The current widespread concern with patient safety started with the published US Institute of Medicine (IOM) report ‘To Err is Human’. This report was quickly followed by a series of national reports that focused on the life lost and the money that was wasted. Around 1 in 10 patients in acute care hospitals were killed or injured as a consequence of a medical error. These medical errors are not truly medical, they are errors that occur in a medical setting.

The Paradox of health care

All paradoxes have at least two contradictory elements.

The first part of the paradox is that health care training is based on a belief that training leads to perfectibility. After the education in medical school they expect you to get it right. That means that errors are equated to incompetence and as a result there has been no tradition of reporting or learning from errors. It is also the case that students do not learn about error-producing situations.

The second part of the paradox focuses on the two most error-producing activities: aircraft maintenance and delivering health care. This is not a very useful model, but it does show the large differences between the two domains:

• Huge diversity of activities and equipment: Long-distance pilots only fly a few types of airplanes whereas health professionals have to work with a wide variety of equipment.
• Hands-on work with limited safeguards: Pilots are not encouraged to touch the flying controls (most planes fly on automatic pilot) but health care work is very hands-on. But the more you touch, the more mistakes you can make.
• Vulnerable and needy patients: Unlike passengers of an airplane, patients are sick or injured. There is also the lethal convergence of benevolence, that means that because the health professionals care about their patients, they do whatever necessary to safe them even if that man’s breaking the rules of bypassing safeguards.
• Local event investigation: Adverse events are mostly investigated locally. This means that the lessons learned are not widely published.
• One-to-one or few-to-one delivery: Health care is very personal.

In short, health care is very error provoking in nature, yet the training in error is non-exiting and the fallibility is stigmatized (this is the paradox).

Models of Medical Errors

There are two unhelpful models: the Plague Model and the Legal Model. And there are two useful models: the person model and the system model.

The Plague Model: One reaction to the high incidence of errors in health care was that it is a ‘national problem of epidemic proportions’. This model want to eliminate the human error, but unlike an actual disease there is no cure. So, we can’t fundamentally change the condition but we can change the conditions under which the errors are made. Another view is that human errors are the product of deficiencies in the human condition. If you hold on to that view the only solution would be to advance the levels of automation to keep humans out of the loop. The problem with this model is that it sees human error as something bad, which is not the case.

The Legal Model: The most important thing from this model is that it has the view that trained professionals should not make mistakes. Errors are rare occurrences, but the ones that cause bad consequences are seen as neglect or recklessness. However, research shows that errors are not rare, and that even highly trained professionals make a lot of mistakes. Especially in health care, errors are very common and are often corrected. Clearly, training does not diminish the chance of making mistakes.

The Person Model: This model sees errors as the product of stubborn mental processes: forgetfulness, inattention, preoccupation, distraction, ignorance, carelessness etc. Remedial measures focus on the patient-professional interface and include: naming, shaming, blaming, retraining, fear appeals and writing another procedure. Although blaming is very common, it is very ineffective and sometimes even counterproductive. It isolates a person from the context, and sits in the way of identifying the errors traps. We should be talking about error-prone situations instead of error-prone people.

The System Model: The base of this model is that humans are fallible and errors are to be expected. Errors are sees as consequences and not as causes. When an adverse event occurs, you should be focused on why did the defences fail instead of who did it.

Conclusion

Since the 1990s there is a great improvement in the patient safety issue that is made more salient and seems to be more significant nowadays. The success can’t be seen in the dramatic decrease in errors, but there are some small progresses in some areas (like ward design, use of checklists). But professionals in health care are still very fallible and the delivery of health care is very error provoking. The changes that are made with this movement are not in lowering the errors but changing the way we look at the errors and that we understand the nature of medical errors that much better.