Quality Improvement & Patient Safety Hot Topics

December 2022 MITE Hot Topic: P-Charts for Statistical Process Control

Author: Hilary Perrey, MHA, LSSBB, CPHQ

MHPI, Improvement Specialist II

Learning Objectives:

1. Describe what p-charts are and how we use them in performance improvement.
2. Explain how to create p-charts.
3. Illustrate a helpful p-chart example.

What is statistical process control (SPC) and how can this be helpful in my performance improvement or quality improvement project? SPC is a standard methodology for measuring, monitoring, and controlling quality during a process using control charts (Minitab 2022)¹. In 1924, Walter Shewhart of Bell Laboratories developed control charts, also called Shewhart charts, to describe whether a process is in statistical process control².  In the November 2020 edition of MITE Hot Topic, Dr. Parker described the concepts of common cause and special cause variation. SPC allows us to distinguish whether a system change represents common cause or special cause variation. The p-chart is an attribute control chart and is used to monitor the proportion of defective items where each item can be classified into one of two categories, like pass or fail (Minitab 2022)³. It is one of the most common SPC used in healthcare. Attribute data can be counted and plotted as discrete events.  The subgroup size, or sample size, is variable and is usually >= 50 (The Memory Jogger Healthcare Edition)⁴.

Examples:

• % of patients with post-operative infections
• % of patients who failed to keep medical appointments
• % of patients who fell on a unit

The p-chart below shows FY22 post-operative infections at Hospital A and was created using SigmaXL, a Microsoft Excel add-in. Please note this data was developed for the purpose of this exercise.

Why do we use p-charts here?

The p-chart is an appropriate statistical process control chart to use when analyzing post-operative infections. Post-operative infections vary monthly and this data shows the proportion of infections relative to total number of surgeries.

How do we create and interpret the p-chart?

There are several statistical tools in addition to SigmaXL which can be used to create p-charts and other control charts including R, a free statistical software program, SAS Enterprise Guide, and Minitab.

In order to create a p-chart using SigmaXL, data must be entered in columns in Microsoft Excel. The example data below was graphed to create the p-chart: month, number of post-operative infections (numerator), and the number of surgeries (denominator). By selecting these columns in SigmaXL, this will create a p-chart. The fraction of infections are calculated below as a percentage of the monthly total surgeries (# of post-operative infections/# of surgeries * 100), so you can see the monthly variation.

The dots which are graphed in the p-chart correspond to the monthly proportion of post-operative infections which are also seen as percentages in the column below. Simply, multiply the proportion on the p-chart by 100 to see what the monthly % defective is. In March 2022, the % of post-operative infections peaked at 5.85% (note: this corresponds to 0.0585 on the p-chart). The mean is 0.016 or 1.6%. The upper control limit (UCL) is 0.028 or 2.8% while the lower control limit (LCL) is 0.004 or 0.4%. Control limits are generated by the statistical package and are used to apply rules for special cause variation.

A process is considered to be out of statistical control when one or more points fall outside of the control limits. There are several other rules used to determine if your process is out of statistical control beyond this particular rule. In February and March 2022, post-operative infections are at an all-time high and are out of statistical control –  the white open circle dot is used to indicate that they are out of statistical control. After initiating improvement work in March, post-operative infection rates fell between April – September 2022 with an increase in post-operative infections just above the mean in July 2022. Hospital A’s improvement work was successful in reducing post-operative infections between April – September 2022 and the process stayed in statistical control.

References:

1. Real-Time SPC | Statistical Process Control Software | Minitab
2. Walter A Shewhart, 1924, and the Hawthorne factory – PMC (nih.gov)
3. Overview for P Chart – Minitab
4. The Memory Jogger II Healthcare Edition: A Pocket Guide of Tools for Continuous Improvement and Effective Planning. org
5. p-Charts – SPC Charts Online

Want to earn CE Credits? Go here:

QI/PS Hot Topic – November 2022

Informing healthcare quality through patient focus groups

Angie Marshall, MS, LSSBB – Improvement Specialist II, MMC Performance Improvement

Learning Objectives:

1. Discuss the utility of patient focus groups in quality improvement
2. Describe how to plan and conduct a patient focus group
3. Identify ways to analyze and visualize qualitative data

Focus groups are a form of qualitative research used to gain a deeper understanding of experience, opinions, behaviors and needs of the customer through a moderated, open discussion. In a healthcare setting it is important that patients can provide feedback about their experience in order to continually improve care. There are several ways to collect patient experience data, and as depicted in Figure 1. These methods vary in description and generalizability. To understand patient experience within a large healthcare system, the focus is often on collecting and reporting quantitative (or “countable”) data because it can be easily organized, analyzed and understood. Quantitative information is more generalizable to the broader population, but less descriptive about individual patients.

