In this video, we will learn about the learning from defects tool. What is a defect? A defect is any clinical or operational event or situation that you don't want to happen again. The learning from defects tool provides a structured approach to identify system factors that contribute to defects, plan improvements and sustain improvements. The tool is a PDF document that has been linked for you in the course of resources. This is the learning from defects tool. As you can see the tool outlines what is a defect, how the tool can be used. It highlights the four basic questions that we discussed earlier: What happened, why it happened, how will you reduce the risk of the defect happening again and how will you know the risk is reduced. Each section of the tool walks you through different questions to help answer those questions. If we think back to the sample defect of that report that we had submitted, we can dive into how this tool can be used. This was the event that we submitted. When registering Sandra Johnson (Date of Birth 10/31/56), I accidentally selected Sandy Johnson (Date of Birth 10/13/66). Luckily, when applying the patient's identification armband the patient told me her date of birth was wrong! It's important to always check two patient identifiers when registering the patient. The medical records system should highlight patients with similar names and dates of birth. In the next few minutes, we will dive deeper into this defect to understand how the tool can be used to help us identify system issues and strategies that we can put in place to prevent this from happening again. The first part of this tool is designed to assess what happened. The tool should be filled out as part of a multidisciplinary event discussion, perhaps something like a CUSP team or at a patient identification meeting. In this case, the first question that the tool prompts you to respond to was who was involved? In this case, it was the registration staff member as well as two patients with similar names. The second question that the tool asks you is, what actions occurred? In this event, the wrong patient was actually selected from a dropdown list and registered. The third question the tool asks was, what were the care team members thinking at the time of the event? Well, through further investigation we know that the registration staff reported feeling rushed, it was a very busy time of the day. In addition, the tool asks you, what were patients thinking and feeling? Well, we don't really know what the patient was thinking in this case, but we can assume that the patient wasn't feeling well since they came to the emergency room, which is where this event occurred. The next question the tool prompts you to answer is, what was happening at the same time? This event happened to happen during change of shift. The tool then asks you, what happened that had a good outcome? Often in events there are things that actually went well. In this case, the patient spoke up that their birthday was wrong, which was really what alerted staff to the fact that there was an error. What had happened that had a bad outcome? In this case, medical records had to be called to correct the charts once the error was identified. This caused a delay in registering the patient as well as providing care. The last question in this section of what happened is, what tools or technologies were being used and how were they being used? In this event, the electronic medical record was being used to register the patient for care. In section two of the tool, we will identify why the defect happened. By drilling down on contributing factors and evaluating whether each factor increased the harm or risk for harm or decrease the harm or risk for harm, we can identify factors with the greatest impact. So, for example, in the first section, the tool asks you to identify a patient or family characteristics. In our defect, the patient was sick but not disoriented. The group felt when filling this out that the patient being sick did not really have an impact on the overall defect. The majority of patients in our organizations are sick. Another example of a factor, patient or family characteristic factor, was that Sandy can also be a nickname for Sandra. This likely increased the risk for harm. Another patient or family characteristic is that at a glance, the birthdays are similar for the two patients. One patient's birthday was October 31st and the other was October 13th. When we think about task factors, for example, there were no education materials for the registration staff on how to look up the patients, this likely contributed or increased the risk of harm. In addition, the policy of checking two patient identifiers was not followed, also increasing the risk for harm. When we think about caregiver factors in this situation, the registration staff were rushed, the group felt that this could have impact the risk for harm. It was also change of shift, so it was incredibly busy. The team felt that that did not actually have an impact in this situation. When we think about team factors, we really didn't have any listed in this example, but you could say that perhaps the team factor was that the patient felt comfortable speaking up, which may have decreased the harm or risk for any as the family member was able to help us cut the events quicker. Knowledge and skills; The registration staff have not been trained on how to search for patients, and every registration staff member searches differently, making it hard to have a consistent standardized process. When we think about technology factors that may have contributed to this defect there are several of them. The medical record does not include patient's pictures. So, there's not a visual cue