People who have been following me for a while know that I occasionally talk about the hierarchy of controls, which I use in assessing risk and establishing appropriate precautions in industrial environments. 1/12 https://www.cdc.gov/niosh/topics/hierarchy/default.html

I was excited to read that guidance from @SickKidsNews on #SafeSeptember makes reference to this theory, but I observed some controls have been misclassified in the document and I wanted to point a few of them out. 2/12

The first thing I'll mention is that not all of these things are actually controls. Example:

"testing capabilities" not a control
"weekly testing of all students and staff" weak control
"daily rapid test before entering" good control
3/12 https://www.thestar.com/news/gta/2020/07/14/richmond-hill-firm-develops-rapid-covid-19-test-to-get-results-in-less-than-5-minutes.html

Screening and exclusion policies are certainly at best administrative controls. I don't see any elimination or substitution controls presented here, not ideal. 4/12

Under engineering controls we again see a mix, ventilation is vague. If the guidance is to validate that the HVAC system achieves the required air exchanges & filtration spec is listed, ok sure, engineering control. If the teacher is asked to open a window? Not so much. 5/12

Furniture placement, break schedules, cohorting plans are all procedural controls, but only if they are clearly specified and controlled. If there is some risk of people not following the rules it's not an engineering control. 6/12

Even the hand hygiene station at classroom entry is not an engineering control. It is important to do as it supports the procedural control of maintaining a frequent hand washing but it does not control risk just for being installed. 7/12

So we have a lot of administrative controls, which is not great. When assessing risk in a factory, if there is a possibility of serious injury, illness or death, we would want to be using higher order controls. 8/12

But higher order controls are expensive! Sometimes for short term risks we accept weaker controls, but then we add many layers and we hope the holes in each don't line up. Think of layers of Swiss cheese. 9/12

When we uncover new risks in workplaces, sometimes a process cannot be stopped so our first step is to leverage PPE and administrative controls while we work on something better. We should never leave a worker unprotected to a known risk. 10/12

And all this assumes that the prescribed controls are effective. PPE that mitigates the risk is certainly better than than an ineffective procedural control. 11/12

Key takeaway? This stuff needs to be specific and all the details are tedious. Effective controls will be expensive but I can't think of much that's more important right now than the safety of kids and our community. #SafeSeptember 12/12

Someone just sent me this article, the authors do an excellent job of explaining the different types of controls. @AshTuite @AmyGreerKalisz @NishaOttawa https://www.theglobeandmail.com/opinion/article-we-can-get-children-back-to-school-full-time-if-we-put-the-right/