An effective method for collecting descriptive patient stories is through a focus group – a guided discussion about a particular topic. A group discussion with patients is a great way to delve into detail on specific topics, to understand participant’s opinions and to encourage new thoughts and ideas. Further, it gives patients an opportunity to play a role in positive change within the health system.

There are limitations to focus groups. They require a significant amount of time to organize and execute, the data can be complex to analyze and there is a potential for bias. Despite these challenges, it is still an effective tool for sampling a subset of a population about targeted topics. Below are guidelines and tips for planning and conducting an effective focus group.

How to plan for a patient focus group:

• Define the objective of the focus group. What research question needs to be informed?
• Determine the session details including date(s), location and length (1 hour is recommended). An incentive can be offered to participants to encourage participation and as a thank you for their involvement.
• Prepare a list of 8-12 open-ended questions to guide the discussion that align with the session objectives. The most important questions should be asked first to ensure they aren’t missed. An ice breaker question can be is posed at the beginning to encourage conversation.
• Select a moderator to lead the group through the discussion questions. This moderator must remain neutral throughout the session so often an outside facilitator is hired for this role.
• Determine who will be the note taker and time keeper to keep the session on track and to record important ideas. Gather consent from the participants if the session is recorded.
• To allow everyone to participate, it is recommended to have 8-12 participants join the session.
• Participants should be invited to participate at random from a pool of eligible patients.
• Invite participants to the event at least two weeks prior to the session, and give the participants a reminder call/email a few days prior to the event.

How to conduct a focus group session:

• The moderator will begin the session by stating the objectives, providing logistical details and asking the ice breaker question.
• The moderator will ask each question in order on the question list and ensure everyone has had an opportunity to respond. Follow-up questions can be asked to gather more information.
• The note taker will record important ideas and observed conversation patterns.
• At the end of the session, the moderator will ask the participants for additional input or comments and will thank them for participating.

Once the patient focus group is complete and the conversation is documented, it’s time to analyze the data and identify trends. Group the discussion notes by question and identify themes from each of the topics. Consider analyzing the results based on demographics or sub-groups. Pull out insightful quotes or stories that exemplify each of the identified trends. A word cloud (figure 2) and/or a Pareto chart can be used to visually display common themes that arose. Tie the findings back to the session objectives and goals, and identify actionable opportunities for quality improvement and/or further investigation.

Organizing and executing a patient focus group is time intensive as compared to other research methods. However, the insights and patient stories that are gleaned from the discussion provide in-depth understanding of patient interaction with a health system. Applying data trends with patient stories allows for a holistic approach to information gathering and makes a strong case for quality improvement initiatives that impact patient experience.

References

1. Barbour, RS, 1999. The Use of Focus Groups to Define Patient Needs. Journal of Pediatric Gastroenterology & Nutrition; April 1999, Volume 28, Issue 4, pp.S19-S22. https://journals.lww.com/jpgn/Pages/ArticleViewer.aspx?year=1999&issue=04001&article=00002&type=Fulltext
2. Berger, S, Saut, AM, Berssaneti, FT, 2020. Using patient feedback to drive quality improvement in hospitals: a qualitative study. BMJ Open 2020; 10: e037641. https://bmjopen.bmj.com/content/10/10/e037641
3. Bombard, Y, et al., 2018. Engaging patients to improve quality of care: a systemic review. Implementation Science; 13:98. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6060529/pdf/13012_2018_Article_784.pdf
4. NHS England, 2016. Guide 09: A bite-size guide to run focus groups for patient and public engagement. NHS England; 05422, pp.1-9. https://www.england.nhs.uk/wp-content/uploads/2016/07/bitesize-guide-focus-groups.pdf
5. Silva, D, 2013. Measuring the patient experience. The Health Foundation; 18^th, pp.1-15. https://www.health.org.uk/sites/default/files/MeasuringPatientExperience.pdf

Want to earn CE Credits? go here: https://mainehealth.cloud-cme.com/Nov22HT

QI/PS Hot Topic- August 2022

Value Stream Mapping

Sarah Chang-Improvement Specialist II, MaineHealth Performance Improvement

Learning Objectives

1. Recognize when to use a value stream map.
2. Identify the key information to include in a value stream map.
3. Apply the value stream map to identify bottlenecks for further evaluation.

