Why It Matters
"Removing the focus on 100 percent reliability can be energizing because it frees us from a nearly unattainable goal."
Processing ...

Stop Wasting Time on 100 Percent Reliability

By Roger Resar | Monday, August 17, 2015
Whether it’s patient safety, value-based payments, population health, or reducing avoidable readmissions, all of the challenges health care providers face today require reliable processes – that is, processes that function exactly as designed, all of the time. What may come as a surprise to some is that these processes don’t have to be 100 percent reliable.

Instead, they should be good enough to give you world-class outcomes.

(Important caveat: Perfection does not have to be the process goal, but it should always be the outcome goal because preventing harm to patients is the ultimate aim.)

Misunderstandings about High Reliability in Health Care
The term “high reliability” is deceptive. Nuclear power plants, aircraft carriers, and the airline industry should aim for 100 percent reliability. It’s not the same for health care. Health care processes have to be reliable, but for the most part they don’t have to be at the same level of reliability as air traffic control.

When I speak at clinics or hospitals, the audience often asks, “How do we get 100 percent reliability?” The concept that IHI Executive Director Frank Federico and I have developed is that in health care, processes need to be reliable enough. That might sound vague, but we’ve determined from our combined almost 50 years of improvement experience that when a process clearly connects to an outcome, the process does not have to happen 100 percent of the time to get an outcome that is 100 percent. This is because in health care we’re often dealing with a biologic system, and a biologic system has resilience.

For example, if I don’t wash my hands 100 percent of the time when I examine a patient, the patient doesn’t get an infection every time. That’s because the human body has resistance. If we could get handwashing to happen 95 percent of the time, we would be at a world-class level of handwashing. When we teach about reliable design, we discourage people from trying to achieve 100 percent reliability. Instead, we teach them how to align processes with outcomes. The question then becomes, “How do you achieve 95 percent process reliability?” In health care, that is more than enough to give you outstanding results.

How do we apply this to the current health care environment? Let’s use population health as an example. To improve the health of a population, let’s say you want to administer the pneumococcal vaccine to all people over 65 and those who have immunodeficiency diseases. You don’t have to vaccinate 100 percent of these people. It would be nice to vaccinate 100 percent of your chosen population, but it’s nearly impossible, so we design our process with the goal of vaccinating 95 percent of these individuals. You might ask, “What about the other 5 percent?” In health care, because the biologic system is resilient, you won’t automatically get pneumonia if you don’t get the vaccine.

The Importance of Imperfect Design
When we teach about reliable design, we ask people to accept less-than-perfect design.

Removing the focus on 100 percent reliability can be energizing because it frees us from a nearly unattainable goal. When we expect a team to reach 100 percent reliability of a process, we set them up for failure and disillusionment.

In addition to being liberating, less-than-perfect design allows us to simplify. The more you strive for 100 percent reliability, the more complex a process becomes. At 95 percent reliability, the process can stay relatively simple. If you’re not sure your process is simple enough, conduct this easy complexity test: If five frontline people cannot explain the process you’ve designed, it’s probably too complicated to implement effectively.

Keys to Sustainability
“Reliable enough” does not mean “anything goes.” Based on our experience, if the reliability of the process is at 80 percent or less, the process is unlikely to be sustainable – that is, to sustain performance at 80 percent reliability. A process needs to be about 95 percent reliable to be sustainable over time.

Sustainability of the process depends upon efficacy and simplicity. Ask yourself these questions about a process you want to implement reliably:
  • Does it work? (Efficacy): The first step in designing a reliable process is showing that its implementation produces the outcome you want. If you don’t have that evidence, the process won’t be sustained in the long run.
  • Is it simple enough? (Simplicity): If the process is too complicated, people won’t be able to consistently do it correctly or they will find ways to work around it.

Getting to 95 percent reliability is still a big challenge, so focus your efforts where they will have the biggest impact. Stop wasting time and effort on the nearly impossible goal of 100 percent reliability. Instead, make sure you have evidence that a process really works to produce your intended outcome and keep it simple.

IHI Senior Fellow Roger Resar is faculty for IHI's How to Design Reliable Processes in Health Care seminar.

You may also be interested in:

Improving the Reliability of Health Care

Video: The Benefits of Reliable Care for Clinicians

first last

Average Content Rating
(0 user)
Please login to rate or comment on this content.
User Comments

© 2023 Institute for Healthcare Improvement. All rights reserved.