Hospital-acquired Infections – Adding Process Metrics

Posted by

It’s been quite a February for many of us, I think.  I have a theory that my blog productivity is directly inversely proportional to the number of snow days at my daughter’s daycare!  I’ve been overjoyed to see the grass peeking out from under the melting snow in the last couple of days however.  Spring, here we come!

Hospital-acquired Infections
Anyways, on to the topic at hand.  We’ve discussed Surgical Site Infections (SSIs) in quite some depth (pt 1pt 2pt 3).  This is only one piece of the overall issue of Hospital-acquired Infections (HAIs).  SSIs are an important type of HAI, but it’s equally important to stay on top of the other types of HAIs as well.

We’ve previously discussed the importance of using rates and risk standardization.  To briefly recap, the number of infections is obviously important.  However, we need to balance holding people accountable with not wanting to punish bigger units for simply having more patients.  By adding in denominator information, we can add another layer of context and meaning: this is the Infection Rate.  Even better is to risk adjust based on historical national data.  The CDC provides the Standardized Infection Ratio (SIR) as a means of comparing your number of observed infections to the number of predicted infections.  An SIR > 1.0 is not good (more observed than predicted), while an SIR < 1.0 is good (less observed than predicted).

Here is a quick dashboard comparing Number of Infections, Infection Rate, and SIR for Non-MBI CLABSIs (central-line associated bloodstream infections, without mucosal barrier injuries):1 CLABSI num, rt, sir

They are all pretty similar in many ways, but they tell subtly different stories.  The differences become more important when comparing different units, each with different characteristics:
CLABSI by Location.png

For instance, ICUs A and C look about the same based on the pure number of infections, but when we look at the infection rate or SIR, we see that ICU A is doing worse.

Side note: I chose red for CLABSIs because they involve blood.  I try to not go overboard with color, but this was a great instance where color could add additional meaning.

Process metrics
So far, we’ve only been looking at metrics that are really big and important.  This is certainly a good place to start.  To move beyond this, I’d like to introduce the concept of Outcome vs Process Metrics.  Outcome Metrics are the big, huge metrics we really care about.  The problem is they’re really hard to impact.  Process Metrics are things that aren’t intrinsically important, but are important because they drive Outcome Metrics.  For instance, let’s say weight is a personal Outcome Metric of mine.  It’s not helpful to just say “lose weight.”  It’s more helpful to say “limit your portion size” or “walk 10,000 steps a day” or “go to the gym twice a week.”  These would be my Process Metrics, as they drive the Outcome Metric I want, namely losing weight.

We’ve been working quite a bit over the last year or two on adding in more Process Metrics to our dashboards.  This helps staff feel more empowered to make meaningful change, which leads to improvement.

For HAIs, the main Process Metrics are Hand Hygiene (i.e. hand washing) and CHG (a chemical treatment that kills bacteria).  By adding these to a dashboard with Infection Rate and SIR for the main HAI types, we get a very functional operational dashboard:
HAI Dashboard

Note the relatively big dip in Hand Hygiene Compliance in FY17 Q2 – this lines up shockingly well with a big spike in Non-MBI CLABSIs.  Not exactly the outcome we want, but it showcases the importance of Process Metrics!

Another side note on color: I had some fun with this, making CAUTIs yellow (they involve urine) and Hand Hygiene blue (hand washing traditionally involves water).  This adds subtle encoding that helps separate out sections of the dashboard for quick comparison.

There’s lots more to do with HAIs, so stay tuned!  Click here to see an interactive version of this dashboard.


One comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.