Last week’s blog discussed some observations we had upon coming back from the National Safety Council Congress and Expo in Chicago. They were related to the hierarchy of controls, which is a concept a lot of people don’t get or get wrong. So we thought it would be a good idea to go back and look a little deeper at the hierarchy of controls so we understand why it is so important.
So, first of all, the hierarchy of controls is a decision-making model that helps us when trying to reduce risks from a hazard. This of course implies that we’ve done our homework by identifying the hazard(s) and assessing the risk(s). The hierarchy of controls is meaningless without those steps happening first.
With that out of the way, lets take a look at the hierarchy of controls found in the ANSI/ASSE/AIHA Z10-2012 standard:
- Engineering Controls
- Administrative Controls
- Personal Protective Equipment (PPE)
Now, lets assume that we wanted to protect ourselves and our employees from a hypothetical hazard – a tiger. Tigers are obviously dangerous and present a serious risk if not controlled. So, what’s the most effective way to protect ourselves from a tiger?
If you’re like most people, your first thought was to use a cage. But consider this – wouldn’t it be safer to eliminate the tiger by either removing it from the work place, or, better, not bringing in a tiger to begin with? Yes! As we’ve seen from new stories of tigers getting out of cages or cages being left open and people being hurt or killed by the tiger, cages don’t always work. But elimination, the top of the hierarchy of controls, always works!
If we can’t eliminate the tiger, because we need an animal, the next question should be – can we get by with a house cat? House cats can still be dangerous, but not nearly as dangerous as a tiger. This is substitution of a risk with a lower risk. Again, this is a very effective control because substitution always works, if the substituted risk is well chosen (i.e. you don’t accidently introduce a worse risk).
Sometimes though we really need to have a tiger, so the next best thing is a cage, also known as an engineering control. Engineering controls often require some work on our part to ensure that they work, but less work than the below controls.
If we, for some silly reason, decide that we can’t have a tiger in a cage, then the next best thing is to keep people away from the tiger with warnings. Warnings appeal the innate human psychological need to not die! If a tiger comes your way you probably don’t need someone to tell you that its time to leave the area, you just need to ensure that you are aware when the tiger comes around. Hence a warning, such as an alarm system would work.
If, however, we want our people to interact with the tiger, but we don’t want the tiger to attack them, we likely should provide them with training on handling tigers. This is an administrative control. These controls attempt to control those behaviors that are not innate – i.e. they must be trained into the person and controlled by supervision.
Finally, if we have to have a tiger, and the tiger has to be out of the cage sometimes, and we have to have people around the tiger, and the tiger will sometimes attack the people then we have to rely on our last line of defense – PPE. We will put everyone in some sort of tiger protective suit (if there is such a thing).
Obviously the above example with the tiger is a little silly, but substitute any other hazard in there yourself and see how the hierarchy works. The principles are the same, even though the specifics change based on the hazards and the situations.
Why have a hierarchy of controls? Because it reminds us to get out of the rut of always defaulting to the “easiest” controls. When given a respiratory hazard most of us first think about respirators. When given a fall hazard most of us think about fall arrest harnesses. But the hierarchy reminds us that our first questions should be – Do we have to have that dangerous chemical here? Do we have to have people go up there? The higher up we go on the hierarchy we more the risks are reduced (or eliminated in some cases), the less human error we often get, the less training that is required, the less supervision is required, and the less life-cycle costs are involved.
Now, things are always more complex than we make it in a short blog. For example, many times we need to have multiple layers of controls to ensure that the risks are acceptable. But if we can start thinking deeper about the controls we implement and move our organizations higher up on the hierarchy the safer we’re going to make ourselves and our organizations.