Value stream mapping (VSM) is a useful tool that seldom sees use outside of the learning arena. It is one of the fundamental tools of the Lean Six Sigma method and has a very practical application in healthcare. One challenge of its implementation is the blurred lines between VSM and a more popular improvement tool, process mapping. Scoping and kicking off a new quality improvement project could benefit from the use of VSM and can ensure that all aspects of the process centers the patient.

The purpose of Value Stream Mapping is to map out the entire process from start to finish of a macro-level patient encounter. It is a process map viewed from the perspective of the patient and highlights all the individual steps that patient goes through from entering the system to existing the system. Common examples might include:

• Outpatient specialty consult clinic visit
• Primary care visit
• Routine diagnostic visits
• Inpatient or outpatient procedures
• Emergency room triage

VSM generally proves its value as a means of mapping out the current state, including process times, of a macro-level process and identifying whether there are opportunities to make improvements by reducing waste. The VSM will allow the team to organize the process, identify problems on the map, and select where a waste walk (or several) might be a valuable exercise to observe certain steps. The map and waste walks serve as a jumping off point to dive deeper into specific process mapping of, for example, the registration process or the discharge process. It is reasonable to have additional process maps that detail the steps identified in the VSM.

Mapping Components

The VSM contains several pieces of information in one diagram that represent the value stream, keeping the patient’s perspective in mind.

Putting it all together, the resulting VSM, using the process of admitting a patient to the hospital from the ED, looks like:

Figure 1: Value stream map of patient admission from the ED to inpatient unit.

Using the VSM to Guide Improvement

Once a VSM is completed, it will be the most useful reference tool throughout the rest of the improvement work. Once process metrics are filled in (through waste walks or reporting if there is capability), work with the improvement team to identify key barriers. Further process mapping to detail the smaller steps within process groups may be necessary. The team should prioritize both reducing the non-value add time (delays/wastes) and reducing process time where there may be extreme variation.

Conclusion

As with any improvement methodology tool, understanding how to use VSM, selecting the appropriate process or part of the process to map with VSM, and understanding the goals are important for effective implementation. It takes practice to adapt the VSM to healthcare processes, but can be extremely useful in lean implementation. There are key differences between the VSM and traditional process mapping – VSM is used to identify and consolidate macro level barriers. A good rule of thumb is if there is a need for outlining the “patient timeline”, likely a VSM will be an invaluable tool to do so.

References

1. Worth, J., Shuker, T., Keyte, B., Ohaus, K., Luckman, J., Verble, D., Paluska, K., Nickel, T. (2012). Perfecting Patient Journeys: Improving patient safety, quality, and satisfaction while building problem-solving skills (Version 1.0). Lean Enterprise Institute.
2. Marin-Garcia, J., Vidal-Carreras, P., Garcia-Sabater, j. (2021). The Role of Value Stream Mapping in Healthcare Services: A Scoping Review. International Journal of Environmental Research and Public Health. 18(3), 951. https://dx.doi.org/10.3390%2Fijerph18030951
3. Anderson, S. (2020). How to Use Value Stream Maps in Healthcare. Minitab Blog. https://blog.minitab.com/en/value-stream-mapping-in-healthcare
4. Gellad, Z., Day, T., (2016). What is Value Stream Mapping and How Can It Help My Practice?. American Journal of Gastroenterology. 111(4), 447-448. https://dx.doi.org/10.1038%2Fajg.2016.38
5. Digital Healthcare Research. Value Stream Mapping. Agency for Healthcare Research and Quality. https://digital.ahrq.gov/health-it-tools-and-resources/evaluation-resources/workflow-assessment-health-it-toolkit/all-workflow-tools/value-stream-mapping

Cultivating Loyal Relationships with High Reliability Organizing

Melissa Retter, MA, CPHQ, CPHRM, CPSO, CPXP

Patient Experience Director

Maine Medical Center

Learning Objectives:

