Note: The following excerpt is taken from Chapter 4 of the Joint Commission Resources book, Advanced Lean Thinking: Proven Methods to Reduce Waste and Improve Quality in Health Care (©2009 Joint Commission Resources). For more information or to order the book, click here .

Error Proofing
Steven Spear, senior fellow at the Institute for Healthcare Improvement, puts it very well when he describes the problem of medical errors and mistakes: Out of approximately 33.6 million hospitalizations in the United States each year, as many as 88 people out of every 1,000 will suffer injury or illness as a consequence of treatment, and about 6 of them will die. Every 15 to 20 minutes, 5 to 7 patients will die as a result of medical errors and infections acquired in health care organizations, and 85 to 133 patients will be injured.¹

It has also been estimated that for every death due to a medical error, there were 10 injuries and 100 instances where injury was prevented.² Most accidents are avoided, but these “near misses” can provide rich information to lead to systems improvements. The use of Lean thinking can rout out those pitfalls to prevent harm and sustain patient safety and quality. Error proofing, known in Japanese as poka-yoke, applies concepts that can prevent mistakes in a process. Error proofing specifically targets processes that are most likely to generate mistakes.

It seeks out risky conditions and revises them to reduce risk. It encompasses the ideas from knowledgeable staff to improve a process’s or area’s flow. Clinical examples of error-prone processes include medication administration errors, wrong site surgery, incorrect patient identification, delays to diagnosis, equipment and device visual controls, and hand washing/infection control. Patient safety alerts exemplify error proofing at its best, used in a clinical context. Efficiency and error prevention can be increased in the nonclinical areas of billing, medical records, and appointment scheduling. Error proofing reduces costs to the organization by reducing duplication of work and has advantages in saving waste in all areas.

Poka-yoke, or error proofing, is one of the main components of the Toyota Production System pioneer Shigeo Shingo's Zero Quality Control system. Anyone from a front-line nurse to a CEO can develop a poka-yoke, and it may look very different in various health care situations.

Although it might seem obvious that anyone would want to stop and prevent mistakes from happening, skeptics might want to adhere to the familiar and what they have always done. Clear and concise data regarding variations and successful outcomes can help convert the reticent. Physicians are traditionally slow to change and may be especially reluctant to apply standard work and error proofing to the area of diagnostics. Clinicians who are approached to participate in Lean thinking may feel threatened and misinterpret Lean to mean that they are being asked to abandon the art of medicine, a cherished tenet. Quite the contrary: Methods, techniques, or equipment that can help standardize a high caliber of cognitive thinking and promote greater clinical decision making will help increase diagnostic accuracy, reduce diagnostic delay, and improve treatment outcomes. Because Lean thinking is a fluid and additive concept best learned through experience, physicians may best see the benefits of Lean in all areas by being shown data from successful project outcomes.

Steps in the Process
Developing a Lean mistake-proofing system or device involves five steps. These steps are as follows:
1. Define the mistake or defect and identify the conditions that may lead to mistakes (perceive).
2. Conduct a root cause analysis (analyze).
3. Generate ideas to eliminate mistakes from happening in the future and to choose standards to follow (standardize).
4. Communicate the conditions that may cause errors (alert).
5. Develop and implement the methods or devices to prevent mistakes in the future (error proof).

Perceive
This stage involves changing your perspective to begin looking for errors and their sources. Then, adopt a habit of looking for opportunities to eliminate the sources. A mistake occurs when there is an intended planned action but the result fails to meet that intention. Also, the plan itself might be a source of the mistake. For example, not checking the identity of a patient before administering a drug is a faulty plan even though the intended action is to give the correct drug. Another example is using a weight on a patient’s chart to calculate dosage, not realizing that the weight was estimated in the emergency department before the patient got to the floor.

Mistakes can result from situations that include confusing instructions or directions, errors in calculations or the data used for calculating, interruptions and distractions, carelessness, hurrying, poor training, and poor follow-up or confirmation techniques. Defects are the end result of mistakes from any source.

