The Model for Improvement

Transcription

1 The Model for Improvement Women s Cancer Screening Collaborative Prepared by Improvement Foundation (Australia) Ltd PO Box 3645 Rundle Mall SA 5000 Date/Time Last Saved 7/03/ :49:00 AM 1 P a g e

2 Contents Background... 3 Overview of the Model... 3 How to use the Model for Improvement... 5 Testing your Ideas for Change using Plan, Do, Study, Act Cycles Implementing a Change P a g e

3 Background Making improvements to products, systems or services requires change. Although change can seem threatening or overwhelming, it can be successfully managed if it is well planned. In primary health care, you will be working on a live system and you need to be sure that a change is an improvement before implementing system wide changes. The Model for Improvement 1 provides a framework for developing, testing and implementing changes in any setting or system, and on any scale. It involves choosing specific and measureable goals, selecting objective measures of improvement that can be tracked over time, and identifying key changes that will result in an improvement. The change effort is broken down into manageable steps, which are tested to determine whether improvement was achieved. The Model for Improvement is an efficient method that avoids wasting effort, and has been demonstrated to be successful in healthcare systems. Overview of the Model The Model for Improvement (MfI) consists of two equally important parts (Figure 1.1): 1. The thinking part consists of three fundamental questions that are essential for guiding your improvement work: What are we trying to accomplish? (to identify the goal) How will we know that a change is an improvement? (to identify measures) What changes can we make that will result in an improvement? (to identify change ideas) 2. The doing / testing part is made up of Plan, Do, Study, Act (PDSA) cycles that will help you test and implement change. Not every change is an improvement. By making small, incremental changes, you have the opportunity to test the change on a small scale and learn about the risks and benefits before implementing the change more widely. A number of PDSA cycles may be required to achieve your improvement goal. 1 Langley GL, Nolan KM, Nolan TW, Norman CL, Provost LP. The improvement guide: a practical approach to enhancing organisational performance. San Francisco: Jossey-Bass; P a g e

4 Figure 1.1 The Model for Improvement (MfI) For an overview of the Model for Improvement, take a look at the video on YouTube: 4 P a g e

5 How to use the Model for Improvement Quality Improvement Tools Answering each question thoughtfully and carefully allows you to plan your improvement effort, step by step. Quality Improvement (QI) tools can help you work through this process. The table below contains a list of QI tools. There are many QI tools that you can use to help generate ideas or solve problems and these are discussed in more detail in Quality Improvement Tools. Try using different tools with your team as they can be fun to use and help break through traditional thinking to identify alternative ideas. Tool Brainstorming Nominal group technique Affinity Diagrams Five Whys Fishbone (Ishikawa) Pareto Process Mapping Force Field Analysis Usage Generating ideas Generating ideas Organising and categorising ideas Defining a problem (determining root cause) Exploring and determining factors that cause or influence a problem (cause and effect) Identifying the causes that have the greatest impact on a problem, which assists with determining where to focus an improvement effort Visualising and understanding what happens in a process (for service/process redesign) Identifying factors that can support or hinder a potential solution. Understanding these factors and using strategies to reduce barriers can increase the chances of the solution being successfully implemented 5 P a g e

6 The Three Fundamental Questions Question 1: What are we trying to accomplish? We often launch into change without stopping to think about what we are trying to achieve. The first question provides an opportunity to consider exactly what it is you are seeking to change. Once you and your team have agreed on the goal, it will guide you, keep you focused and motivate the team. To answer the first of the three fundamental questions, you will need to write a clear and concise goal for improvement. Key messages Define the problem. Success is all in the preparation! Understanding what the problem actually is and thinking about why there is a problem helps in developing your goals. Set bold but realistic goals that are specific and have a defined timeframe. Use plain language and avoid jargon, so that the meaning is clear to everyone. Include information that will help keep the team focused (for example, the location of the test and a focus on certain patient populations). Example Q1. What are we trying to accomplish? (Goal) Your team has decided to work on Breast Cancer Screening. You might state your goal like this: Our goal is to: Increase the number of Aboriginal women who ve had a breast screen This is a good start, but how will you measure whether you have achieved this goal? Your team will be more likely to embrace change if the goal is more specific and has a time limit. So, for this example a better goal statement would be this: Q1. What are we trying to accomplish? (Goal) Our goal is to: Increase the number of Aboriginal women, aged between 50 and 74, who ve had a breast screen mammogram by 10% by May 2017 Tip Use the SMART principle as a guide: an effective goal is: Specific, Measurable, Achievable, Relevant and Time bound. (Your goal will be less effective if it is vague, hard to measure, unrealistic, irrelevant or open-ended.) 6 P a g e

