Our next topic is Failure Mode and Effects Analysis, for short is an FMEA. Here's a really great definition that the joint commission provides, and that is a failure mode and effects analysis is a systematic method for examining a process prospectively, by the way a very important word, because it makes this different than an RCA that we talked about previously. So, it's a systematic method of examining a process prospectively for possible ways in which failures can occur and then redesigning the process to eliminate the possibility of the failure, to stop the failure before it harms an individual, or to minimize the consequences of failure. Failure mode and effect analysis has been with us for quite a while. Was actually developed by the military back in the 1940s, to help us guide missiles and to make the missiles more accurate. Some years later, it was used by NASA to help get the man on the moon. So what they had to do, since they had never done that before is anticipate, what are the failures that might happen, and how might we prepare for those failures, if they do happen. Some years went by, and the automotive industry got involved following the Ford Pinto tragedy. The Ford Pinto was designed very poorly and actually was quite hazardous. The gas tank was way toward the back. So, if you had a rear end collision, the car would explode. Well, that should have been able to be detected, before they ever put them into production by using an FMEA. So, that wasn't done. But since 1979, the automotive industry had become quite strong advocates of failure mode and effect analysis techniques. It wasn't until the 1990s, that health care started using the failure mode and effect analysis, and we should be able to use this technique, and we do to our benefit in making our healthcare systems safer and more efficient. It was... Now currently the Joint Commission requires any institution accredited by them, to conduct a proactive risk assessment, but use the failure mode and effect analysis methodology. And, each institution must do at least one of those every 18 months, involving a potentially dangerous situation. So, here are the major steps in failure mode and effect analysis, as an aside, if you can join a team and actually do a failure mode and effect analysis, because that's really the only way to learn it. But here are the major steps. First, we identify error prone process. An example, preparing chemotherapy. Next, we define all of the process steps involved with that. Next, for each process step, a group of us, a multi-disciplinary group of us sit around a table, and try to figure out how might that fail and we can use some imagination. And so, for each process step that we've identified, we could very easily come up with three or four potential failure modes, that we have to anticipate, that we have to plan for, to prevent those from causing problem. Next, is we estimate the severity and the likelihood of each failure. Then after that, we prioritize which failure modes we're going to address. When you conduct an FMEA, it's not unusual to come up with 50, maybe even 100 different failure modes. And it's often not possible to address all of those simultaneously. So, what you want to do is look at the frequency. If something has moderate severity and it's occurring twice a day, that might get priority over a severe- a failure mode with severe harm, that only occurs once every 10 years. So, we balance frequency with severity to come up with a prioritization scheme, to address a doable number of these fixes as we try to improve our process. The next one is, to change the system. And, as already alluded to the way we change it, is to eliminate the possibility of the failure, or to make failures less likely or to mitigate harm if the failure occurs. Now, as with many things that has benefits and challenges, here are the benefits. We can anticipate and eliminate failure modes before they occur. Multi-disciplinary in nature, that's the failure mode and effect analysis, is a multi-disciplinary process. It seeks input from front line workers. When we're talking about safety, sharp end are those people actually doing the work, blunt end are the managers, the support, the Presidents. Those are the people behind the scenes. Their job is to allow the sharp end people to do their job. So, this is a process where we go to the front line, as well as the support roles to find out what's going on, and how might we fix it. And it provides... The FMEA provides a systematic method for improvement. It's got a lot of challenges and the main one is it can be very time consuming. So, I have now talked about FMEA and in one of my previous lessons, I've talked about RCA. So let's compare these two, because they are similar but there are some important differences. And FMEA is prospective, it's proactive, and it asks the question what if. What if this happens, what should we do? What if that happens, what should we do? It can help prevent harm. And, really important, when you're conducting an FMEA, it should be safe, for the members of the group to be very frank, open, and honest. So, it should be safe for them to be there, and to be very frank. We don't want a boss say, listening to them, and they say would be one of the failure modes is not following the policy. And, the boss might be upset that you're not following a policy. Well, you need to find out why that policy isn't being followed, and you need to encourage people to be very frank and open. In a root cause analysis, that's retrospective, it's reactive. So, in an RCA we're looking at one specific kind of failure, trying to prevent that from happening again, whereas in FMEA, we're looking at a whole host of potential failures, trying to prevent them from happening again for the first time. But in RCA, we're focused on one particular failure that already occurred. We're asking why did it happen? Who was involved, and why they were set up to make the error? It can- what we're trying to do with an RCA, is to try to figure out how can that harm be prevented in the future. But like in FMEA, it also needs frank, open discussions and it should be safe for our employees to have those conversations.