1.Recognize the distinction between patient experience and human experience

2.Describe how harm can erode human experiences and negatively impact trust and loyalty

3.Identify principles of high reliability organizing that facilitate positively memorable human experiences

When we ponder the essence of relationships that are established in healthcare settings, we are unable to do so without succumbing to the humble and simple realization that we are human beings caring for human beings. Our shared human condition results in a mutual vulnerability to the myriad of circumstances that we encounter in life. Our patients experience a wide range of influences that impact their well-being just as our care team members do. The Beryl Institute defines patient experience as, ‘the sum of all interactions shaped by an organization’s culture that influence patient perceptions across the continuum of care’ (2021). Beyond the sum of all interactions, patient experiences are defined by individual interactions they have with all other human beings. Together, we are empowered to co-create a consistently safe, compassionate, and positively memorable culture that arises when trusting relationships are formed. It behooves us to expand our mindset to think about relationship-centered care. Our engagement has a profound impact on the caliber of the relationships that we cultivate with our patients and their families.

Trusting and loyal relationships with our patients, colleagues and ultimately, organization, are eroded when we experience harm. While acknowledging the importance of physical harm, we focus our attention now on a few examples of the effects of psychological and emotional harm (Figure 1). Our patients may experience harm if they feel that we are discriminating against them and/or they perceive a lack compassion, courtesy, or respect. As a result, they may not receive indicated care or they may articulate anger towards caregivers. They may experience a desire to seek care elsewhere in the future, and share their negative experience with others, effectively harming our organizational reputation. Our colleagues may experience harm if they perceive a leader in the organization is controlling or condescending. They may feel unimportant and be less prone to speak up in the future when they identify safety concerns or have ideas about how to improve care. They may decide to seek employment elsewhere. Even worse, they may become disengaged and reckless during the delivery of patient care.  It is essential to note that nurturing relationships requires more than the mere absence of harm. We must prevent harm and consistently deploy best practices that are grounded in the science of organizing for high reliability.

When it comes to preventing harm and cultivating a care environment and work environment, there are practices within the realm of high reliability organizing that promote establishment of compassionate and positively memorable human experiences. Operating in a highly reliable manner does not imply that we are error free! Weick and Sutcliffe (2001) aptly reflect on how errors occur in complex systems. They highlight the processes and structures coupled with increased workload, distractions, over-confidence, and time pressures that impact the behavior of individuals and groups at the front line. Key tenants of high reliability include a commitment to the right culture, deferent and present leadership, continuous learning, organizational and individual resiliency, prevention and a zero tolerance for human harm (Figure 2). Reduction of power distances between patients, families, and colleagues by making eye contact, smiling, and greeting others by their preferred name creates a welcoming and inclusive culture. Commitment to the right culture also involves holding ourselves and our team members accountable for compassionate treatment. Noteworthy hallmarks of high reliability include a focus on preventing harm, a reluctance to simplify, a preoccupation with learning and deference to expertise. We seek to mitigate damage that can fracture trusting relationships with others by sharing human experience stories. We must go to the sites of care delivery to host listening labs, attend team huddles, and humbly defer to the knowledge of the workers at the front line. Lastly, our well-being as care team members is of paramount importance. When we take care of ourselves and we feel appreciated, we become resilient. When we are resilient, we can be mindful and present in each fleeting moment that we are blessed to share with others. We are empowered to tap into our passion as caregivers, and co-author consistent and mutually delightful human experiences. The result is devoutly loyal patients and care team members.

Figure 1. Harmful Communication and Behavior Outcomes*

Retter, M. 2021

*Examples above are not all-inclusive as additional ways harm impacts patients and team members exist.

References

1. Cook, R. and Woods, D. 1994. Operating at the Sharp End: The Complexity of Human Error. Healthcare Performance Improvement, LLC, All Rights Reserved.
2. Schultz, E. A. and Lavenda, R. H. 2005. Cultural Anthropology: A Perspective on the Human Condition. 7^th New York: Oxford University Press.
3. Weick, K.E. and Sutcliffe. 2001. Managing the Unexpected: Assuring High Performance in an Age of Complexity. San Francisco: Jossey-Bass Publishers.

