) and then be able to
explore the contributing factors in a variety
of ways. This investigation needs to be very
flexible AND easy to use.
Let's take a look at how a Process Expert can
view high level performance of the indicators
they are responsible for, and how they can use
some Root Cause Analysis techniques to gain
a better understanding of the factors controlling
the outcomes. We'll use some Adverse Drug Event
(ADE) indicators to see how we're doing. Let's
start at a Scorecard view to see the high level
performance picture:
From the Scorecard, we can see that we are
not meeting our performance goals for those
Adverse Drug Events attributed to wrong patient
identification. By clicking on the link, we
can take a close look at the performance of
this indicator over time:
The trend here is represented by a P Chart,
which helps us understand unusual variation.
Notice that for Q2 2007, we have a value which
is outside of the control limits - we need to
take a close look to understand what is going
on here. Mousing over this period will give
us some details and clicking on it will bring
up some more choices to help us look at the
details for this period. Let's go ahead and
look at a summary of the Adverse Drug Events
for this period:
Here is a summary of the specific ADE cases
for Q2 2007. Since we want to reduce ADE associated
with Wrong Patient identification, we want to
focus on those events attributed to Wrong Patient
Identification. Each of the ADE Events associated
with this are highlighted with a light red background
making them easy to identify:
So this gives us the summary of each of the
Adverse Drug Events for Q2 2007. We can now
look at the specific events to get at the detailed
information for each by simply clicking on the
link on the left. For example, clicking on 16701377
will display the following detailed info for
that Adverse Drug Event:
This mechanism provides basic Root Cause Analysis
by allowing us to see the big picture and then
to drill into the specific reasons of why an
indicator did not meet our goals or what contributed
to unusual variation in the process. This provided
access to the details, but we'd also like to
get a much deeper understanding of Adverse Drug
Events and what factors are creating them. We
would also like to look at Adverse Drug Events
from many different angles to see what the contributing
factors could be. We would want to look at a
variety of different metrics (ADE Count, % Patients
Harmed, % Preventable, Time to Report, etc.)
and be able to break these down by various factors
(Department, Shift, Drug Involved, Severity,
etc.)
By clicking on the Root Cause Analysis link
from the Indicator trend, you can drill into
this data to explore what the various contributing
factors are. Let's look at % Wrong Patient by
Stage of Event and by Department for example.
Here's the analysis for the same overall time
period of our Indicator Trend (Q4 2004 thru
Q3 2007):
We can see that the highest percentage of time,
this occurs during Prescribing. By clicking
on that bar, we can see the rate for each department:
By clicking on the Family Medicine Clinic,
we can see a summary of the events. And if we
want to take a close look at them, we can simply
click on the Record.
This technique allows the Process Expert to
easily look at a variety of factors that may
control Adverse Drug Events to look for specific
areas for improvement. Process Experts can quickly
access detailed information when there is unusual
variation as well as step back and look at the
contributing factors.
