# Lesson 2: Applying the Model for Improvement

### Lesson 2: Applying the Model for Improvement

#### Summaries

• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Introduction to Measurement
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Finding Changes that Lead to Improvement
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Chartering Your Project
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Don Berwick's Dinner Project and Big Changes in Scotland
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Setting Up a Successful Project
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Faculty Footnotes
• Lesson 2 - Applying the Model for Improvement > Lesson 2 Lectures > Your Turn

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Introduction to Measurement

• Now let’s talk about a second part of the model for improvement, establishing measures.
• So how can we tell if a change is or isn’t an improvement? Well the answer’s obvious, track the effect of the change.
• Let me first try to debug a common misconception about measuring for improvement.
• People say, you cannot improve what you do not measure.
• You and I all the time get better at things without numerical assessment.
• I’m pretty sure that we got better and better at parenting over the years.
• On closer inspection, we actually did, after all, measure our parenting as we went along.
• Something would go wrong, and we’d sit down in the kitchen, and we’d ask each other, what just happened? Why is Dan so mad? Is Jessica feeling better? And so this is my point.
• Use every available way to notice and track your progress.
• It’s best of all when those metrics are tracked and displayed on a run chart, a chart over time.
• I want to make just one key point about measurement over time.
• Remember my assertion that formal experiments with randomization or formal control groups, blinding and so on, that they aren’t all that helpful for learning in the complex, interdependent context sensitive process of real world improvement? So what’s the alternative for inference? How can you learn without, say, a control group? Well comparison groups can help.
• In the model for improvement, keeping track of things over time, things in sequence, that allows you to assess the effects of changes.
• Unless she’s committed to a randomized trial, which would be rather silly and wasteful, she can improve but only if she notices both process and outcome, and if she and her coach retain awareness of the sequence, which swing happened when and in what relationship to the grip.
• Keep track of the outcome sought and keep track of the possibly relevant processes, the causes you suspect and that you’re working on, the changes that you make, and then remember the sequence.
• It’s called a balancing measure, or maybe we should just say a side effect measure.
• These measures refer to things that might be inadvertently, undesirably affected by the change.
• Maybe our golfer needs to keep track of aches and pains as she alters her swing.
• Finding a good set of measures, outcomes, processes, balancing measures, that actually could be a project itself.
• It’s really not uncommon for an improvement team to spend some time developing ways that it’s going to assess its progress to find the measures that feel right, that feel practical, meaningful, inexpensive, user friendly.
• I remind you, seeking measures that have heart, that you care about.
• Now how would you know if a change is an improvement or not? Jot down some possible measurements if you like, but know, we’re going to spend some more time focused on measurement in lesson three.
• Back to the projects I mentioned to you earlier, for example, my student who was working on his backpack weight, he used an outcome measure, pounds, and some process measures, the number of books in the backpack.
• He was measuring whether he ever was missing an item that he actually needed.
• I spoke with my wife and my kids to help figure out what I should track over time.
• They told me really clearly that they really missed me at dinner time and that weekend days were much more important to them than weekdays.
• I counted and I tracked the number of dinners missed and the number of weekend days that I was away.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Finding Changes that Lead to Improvement

