A Christchurch business was evacuated this week over a hazardous, unidentified substance, with one person taken to hospital in a moderate condition, as reported by RNZ. That's what a workplace injury looks like in our heads. Sirens. Cordons. A fire crew in breathing apparatus.

Now here's the quieter number. In 2024, the most recent year of available figures, New Zealand recorded 486 work-related compression syndrome injuries (Source: Stats NZ, work-injuries-by-industry). That's the highest count in 15 years. You have to go back to 2009 to find a worse year.

Compression syndrome is the opposite of a fire alarm. It's gradual onset, which means nothing happens on any given Tuesday. Pressure builds on a nerve over weeks and months of the same movement, the same grip, the same posture. There is no incident to report. There is just the day you realise your hand doesn't work properly anymore.

And the trend is ugly. In 2013, there were just 150 of these injuries. That was the floor of a remarkable run: 324 in 2011, 153 in 2012, 150 in 2013. Whatever New Zealand workplaces were doing back then, it looked like a genuine safety win.

It didn't hold. By 2015 the count had jumped to 393. It wobbled through the late 2010s, sitting at 315 in 2017 and 324 in 2019. Then it took off. 453 in 2022. 444 in 2023. 486 in 2024. That's three consecutive years above 440, something that hasn't happened in the entire 15-year run of this series.

Put the two ends together and the scale of it lands: compression syndrome injuries have more than tripled since 2013, up 224 percent. It built gradually, year on year, with no single dramatic moment to hang a story on.

Why does this matter more than a one-off evacuation? Because gradual onset injuries are the ones a workplace has the most control over. A hazardous substance leak is a failure of containment, often sudden. Nerve compression is a failure of design, repeated hundreds of times a day, in plain sight, for months, until someone files a claim. Every one of those 486 is a person who kept doing the job while it slowly stopped working for them.

There's a fairer reading available too, and it deserves airtime. Part of any rise in gradual onset claims can be better recognition: workers who once quietly wore the pain now getting it recorded. If that's what's happening, a rising number is partly a sign the system is catching injuries it used to miss. Good. But recognition alone doesn't explain a jump from 150 to 486 while employment growth was nothing like that steep.

The honest position is this. The 2009 figure of 555 shows we've been worse. The 2013 figure of 150 shows we've been dramatically better, and recently. Both of those things are true, and the second one is the important one, because it means the current number is not inevitable.

We know how to make this go down. We did it once, inside the last dozen years. Then we stopped.