"In God we trust. All others bring data."

W. Edwards Deming, an industrial engineer and arguably one of the founders of the Total Quality Management movement, reportedly once said, "In God we trust.  All others bring data."  In other words, if you are both omnipotent and omniscient, you probably can make a convincing argument on your own.  If not, you better have the data to back it up.

There is absolutely no question that data can be powerful.  The right data can motivate change, and change is the essence of continuous improvement.  You may already be convinced, but just to prove it to you, I want to briefly talk about two important and recently published studies.  

As Deming himself stated, "Uncontrolled variation is the enemy of quality."  Clinical protocols and care guidelines exist to minimize variation in care as much as possible.  Once variation is controlled, providers can monitor outcomes and adjust the protocol accordingly.  However, developing these clinical protocols and care guidelines is not easy - it takes time and money (and most importantly, it takes buy-in from the key stakeholders).  Here is where these two aforementioned studies come into play.  Both studies involve critically ill children who have undergone surgery for congenital heart disease.

Using a technique that they called "target-based care", investigators at Primary Children's Hospital in Salt Lake City, Utah and Lucille Packard Children's Hospital at Stanford in Palo Alto, California significantly reduced the variation in care that children received following cardiac surgery.  The way that they accomplished this is particularly interesting (and relevant to the "power of data").  They did not use clinical care guidelines or protocols.  Instead, they merely posted specific targets for the expected or typical time it would take to remove a patient from the ventilator, discharge from the intensive care unit, and discharge from the hospital.  These targets were based on either published data from national outcome registries or internal data, respectively.  That's all there was to it.

These targets were placed on regular signs and posted at the bedside for all to see.  The targets were visible to both the members of the care team, as well as the patient's family.  As the investigators emphasized, this level of transparency led to both clinical improvement opportunities in real time, as well as influencing expectations and creation of shared mental models to improve the family's understanding of the care being provided.  The care team was not asked to alter their practice, but even without a protocol or guideline, the variation in care (as measured by the standard deviation around quality metrics, including the targets themselves) significantly decreased.  At least at Lucille Packard, the overall length of stay in the ICU and hospital also decreased!  Finally, the patient/family experience improved as well.

The simple act of posting the typical and/or expected benchmarks in the care of the individual patient reduced the variation in care, improved the patient/family experience, and in at least one of the studies, decreased the overall length of stay in the hospital.  Here are two perfect examples of the power of data.  Nothing fancy or costly was required, except having access to data (which is important, but not too difficult) and a piece of paper to write and post the data on for all to see.