Hi Svick
Part of the problem is going to be working out what to do when somebody is exposed to multiple hazardous substances.
Looking at it VERY simplistically suppose employee A is exposed to solvents and lead.
How these enter the body and then impact on it are largely different, so you could mostly assess each individually. But I wrote "largely" and if you are in borderline territory perhaps you need to consider the potential interaction much more closely.
However, if we change the scenario such that employee A is exposed to solvent G and solvent H, it is likely that the impacts are additive, so you need to work out what proportion of the Workplace Exposure Limit (or STEL etc etc) they are liable to get from G, the same for H.
So suppose their exposure is at 0.75 of the WEL for solvent G, and 0.5 of the WEL for solvent H, you might be within the WEL for each, but would exceed the notional WEL for the mix of hazardous substances if you add up the fractional exposures in each case as a proportion of the WEL for each. 0.75 + 0.5 = 1.25 = problem!!
Then there is the even more complex scenario of "synergistic" mixtures of exposure, where rather than adding each fraction together you apply a multiplier.
Perhaps the most commonly quoted example is someone exposed to asbestos who also smokes cigarettes (or similar).
Lots of research to indicate what proportion of workers exposed to V fibres/ml asbestos in air for 40 hours a week for many years are likely to then sustain one or more respiratory health conditions.
Ditto research to tell you the likelihood of someone smoking say 20 cigarettes a day for many years sustaining one or more respiratory health conditions.
However, if the exposure is to both the risk is much greater than simply adding together the numbers for each individual contaminant.
If you have such a scenario then the former COSHH e-tool probably wouldn't have given you an adequate assessment of the overall risk.
That took might have been OK for very simple scenarios preferably with only one contaminant to consider, but would be unlikely to produce the required "suitable and sufficient" assessment in more complex situations.