Human errors and education - Reason - 2000 - Article

There are two approaches when it comes to the human error problem. They differ in their opinion on causation and error management.

Human error approaches

The person approach

The person approach is very similar to what has earlier been called the ‘sharp end’. It focuses on the unsafe acts, errors and procedural violations by people such as nurses, physicians and surgeons. They are being caused by deviant mental processes such as forgetfulness, lack of motivation and recklessness. Error management is aimed at reducing unwanted variability in human behavior. Think about posters, procedures, punishments and blaming. Errors are often seen as moral issues, where bad things happen to those who are not good at their job. This is sometimes called ‘the just world hypothesis’.

This approach is still dominant in the medical field. There are several reasons why people choose this approach. First of all, it is much more satisfying to blame a person, than to blame an entire institution. Secondly, taking distance from the error and blaming another individual is in the best interest of those in charge.

The person approach has serious shortcomings however. There are many errors for which identifiable individuals cannot be blamed. Also, it can be the best people who make mistakes. It is not only the bad, lazy and unmotivated workers who make mistakes. Finally, work conditions can provoke mistakes. They can be in such a way that it doesn’t matter who is working, the mistake is nearly inevitable or very likely to be made.

The system approach

This approach doesn’t ‘blame’ the person, but seeks for a cause within the system. Error management is based on the assumption that we cannot change humans, but that we can change the conditions that they work in. There is a big focus on defenses. When an error occurs, they do not look for which human to blaEffective risk management should focus on enhancing human performance in all the levels of the system, and not just on minimizing certain errors.me, but they investigate why the defenses weren’t sufficient.

The Swiss cheese model

The Swiss cheese model shows us how defenses can be penetrated by an accident trajectory. The system approach focuses on creating defenses. In the reality, these defensive layers are not completely intact and have holes in them. Just a single hole in one of the defensive layers does not immediately cause an error. But if the holes align and overlap, they create an accident opportunity.

There are two reasons these holes exist. Active failures are the unsafe acts that are committed by people who are in direct contact with the system. Examples are slips, lapses, mistakes, fumbles, and procedural violations. They usually have a direct and short impact on the defenses. The person approach usually does not look any further than this. However, in the Swiss cheese model there is another reason that is causing holes in the defenses and that therefore helped creating the error. Latent conditions are the inevitable resident pathogens within the system. They are the consequence of decisions by top level management. They have two kinds of adverse effects. They can create error provoking conditions in the workplace (e.g. timepressure), and they can create weaknesses in the defenses (e.g. bad material).

These latent conditions can exist for years without problems. The error arises when they latent conditions align with the active failures. However, it is not only the active failures that can be blamed then.

Managing errors and high reliability

Error management focuses on two aspects. It tries to prevent dangerous errors and tries to create systems that are better able to deal with errors and their effects. A system has ‘safety health’ when it is able to deal with the operational dangers and still achieve its goals.

High reliability?

Reliability is defined as a dynamic non-event. It is dynamic because safety is guarded by human adjustments. It is a non-event because successful outcomes rarely call attention to themselves.

High reliability organization are systems that operate in hazardous conditions that have fewer than their fair share of adverse events. They can adapt themselves to match with local circumstances. In their core, they are being managed in the conventional hierarchical manner. But in emergency situations control is in the hands of the employee on the spot.