IHI Module QI 102- How to Improve with Model for Improvement PDF

Preview

Summary

IHI Module QI 102- How to Improve with Model for Improvement SUMMARY ...

Description

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

Lesson 1: How Can we Improve? Central line-associated bloodstream infections (CLABSIs) are a nasty problem in hospitals around the world. A central line is a tube that is surgically inserted into a major blood vessel, making it faster and easier to deliver medication in an intensive care unit. But when the tube is not put in correctly or kept clean, it can become a way for germs to enter the body and cause deadly infections. According to the Centers for Disease Control and Prevention, more than 30,000 of these types of infections occur in US acute care facilities each year. 1 In 2005, the Saudi Aramco Medical Services Organization (SAMSO), a complex of hospitals in Saudi Arabia, set out to reduce the rate of central line infections in its neonatal intensive care unit — a unit dedicated to treating very sick newborn babies. For every 1,000 days that patients were equipped with central lines — or, in medical terminology, “catheter days” — there were about 11 related infections. SAMSO used the Model for Improvement to reduce that number. Within a year, the rate of central line infections in the unit was down to zero — and it stayed that way for months in a row. Will, Ideas, Execution Whether your goal is to reduce central line infections in the neonatal intensive care unit or to get more sleep and exercise in your daily life, the basic principles of improvement are the same. Let’s hear more about this from Dr. Mike Evans, Associate Professor of Family Medicine and Public Health at the University of Toronto:2 As we discussed in Lesson 3 of QI 101: Introduction to Health Care Improvement, every system is perfectly designed to get the results it gets — the only way to get different results is to change the system. IHI uses a simple mantra to describe the essential elements for systems improvement: “Will, ideas, and execution.”

The engine for the execution step is something we call the Model for Improvement. This lesson presents an overview of the Model for Improvement; subsequent lessons in this course will look at each component of the model in greater detail.3

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

MFI: Three Questions and a Cycle The Model for Improvement, developed by a group called Associates in Process Improvement, begins with three fundamental questions:3  What are we trying to accomplish? (You ask this to establish your aim.)  How will we know a change is an improvement? (You ask this to establish

your measures.)  What change can we make that will result in improvement? (You ask this to determine

the changes you will test.) Next comes the PDSA cycle — Plan, Do, Study, Act. This four-phase process was originally developed by our old friend Walter Shewhart (whom we discussed in QI 101: Introduction to Health Care Improvement). Then Shewhart’s colleague, W. Edwards Deming, popularized Shewhart’s cycle as a way to develop, test, and implement changes. The PDSA cycle gives us a way to quickly test changes on a small scale, observe what happens, tweak the changes as necessary, and then test again (perhaps with a larger or broader test group, if our confidence in the idea has grown). Instead of spending weeks or months planning out a comprehensive change, then putting it into practice only to find that it’s fundamentally flawed, the PDSA cycle enables rapid testing and learning. You actually conduct PDSA cycles every day. We all test out new ideas constantly. Consider these tests, for example:  “I ate breakfast this morning instead of skipping it like I usually do. That worked out

pretty well for me.”  “I took the train to class this morning rather than ride the bus. It ended up taking longer

on the train and was more expensive.”  “Instead of writing out notes while in class, I tried to type the notes into my laptop. That

was a total disaster! I missed half the lecture trying to type faster than my skills allowed.”

Using the MFI to Improve Care Before we return to our real-life example (remember SAMSO?), let’s run through the steps an improvement team might go through when applying the Model for Improvement in a clinical setting:

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

1. Set an aim. A general statement — something like, “We will improve our infection rate” — isn’t good enough. The aim statement should be time-specific and measurable, stating exactly: “How good?” “By when?” and “For whom?” 2. Establish measures. You need feedback to know if a specific change actually leads to an improvement, and quantitative measures can often provide the best feedback. 3. Identify changes. So, how are you going to achieve your aim? Where do new ideas come from? You can spark creative thinking in various ways, and there are tools that can help. 4. Test changes. This is where the PDSA cycle portion of the Model for Improvement comes in. By planning a test of change, trying the plan, observing the results, and acting on what you learn, you will progressively move toward your aim. 5. Implement changes. After you have a change that results in improvement under many conditions, the logical next step is to implement it — meaning, make the change the new standard process in one defined setting.

Improving Care in the Neonatal ICU Remember SAMSO’s NICU? Their improvement team consisted of the following individuals:  Dr. George Cheriyan, the head neonatologist  The nurse in charge of the unit  A data collection clerk  The infection control coordinator  The unit’s quality improvement coordinator

(Learn more about forming an improvement team and collaborating with multidisciplinary team members in QI 105: Leading Quality Improvement.) Here's how the team used the Model for Improvement to cut their central line infections to zero. Their improvement journey consisted of the following steps: 1. Set an aim. SAMSO's aim was to decrease the infection rate in the NICU from 11 infections per 1,000 catheter days to fewer than 5 infections per 1,000 catheter days in nine months.