Want to earn CE Credits? Click Here

Affinity Diagrams

Ghassan Saleh, DMD, DS

Director, MaineHealth Performance Improvement

 Learning Objectives:

1. Define the affinity diagram and clarify when best to use.
2. Illustrate the steps in generating affinity diagrams.
3. Describe best practices when developing affinity diagrams.

Let’s imagine that you are in a grocery store. You are taking care of your weekly shopping and trying to go through your list. You start by grabbing some oranges in the fruit and vegetable department, a gallon of Greek yogurt from the dairy department and some Cheetos from the chip aisle. As you go further down the list you realize that you also need some peaches – back to the fruit and vegetable area. And guess what? You also need some milk and cheese. You cross the store again to dairy section. Before you know it, you go back and forth several times. Wouldn’t it be a lot easier and more effective if you had grouped everything you needed by categories? That’s the idea of the affinity diagram.

An affinity diagram is a visual tool that helps improvement specialists organize the information they come up with during a brainstorming session. This organization take place by grouping those ideas to their affinity, or similarity. This way, the generated ideas become easier to act upon. The affinity diagram can also help stimulate new patterns of thinking, sparked by groups formed to find creative solutions to difficult problems and involving people from diverse backgrounds.

When to Use the Affinity Diagram?

Simply put, when your brainstorming session is over. Affinity Diagrams aren’t a brainstorming tool themselves but rather a way to organize, consolidate and act on ideas. They are specifically helpful when:

• Brainstorming with a large group of people
• You end up with a big number of ideas (or lot of data points)
• Dealing with complex problems
• Group consensus is needed

Tips: Good Practice when Developing Affinity Diagrams:

• Always start your brainstorming session with a clear objective – What is the problem at hand? What are you trying to solve?
• Assign a facilitator to help keep the conversation focused.
• Don’t come up with predetermined categories for your affinity diagram. You decide on the categories after all ideas are out on the white board (or virtual board).
• Lastly, affinity diagrams helps you solve for the three killers of a Kaizen meeting: 1) Meet, not discuss. 2) Discuss, not decide and 3) Decide, not do. Affinity diagrams can certainly help you decide but you should always follow that decision with a clear action plan: who is going to do what, when and how to make sure that the “do” is happening.

Steps Taken in Generating Affinity Diagrams

There are four steps in developing affinity diagrams. They are:

Step 1:  Display ideas you generated during a brainstorming session.

MaineHealth OpEx program’s preferred way of brainstorming is “brain-writing” where each idea is written down on one sticky note (or virtual sticky). Figure #1 is an example of displayed ideas for issues in implementing Continuous Process Improvement in healthcare

Figure 1: Issues in Implementing Continuous Process Improvement in Healthcare

Step 2: Sort ideas into similar groups.

Figure #2 shows the same ideas as figure #1 organized into similar themes.

Figure 2: Issues Organized into Similar Groups

 Step 3: Create header cards.

Header cards are created for each of the groups choosing a title that best describe the theme of each group. See Figure #3

Figure 3: Header Cards for Each Group of Ideas Categorized in Step #2.

*TQL in the far right box = Total Quality Logistics.

Step 4:  Draw finished diagram.

The finished diagram displays each group of ideas with their respective header card at the top of the group. Figure #4 shows the final product of the affinity diagram for issues in implementing continuous improvement in healthcare.

Figure 4: Finished Diagram

Resources:

1. 6^th edition of the Project Management Body of Knowledge (PMBOK Guide).
2. Minnesota Department of Health https://www.health.state.mn.us/communities/practice/resources/phqitoolbox/affinitydiagram.html
3. Pictures of the example is adopted from 2013 American College of Cardiology, CQI Knowledge Assessment Quiz: Answer Key https://cvquality.acc.org/docs/default-source/qi-toolkit/03_knowledgeassessmentquiz_answerkey_12-10-13new.pdf?sfvrsn=55478fbf_2

Want to earn CE credits? Click here

Integration of Leading and Lagging Indicators in Healthcare for Quality Improvement

Vijayakrishnan Poondi Srinivasan, MS, LSSBB

Quality Management Engineer

Maine Medical Center

Learning Objectives:

1. Introduction to Performance Measurement System (PMS) and Hierarchical Levels
2. Alignment of Leading and Lagging Indicators in the PMS
3. Illustrate integration of Leading and Lagging Indicators in Healthcare for Quality Improvement

An effective Performance Measurement System (PMS) enables an organization to assess whether goals are being achieved and facilitates improvement by clarifying goals, highlighting gaps, and facilitating reliable forecasts. Therefore, a strong PMS enables an organization to align its process level performance with management level goals.

