Incident Data
Knockin’ Down the Little Guy
As someone who has had to evaluate contractors on a regular basis, I can tell you that when injury statistics are used to “Grade” contractors....
Anyone who has been in health and safety, or had it as part of their “performance measurement”, knows what the TRIF (or TRIR) is.
Acronyms in themselves are not good or evil…light or dark. However, overuse can certainly cause confusion. And that’s bad.
Anyone who has been in health and safety, or had it as part of their “performance measurement”, knows what the TRIF (or TRIR) is. For those of you who don’t know, this is the Total Recordable Injury Frequency (Rate). These are used interchangeably in North America.
The origin of this measurement tool was to normalize the number of injuries across companies of different sizes. It was invented with good intentions. However, as the saying goes, the road to hell is paved with good intentions. Or, in this case, the road to the dark side.
I have worked on both the contractor (evaluatee) and client (evaluator) side, and I can say with certainty that it is used to assess a safety program, both internally and externally. With all the best intentions, of course. But what is the point of this assessment? And is this really the best method? Many have questioned this for a long, long time (in a galaxy not so far away).
The short answer to this question is NO. The long answer is, the data proves it. Here is why.
A study performed by the Construction Safety Research Alliance and the University of Colorado (Boulder) sought to answer these questions (note: they use TRIR instead of TRIF, but it is the same).The questions they asked were: Is TRIR statistically valid, meaningful, comparable across injuries and companies or at all predictive?
The results of the study,(which can be found here), were fascinating and are summarized as follows:
- TRIR is not predictive of future TRIR. The whole point in using a measure such as this is to attempt to predict the performance of a contractor, business unit, or company as a whole. This is not possible.
- TRIR is not predictive of fatalities. There is NO statistical correlation between TRIR and fatalities. There are obviously other factors in play that influence a safety management system.
- TRIR is approximately 98% random. Statistically, TRIR is almost the same as spinning a roulette wheel.
The force is not with us in this attempt at using the TRIR to predict. So what now?
Obviously we have to track incidents, and normalizing them with hours is still necessary. But if we are trying to move away from TRIR in assessing a safety management system, what do we replace it with?
Hint: Leading Indicators and Positive Safety Activities…This is the way…
Stay tuned for our next post from a guest author who will go into more detail about the dangers of using these statistics for evaluating contractors.