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

2. Establish measures. Dr. Cheriyan’s team measured the rate at which babies were getting bloodstream infections — their outcome measure. They also measured how often staff members were actually doing the things they were asked to do to prevent infection — their process measures. 3. Identify changes. Dr. Cheriyan’s team decided to test several changes, including:  Having staff implement a set of interventions (also known as a “bundle”) that evidence

showed would reduce the rate of central line infections. Some of the interventions included improving hand hygiene and checking the lines daily.  Instituting a checklist that staff would review during a time-out before inserting the central line.  Requiring staff to place red tags on the beds of the babies with central lines, so everyone would remember to take extra precautions. 4. Test changes. Here’s where the PDSA cycle came in. Dr. Cheriyan’s team trained the unit’s staff on all the changes and testing each change on a small scale. They learned from each test and measured how well everyone stuck to the new protocol. They achieved the following results, which they analyzed in the “study” portion of the PDSA cycle:

The two process measures indicate the staff was doing a pretty good job of using the use of the checklist (blue line), but not as good of a job using the central line bundle (orange line).

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

The blue line shows the outcome measure, how often newborns were getting bloodstream infections. The infection rate dropped after the team began testing changes, as marked by PDSA 1, 2, and 3. After studying the data, the team used additional PDSA cycles to tweak the process so that staff would comply with the five-step bundle more often, and continued to see improvement. 5. Implement changes. Once they had refined the process changes through several PDSA cycles, the team created a new policy governing the placement and care of central lines. They started training everyone who rotated through the unit on the new protocol. In other words, they made the new procedures part of everyday work life. The team successfully changed the way the NICU handled central lines. Eventually, SAMSO spread the same protocol to its pediatric intensive care unit.

Lean, 6 Sigma, Other Models The Model for Improvement is appropriate for all types of improvement efforts and has been used successfully by thousands of health care organizations in many countries to improve a variety of health care processes and outcomes. However, there are other useful models to guide improvement out there. Dr. Don Berwick, IHI President Emeritus and Senior Fellow, will explain why having a standard road map for improvement — rather than the specific framework you decide to use — is the most important thing: For comparison, here’s a high-level look at the basic components of two other improvement frameworks, Six Sigma and Lean: (To learn more, check out IHI's white paper "Comparing Lean and Quality Improvement"):4 Six Sigma

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

Six Sigma focuses on reducing variation, or the defect rate, measured by Sigma level, or “Defects per Million Opportunities.” The Six Sigma improvement framework consists of five basic steps, known as "DMAIC":  Define the problem in detail.  Measure defects (in terms of “defects per million,” or Sigma level).  Analyze under what conditions defects occur by using process measures, flow charts,

and defect analysis.  Improve by defining and testing changes aimed at reducing defects.  Control your results by determining what steps you will take to maintain performance.

Lean Lean focuses on improving value from the customer’s point of view, by reducing waste of time and resources. The basic steps for Lean include:  Specify the value desired by the customer.  Identify the value stream for each product providing that value and challenge all of the

wasted steps currently necessary to provide it.  Make the product flow continuously through the remaining value-added steps.  Introduce pull between all steps where continuous flow is not possible.  Manage toward perfection so that the number of steps and the amount of time and

information needed to serve the customer continually fall. Again, all of these frameworks are useful, and the important thing isn’t which framework you choose; it’s choosing one that works for you and getting to work on improving!5 The Model for Improvement begins with three questions designed to clarify the following concepts: Your Answer: Aims, measures, changes The Model for Improvement begins with three fundamental questions about any given improvement, designed to address the aim (what are we trying to accomplish?), the measures to be used (how will we know a change is an improvement?) and the changes to be used (what changes can we make that will result in an improvement?).

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

Applying the Model for Improvement to the clinic’s improvement goal, which of the following is the most reasonable aim statement? Your Answer: Increase the number of patients reporting they are “very satisfied” with the clinic’s scheduling by 50 percent within six months. An aim statement must specify “how good, by when.” Improving patient satisfaction with scheduling is a reasonable goal. Answer D is best described as an opportunity statement, as it contains no specifics about how much the clinic must improve, nor by when. Answer C is more of a “change” statement than an aim statement. What is the team's next step? Your Answer: Test their change plan using the PDSA cycle. Once you have worked through the first three questions of the Model for Improvement — the questions about aims, measures, and changes — it’s time to do a small test of change using the PDSA cycle. The clinic should have already developed their measures, and now is not the time for a break — because the hard work of improvement is just beginning! The orthopedic clinic plans the change to improve scheduling, and then it carries out a small test of change with three patients on Tuesday morning. What’s the next thing the clinic’s improvement team should do? Your Answer: Measure to see if the change led to improvement. The team has planned a test of change and now they’ve done the test. The team must now study how the test went (the “S” part of the PDSA cycle). They can look at a mix of process measures (such as how often appointments started on time) and outcome measures (such as how satisfied the patients were with the new process). When trying to improve a process, one reason to use PDSA cycles rather than a more traditional version of the scientific method (such as a randomized, controlled trial) is that: Your Answer: Both C and D PDSA cycles allow for rapid and frequent review of data and then adjusting the test of change based upon those findings. For example, if a new guideline that’s meant to improve pneumonia care isn’t working, PDSA cycles allow you to change the guideline quickly and test its efficacy, rather than waiting until the end of a long study period.

