This transcript is intended to provide webinar content in an alternate format to aid accessibility. We apologize for any inaudible or unclear content as a result of audio quality.
Model for Improvement; Plan, Do, Study Act
Presented by [Kristi Wergin] (50-minute Webinar) [4-01- 2013]
Kristi Wergin: Hello everyone, this is Kristi Wergin with Stratis Health, the Quality Improvement Organization in Minnesota. I would like to welcome you all to this webinar for the Acting to Achieve Excellence Campaign.
For those unable to participate in the introductory webinar I would like to introduce myself. Again, I’m Kristi Wergin, a Program Manager with Stratis Health. I’m leading the Stratis Health Acting Together to Improve Excellence Campaign team. I provide quality improvement leadership for Minnesota nursing homes as part of the Centers for Medicare and Medicaid Services Quality Improvement Organization contract. My experience as a registered nurse in elder care includes: direct resident care as a nursing assistant and staff nurse, as well as work in quality improvement, staff development, leadership management and customer employee satisfaction.
For those of you that participated in the National Nursing Home Quality Care Collaborative Virtual Learning Session, you heard about a quality improvement practice called ‘PDSA’- Plan Do Study Act. Although some of you may be familiar with PDSA this practice may be new to others. Following this learning opportunity, our goal this afternoon are:
- That all participants will be able to describe the fundamentals of the model for improvement.
- Identify key concepts about testing changes on a small scale using PDSA cycles.
- Execute a small test of change within your nursing home.
The Institute for Healthcare Improvement (IHI) uses the model for improvement as the framework to guide improvement work. IHI is an independent, non-profit organization that is a leading innovator in health and healthcare improvement worldwide. The model for improvement was developed by associates in process improvement as a simple yet powerful tool for accelerating improvement.
The model for improvement consists of two parts:
- Thinking
- Doing
The thinking part consists of the three fundamental questions for achieving improvement. These three questions area:
- What are we trying to accomplish?
- How will we know that change is an improvement?
- What change can we make that will result in an improvement?
People tend to want to jump straight to the solution rather than working on the root of the problem. If you answer these three fundamental questions, it will help to ensure that you are dealing with the issue that needs to be addressed. These questions guide the work and lead to the doing or action part of the model for improvement cycle known as PDSA.
Start by defining what you are trying to accomplish. This is your aim statement or goal. The aim should be time specific and measurable. It should also define a specific population of residents or other systems that will be affected. You will need to write a clear and concise goal for your improvement and identify the objectives in simple language that’s easy to understand.
A good way to determine if you have set a clear and concise goal is to use the SMART acronym.
Specific
Measurable
Achievable
Relevant
Time bound
For example, using this guideline if you’ve decided to work on decreasing falls your statement might be to decrease the number of falls by 20% in the Lakeside neighborhood by August 1, 2013. Some examples from homes participating in the Attacks Campaign, working on decreasing antipsychotic drug use are:
reducing antipsychotic drug rate 5% by the end of the year 2013,
to reduce the use of antipsychotic drug use by 50% by March 31, 2014, and
reduce the use of psycho active medications in the absence of psychotic or related conditions by 10% from the baseline of 6.7% to equal or be less than 6.03%.
Those are nice specific goals that some of you chose.
Next, think about and decide how you will note if change is an improvement. While some improvements may feel intuitively to be a good thing, consider how you will prove that the change is actually an improvement. Measurement is fundamental to answering this question. Without measurement how do you know if the change has led to an improvement? All changes do not lead to improvement; however, all improvement does require change.
For example, you may decide to work on decreasing falls by implementing the use of personal alarms for residents with a history of falls. To see if this is effective it is important to track the fall rates on the residents with alarms to determine if these alarms have reduced the number of falls. You may find that increasing the number of personal alarms did not reduce the number of falls. In addition, if you’re tracking quality of life measures this may reveal that the alarm use was decreasing the resident’s quality of life. This is an example of a change that did not lead to an improvement.
To make improvements it’s necessary to make changes in your systems or processes. Think about what change(s) you can make that will result in improvement. This is where you decide what idea or ideas you will test in order to obtain your goal. Start by studying how the current process works. This can be done by listing the steps involved in a process using direct observation, brainstorming or consultation with the people responsible for each step. Compare the actual process to what is desired and identify areas in which errors or problems could or do occur.