7 Question 2: How will we know that a change is an improvement? Without measuring it is impossible to know whether the change you are testing is an improvement. Measurements should be taken before you make the change (baseline data), and at regular intervals after making the change. Key messages Try to find measures that show progress towards the goal; however you may have to accept that your measures are not perfect. Spending too much time trying to create the perfect measurement system is a common pitfall. Don t collect more measures than you need. Make measure collection as simple as possible. If you choose a measure that must be taken manually, create a simple data collection form and make it someone s job to fill it in. Everyone in your team needs to know what you are measuring, how, when, and who is responsible for collecting the data. Use diagrams and charts to show your measures to the team. Example Q2. How will we know that a change is an improvement? (Measures) Continuing with our example of the Breast Cancer Screening project, your answer to the second fundamental question might be this: We will measure: The number of Aboriginal women in our practice The number of Aboriginal women aged 50 to 74 in our practice The number of Aboriginal women who have had a mammogram performed Tip Consider choosing a combination of process measures (measures that tell you whether the actions you have chosen are actually happening) and outcome measures (measures that show you the effects of your actions). 7 P a g e

8 Question 3: What changes can we make that will result in improvement? By the time you answer this question, you should know your goal and also how you will measure progress towards achieving your goal. Here s when you become creative. Encourage the whole team to contribute ideas. After answering this question, you will have a range of useful ideas, which can be tested in PDSA cycles. Key messages You know your own healthcare service best, so keep your goal in mind and use your knowledge and experience to guide you to the ideas that suit your unique situation. Adapt ideas from others. Resources and discussions on the Women s Cancer Screening Collaborative (WCSC) site and in qiconnect can help you. Think small and test. Think about testing a change with one GP, one health worker, or a select group of patients. The knowledge gained from this small test will help you determine if the change had the desired effect and is suitable for wider implementation, or whether other changes should be tested. This process will improve your chances of making a successful change. Example Q3. What changes can we make that will lead to an improvement? (Ideas) Using the Breast Cancer Screening example, your answer to the third Fundamental Question might be this: Ideas for change: Ensure all women who identify as Aboriginal have their status recorded in the correct field in the software Liaise with private radiography providers to request a list of your Indigenous female patients who ve had a mammogram Check free text fields and results to identify Aboriginal women that have had a mammogram recently and are not captured correctly in the system Ask Aboriginal women whether they have had a mammogram, when they had the screen and where Promote mammography services using culturally appropriate resources Discuss mammography services with Aboriginal women when they attend the practice/service 8 P a g e

9 Tips Consider generating ideas by: running brainstorming sessions with the team chatting informally when you have the chance (e.g. during lunch) asking for ideas during a team meeting setting up a suggestion box for staff (or even one in the waiting room for patients) ing everyone in the team considering the steps normally taken to complete a task and removing one to see what happens. For example, if you couldn t use the telephone to contact patients, how would you communicate with them? If the nurse left, how would you run your clinics? 9 P a g e

