How Prudence Rehab client cut its highest-risk bending by 37% in eight weeks

IntelliCoach pilot | April to June 2026

Prudence Rehab client repairs and replaces glass, working across both residential and commercial sites. One job might involve a rusted sliding door runner in a strata unit; the next, a full window replacement in a high-rise tower they helped build years earlier. With around 30 to 40 glaziers across both sides of the business, much of that work takes place on other people’s property, in spaces that were rarely designed with a glazier in mind.

Prudence Rehab had already started looking more closely at how to protect that workforce. Their General Manager met the Prudence Rehab team at the Workplace Health and Safety Show, saw the wearable sensor belts behind our IntelliCoach Program and took the idea back to the company’s health and safety committee.

The challenge

Glazing is repetitive, awkward and often improvised. Bending into a low-sliding door track. Twisting to lift an 80-kilogram frame out of a tight doorway. Working above shoulder height because the ladder is back at the ute and it’s easier not to bother.

Lower-back injuries are the most common workplace injury in Australia and among the most expensive to claim. But what worries us most is how invisible the build up can be. Nobody logs an incident report for ‘bent awkwardly for the two-hundredth time this month’. It accumulates quietly, often over years, until eventually it doesn’t.

‘When you speak to a tradesman or a construction worker, it’s always late 30s or early 40s and their back’s gone, their shoulder’s gone. So where have we gone wrong as an industry, that a 35 year old is saying that and still has at least 20 years of work ahead of him?’ Raj Sampelly, Head of Injury Prevention, Prudence Rehab

Our client wanted to answer two questions: where does that risk sit across their teams, and can habits formed over years on the job genuinely change?

Data first, then coaching that fits the job

IntelliCoach pairs wearable movement sensors with onsite coaching from a Prudence Rehab consultant. The sensors pick up how people are actually moving through a shift: lifting, twisting and loading up one side. Then the consultant works through it with them on the floor, in the job they’re doing. Most wearables stop at the data, and that’s usually where the useful part starts.

The pilot ran for eight weeks with ten team members.

Weeks one and two: establish the baseline. The team wore the belts and worked as normal, with no coaching involved. This created an individual movement profile for each worker, tracking total bends, bends beyond 50 degrees and the deeper bends beyond 60 degrees that carry the greatest injury risk.

Weeks three to eight: take the data onto the job. Once the data showed where risk was concentrated, our consultant spent a day per week with the crews onsite, not in a meeting room. They were in utes, on driveways and, occasionally, introducing themselves to homeowners partway through a job.

That time alongside the team surfaced details the data alone can’t show.

One glazier held an above-shoulder position for ten minutes rather than walk back to the ute for a ladder. A toolbox was mounted so every tool had to be twisted out sideways from under a Ute bar. Two workers of very different heights were sharing one scaffold setting, making the setup uncomfortable for at least one of them all day.

‘Some of these things we see immediately, and another organisation might not. They’ve done it that way for so long that it’s just become part of how the business runs. That doesn’t mean it’s right.’ Raj Sampelly, Head of Injury Prevention, Prudence Rehab

The coaching was just as practical. Kneel to do the siliconing rather than bending over it. If anything above shoulder height takes more than a minute, get the ladder out. On the repetitive two-person lifts, pair workers of similar height, which sounds like a small thing until you see the difference play out over a full shift. And plan the route before you pick up the glass.

None of it was new thinking. It came from watching the job in front of them, which is more than you get out of a generic safety briefing.

What the data showed

Across 1,075 wear hours, the numbers moved in the right direction:

• Bends beyond 60 degrees fell 37.2% per hour worked, the movement most closely linked to lower-back injury.
• Bends beyond 50 degrees fell 29.3%.
• Total bends fell 23.6%.
• An estimated 7,800 risky bends were avoided across all categories, measured against what the team’s baseline rate would have produced over the same coaching hours.

The pattern is particularly important. The deeper and more dangerous the bend, the larger the reduction. That’s consistent with coaching landing on the movements that actually matter.

Two workers reduced their deep bends by more than 80% per hour worked. A handful of others had thin baseline data, or appeared to shift into different tasks between the two phases, which can skew a percentage change either way. Those individuals were flagged for a closer one-to-one review rather than folded into the headline figures.

What comes next

A pilot exists to answer one question: Is there something worth continuing?

For our client, the data says yes.

Off the back of it, our client has been named a finalist in the Australian Glass and Window Association Design and Industry Awards.

The next phase is still to be scoped. Broadly, it would bring in the team members who haven’t been through the coaching yet and set a minimum wear-time target so the data stays reliable. The bigger piece is connecting movement data to recordable injuries and lost time over a longer stretch.

That’s the part that matters most. Fewer risky bends is the leading indicator. The outcome we’re after is fewer workers hitting their late 30s with a back that’s already “given out”.

A different way to think about prevention

IntelliCoach is how Prudence Rehab approaches injury prevention. Look at the whole person, start from the job they actually do every day and aim for changes that hold.

It isn’t tied to one device or technology, either. It’s a program built around our consultants’ coaching, paired with whichever wearable technology suits the workforce and the risk.

If your business carries manual handling risk in field based or trades work, we can talk through what a pilot could look like for your team.

Results reflect one pilot with one workforce over an 8 week period. Individual and team outcomes will vary depending on tasks, environment and participation.