Hi Kelly

I should possibly open by saying that I have investigated 10 fatal accidents involving people falling whilst accessing or doing roofwork. All different in many ways, which is a feature of roofwork. The only common factor was gravity.

Because of all those differences, each case needs to be thought out on its own merits. What precautions are reasonably practicable, what precautions are not?

Way back in 1987 [so long before anyone had thought of the acronym CONDAM, later CDM!!]  I wrote the prosecution report that resulted in Client, Main Contractor and Sub-Contractor each being convicted after a scaffolder fell through a fragile roof light on a flat roof.

He didn't die, but sustained a very serious head injury. It happened on a July Wednesday. On the Monday I had completed the investigation of a fatal accident in which a man had fallen down a stairwell when attempting to access the roofspace. On the Tuesday I investigated the fatal accident in which a man fell off the roof of the 4 storey building in which he lived.

As I recall the work was sorting out the rendering of a wall alongside the roof so that a scaffold was being erected to enable this work to be done.

The Client did actually realise that the rooflights were fragile and had provided 10 guard-rail frames which could be deployed where appropriate to protect against which of 36 rooflights people might be working near., so a situation not entirely dissimilar to the surveyor's proposal that you could use a moveable crash deck under the roof lights at your sports centre, so you have the same puzzle as to where people will be accessing and working at any time and how to make sure that adequate control is in place near to where those people will be moving or working. In my scenario the evidence showed that the deployment of only 10 guard-rail frames would not be sufficient.

Your narrative presents a clear example of the benefit of Early Contractor Involvement.

First, I am by no means convinced that providing edge protection around the rooflights might not be appropriate. It could be possible to leave just the "leading edge" unprotected [possibly using  restraint to stop someone falling over that edge], or even to work kneeling to do some work at that leading edge with the protection still in place.

There has been research done on how "soft landing systems" [typically nets or airbags] perform and this depends on variables including fall distance and whether the positioning of the equipment is such as to actually catch the whole of the person falling. With a proposal for decking this could not be described as a "soft landing system", so you would want the fall distance to be very small, noting that you do not have to fall very far to sustain very serious injury - I have investigated three fatal accidents involving falls of less than 2m [a number which used to be a key threshold].

Then there is the issue of the rest of the roof which you say is in poor condition. Are you confident that none of this has deteriorated to the point where it should be considered "fragile". The National Federation of Roofing Contractors have published lots of guidance, much of which is freely downloadable from the NFRC website. This includes guidance on what forces you should be planning for.

So, my suggestion is that you bring in one or more industrial roofing contractors to advise you on how you THEY think this work should be done. On your part [meaning also that of the Client] you need to provide them with the parameters e.g. as to how much of the sportshall can be vacated at any time and for how long.

Might need to pay them just for the advice so that you can then embody appropriate precautions into the Bill of Quantities [and of course CDM Pre Construction Information] so that you can try and set a level playing field for tenderers. If they are going to be on the tender list you might get some of that money back as they should be in a better place to understand the project and price accordingly.

Now, of course, there is rarely only one solution to a puzzle. So you might want to qualify any requirements in the BoQ [and PCI] with some get out clause that allows a tenderer to come up with an alternative in their technical submission.

Good luck, Peter

Originally Posted by: peter gotch

First, I am by no means convinced that providing edge protection around the rooflights might not be appropriate. It could be possible to leave just the "leading edge" unprotected [possibly using  restraint to stop someone falling over that edge], or even to work kneeling to do some work at that leading edge with the protection still in place.

There has been research done on how "soft landing systems" [typically nets or airbags] perform and this depends on variables including fall distance and whether the positioning of the equipment is such as to actually catch the whole of the person falling. With a proposal for decking this could not be described as a "soft landing system", so you would want the fall distance to be very small, noting that you do not have to fall very far to sustain very serious injury - I have investigated three fatal accidents involving falls of less than 2m [a number which used to be a key threshold].

Then there is the issue of the rest of the roof which you say is in poor condition. Are you confident that none of this has deteriorated to the point where it should be considered "fragile". The National Federation of Roofing Contractors have published lots of guidance, much of which is freely downloadable from the NFRC website. This includes guidance on what forces you should be planning for.

Peter, I cannot thank you enough for this detailed response. It has been incredibly helpful and will allow me to determine how we move forward from here.