The first step for the error-proofing team will be to define mistakes and to differentiate between mistakes and defects. Allow the group to brainstorm all the possible sources of mistakes. Next, give examples of these mistakes and defects from the work environment. Then, categorize the different types of mistakes (rushing, poor directions, etc.).

Analyze
The team will then work together to identify the defect, the resulting condition, and where the defect is found. You must be able to identify and describe the problem or defect in depth, including the rate of defect over time. Documenting the details of the mistake may provide a means for a group to discuss the mistake or defect. Ask the following questions as you record the error:

• When does it occur?
• Where does it take place and what is the defect that resulted?
• Who identified the mistake or defect?
• Who made the mistake?
• Did it happen because of variation of the standard operating process?
• What was the variation that occurred?
• Why was the standard varied?

When performing any analyses, keep from assigning any blame. Simply identify “what happened” as fact rather than put any interpretation to the events or activities. Using the 5 Whys tool can help uproot the answers to these questions. Although the cause of an error may seem obvious, asking “Why?” five times can help reveal source upon source of the mistake or error.

During the analyze phase, don’t jump to conclusions. Keep seeking roots and the roots of roots. Any information that pertains to the mistake should be collected and analyzed. Identify the “red flag” conditions, which include repetition or routine, processes that are only seldom conducted, making faulty assumptions (such as similarity of patient names and drug names), and relying on memory to execute a complicated process. Other red flag conditions include (but are not limited to) the following:
• High patient or client volume
• Poorly defined standards
• Poorly defined sequences of processes
• Poorly designed devices, equipment, supplies, and packaging
• Computerized alerts that can (and often do) get overridden by busy providers
• Lack of methods to confirm accurate information
• Ambiguous inventory systems where location, structure, or setup lead to confusion
• Reliance on illegible handwriting and medical abbreviations in physician orders
• Situations in which providers can be interrupted or distracted when administering medications or caring for patients
• Poor communications between providers at shift transitions and from hospital physicians to primary (outpatient) caregivers
• Lack of confirmation systems when using look-alike/sound-alike medications
• Lack of training for nonproviders who interact with patients
• Reliance on word of mouth of the patient or client in regard to providing medication and other health care history
• Environmentally challenging conditions such as temperature fluctuations or lighting defects

Don’t make assumptions about the validity of the information you are preparing for analysis.

Information must be current, accurate, and verified. Current means as up-to-date as possible. Be precise about the data that are being collected and analyzed or the solutions may not be accurate.

Standardize
This step is for generating ideas. Create standard work instructions and communicate them to staff members who do not follow them. Such communication can be handled through a document that is already in existence. Obtain consensus on any new standard work procedures. Cost allotments should be judged on return on investment. If errors and defects can be eliminated without a major cost, go that route.

However, keeping the patient as the focus, don’t avoid cost that could provide the best outcomes—safety and quality come first.

Alert
Communicate to staff the conditions that may cause an error. Create “alert” conditions, instituting visual controls as much as possible to remind staff of risky red flag conditions.

Error Proof
Create error-proof devices or systems. These may be as simple as standard instructions and forms. Implement the techniques. Create a system to follow up with noncompliant staff.

This last step is the time when the team identifies and implements the level of protection that is most warranted and cost effective. Solutions, devices, and methods are put into place. This is often a difficult step to accomplish because of differences in priorities and contingencies at any given time. Each priority should be followed up by the team leader so that momentum will not be lost. Management should also work to help the team receive the resources it will need to effectively put this step into practice. The list should be posted where everyone can see continual progress improvement. A visual chart to provide at-a-glance progress is helpful to keep on track.

A patient safety alert (PSA) system is a means to implement error proofing. The PSA system involves calling out anything that may harm a patient before injury is incurred.

References
1. Spear S.J.: Fixing health care from the inside, today. Harv Bus Rev 83(9):78–91, 158, 2005.
2. Printezis A., Gopalakrishnan M.: Current pulse: Can a production system reduce medical errors in health care? Qual Manag Health Care 16(3):226–238, 2007.