Why develop a back-up plan for a reliable process? (Part 4 of 5)

Frank Federico, RPh, IHI Executive Director

Welcome to session four of our five-part series on developing reliable design. So, we’ve talked about developing a high-level flow diagram. We’ve talked about standardization. Now we are going to talk about developing a back-up plan, or a redundancy. We like to use the word “back-up plan” because “redundancy” might have a negative connotation for some people, but ultimately that’s what it is.

In order to develop a back-up plan, you have to think of the reason why we want a back-up plan because it feels like a use of resources that could be used in a different way. Well, we know that even with the best designed reliable process in the first step of standardization, we don’t always achieve 100 percent, and sometimes we don’t achieve even 90 percent, so it’s important for us to have something in place to mitigate when that happens, to be able to ensure that our patients get appropriate care.

The first step is a process by which we say, if you can get the first standardization step to at least 80 percent of reliability (that’s “at least” — it doesn’t mean it can’t be more, of course it can), but when you get to about 80 percent and you continue to work and try to get it better, and it’s just maybe 85 percent — it’s tough to get any better — then it’s time to think about, “What is the back-up plan? What do we need to do?”

So think of the math. If you have 100 patients and 80 percent of them get the reliable care, that means you have 20 who don’t get it yet. You then develop a back-up plan, which is also 80 percent reliable; now of that 20, 16 patients get the care. When you add 80 and 16, you have 96. You now have 96 percent of your patients getting reliable care. That’s the concept of “at least 95 percent or better.” So remember, the first step “at least 80 percent reliable.” (If you do better, that’s great.) Have a back-up plan to then catch all of the patients who aren’t caught in the first plan, and that will get you to your higher level of reliability. Now this is a safeguard that you will have to have in place in order to ensure that your patients get the right care.

What are examples of back-up plans? One hospital had a protocol for all patients with congestive heart failure. However, sometimes the patients were admitted to the hospital and were not put on that protocol, or sometimes patients were admitted but did not develop congestive heart failure until they were in the hospital. One thing that the hospital looked at is, “What’s a common element in the care of all of our patients?” They determined that pharmacy, in their role of checking medication orders and distributing the medications, could be a second back-up step. It could be someone who could check those orders — when a dose for ¬¬¬¬¬¬¬furosemide or another diuretic reached the pharmacy, their job was to call the unit and ask the nurse, “Is this patient on the congestive heart failure protocol?” Because if the patient was, then you’re done. If the patient is not, then that was an opportunity to mitigate the failure of the first step: Put the patient on the protocol — now the patient is going to receive that care.

Another method is to engage patients. Can you call a patient after discharge and ask them, “Did you receive your Pneumovax while you were in the hospital?” And if the patient says “yes,” then you’re done. If the patient says “no,” then that’s an opportunity to mitigate: Either contact the primary care provider or have the nurse go to the house to give that vaccine.

Another one is an example that we use in primary care, where a group is looking at how to provide nutrition counselling for their teenage patients. They had a very standardized process to ensure that information was made available to the teenager while the doctor was in the room. But sometimes the process failed. They were busy, other things got in the way. What they did is develop a back-up plan, by which every time one of the patients — in this case, the subset of the population “teenagers” — checked out, the person at checkout would ask them, “Did you receive nutrition counseling during your visit?” If the patient said “yes,” they’re done. If the patient said “no,” then that person at checkout would make sure that a nurse would provide nutrition counseling for that patient.

That back-up plan also needs to be standardized. And as much as possible, do not create another process for the back-up plan, but incorporate it into the work of what people are already doing, just as I explained in my examples. This was not add-on work, it was part of what they did as part of their jobs.

When you develop the back-up plan, think of it using the same processes for the standardization of who, what, when, where, how, and with what. You have to be able to answer those questions.

Now, when should you move to the back-up plan? The back-up plan you should begin developing when your standardized process is at least at 80 percent reliability, if not more. But do not start your back-up plan until you’re at that point. If you are not that reliable, having a back-up plan just creates more confusion, creates more chaos, and, again, that goes back to the variation that we were trying to eliminate from the very beginning.

When will the back-up plan fail? The back-up plan will fail when it becomes the primary process. Now what does that mean? So, if you develop a process to ensure that the head of the bed is always elevated for a ventilator patient or that a patient always receives an antibiotic before they get to the operating room, and that continues to fail, and the back-up plan becomes the only way that the patient gets appropriate care, you then have the likelihood of putting pressure on the wrong part of the system — at which point, that back-up plan will fail. It is no longer viable. It is important to go back to the original process and understand why it’s failing because that’s the primary process. That’s the one that should work at least 80 percent of the time.

It will also fail if it’s never used because people will say, “Well, why should I check it? One hundred percent of the time it’s working,” and they’ll stop checking and being the back-up plan, and that results. Now when there is a failure, somebody doesn’t follow up on it.

It also may fail when it becomes just another function in a really busy day. That’s why it’s so important to incorporate it into work that’s already being done.