Preventing Harm to Patients: 4 Different Ways to Put the Right “Barrier”

Understanding and implementing safer systems of work can reduce the risk to patients. We are all looking for safer systems of work and often we put in place what seems like a good preventive step, maybe following on from an incident. There are a number of ways to ensure you are putting in the right “barrier”, and ways of ensuring it is will do what you are expecting it to.


4 types of barriers

Barriers can be grouped into four types:

  1. Human action – checking a drug dose prior to administration
  2. Administrative barriers – training, supervision, and procedures
  3. Physical – protective equipment/storage
  4. Natural – place time or distance

In practice, the easiest barriers to put in place are the human actions, but it is easy to see then why mistakes keep occurring. Barriers can be evaluated as being strong average or weak to counter a hazard. Any barrier involving human action is marked down.

Administrative barriers are only as good as their promulgation and need to be current and constantly reinforced.

Physical barriers are harder to implement as they require investment and often add in a time factor.

Natural barriers can again cause delay and are therefore often overlooked as cumbersome, but if you look at the measures recommended for safer use of potassium solutions, it is the physical separation of potassium, the requirement to treat it differently that makes the barrier effective. Recommendations for intrathecal chemo also uses “time” and separation (“place”) barriers to counter the hazard.


How to evaluate barriers

Barrier analysis can be used retrospectively when an incident has occurred, or prospectively, as part of a risk assessment strategy. A useful tool. Have a go.

Anyone interested in patient safety should look out the DoH/Design Council report – Design for Patient Safety. This is a document that is eye catching and makes a good read.

Leave a Reply

Your email address will not be published. Required fields are marked *