A comprehensive PMS is constructed at three levels based on the hierarchy of the organization:

• Strategic Level: The main objective of this level is to translate the needs of the customer and stakeholder into defined goals and objectives.
• Tactical Level: This level supports the strategic goals and objectives developed in the strategic level and defines the drivers for achieving those goals.
• Operational Level: This level regulates the day-to-day output relative to schedules, specifications, and other aspects. The main scope at this level is streamlined process to work as quickly and efficiently as possible.

The PMS is characterized by the mixture of two types of performance measures. They are leading (cause) and lagging (effect) indicators respectively. The leading indicators are performance drivers in the operational level. Also, leading indicators are the operational inputs to the process. The lagging indicators are core outcomes that have a serious impact on the strategic level of the PMS. A balanced PMS should have a mix of outcome measures (lagging indicators) and performance drivers (leading indicators) which yield a cause-and-effect relationship.

Leading Indicators are also termed Performance Indicators (PI) and are present at the operational level and tactical level of the system. PI are the fundamental set of indicators defined for a process. These indicators include the input provided for each process. Strategies formulated at the management level are applied to PI because they serve as the input to the system.

Lagging Indicators are termed Key Performance Indicators (KPI) and are present at the strategic level and tactical level of the system. KPI are derived from the fundamental performance measures of a process. They are very useful to collect information on day-to-day activities and progress of a process. These indicators guide strategies for achieving the objective of an organization. The frequency of measurement differs based on the nature of the indicator, but they are often collected on a daily basis. It is important to establish a cause-and-effect relationship by linking the performance measures (PI and KPI) within and between different levels of the performance measurement system.

Figure 1: Alignment of Leading & Lagging Indicators

The application of this concept in Healthcare provides clarity in tracking the performance of the system at various levels of the organization. The example provided below, from the Orthopedics Service Line Clinical Transformation Project, demonstrates how efforts made in the operational level flow into the tactical level, and ultimately impact the strategic goals of the organization.

Problem Statement: Until January of 2018, Centers for Medicare & Medicaid Services (CMS) had Total Knee Replacement (TKR) on the inpatient procedure only list.  This requirement changed and CMS expected institutions to classify TKR patients as inpatient or outpatient and support this with appropriate documentation.

Goal: To have same day or one (1) overnight length of stay (LOS), making the procedure truly outpatient with respect to level of care.

Strategic Level Measures:

• Reduce average LOS ≤ 1.3 days
• Reduce reoperation rate (90 days)
• Reduce readmission rate (90 days)
• Reduce post-operative ED Visit rate (90 days)
• Maintain or improve patient satisfaction scores
• Reduce the variable direct cost per case for the episode of care by “X%”

Tactical Level Measures:

• Increase % of patients receiving Tranexamic Acid (TXA)
• Increase % of patients receiving Spinal Anesthesia 

Operational Level Measures:

• Implement patient inclusion and exclusion criteria for one-day knee replacement procedure
  • Develop and implement a clinical pathway for patients eligible for one-day knee replacement surgery
• Educate patients on the entire episode of care to prepare them for surgery and avoid any delays with on-time discharge
  • Increase the % of patients attend the educational session prior to surgery

References:

1. Daniels RC, Burns N. A framework for proactive performance measurement system introduction. International Journal of Operations & Production Management. 1997.
2. Bourne M, Mills J, Wilcox M, Neely A, Platts K. Designing, implementing and updating performance measurement systems. International journal of operations & production management. 2000.
3. Kueng P. Process performance measurement system: a tool to support process based organizations. Total quality management. 2000.
4. Rodriguez RR, Saiz JJA, Bas AO. Quantitative relationships between key performance indicators for supporting decision-making processes. Computers in Industry. 2009.

5. Tangen S. Performance measurement: from philosophy to practice. International journal of productivity and performance management. 2004.

Want to earn CE Credits? Click here

Frequency Plots:

How to tell a quality improvement story utilizing “plots”?

