The hierarchy of controls is one of the best-known ideas in occupational safety. NIOSH presents five levels, from most to least effective: elimination, substitution, engineering controls, administrative controls and personal protective equipment. NIOSH notes that the top three are more effective because they control exposures without significant human interaction.
For many organizations, knowing the hierarchy is not the hard part. Applying it is. And many of the reasons controls settle at the bottom of the hierarchy sit outside the safety department’s authority.
Why controls tend to drift down
Speed is part of it. A new procedure or a box of gloves can be issued today, while an engineering change needs design, budget approval and downtime. Budget ownership matters too: a safety team can usually buy PPE, but elimination and engineering controls are typically decided by engineering, operations or capital projects, often long before safety is involved.
There are subtler drivers as well. PPE, signs and training records are easy to show an auditor, while an eliminated hazard leaves nothing to inspect. Contractor and short-duration work rarely seems to justify redesign. And when a control relies on behavior, a failure can be recorded as individual error rather than as a design problem, which removes pressure to change the design.
Put simply, programs often default to PPE not because anyone believes it is the best control, but because it is frequently the only control a safety department can implement on its own, this week, without a capital request.
Interim controls that quietly become permanent
OSHA’s recommended practices on hazard prevention and control advise using interim controls while longer-term solutions are developed. That is sound advice. The risk is that the interim control is where the effort stops. A useful governance habit is to give every interim control on a critical risk an owner and an expiry date, by which a higher control is implemented or the decision is formally re-approved.
Practical moves for leaders
- Review critical risks by control level. For each high-severity risk, record which level each control sits at. A register dominated by administrative controls and PPE is a signal worth discussing.
- Challenge at the design stage. Build hierarchy questions into project reviews, procurement specifications and management of change, where elimination and substitution tend to be cheapest.
- Share ownership. Make hierarchy decisions on critical risks a joint responsibility of safety, engineering and operations, with budget holders accountable for control choices.
- Verify, don’t assume. Check in the field that engineering controls remain in place and administrative controls are actually followed.
- Learn from near misses. Each near miss involving a behavioral control is a chance to ask whether a higher control is possible.
Leaders should also watch for controls that introduce new hazards. OSHA gives examples such as hearing protection that makes it difficult to hear backup alarms.
The same logic in other jurisdictions
The EU’s Framework Directive 89/391/EEC reflects the same preference, listing combating risks at source, replacing the dangerous with the non-dangerous or less dangerous, and giving collective protective measures priority over individual ones among its general principles of prevention. In Mexico, NOM-017-STPS covers the selection, use and handling of personal protective equipment. Meeting a PPE standard is compliance at the bottom of the hierarchy, not a substitute for the controls above it.
Some standards and training programs use more than five levels. CCOHS notes that CSA Standard Z1002 adds a layer for systems that increase awareness of hazards, such as alarms and warning signs, between engineering and administrative controls. Whatever the number of layers, the order starts with elimination.
Any worked control example should be treated as illustrative. Controls are site- and equipment-specific, and they are best validated through a competent hazard assessment of the actual equipment and process.
Closing thought
Moving up the hierarchy usually requires decisions about design, procurement and budget. That makes it a leadership agenda item, not just a safety department task. For the five levels, illustrative examples by hazard and the hidden weakness of administrative controls, read this in-depth look at the hierarchy of controls.