• Now, let’s examine a third part of the Model for Improvement, finding good changes to test.
• What change? So the point of the model for improvement is to test changes that might be improvements, find out if they really are, and then to act on what we learn.
• So part of the process is to find changes that are worth testing.
• That’s one way, find someone who’s better than you are, and ask them for suggestions.
• She could read books, she could read papers to find ideas, she could ask her friends.
• She could go find a video of a great golfer and watch it over and over again.
• She could derive an idea from her understanding of that theory.
• She could think of sports and activities that aren’t golf, but that have some analogues to golf, and see if those give her some ideas.
• Remember Don Goldmann and his story about cucumbers in the last lesson? He identified and he tested a number of different changes in the hope of improving his vegetable garden.
• Ask the Centers for Disease Control, CDC, check their website for ideas.
• Find out, if you can, what hospital in your state, or in the nation, or in the whole world, has the lowest infection rates, and send a team there to watch and ask questions.
• Understand the science of infections, the theory and how they are controlled, and see what ideas that prompts.
• Two other sources of ideas for change require some new thinking and maybe even a new culture.
• It is remarkable how often someone right on the front line, right at the coal face, as they say, has a really good idea for a change, it’s just that nobody ever asked.
• They encourage, they harvest suggestions for change from their front line workers.
• In health care, the front line, remember, isn’t just clinicians and employees.
• That means involving patients, families, communities right into the center of management and improvement processes.
• A second, look into industries and enterprises that are far, far away from health care as sources of ideas for health care changes.
• A lot of IHI’s work on smoothing patient flow, for example, came from very detailed study of food service systems in hospitals and restaurants.
• Patient safety took a leap forward in health care when we finally undertook studies of aviation safety systems and brought aviation experts into hospitals, literally, to observe and to comment.
• So where is all this headed? It’s headed to identify a portfolio of changes that are worth testing.
• Just hearing about an idea is enough to make you excited about testing it.
• These have scientific or logical pedigrees, and then notice which of those general concepts spark specific ideas for use in your own system.
• So keep an eye out for specific changes that represent that concept.
• That’s a change concept I first encountered in the book called The Improvement Guide by Langley, Nolan, and Nolan.
• It has alerted me to watch for ideas that help or cause people to understand for the first time sometimes, that they’re interdependent.
• That book The Improvement Guide, by the way, that’s an unparalleled source of ideas for change concepts.
• It lists over 100 generalized change concepts in its appendix to prompt your thinking.
• There’s even an app now that you can download that has over 70 of those change concepts ready and waiting for you.
• The model for improvement counsels a search for ideas that are worth testing.
• Be prepared in an organization or a group, for lots of push back on those ideas.
• So you’ve got a personal improvement aim now to work on and some idea of how you would find out if a change is an improvement or not.
• Let’s say, three changes, that you’d be willing to test.
• So I decided to try a one month test of simply declining every single invitation and then seeing what the effects were.
• What is the largest informative test of change that you can run by next Tuesday? What’s the largest informative test of change that you can run next Tuesday? So here are the rules.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Chartering Your Project

• In projects like that, you may be a little dependent on other people for resources, or ideas, or supplies, and so on, but to a great extent, improving in those areas is a solo sport.
• Aims- whose aims? Measurements- what measurements? What changes are promising? Where and how are we going to test them? What did we just learn from the test? It can all get pretty messy pretty quickly.
• So just choosing an aim can be a serious exercise in discipline, group process, and in conflict resolution.
• A well prepared improvement charter is not a panacea, but it’s one asset that can help navigate through some of the complexity of getting started on improvement.
• It’ll include aims, measures, change concepts that might be relevant, and some statement of rationale saying why is this aim important.
• By the way, a charter can also help communicate with people who aren’t going to be directly involved in the project, so they understand what’s going on.
• Some charters will designate as team members people from three groups- people from the system that’s going to be worked on, the processors, from key suppliers to that system, people they depend on, and from key customers to that system, the people who depend on them.
• A team to improve an operating room flow might include, for example, a surgeon, an operating room nurse, and a scheduler, process participants, maybe people from the hospital admitting office and the equipment warehouse, they’ll be suppliers.
• Importantly, patients, family members, and maybe nurses from the surgical ward, customers of the operating room.
• Most crucially the sponsor is going to be there to remove obstacles to testing, obstacles to change that the team can’t remove, isn’t equipped or authorized to remove.
• It pulls people together and invites them together under a common framework of aims and conditions that’ll help them become the team that they’re going to need to be if the model for improvement is going to do its work in the service of learning.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Don Berwick’s Dinner Project and Big Changes in Scotland

• DON BERWICK: Now, the Model For Improvement offers a very useful roadmap to establish and guide your improvement efforts.
• It does help in an organization if some model is used and stabilized as a support to concerted action and to make conversations easier.
• ” Remember my decision to try to make my family and myself happier by spending more nights at home having dinner instead of on the road? I set an aim, with the help of my family, deciding that the two key goals were to have dinners with my family, and to have weekends that we’re not interfered with by travel.
• I set up a metric system, a run chart – actual graph – on my desk, over time of dinners missed, and weekend days missed, and I began to track those with month by month scoring.
• They had done a careful analysis of risks and opportunities in patient safety, which led to about a dozen work streams, or maybe a few more, each of which was directed at reducing one type of patient injury- infections postoperatively, pressure ulcers in the intensive care unit, ventilator pneumonias, and so on.
• Every hospital in Scotland was involved, and they began working together using IHI’s so-called Breakthrough Series Collaborative Model.
• That’s a complete improvement system organized around the Model For Improvement, with appropriate senior leadership support, and appropriate infrastructure so that the thing could act like a system.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Setting Up a Successful Project

