There’s a fundamental problem with measuring safety by injury rate. And I know it’s a constant frustration for many of you as it often crops up when I talk to people in our industry.

So why do we continue to measure safety this way?

Everybody does it

All businesses measure their safety performance by counting how many people get injured. And it’s an instrument to benchmark your performance against other businesses, so it’s a particularly tempting way to measure. Senior management teams tend to drive this kind of measurement too so it’s much more high profile in annual reports these days.

There’s a legal requirement to report certain accidents/incidents

This varies between countries but there tends to be a reporting requirement for significant incidents which makes it an essential metric.

It’s relatively easy

Some businesses are better at reporting accidents/incidents than others but generally it’s straightforward for employees to get the importance of reporting them, especially the significant ones. In a lot of cases it would be impossible to hide anyway so the information is readily available.

My issues with measuring safety by accident/incident rate

In this case the old adage ‘what gets measured gets managed’ remains true and this means significant incidents tend to have a distorting effect on what’s important to a business.

Broken compass

But what tends to make the difference in outcome between someone behaving unsafely and something more extreme such as a fatality? Well, it’s often luck because a business is managing by outcome and not risk potential – possibly missing something much more serious around the corner.

Another flaw is that a business actually becomes a victim of their own success. As numbers of accidents/incidents are reduced there’s less information to learn from to prevent future events happening.

Remember, accidents/incidents are simply events with more serious outcomes than near misses, unsafe behaviours or unsafe attitudes, values and beliefs. So rather than just learning from and measuring those events with more serious outcomes, why wouldn’t you track and learn from events which don’t have a serious outcome i.e. hazards, near misses and unsafe behaviour?

This way you stop serious outcomes before they happen.

You’re also using all the information at your disposal to make a better choice about what your safety priorities are, rather than one or two incidents with a serious outcome. No matter how severe the outcomes of serious incidents are, they might not actually relate to the biggest risk currently in your business.

Finally, by learning from safety situations before an incident happens, it also becomes much easier to highlight good practice – praising and reinforcing where possible and changing people’s view of safety in the process.

So, what measures should you use to learn about safety before an incident happens? I’ll tackle a range of ideas next month.