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

Lesson 2: The Key Elements of an Aim Statement: If improvement were simple, everyone would be at a healthy weight, speak many languages fluently, and conduct their work and relationships with brilliant ease. No one would be late. Nothing would be mislabeled. Every patient would receive exactly the right care at the right time. It’s a testament to the complexity of improvement that things generally don’t work like this. An organization — or an individual — just won’t improve without a clear and firm intention to do so. The first thing you need is an aim. An aim statement is the answer to the first question in the Model for Improvement: What are we trying to accomplish?1 A good aim addresses an issue that is important to those involved; it is specific, measurable, and addresses these questions:  How good?  By when?  For whom (or what system)?

Here are a few examples of some not-so-useful aim statements:  We’ll do better on tests.  I will lose weight.  I’m going to exercise more often.

How good? By when? Who knows? We could try forever and never get there, because it’s not clear where “there” is. Now, here are a few examples of good aim statements:  Every member of our study group will increase his or her grade point average by 10

percent or more within eight months.  I will weigh 160 pounds or less by February 1.  I’m going to run at least 10 miles per week by July 4. All of these aims are specific about measure, time, and population — in other words, how good, by when, and for whom. Some aim statements also contain information about key strategies and other guidance for the improvement project. But the aim should be succinct and not contain extraneous background

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

information or side issues. It should be easy to understand exactly where we’re going and what we expect to accomplish.

Crafting an Aim Statement: Running 10 miles per week is a lot simpler than getting everyone to wash their hands in the intensive care unit. So how do you begin developing an aim statement in a complex clinical setting? And how high should you set your ambitions? Setting incremental goals can help. And of course, the level of improvement you're seeking will generally be related to how far up the organizational ladder you are. Four Guiding Principles for Setting Aims  Meaningful  Measurable  Be Ambitious  Don’t wait to get Started Having a clear aim statement is important in quality improvement work because: Your Answer: Aim statements provide a clear and specific goal for the organization to reach. Whether you’re trying to reduce your commute time or cut down on the incidence of surgical-site infections, having a clear and specific aim statement makes your project more likely to succeed. Good aim statements include a specific, measurable goal, a deadline for achieving the goal, and information about which population will be affected: how good, by when, for whom. They do not, however, remove all obstacles from the process. And while many funding requests and leaders require strong aims, it's not always a requirement.

An aim statement should include the following: Your Answer: Numeric goals, specific time frame, and the patient population or system affected Aim statements should specify measurable numeric goals, a time frame for attainment, and the group or system affected. Costs and team members, while important to the success of the quality improvement project, are not part of the aim statement itself.

Which of the following is the most effective aim statement for this project? Your Answer: Within three months, the emergency department will administer all pain medications within 45 minutes of order time. Effective aim statements contain a time frame, a definition of the population to be affected, and specific, measurable goals. Answer B meets all three of these criteria. While answers A and C may be useful process changes to reduce the delay between the ordering and administration of medications, they are not aims in and of themselves. Option D is not specific enough, as it does not contain information about how much the department should improve.

HLSC 3P94

IHI Module QI 102: How to Improve with Model for Improvement

The charge nurse in the emergency room asks Brenda to assemble a team to improve the delivery of pain medication. As she considers who to place on the team, Brenda should: Your Answer: Review the aim statement to make sure the team includes representatives of all processes affected by the team’s aim. Including the right people on the change team is crucial to a project’s success. The team should include representatives of all processes affected by the team’s aim, which is why Brenda should review the aim statement. Further, it should include people with enough authority in the system to remove barriers and implement changes; people with clinical or technical expertise; and people who can drive the project on a day-to-day basis. A team representing just one profession is rarely as effective as an interprofessional team.

During Brenda’s first group meeting, the members ask to review the aim statement to make sure they agree it addresses the current problem. With Brenda’s approval, they all decide to rewrite it. However, when they meet to consider what would be a better aim statement, the group loses direction. In order to help them, Brenda might want to: Your Answer: Remind the team of the Institute of Medicine’s dimensions of health care quality. Writing an effective aim, especially when it comes to being specific about the improvement desired, can be surprisingly difficult. The Institute of Medicine’s six dimensions of health care quality can often provide guidance and direction when a team is struggling to formulate an effective aim statement. (Reminder: A handy way to remember the six dimensions is the mnemonic “STEEEP”: safety, timeliness, effectiveness, equity, efficiency, and patient-centeredness.)

Lesson 3: Why do we need Measures? It's common sense that without some type of feedback, we have no way of knowing whether the changes we are making are leading to improvement. In health care systems, measures usually provide the most useful form of feedback. Specifically, data will help you:  Understand current performance. (People doing improvement usually call...