David Slater May 5, 2020
Now that many countries are starting to emerge, blinking, into the dim light at the end of the first, (hopefully the last!), tunnel, we are all faced with an impossibly difficult decision.
How do we decide what measures are going to be effective enough and sufficient to protect workers and the general population from this new, unforeseen (major?) hazard, the Coronavirus?
Our community of safety professionals is currently mostly self-isolated and disengaged with the occupational exposure problems being tackled and answered by the healthcare professionals and epidemiologists on past and emerging experiences. So to us it seems a bit of an academic question, we know everybody will do as we’re told, as the powers that be decide.
But how will they decide? Shouldn’t we as professionals, recognise and sympathise with them as to the difficulty of the task, that in the end will be made as much on socio-politico- economic grounds as on traditional “Health and Safety” principles? But on what principles could / should such an estimate be based? How would we as risk and safety professionals do it in our normal, personal contexts and advise our managements on such a decision?
I realise we haven’t got the time to do the full “Cullen” safety case, or write the “normally required” Safety Management Systems and Procedures, but there are some techniques we use that could at least appraise and hopefully endorse the measures that will be mandated. Confirmation could help both better communication and confidence.
Some of us might look at reasoning with cheese slices, or drawing simple Bow Ties, with face masks as the first barrier, followed by “distancing” and then backed up by “Training and procedures” to ensure a disciplined behavioural response.
Some might then use a Layers of Protection Analysis approach to put probability of failure estimates on each of these barriers to get a probability of exposure. After the knot, the clinical data on the development of the event tree is sadly much more accurate and hence a probability of fatalities can be obtained.
Traditionalists might use fault trees to flesh out the front end and Behavioural Based Safety techniques might give estimates of the effectiveness of the Human barriers (and human frailties / realities?). More advanced thinkers might try to build STAMP and FRAM models, but it will / must be possible to come up with a risk estimate for any combination of safety measures that may be required for easing the lockdowns.
So we’ve got an idea of the relative effectiveness of individual and combinations of “barriers”, what do we do with the results?
Presumably the governments are already using the numbers in their Cost Benefit Analyses on the economic and human value of a life.
Are the health and Safety Regulators using them in assessing their “Tolerability”?
Or, because of the complexity and urgency of the challenge, are we going to have to resort to and recognise the Victorian compromise of “As low as reasonably Practicable / Achievable”. (In environmental regulation this was BATNEEC, or its irreverent alternative formulation CATNIP).
Most of the advice on setting these criteria and standards (and hence implicitly accepting, but not expressing quantitative risk levels) seems to be based understandably, on the advice from committees of academic epidemiologists and Medical practitioners. Risk based predictions don’t seem to have been taken very seriously in the past.
But seriously shouldn’t we challenge all this under-utilised (at the moment) talent and experience of the safety professionals to produce some useful formal approaches as reference risk assessments and safety case numbers. This could be a small (but helpful?) contribution towards assisting our decision makers to get it right. We all know that there will be plenty of people ready, qualified and quick to point out where they might have got it wrong!