Sonja C. Orff, RN, MS, CNL, CSCT

Quality and Safety Coordinator

Operative and Perioperative Services

Maine Medical Center

September 2021

Learning Objectives:

1. Describe how to utilize a frequency plot/graphic to display quality improvement data
2. Identify the benefits of the different types of frequency plots

The adage, “a picture is worth a thousand words” carries significant weight when embarking on a quality improvement opportunity. Graphic displays of data offer insights that lists of numbers alone cannot. Visual tools to analyze trends and patterns in quality are powerful aids to achieve continuous improvement. Frequency plots provide graphical display of data sets that reveal associations and relationships. There are at least six frequency plot methods which one can choose.

The frequency graph one chooses depends on the type of data to be analyzed. There are graphics for continuous data, which is data that can take any value (e.g. height, weight, temperature, length) and graphics for attribute data, which is data that can be counted and given a whole numerical value (e.g. surgical case volumes).  The goal is to compare the differences between groups, and/or to study the relationships between variables and values. Below are examples and applications of frequency plot graphics.

Histogram

Histograms showcase the frequency of continuous data values (y axis) by displaying the distribution or “shape” of a data set. It also shows the spread of the data set (x axis) while capturing the presence of outliers or gaps in the data points. The histogram utilizes rectangular vertical bars to depict where most of the data occurs. These graphics should be constructed utilizing a sample size of at least 30 data points. If the data size is too small, the histogram may not accurately display the distribution.

Dot Plot

If the sample size is less than 30, a dot plot is preferred. A dot plot is a graphical representation of data utilizing dots plotted on a simple scale. Dot plot, when applied for small data sets, can be used for both continuous and discrete data sets. Above is a simple example of how a dot plot can be applied.

Histogram vs. Bar Chart

Each column or bar of the histogram represents the frequency of occurrence of quantitative continuous data (y axis). The columns or bars in a histogram and bar chart can vary in height and shape. However, as depicted above, the histogram has no spaces between the bars. What’s more, a bar chart shows the comparison of categorical discrete variables as opposed to number ranges.

Pareto Chart

When a bar chart presents the categories of data in a descending order of frequency and the cumulative total is represented by the line, this is known as a Pareto Chart. Above is an example of a Pareto Chart exemplifying the descending order of category data and the cumulative total trend line. Pareto Charts are discussed in more detail by Alan Picarillo in the June, 2020 edition of MITE QI/PS Hot Topic.

Stem and Leaf Plot

Stem and leaf plot is a frequency graphic less utilized but worth considering as a rapid approach to analyze and display data. The key difference between this plot and a histogram is that a stem and leaf plot can be constructed manually without the use of analytical software. Furthermore, a stem and leaf plot shows individual data points, resembling a table, whereas a histogram does not. The stem on the left displays the first digit(s) and the leaf on the right displays the last digit. Above, one can see that the individual data points in the 20-29 range are represented most often (four times) in the stem and leaf plot below.

Box and Whisker Plot

The box and whisker plot shows the following noteworthy statistics of data: median, maximum and minimum values, and upper and lower quartiles. The data are plotted in a way that the top 25% and the bottom 25% of the data points are represented by two whiskers. The box in the middle represents the remaining 50% of the data. Box and whisker plots are especially worthy when performing a comparison analysis between several data sets. This frequency plot allows for the visual comparision of central tendency, the variability of multiple data sets, and the presents of outliers.  Above is an example of a horizontal box and whisker plot.

Selecting a frequency plot that best tells the quality improvement story has many benefits. Visual presentation of data can motivate and provide an opportunity to elicit contribution and buy in from stakeholders. These graphics also provide a method and means in which to monitor for success, change, and opportunities for improvement.

References

CIToolKit. (2020). Graphic Analysis. Retrieved September 12, 2021, https://citoolkit.com/articles/graphical-analysis/

CIToolKit. (2020). Histograms-and-Boxplots. Retrieved September 12, 2021, https://citoolkit.com/articles/histograms-and-boxplots/

Hessing, Ted. (n.d.). Frequency Plots. Retrieved: September 10, 2021, https://sixsigmastudyguide.com/frequency-plots/

Model Systems Knowledge Translation Center (MSKTC). (n.d.). Effective Use of Histograms. Retrieved: August 19, 2021, https://msktc.org/lib/docs/KT_Toolkit/Charts_and_Graphs/Charts_Tool_Histograms_508c.pdf

Stem-and-leaf display. (2021, August 14). In Wikipedia. https://en.wikipedia.org/wiki/Stem-and-leaf_display

Want to earn CE credits? Click here!

