Ron H

You are right about sanity checks, but I am concerned that your suggestions are actually moving away from a risk based approach, which requires likelihood and severity to be considered together. 

The "most statistically likely outcome" suggests that the severity of the outcome is not a concern.  Whilst, the "hazards and controls affecting higher populations" suggests that the likelihood is not a concern.

I agree that scoring matrices are only a tool with limited use.  However, they are good at illustrating the concept of risk (severity x liklihood).  The problem is they are often misused.

This is proving to be a very interesting discussion.  It does suggest that our whole approach to risk assessment needs to be reviewed.  It is interesting (worrying) that there is so little consensus on such an important principle in H&S.

Originally Posted by: Ron Hunter

Ensure sanity by sticking with the most statistically likely outcome. 

I 'd advise caution using this approach for all but the most inane of activities - by your method you are relying heavily on past outcome as an indicator of risk.  You will miss the risks that have low frequency / massive impact.    The results of your RA also feeds your emergency response, business continuity, crisis mgmt processes - so relying on past outcomes may not prepare the organisation for what might happen

Separately, when used correctly, the risk matrix is also  useful for classifying incidents (actual/potential) and therefore deterimining the investigation levels.

As neil88    say’s you have to be careful about this approach since it relies very heavily on assumed outcomes and that enemy of H&S “good old common sense”.

The point of risk assessment is not to create 5 x 5 matrices of varying degrees of complexity but to decide whether a particular process is safe(just that).

Post #42 suggestion is that you should avoid the worst case scenario as they are unlikely to happen, giving an example of the likelihood of someone falling and hitting their head on a desk and dying or being seriously injured. The fact is such scenarios do happen- the worse injury I had to deal with (in an organisation that dealt with biological agents, toxic chemicals, radioactive materials etc) was someone falling over a box of files and severely injuring their shoulder. The unfortunate lady required reconstructive surgery and ended up taking early retirement.

I am not saying that a suitable and sufficient risk assessment would have magically saved her shoulder but the process would have highlighted the issues in that area and focused the minds of some very complaisant staff.

Similarly the issue of paper cuts. We had staff working in registry where they handled hundreds of documents every day. We could have treated this as just an occupational risk but it was having an effect on morale and efficiency, so we decided that staff working in that area should wear cheap disposable cotton gloves. Problem solved, but note the cotton gloves approach only applied in the registry, not in any office. There was no blanket policy as the risk assessment process identified clearly the area which would benefit from the approach to most and enabled us to justify   not applying it elsewhere.