if you have registered the wrong patient. In addition, the medical record does not highlight similar names and dates of birth. There's no electronic notification alerting you to that. The medical record also requires the staff member registering the patient to pick from a very long list of patients if it had perhaps narrowed the list down in some way, perhaps the staff member would have been in a better position to select the correct patient. When we think about the local environment, this happened on a Friday night and Friday night is usually a very busy time in an emergency room. We did not identify any institutional factors or any other factors involved in this event, but you can see that this tool has spaces for those as well. So, when using this tool in your groups, the goal is that you really take one of these contributing factors and you keep asking why, what contributed to that fact, why did it happen, why did it happen, why did it happen? So you can use the five whys approach to make sure you're really understanding the root causes in going deep into what contributed to the defect. In this section, we will identify how to reduce the likelihood of the defect from happening again. A way to do this is to plot contributing factors based on the frequency with which they occur, and whether they are major or minor reason for why the defect occurred. This is a great activity to do as part of a group. The goal is that a good targeted intervention should be something that addresses factors in the upper right hand quadrant. You can see on this example, the patient being sick was something that occurs often but was a minor reason contributing to why the defect occurred. However, no education or having a standard or not having a standardized process for patient lookups is something that occurs frequently, and is a major reason why the defect occurred. Similarly, having similar names and dates of birth is something that occurs fairly frequently, and was also a major contributing factor in this defect. In this next section, we will identify who our key stakeholders are and what is expected of them. In this case, medical records will help us develop a standardized procedure for searching and creating education documents for registration. Jackie is a staff member in medical records and will be assigned to help with this. We will also check back in on this action item in one month. Another stakeholder in this is our IT colleagues. They will help develop an algorithm that highlights patients with similar names and dates of births. In addition, they will help us explore if patient pictures could be added to the electronic medical record or patients wrist band. Molly is our contact from IT who will help us with this. As this is a fairly large request, we'll check back in on this one in two months. The third stakeholder that we've identified is the registration staff manager. In the short-term, this registration manager will re-educate staff on the current patient identification policy that requires name and date of birth be verified when registering. Jen is the manager for this and we will follow back up with her in two weeks to ensure that this has been done. The last section of this tool helps you know how the risk will be reduced. A good part of any intervention is making sure that you have a means to evaluate it. So in this case, we have decided that we will create a standardized procedure for searching, and create name alerts for patients with similar names and date of birth. There are two ways that our group will work to measure the success of this intervention. The first is to count the number of times the wrong patients information needs to be corrected by medical records. This record is when an event happens and the wrong patient is selected, medical records has to fix the charts and so this is a way we can identify when it is happening, provided somebody catches it. This will be recorded on a safety dashboard, and we will follow up in one month to make sure that this process is in place. Another way that we will attempt to measure the success of this intervention is through a survey to registration staff about the new search functionality and procedures. The patient identification work group for the organization will help design and implement the survey. All data will be stored on a secure shared drive and we will follow up in three months on the status of the survey. It's now time to brainstorm some potential solutions. This is another activity that should be done as part of a group. The first solution could be to tell staff to be more careful. Another solution could be to re-educate on the importance of verifying two patient identifiers as outlined in the organizations policy. Another solution could be to develop standardized procedures for patient look-up that maximizes the chance of the registration staff actually finding the correct patient. Another solution could be to create electronic notifications alerting staff to patients with similar names. And another solution could be to consider including patients pictures in the medical record and on their identification wristband. When we think about these solutions, it's important to strive for solutions with stronger air proofing strategies and forcing functions making the defect less likely to happen again. You can see that including patients pictures in a medical record or creating notifications to staff alerting them to patients with similar names is a much stronger intervention than just telling staff to be more careful. Lastly, as the learning from defects tool is really in the spirit of high reliability, it's important to remember that reliability is not bankable, nor contingent on how many failure-free performances lie behind.
