Hey everybody. Welcome back to the Substack.
I am putting the finishing touches on Chapter 2 of Measuring the Unmeasurable, and before it officially drops in a couple of weeks, I wanted to give you a quick preview of what we are about to get into.
Let me ask you a question. Have you ever had a moment where your organization’s safety dashboard is glowing perfectly green, your leadership is celebrating, maybe your EMR just dropped, but you and your safety team are sitting in the office, completely exhausted, because you know for a fact that your workers are bleeding on the floor?
I have. I experienced that exact scenario early in my career working for a rapidly growing contractor. Our EMR went down, the project managers were thrilled, and meanwhile, I was dealing with crushed limbs, lost eyes, and an endless stream of strains and sprains. We had a classic watermelon dashboard: green on the outside, bleeding red on the inside.
And it forced me to ask: Did we all just suck at our jobs, or were we looking at the completely wrong metric?
In Chapter 2, we are going to look at the exact math behind why our metrics are failing us.
We are going to talk about the SIF illusion. You’ve all seen the charts: over the last thirty years, we have driven non-fatal, recordable injury rates into the dirt. But fatality rates? They have barely budged.
And the primary reason for that is our industry’s absolute obsession with TRIR.
In this next chapter, we are going to break down some groundbreaking 2021 research from Matthew Hallowell and the Construction Safety Research Alliance. They analyzed 3.26 trillion work hours. Trillion. With a T.
Do you want to know what they found out about TRIR?
It is mathematically random.
It is not associated with fatalities, it has zero precision, and it is not a valid indicator of the impact of your safety interventions. Using TRIR to measure the success of a new safety program is like taking a new vitamin supplement, rolling a pair of dice every morning, noticing you are rolling fewer snake eyes, and claiming your vitamin is controlling the dice.
It’s not science. It’s statistical noise.
But wait, it gets worse. Because it’s not just that the metric is random. The data we are feeding into it is practically made of floppy cheese.
We are going to look at some brilliant work by Dr. Todd Loushine and Mike Wolff that proves OSHA’s ITA data is riddled with massive inconsistencies.
And finally, we are going to look at a massive, multi-year study of a 21st-century safety management system implemented by the Army Corps of Engineers. We’ll talk about the absolute irony of validating a proactive system using something called an SBLI (a Severity-Based Lagging Indicator). You literally cannot make this stuff up.
But here is the thought I want to leave you with today, and the question Chapter 2 is going to force you to answer about your own organization.
In that Army Corps study, the new management system showed a significant drop in contractor incident rates. But government civilian employee rates didn’t drop at all.
Why would a safety management system only protect the contractors?
It wouldn’t.
But you know what affects a contractor differently than a government employee? The fear of losing a 50-million-dollar bid. Government employees don’t lose their contracts if they report a twisted ankle. Contractors do.
So I leave you with this: When you implement a new safety program and your TRIR goes down... did you actually reduce the occurrence of injuries, or did you just increase the suppression of reporting?
When you rely on mathematically broken lagging indicators, you can’t tell the difference.
Chapter 2 drops in just a couple of weeks. Keep an eye on your inbox, get your highlighters ready, and let’s get to work.