10 Testing your Ideas for Change using Plan, Do, Study, Act Cycles The three fundamental questions will have prepared you for the next stage of quality improvement. It is now time to test the ideas for change that you developed for the third fundamental question. This involves testing each idea using PDSA cycles. Through the PDSA cycles you will find that some changes lead to improvements. If so, these changes can be implemented across the healthcare service and may generate further effective ideas. You may also find that some improvement ideas aren t successful. Analyse why they didn t work and learn from this. By carrying out small tests in PDSA cycles, you have avoided implementing this unsuccessful change on a wide scale. PDSA Cycles in More Detail Plan A well-developed plan includes the what, who, when and where, predictions on the possible outcome, and the data to be collected. For best results, make your plan as clear and detailed as possible. Ask yourselves: What exactly will we do? (Remember to only test one small idea) Who will carry out the plan? When will it take place? This should be in a short timeframe, generally no more than a week. Where will it take place? What do you predict will happen? What data/information will we collect to know whether there is an improvement? These data should be specific to the change that you are testing. Note: the data you plan to collect to test your change idea may be different from the measures you chose in your answer to the second fundamental question (to measure progress towards your overall goal). Do Write down what happens when the plan is implemented. Also document any other observations. Sometimes there are unintended consequences, positive or negative, and these should be captured. Study Reflect upon what happened. Think about and summarise what you have learnt, analysing the data collected and comparing the data with your predictions. If there was a difference between your predictions and what happened, consider why. At this point you should be confident about the outcome and whether the idea will contribute towards achieving your goal. Act In light of the results from your test, will you implement the tested change, or amend and test again or try something else? Write down the next idea you will test. What will you do differently? Be sure to start planning the next cycle as soon as you can, to keep up the momentum. 10 P a g e

11 Continuing the example and using the ideas selected at the third fundamental question, two PDSA cycles might look like the examples below. Cycle 1 Cycle 1: Ensure all women who identify as Aboriginal have their status recorded in the correct field in the software Plan Do Study Act What: Run a search in our software to identify women currently coded as Aboriginal and/or Torres Strait Islander Who: Practice manager When: Friday 17 th March Where: At the practice Prediction: We will have a small number of women who are coded as Aboriginal and/or Torres Strait Islander; perhaps only about 6 Data to be collected: Number of women coded as Aboriginal and/or Torres Strait Islander This was a quicker process than we expected; only taking 20 minutes. 8 women have been recorded as Aboriginal and/or Torres Strait Islander. This is a little more than expected, however it reinforces the need to do more work in this area. We will check free text fields and results to identify Aboriginal women that have had a mammogram recently and are not captured correctly in the system. Cycle 2 Cycle 2: Liaise with private radiography providers to request a list of Aboriginal women who ve had a mammogram Plan Do Study Act What: Call xx radiology to ask if they can inform us whether any patients have had a mammogram at their service in the past two years. Produce a list of the for the practice manager to review Who: Receptionist When: Wednesday 23rd March Where: At the practice Prediction: There will be 1 Aboriginal woman that has had a mammogram Data Collected: Number of Aboriginal women that have had a mammogram in the last two years Done Does every idea require a PDSA cycle? No Aboriginal women have had a mammogram at xx Radiology in the past 2 years Results of this work will be discussed at next team meeting. We will ask Aboriginal women whether they have had a mammogram, when they had the screen and where. 11 P a g e