When studying the process, look for current steps or tasks that may be redundant or ineffective. Look for logical placement of the tests that are performed. Forgotten tests that could impact the outcome. Continuity of care across disciplines, for example, with a registered nurse and a physical therapist or within groups of the same discipline, for example, with same groups of nurses and steps in the process that could be added to improve quality and resident outcome.
Always look for ways to limit variation in the process. Streamline and simplify the process. Remember the simpler it is the less likely you’ll be to have an error. When deciding about what change can be made that will result in an improvement, be sure to review best practices. For example, what has worked in other facilities? You can refer to the change package that CMS has developed by studying best practice nursing homes. You can look to see what the literature says and don’t be afraid to copy. You can copy things that have worked for other homes, but don’t wait for perfection just do it. Start somewhere, learn as you go and as long as something is safe if you wait for all the answers or the perfect method you’ll never get started.
The second part of the model is PDSA which represents the doing part. PDSA is a rapid quality improvement activity. It was developed in the early 1990s as a way to approach and break down change into manageable components and testing each small part to make sure services are improvement and no effort is wasted. PDSA is the part of the IHI model for improvement which walks the team through the four steps of planning the intervention, doing or putting the intervention into place generally on a small scale, studying the results and acting or deciding on what actions to take next.
Consecutive PDSA cycles are trials of change on a small scale to collect information about the effectiveness of incremental changes. By working small in terms of scope the change can be managed. For example, if what is tried does not work as well as planned you can always go back to the way things were done and try something different.
The first step is plan– when developing a plan your team should ask the following questions and record responses.
- What is the objective of the test? Here you should state what you want to accomplish by doing this test. Once you predict what will happen, state what you think will happen as a result of your planned action.
- What exactly will you do? Clearly define the tasks activities and so on that will be undertaken to achieve your idea.
- Who will it involve?
- Will you start small with just one neighborhood department or one group of residents?
- How long will it take to implement? It’s important to define a time period and make it as specific as possible, and it should occur over a short period of time.
Consider what resources you will need. For example, will you need extra time, staff or equipment? What data or information will you collect to know whether there is an improvement? Think hard about what data you will need to collect to check the outcome of the change to know where there is an improvement.
The second step is do– this is actually where you’re testing a change. This is the fun part.
It’s important to do the activity and not get caught in the planning phase for too long. Again, implement the change on a small scale. When first using the PDSA cycle start with a very small project so you can complete it in a relatively short period of time. The first time you will likely discover changes you want to make on the next cycle.
Next, carry out the test and work with whatever plan you made. Document the problems and unexpected observations and not any problems encountered along the way. This will assist you in analyzing the cycle and avoiding problems in the future. Begin analysis of the data. You might want to complete chart audits, collect whatever data you need to help complete the quality picture, begin analyzing the findings, continue to document expected and unexpected observations along the way and as you do this you’ll start to see if your test of change is providing the desired results.
The third step is study– this is a time for reviewing your test of change.
Use this phase to quantify and measure what happened. Study the data collect and think about its impacts. Think about what could have been done differently. Complete the analysis of the data. Think about where you were and compare that to where we are now. Has it made a difference? Compare the data to your predictions. Were the goals and aims met? Are you satisfied with the results? What were the results of the best practice that you chose?
Summarize and reflect on what was learned. What did you learn? What could be done differently? Detail problems, success, surprises and list them specifically so you can substantiate your conclusions and also have them as a resource for future quality improvement cycles.
The fourth step is act– this is when you need to think about what will be taken forward and what opportunities may have arisen. Prepare a plan for changes needed to be made before the next cycle. Document what you will move to and when. There are three responses to guide you in this act phase.
Adapt what works partially. Make revisions to make it work well.
Adopt what worked well and consider expanding the changes to residents, staff and other neighborhoods, or
Abandon what didn’t work.
Once you’ve changed your approach you can repeat the PDSA cycle with the new ideas to see how that works out.
I know this is a lot of information at once and believe it may be helpful to look at a few examples and work the PDSA cycle through some examples and hopefully some of these will sound familiar with the things you’re currently working on.
