Just Culture

To collect productive investigative data, we must promote a culture in which employees are willing to come forward in the interests of system safety. Yet, no one can afford to offer a “blame-free” system in which all conduct has impunity — society rightly requires that some actions warrant disciplinary or enforcement action. It is the balancing of the need to learn from our mistakes and the need to take disciplinary action that motivates adoption of a just culture.

A just culture recognizes that competent professionals make mistakes and acknowledges that even competent professionals will develop unhealthy norms (shortcuts, “routine rule violations”), but has zero tolerance for reckless behavior.

Brief yes/no questionnaire for senior managers assessing the organization’s commitment to a just culture.

Instituting a Just Culture

In many situations, the correct action promoting a just culture is not intuitive. It is human nature to become upset and seek “justice” when a patient is severely injured, yet our response to events and near misses should not be based upon actual outcomes, rather upon the potential for patient harm.

Individual responsibility for the problem needs to be decoupled from injury severity. Sometimes an event without any harm should lead to significant employee discipline. Other times an employee may severely injure a patient, yet an analysis of the situation shows that it was the organization’s equipment, processes, systems, or work environment, not the employee, that was at the root of the problem.

To assist in changing event investigation techniques, several algorithms have been developed. On this page we describe two: one by David Marx that is in use in many California hospitals, and the UK National Health Service algorithm used throughout the United Kingdom. These algorithms function differently. To avoid confusion organizations should not simultaneously use both.

Just Culture Algorithm from Outcome Engineering

This algorithm, developed by Outcome Engineering and David Marx, focuses on duty and mechanism of error.

According to this algorithm, there are three basic duties:

1. Duty to produce an outcome. If an individual knows the desired outcome and should be able to produce it (e.g., safe removal of an inflamed appendix), failure to do so represents breach of this duty.
2. Duty to follow a procedural rule. If the individual knows the proper procedure and it is possible to follow the rule (e.g., the procedure for inserting a central venous catheter), failure to do so represents a breach of this duty.
3. Duty to avoid causing unjustifiable risk or harm. Breach of this duty occurs when an individual intentionally harms the patient or acts recklessly.

If a duty has been breached, then the mechanism of the breach is identified. There are three identified causes:

1. Human error. This is an inadvertent act (“slip,” “lapse” or “mistake”).
2. At-risk behavior. Typically, this is a conscious drift from safe behavior, occurring when an individual believes that drift doesn’t cause any harm. An everyday example is the willingness of some drivers to roll through stop signs. Those drivers do not see that as risk-taking behavior, as, in their experience, nothing bad happened consequently.
3. Reckless behavior. In this case, the individual has chosen conduct that he knows poses a substantial and unjustifiable risk.

The response to an event (or near miss) is tied to the mechanism of error. An isolated human error is an opportunity to correct system weaknesses (e.g., confusing drug labels). The individual making the error should be consoled, rather than disciplined. At-risk behavior may also indicate a system vulnerability that should be fixed. However, the individual should be coached so that he understands the risks he has taken. Reckless behavior may be grounds for disciplinary action. The intent is to reduce the risk of future reckless conduct, and may include removing the individual from the organization.

Repetitive problems are often caused by system weaknesses, but sometimes are individual performance issues, particularly when coaching or additional training has not improved the problem. For example, repetitive human errors may be an indication that the individual is not capable of performing safely in his current job. Repetitive at-risk behaviors may be due to impairment (e.g., drug abuse) or unwillingness to follow proper protocols.

Just Culture Algorithm from the UK National Health Service

Incident decision tree adapted from the UK National Health Service’s National Patient Safety Agency (NPSA), and a Guide to Use (from the 2003 version). The NHS also issued Guidelines for Action Following Patient Safety Incidents. This algorithm focuses on identifying whether an individual was at least partially culpable for the event. The following is a brief summary of the decision tree, from the NPSA web site.

The Deliberate Harm Test

In most patient safety incidents the individual had the patient’s well being at heart.

However, in some cases the intent was to cause physical or emotional harm.

The Deliberate Harm Test asks questions to help identify or eliminate this possibility at the earliest possible stage.

The Physical/Mental Health Test

If intent to harm has been discounted, the Physical/Mental Health Test helps to identify whether the individual’s (not the patient’s) ill health or substance abuse caused or contributed to the patient safety incident.

The Foresight Test

If intent to harm and incapacity have been discounted, the Foresight Test examines whether protocols and safe working practices were adhered to.

The Substitution Test

Finally, if protocols were not in place or proved ineffective, the Substitution Test helps to assess how a peer would have been likely to deal with the situation.

Combination of the UK Model and the Outcome Engineering Model

Allen Frankel proposes a combination of the two models, with some simple questions to help investigators walk through the possibilities:

Was the individual knowingly impaired? Did the individual consciously decide to engage in an unsafe act? Did the caregiver make a mistake that individuals of similar and training would be likely to make under the same circumstances (substitution test)? And does the individual have a history of unsafe acts? This short list of questions provides a clear model of accountability and also tells providers “what the rules are,” so they feel safe to speak up so we can learn and continually improve.

Chart reprinted from Patient Education and Counseling, Volume 80, Issue 3, Michael W. Leonard, Allan Frankel, The path to safe and reliable healthcare, Pages 288–292, Copyright 2010, with permission from Elsevier.