12 You do not have to test every change idea with a PDSA cycle. You might decide to make some simple changes (e.g. generating a list of patients, flagging patient records, or scheduling a meeting) before you undertake a testing cycle, because you know those changes will definitely be needed and you are not interested in studying their effect. On the other hand, if you are interested in whether extra team meetings will make a difference, you might choose to undertake a PDSA cycle to test whether they are contributing to improvement. In general, you would probably choose to undertake a PDSA cycle whenever you aren t sure if the change will be an improvement, or whenever you want to break down a set of changes into manageable steps. Common traps Experience has shown that people tend to fall into some common traps when using PDSA cycles: not reflecting on what they have done in order to learn from it (i.e. Plan, Do, Plan, Do) implementing changes without planning (i.e. Do, Act, Do, Act) not involving team members not setting aside dedicated time to undertake the work not documenting PDSA cycles not completing PDSA cycles thinking too big: beware of cycles longer than one week or involving too many steps. Here is an example of trying to test too many ideas in one PDSA cycle: Example Idea Offering health checks to Aboriginal and Torres Strait Islander women Plan: Search software for women who have been billed for an Aboriginal health assessment/health check and ensure that their ethnicity is recorded appropriately. The practice manager is to search the billing software for women that have been billed for item number 715 and produce a list of the individual patients identified. The practice manager is to send a letter to all patients, asking if they are of Aboriginal and/or Torres Strait Islander origin and explaining the reason for asking this. The practice manager will also flag these patients in the medical software and then code patients who have responded to the letter. In this hypothetical example the healthcare service has several ideas for improvement. However, it is easy to be overwhelmed if a lot of changes are being attempted at once. This health service is likely to have a greater chance of success if they break the PDSA cycle down into smaller steps. For example, the team could break it down into four separate PDSA cycles, e.g.: Search the billing software to identify patients who have been billed item number 715. Produce a list of the women and ensure that their ethnicity is accurately recorded. 12 P a g e

13 The practice manager is to develop a template letter providing the opportunity for patients to self-identify as an Aboriginal or Torres Strait Islander person, and the reason for this request. Send a letter to the women that do not have their ethnicity recorded to ask them to visit the health service. Place a flag in the medical record so that the GP can ask the national standard question during a consultation. This could be a large activity and generate additional work that is not required or could be avoided. Each of the above concepts should be tested using a PDSA and from the test, you ll be able to identify if it worked and the impact on the health service. Tips No PDSA cycle is too small. Making them too big is a more common problem. Keep it simple and just do it (e.g. Think what can we do by next Tuesday? ). A PDSA cycle that is too large can usually be broken down into two or three individual PDSA cycles. You should expect to complete a series of PDSA cycles to achieve your goal don t expect complete success first time. It is by building on your previous cycles that you achieve results. Make sure you involve the right people in your work as improvement is nearly always a team effort. Set aside protected time to do the work and don t try to squeeze it in to a busy schedule. Documenting your PDSA cycles is a great way of motivating the team, by sharing what you ve learned with other people. Don t be discouraged if an idea doesn t work. You will learn as much from something that did not go well as you can from something that did. Submitting Your PDSA Cycles You will be required to submit at two PDSA cycles to qiconnect, IF s web portal, every month. Please use the following link to upload MfIs to the WCSC site: You will be provided with a step-by-step guide that will detail how to upload Models for Improvement and PDSA cycles. The WCSC program manager and the NCPHN support team will also provide direct support and feedback on your Models for Improvement. 13 P a g e

14 Implementing a Change Once you have tested a change and determined that it is effective, and your team supports this new way of working, you will need to implement the change. Implementing a change means making it a permanent process within the organisation. For simple changes, implementation will be relatively straightforward. For other changes, effective implementation will require training and on-going monitoring to ensure that the team does not return to the old way of doing things. When considering implementation, ask yourself these questions: What other changes are needed to support implementation of this change long term? Your new way of working may require alteration in support structures such as job descriptions and standardised procedures. For example, you could add the maintenance of patient registers as a part of the practice nurse s role to ensure this work is carried out consistently and is supported by the organisation. Does the wider team need to be involved in the implementation of the change? Testing a change may have involved a small group of people. However, implementing the change may affect others that were not involved in the testing process. You will need to consider how to engage these individuals, identify any resistance and promote the benefits of the new way of working. You may find that the data you collect through your testing provides valuable information to convince the wider team about the benefits of the proposed change. For instance, you could print monthly graphs showing the improvement achieved after implementing the ideas tested in your PSDA cycles, and display them in the staff room or present them at staff meetings. Will a regular review process be needed to make this change permanent? You may consider that a bi-annual audit of the new system or process is needed. That could be a component of the standardised procedures that are developed to support this new system. 14 P a g e