The rapid growth and proliferation of managed
care organizations in the healthcare industry
has caused many public health agencies, providers,
employers and consumers to question the quality
of healthcare and the consequences for patient
safety. The mantra for healthcare quality can
be categorized into one simple statement: predictable
quality at a predictable cost. Webster defines
"predict" as "to declare or indicate
in advance…foretell on the basis of observation,
experience or scientific reason." With
so many dashboards available today, what's the
big deal? All reporting platforms provide us
with the needed information to "predict
quality." Pick one and you're good to go…right?
Not so fast…
Dashboards include images to help us easily
understand our "current reality."
These images include stop lights, gauges, etc.
and generally report on "averages."
Are we performing well against our internal
and/or external benchmarks? Well, as we have
all heard, patients do not feel averages…patients
feel variation. Continuous quality improvement
(CQI) is an approach to quality management that
builds upon traditional quality assurance methods
by focusing on the "process" and the
type of variation that exists within the process.
The primary threat to our business survival
comes not from events (budget cuts, new hires,
downsizing, credentialing, internal/external
requirements, etc.) but from slow gradual processes
to which we are 90 percent blind. Inadvertently,
we often times perpetuate this gradual decline
by utilizing a wrong approach for the desired
result. I think we all would agree that one
would not use a screwdriver to pound nails!
Why then, do we permit the use of Excel, report
writers or other like "tools" to help
us understand a "process?" The answer
may be in our lack of understanding of CQI.
A process is a series of actions or operations
conducting to an end. Every process contains
some degree of variation. For example, when
I sign my name, the resulting output is never
exactly the same as the one before or the one
after. Explanations for these variations include
being in a hurry, feeling fatigued, the quality
of the writing instrument, etc. We call this
"common cause variation." The bank
will still cash my check with a signature that
shows common cause variation. If only common
causes of variation are present, the output
of a process forms a distribution that is stable
over time. Now let's say I fall off my skateboard
(yes, I can shred with the best of 'em!) and
break my writing hand and am in a cast. The
cast doesn't allow me to grip a pen correctly
and therefore, I sign my name with my non-writing
hand. The variation in this signature is called
"special cause variation", as it is
created by a non-random event leading to an
unexpected change in the process output. The
process output is not stable over time and is
not predictable. The bank probably holds my
checks until it ascertains why my signature
has changed so drastically (or at least I hope
they do!).
OK. So we have common cause and special cause
variation. We expect common cause but not special
cause. How do we tell the difference? This is
where CQI technology comes to your rescue.
Let's use an example of Heart Failure Composite
Scores. The Composite Score is the sum of the
numerators for each HF measure divided by the
sum of the denominators for each measure times
100. First, let's look at the HF Composite Score
for an 18 month period through a bar chart.
Figure 1. Bar Chart of Heart Failure Composite
Scores
The height of each bar represents the percentage
by month of the HF Composite Score. As shown,
we can see a low month of June 06 and a high
a year later in June 07. We can make some inferences
of the process with the early months showing
a gradual decline and then a "general"
increase from June 06. What we don't see is
the volume for each month, an explanation as
to why we began an improvement in the process
or which months exhibited common cause variation
and which showed special cause variation. Now,
contrast this same data with a P chart and add
corrective action and begin a new "phase."
Figure 2. P Chart of Heart Failure Composite
Scores
Through the use of the P chart and CQI technology,
we now see additional information including
sigma limits, a corrective action plan and a
vertical line dividing the pre- and post-corrective
action plan process metrics. We also see "red"
data points. The red data points indicate special
cause variation. These data points violated
rules established by the Joint Commission. Examples
of violations include:
We can determine how stable and predictable
our process is through the use of CQI techniques,
providing us with opportunities to intervene
where necessary. We also can initiate a new
process based on a corrective action and determine
if the desired results or outcomes are taking
place.
Next time you hear, "all healthcare dashboards
are created equal," respond with "your
thinking processes are broken!" Stop lights,
targets and scorecards are very important to
your understanding of what is taking place in
your organization. Yet, if the dashboard does
not include CQI techniques, your processes may
be exhibiting special cause variation without
your awareness. Your report is merely "average."
Focused healthcare organizations are providing
their leadership with continuous feedback on
critical organizational metrics. Move away from
simply generating reports to supporting analyses
for informed decision making.
Statit has more than 25 years of experience
in helping our customers recognize the difference
between common and special cause variation and
then to make corrections to improve the process.
Shouldn't you be practicing predictable quality
at a predictable cost? Your patients will be
glad you are!
Statit is here to help. If you would
like to learn more about this functionality,
give us a call at (541) 752-4500 or send us
an email at
. We will show you how easy it is to create
these reports while we are on the phone together.
