Hi Chris - I am not sure that the qualifying word "root" is neeeded for your question.

There will be some events that (looking back and then forwards, ie. so as to avoid hindsight bias) were not and could not be foreseen - no cause. A "freak" accident.

Then far more which do have a cause but where a risk assessment might reasonably conclude that doing the work by some other method is not reasonably practicable.

BUT.....if the worst comes to pass and the incident ends up in a UK court, the onus of proof as to what was or was not reasonably practicable rests with the primary duty holder, usually the person's employer,

As Kate has alluded to, there will always be a root cause.

However I have done plenty of incident investigations over the years were an assignable root cause has not been concluded.

Usually because the residual risk that casused the incident was known and accepted  and/or because the root cause has no appropriate remedial action.  In the RCA methodology I use, all the branches of the chart would come to a justifyable stop.

Managers hate this scenario at the review stage of the investigation as for years the management of zero mantra has been that all incidents must have a cause and remedial actions put in place to prevent the root cause and therefore the incident.

Morning Chris

Perhaps I have been a little less "lucky" with my investigations!

In the course of 9 months, I investigated three accidents at tenement refurbishment projects in the same post code.

1. Apprentice roofer fell through cupola roof light and landed at the bottom of the stairwell - about 20m. 5 stitches to the front of his head and five to the back.

2. Man fell from stairhead scaffold and landed at the bottom of the stairwell - about 18m. Tore ligaments in his leg.

3. Plumber had a single floor board up in a flat at towards the end of the project. Fell maximum of 340mm but due to a combination of things happening stumbling (with a clear cause-effect chain along the way) died in hospital 10 days later.

In each case, the immediate cause was obvious - slip, trip or fall on the level/from height.

In Nos 1 and 2 a root cause analysis would come up with lots of underlying causes that could and should have been identified to introduce actions that, if they had been taken would have prevented the accident and thence, injury.

I doubt that these actions would include checking that the staircase bannisters were in good condition (though that should have been done for OTHER reasons) - two men's injuries were almost certainly substantially reduced as each bounced off the bannister on their way down.

But, as for Number 3, absolutely no point in doing an RCA - Short of demolishing the building and starting again, you have to remove that single floorboard to do a task required in the refurbishment and I challenge anyone to tell me that it would be reasonably practicable to erect double guard-rails and toe-boards to protect against someone falling 340mm when the floorboard is going to be out of position for an hour or two.

....and even on a new build there are going to be occasions when a floor board is not in position EXCEPT if the most forward Modern Method of Construction ('MMC') are in place, so that all the services are fitted in the factory and then the complete room ("module") is slotted into place, but......introducing other risks, usually inclusive of crane operations. But the designers would probably not be thinking about the remote risk of falling 340mm when choosing an MMC option.

P