• DON BERWICK: There isn’t any one right way to set up an improvement project, but here’s a few general lessons from past successes and failures.
• Second, a clear charter helps so the team can all understand what they’re about, and it can explain it easily and in a regularized way to the rest of the organization.
• They need resources for meetings to do the proper measurements that they’ve set up, to support their tests of change, and in busy organizations back filling the time of team members so they can get the improvement work done without harming the work of their usual production units.
• If they’ve done their work well, the harvest is going to be lessons that could be incorporated into the routine work of the organization and spread everywhere.
• Those are two tasks- stabilization, which is also sometimes called holding the gains, and spread, making what the team learned become the new normal everywhere.
• They’re beyond the means, and they’re beyond the span of control of most project teams.
• Their duties, stabilization and spread, that revert to the senior executive level of an organization.
• Teams and the model for improvement are incubators for learning.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Faculty Footnotes

• DON GOLDMANN: As you know, we like to have a short dialogue after every lesson, and today, I’m delighted that we’ll have a chance to talk with Don Berwick, who just delivered a series of lectures in that lesson.
• Don, you mentioned the importance of setting a measurable aim.
• I think you said what by when, or sometimes you say how much by when.
• By this, I think you mean be aspirational, get out there and really catalyze people to make big change, stretch, and I get that.
• So I don’t think I’ve seen a downside to really aspirational aims and say, why don’t we be the first place that gets to zero.
• You know, the progress that a team, or a group, or a department has made toward that, and never walk away from that.
• It to do with, really, the core idea and quality of meeting social need.
• There’s a need out there, and it is for things we’ve never done before.
• If we keep doing what we’ve done in health care and just keep trying to be more or less within the envelope of current performance, we’re actually not meeting the social need.
• A bold aim comes from that kind of thinking, which is, what does the world need from us, and why don’t we try to get there.
• So I guess I’m after the excitement, and I’m a little less worried about the chance we won’t get all the way there.
• DON GOLDMANN: Does this require a culture where people are perfectly happy to say, well, we didn’t quite make it this year, but we really did this well, and maybe we need to do some other things differently? DON BERWICK: Absolutely.
• I mean, this is- you know, the discouragement, fear all oppose real progress and improvement.
• You know, I’m a pediatrician, so I think about kids.
• DON GOLDMANN: I think I made a mistake in my upbringing of my kids.
• DON BERWICK: I’m sure you did fine, but it is in our nature as human beings to try stuff we love, and then try to do it even more, or even better.
• That’s, in a way, I think improvement’s attempting to build on that human nature instead of discouraging, capping, and downsizing our goals.
• DON GOLDMANN: So the next question’s a little bit technical.
• In other words, what would’ve happened if you had not done your intervention and your quality improvement work.
• There must be ways, at least sometimes, to get comparison groups where you can demonstrate that things really got better, quicker, because we did this, and because we did it, we have an advantage.
• What do you think about that? Are there ways to do this in reality? DON BERWICK: Sure.
• You know, if you’re trying something on six east and not on six west, why not look and see if there’s a difference? I just have learned to become very skeptical about very formalist experiments as kind of the foundational science of improvement.
• The other thing they do is, I think they throw away wisdom, because I think the workforce, for example, involved in work, they could notice something that wasn’t part of the original design.
• We’re after better and better hunches in the search for improvement.
• That’s the thinking of improvement actually that’s what we use in most of our lives.
• I think it will work in our complex systems very well.

#### Lesson 2 – Applying the Model for Improvement > Lesson 2 Lectures > Your Turn

• DON GOLDMANN: Now that you’ve crafted your opportunity statement, it’s time to create a formal charter which should mirror the Model for Improvement.
• For my personal improvement project, I’ve picked a process measure that I’m quite confident is linked to more calories out, because the elliptical machine I use at the gym has a nice display that calculates how many calories I’m expending.
• At the very least, I need to get on that elliptical two times a week.
• Well, how will I know change is an improvement? In other words, what’s my measurement framework? This will be easy to monitor.
• I should gradually be increasing how often I get to the gym, my level on the elliptical, and the time of my exercise routine.
• I found that when I go to the gym, the time I used to take to stretch diminishes.
• So I will need to develop a measurement framework for stretching, frequency of stretching exercises, and time to stretch.