Like you, I sometimes click on the LinkedIn posts I see shared.  Sometimes I run across an interesting topic on that platform (or some other social media outlet) and I click to read well beyond the excerpt.  Recently, I clicked on one about how doctors have to "fight back now" against many issues, including the "tyranny of data" and scanned through.

This feeling isn't new.  In fact, there are some rules and principles taught in coursework on data driven quality improvement that are meant to prevent exactly this situation.

Here are some of those simple rules on how to avoid having staff (including surgeons like me) believe we live under a "tyranny of data" as described by the physician in the excerpt below.  (By the way, as usual, credit the excerpt below to the link associated with it.)  After all, when we feel we live under a tyranny of data, we usually don't want to participate in the quality improvement process!

  • Make data easy to collect.
    • There are techniques to make both continuous and discrete data very straightforward and fast to collect.  None of us, in healthcare, have time to collect data.  Of course, as usual, just when we think we don't have time may be the most important time to get data!  Whether you agree or not, there are some techniques (here) that make data collection rapid and meaningful.  A good sample (one that represents the whole system including work we do at midnight and on weekends!) is very important to minimize time spent collecting.
  • Learn a little about how the data work and educate others.
    • After medical school and an MBA, it took me 2.5 years to learn how to use data for quality improvement.  (And I'm still learning!) You may be faster to learn than me, but the extra training impressed on me how much goes into using data correctly for quality improvement (QI), and how little we understand in healthcare about how to perform data-driven QI.  We are in our infancy, and are about a decade behind in how to use data for QI.  It's a good idea to help spread the word and educate about the use of data in healthcare QI (that's what I'm up to) but also to take a minute to make sure we are up to speed on the topic.
  • Keep the group together in the data.
    • When you perform a QI project, don't separate people or groups.  Let me explain:  are you looking at turn-around times in the OR?  DON'T label them with a doctor's name.  Why?  It gives the illusion that it's personally assignable to the physician.  It ignores that the physician is the tip of a much bigger process of which they are only a part.  Wait a minute!  Doesn't the surgeon sometimes delay turnaround?  Yup, of course!  And if you decide to measure that no problem...just no names attached!  Looking at the expense of performing a hip procedure?  DON'T track cases by doctor.  It easily causes misalignment.  After all, it's easier to use less equipment and adjuncts on sicker, higher risk patients...by not operating on those patients, or at least raising the threshold for which higher risk patients have a procedure done at your hospital.  Unfortunately, at some centers, suddenly patients with comorbid conditions aren't getting their hips done as often.  That's NOT what you intended with your project to promote value in hip replacement.

Regardless of the specifics at your institution, separating groups with data collection, or putting someone's name on a endpoint, not only ignores variation and systems thinking, it makes them forever a non-participant in the QI process.  They won't want to work with it.  They may even sabotage it.  If your name were attached to something you couldn't control, what would you do?

Before you and I exit, one last thing:  I've heard, before, the cry of personal accountability.  "How can there be personal accountability in QI if no one's name is attached to anything?" It's a classic!

Here's my answer:  personal accountability, personal performance, is one factor in the 6M's that cause variation and quality issues.  It's a piece of the puzzle and can't be ignored.  Yet we need to immediately follow that up with the fact that personal performance doesn't function independently of the other causes.  Sometimes people can perform really well and there's still bad quality.  In great systems, people can perform poorly and things still go great.  (I've seen it.) Usually the performance is somewhere in the middle.  Yes, sometimes a root cause analysis shows there was an egregious performance problem.  That happens and must be addressed.  But, unfortunately, we as administrators are often quick to assign things as personal problems that are system issues...sometimes system issues that we ourselves own.  Recognizing how each portion of a system gives an outcome (including personal performance) get us much further than a narrow focus on any one of the 6Ms.

The bottom line:  personal accountability for performance is one piece of a system that gets an outcome.  

And systems reliably get just the outcome they are designed to get.  

In fact, systems set people up for certain outcomes...

Let's bring it 'round here to the tyranny of data:  data seems tyrannical when it's tough to collect, singles out individuals for systems issues, and when we lack education about what it means and how to use it.  Healthcare is years behind in its use of data to improve, and we are experiencing the growing pains.  They come to us as data that have an oppressive face.  Let's help spread the word on the meaningful use of good data...good data helps us much more!