QIPS (Quality Improvement Patient Safety) Hot Topic-Suneela Nyack MS, RN August 2021

Innovations in Improvement: Virtual Improvement Coaching during the Covid-19 Pandemic

Learning Objectives

1. Examine pros and cons of a virtual improvement coaching model necessitated by the Covid-19 pandemic
2. Discuss the value of virtual coaching methods for healthcare improvement teams
3. Determine opportunities to refine a Virtual Improvement Coaching Model

While the Covid-19 pandemic caused much disruption to established performance improvement workflows, it has also created opportunity for innovation and rapid transformational change.  To stay true to our MaineHealth Quality and Safety Mission, “Create a culture of continuous improvement that promotes quality and value in our healthcare system with an alignment of purpose…,” we resolved to innovate and adapt our traditional coaching methods.

To their credit, many MaineHealth care teams continued Key Performance Indicator (KPI) practice throughout the Covid-19 pandemic. Inspired by their engagement we turned to contemporary articles, blogs and webinars to learn more about remote and virtual coaching methods as a compliment to traditional techniques. Many authors underscored the value of establishing teamwork and camaraderie as key ingredients for the success of remote teams measured in terms of achieving strategic goals. Laura Spawn for Forbes (2020)¹ writes, “Overall, achieving goals as a team creates a culture of clear communication, job satisfaction, and motivation”.  She asserts this outcome is best achieved when all employees pitch in to meet strategic goals.  In a similar vein, Graham Kenny, (Harvard Business Review 2020)², discusses the importance of aligned top-down and bottom-up relationships for high value KPIs and strategy deployment, causality, and rapid responsiveness to change. In an earlier HBR publication, Dhawan and Chamorro-Premuzic³ suggest best practices to minimize the challenge of working with remote teams: clarity of written communications, avoiding digital overload, establishing communication norms, customizing approaches for individuals, and lastly, creating intentional space for celebrations.

Six months after we launched virtual improvement coaching, we conducted a survey of department leaders and teams active with KPIs and Operational Excellence to learn about what we could do better. All respondents agreed that their Op Ex Coach adequately addressed their questions and empowered them to advance their improvement work. More than 90% reported they had a strong relationship with their Op Ex Coach and were comfortable reaching out for help. This feedback suggests early success of our emerging Virtual Coaching Model and importantly, offers direction for improvement.  In combination with wisdom gleaned from the literature, we identified opportunities for refining virtual coaching techniques:

1. Establish strong relationships early in the process to foster camaraderie and teamwork as precursors for successful virtual coaching.
2. Invest time in pre-work such as agenda planning to improve productivity during coaching sessions
3. When possible, concurrently engage key stakeholders located in diverse settings to optimize value for the improvement team
4. Develop proficiency to fully leverage screen sharing capabilities on video conferencing platforms.
5. Deploy specific facilitation techniques to empower problem articulation, explain complex processes, and generate improvement ideas in a virtual forum

In summary, in keeping with our core values and mission, we have innovated an effective Virtual Improvement Coaching Model to support care teams and leaders. Guidance from a range of authors and PDSA (Plan-Do-Study-Act) thinking has led to proven virtual coaching workflows, optimized by proficiency in videoconferencing applications. Unexpected wins of a Virtual Coaching Model are expanded capacity, and the opportunity to engage key stakeholders from different locations at the same time.  Lastly, emerging expertise with Virtual Coaching techniques adds to our improvement toolkit, and opens the door to further innovations.

References

1. Spawn, L. (2020). Four Strategies for Setting Measurable Goals in a Remote Work Environment. Forbes. https://www.forbes.com/sites/forbeshumanresourcescouncil/2020/01/29/four-strategies-for-setting-measurable-goals-in-a-remote-work-environment/?sh=7f9441603014.
2. Kenny, G. (2020). What are your KPIs Really Measuring? Harvard Business Review.

 https://hbr.org/2020/09what-are-your-kpis-really-measuring.

3. Dhawan, E. and Chamorro-Premuzic, T. (2018). How to Collaborate Effectively If Your Team Is Remote. Harvard Business Review. https://hbr.org/2018/02/how-to-collaborate-effectively-if-your-team-is-remote.

Want to earn CE credits? Click here!