Let’s start first with IvyCareCenter, who would like to try a small test of change to help them reduce their use of off label antipsychotic drug use, similar to what all of you are doing now. The team needs to do some brainstorming and this is the thinking part. They start by discussing what they want to accomplish by putting together an aim statement or goal. The data may be in the form of a story.
For example you may recall Hattie in our first webinar. Hattie was admitted to the nursing home from the hospital after being treated for pneumonia and dehydration. While at the hospital Hattie was placed and scheduled olanzapineand prn lorazepam. One week after admission, the staff notes Hattie is sleeping much of the day, hasn’t been participating in activities and hasn’t been eating well. Her daughter told the staff that this isn’t normal behavior for her, when she was at home she enjoyed reading and knitting yet shows no interest in these activities.
The nursing home staff suspects that fatigue and disinterest may be due to the olanzapine. She hasn’t needed the lorazepam since admission but she’s getting olanzapine three times a day. If they talk about the need to ask the physician if they can start reducing olanzapine, they suspect they may be able to reduce or discontinue antipsychotic medications on other residents as well.
So they decide on this smart goal, to reduce the number of residents on the neighborhood with an order for an off-label antipsychotic medication by 15% by May 1, 2013. This is a smart goal because it’s specific and clearly states which residents they will be following. It’s measurable and they will count the number of residents with an order for an off-label antipsychotic medication before and after the intervention. The goal is achievable. They’re going for a 15% reduction, which for this group of residents would be 2 of 12. The goal is relevant. There is a current national initiative to decrease the use of antipsychotic medications and Ivy Care Centers rates are higher than state average. The goal is time bound and the date is well spelled out.
Even though intuitively it feels right to decrease or discontinue these medications it’s important to make sure it’s in the best interest of the residents. While doing the reductions this home has decided to monitor activity involvement logs, behavior logs and sleep logs every shift for two weeks before the drug reduction and two weeks after the drug reduction. They will use these logs to determine if the resident is more alert and engaged since the drug reduction and to determine if there has been any change in their behaviors.
For example in Hattie’s case, we would hope that with the reduction or discontinuation of the olanzapine, she would be sleeping less, eating better and being more involved in activities. If so than this change will be an improvement for her. If she continues to be lethargic, doesn’t eat well and they see an increase in difficulty to manage behaviors the change will not be an improvement. However, this doesn’t necessarily mean that the answers to increase the dose of the olanzapine. At this point it’s time to do some detective work.
What are her behaviors telling us?
Is there a way we can meet her needs without the use of an antipsychotic medication?
When studying the systems and processes that are related to antipsychotic drug use, the team notes that they have no system in place to review and trial antipsychotic drug reductions when residents are admitted with these medications, like Hattie. As they brainstorm they realize that they don’t have a process in place to regularly review and attempt the reduction of off-label antipsychotic drug use on any of the residents at IvyCareCenter.
They decide to start by working with the pharmacist and physicians to reduce these medications for all the residents in their memory care neighborhood. They’re now ready to start the second part of the model for improvement trying a small test of change by completing a PDSA.
The first step is plan. They define the objective of this test as a decrease in the number of residents with an order for an off-label antipsychotic medication. They predict that for some residents, decreasing the dose or eliminating the medication will improve the residents’ quality of life because they will be more alert, increase their independence with ADLs and be more engaged in activities.
The team has decided that it would make the most sense to start with the consulting pharmacist since they are already reviewing medications every month. They decide they will ask the pharmacist to make recommendations for reduction of off-label antipsychotic medications that have not been reduced in the last six months to the physician as part of the next monthly medication review.
Since their memory care neighborhood has the highest use of antipsychotic medication they decide to start there. The team is excited about the test and they want to do it quickly, so they decide to carry out the intervention in the month of April. They anticipate they will need to make sure the nursing assistant’s and activity staff has extra time to complete daily behavior, sleep and activity logs.
The data they need to collect are the number of residents with an order for an off-label antipsychotic medication before and after the intervention. They will also need to collect data from their behavior and activity logs to make sure the change resulted in an improvement. They also need to collect data about resident’s sleep to make sure the change resulted in an improvement.