Improvement Insights Blog
RRTs vs Code Blues – What is Your Data Telling You?
Sometimes your data seems to be telling you something counterintuitive. Here’s a healthcare example using Rapid Response Teams (RRTs) and Code Blues. What is your data telling you?
“Hi, I’m Jay Arthur, author of “Lean Six Sigma For Hospitals” and QI Macros [software].
“Last week I got a call from a customer and they said, “Here’s some data and we need your help a little bit.” (This is rare, okay, but they did ask for help.) So anyway, the issue is in a hospital when patients’ respiratory [numbers] and everything else drops to zero, that’s what they call a Code Blue, right? They’re essentially out of it.
“Now a few years back, hospitals started implementing what are called Rapid Response Teams, or RRTs for short. A Rapid Response Team, they look and they notice that respiration and heart rate and blood pressure are all going down; we better get on this, right? The rapid response team’s job is to go and make sure that people don’t go into a Code Blue (they don’t “code,” as they say in the trade). We don’t have to get out the crash cart and try and revive them.
“So what they were noticing was that in their scatter plot, RRTs were going up; now you would think that if RRTs are going up that Code Blue should go down, right? That would seem logical, but that’s not what was happening. Code Blues were also going up. They thought, “Oh my gosh, what’s wrong here?”
“I think this is one of those things where you have two different distributions of data: the more patients you have, it seems the more likely it is that you would have more codes and more Rapid Response. Does that make sense? There’s just “the more… the more,” so this seems logical.
“Now, we do not know if Code Blues are less or more, right? So what I would do is I would come back in there and say, “Okay, well, let’s look at the Code Blues. Out of the Code Blues, how many of those had an RRT and how many didn’t?” If they had no RRT, I would expect that part of the Pareto chart to be high and those with RRTs to be less.
“I think we were showing two different distributions on here, right? So let’s go look at the ones that didn’t get an RRT – a rapid response team. Why did we miss the signals that we should go do a Rapid Response Team? What’s the clue? This is one of the things I see sometimes: people look at their data and they [think], “Well, I think… our code blues should be going down.” Well, no; not happening.
“So we’re looking at the data incorrectly. Let’s go look at it in a way that will tell us if there’s something actually going wrong here. Now if it was reversed and there were all of these people who coded, had an RRT, then what was wrong with our RRT process? How did we miss it and how did we fail? Let’s go problem solve that, right? One of these two things is happening, and I think it’s most likely there was no RRT and they may have missed somehow. Maybe one thing was going down, pulse and respiration was fine but the blood pressure was… you know, maybe one thing or another was going on.
“So that’s my Insight for this week: When you look at your data it may not tell you what you’re hoping for, but it gives you insights about how to look at your data in a different way and get to a problem you can actually solve.
“That’s my Improvement Insight for this week. Let’s create hassle-free healthcare. Let’s go out and improve something this week.”