Statistical Process Improvement
Georgetown University



Introduction to the Course

I am Dr. Alemi.  Throughout my career I have worked in the healthcare industry and in academia to improve quality of care.  This includes a stint at Veterans Administration where I was Chief of Office of Performance Improvement. I have worked with venture capital to design better and more efficient insurance companies and I have worked in academia to design new and more casual methods of quality improvement.  When I look at our healthcare system, I am totally frustrated. I cannot understand why medication errors keep occurring. I am disappointed that most people do not get antidepressant that benefits them. I find it strange that patients in emergency rooms are boarded for more than 6 hours waiting for a hospital room.  All of this tells me that the healthcare system needs to improve. We all deserve something better. The status quo will not do.

This course prepares you to use statistical process control to improve healthcare and inform improvement teams.  Statistics may seem boring, but when you consider what is at stake you quickly realize that it matters. It deserves your attention. Data helps you discipline your insights and intuitions. The course starts with an overview of process improvement and then focus on three topics:  risk assessment, control charts, and benchmarking. Speaking of data, I could not teach this course only a few years ago.  The course relies on CMS’s publicly available data on hospital performance.  A few years ago this data was not publicly available. Now they are and because of its availability, the course is radically more practical.  It is more real.

Throughout the course you will download data from Hospital Compare site and analyze the performance of real organizations.  I imagine from time to time, that a student who has finished this course shows up at an employment interview and informs the employer about the quality of their care.  It turns the table around. Take what you learn in this course to your job interview and show them how much you know about their organization.

In this course,  there are extensive assignments each week.  These assignments are accompanied with web resources that show you how to do the analysis or what are the correct answers.  These assignments are designed to give you skills in doing statistical process control. You will be able to analyze data and evaluate the performance of your organization.  Interested? As T. S. Eliot says, “Let us go and make our visit.”

In order to get you the necessary data analysis skills, we ask you to analyze data.  Yes! We keep asking you to do so, repeatedly. Sure, that is time consuming and repetitive, but practice makes perfect.  Without these repetitions you may not have the confidence that you can do it. To help you, we provide the correct answers and show you videos on how to solve the problem sets.  All you have to do is to copy the steps. Learn the concept and then do it. Keep doing it until you are well versed in it.

Finally, like medical residents, the course is organized on the principle of learn one, do one, and teach one.  You learn from the instructor how to do a control chart, you do what you have learned in assignments, and you also teach at least one of the assignments. Each of you are expected to make a video teaching others how to do one of the assignments.

Introduction to First Week

This week focuses on the big picture.  You will learn about how improvement efforts are organized and what happens in various phases of process improvement. We will talk about 10 steps in improvement efforts.  Some of these steps are carried out by organizational leaders to prepare the organization for change. Other steps are carried out by improvement teams. You will learn details about the culture that is conducive to process improvement.  What do I mean by culture? Well? Organizations that want to change for the better must not blame their employees; they must seek system changes and not more effort by their employees. We are not after bad apples but improving the entire basket of fruit. In improvement steps we talk about defining and solving the right problem.

Together we will learn what are good and poor problem definitions. We talk about how to use the voice of the consumer, I literally mean the recording of patient’s complaints, to get the organization to wake up to the need for change. We go on to talk about how to run efficient meetings, I am sure that you have been in those frustrating, torturous meetings that run on and on.  Lets ditch those meeting and learn how to run meetings more efficiently and connect better with each other. We will discuss what data to collect and how to verify that change has led to improvement. These are just a few steps in the 10 step process.  There is a lot to cover this week, so stay tuned. As they say, come hungry, there is a lot here.


  • Describe the role of organizational leaders in process improvement
  • Describe how interdisciplinary teams can improve care processes
  • Describe five strategies for improving the definition of a problem to be addressed by improvement teams.
  • Describe strategies for better meetings in improvement teams
  • Explain process improvement tools such as fishbone diagrams, flow charts, and storyboards.
  • Discuss the importance of using media to give voice to patient's experiences.
  • Plot a control chart using Excel

Assigned Reading

  1. Introduction (use instructor's last name as password) Read►
  2. Leading change Read►
  3. PDCA cycles Read►  
  4. 92 examples of process improvement efforts Read►



Instruction for Submission of Assignments: Assignments should be submitted directly on Blackboard.  In rare situations assignments can be sent directly by email to the instructor. Submission should follow these rules:

  1.  Submit only one document, usually an Excel file.  All questions should be answered in different sheets.  Each sheet should be labeled with the question number. 
  2. All Excel cells, except the cells containing the data, must have formulas.  Do not paste the value into the cell, it must be calculated using a formula.  Even simple steps, such as adding two numbers, should be done using formulas. 
  3. Make sure that any control charts follow the visual rules below:  (1) Control limits must be in red and without markers, (2) Observed lines must have markers, (3) X and Y axis must be labeled, and (4) Charts must be linked to the data. 
  4. Copy and paste SQL or R code into Excel sheet.  Plot data in Excel. 
  5. The first sheet in the file should be a summary page.  In the summary page you should list how your answers to the question differs from answers provided within the assignment (inside Teach One or other answers).  You must indicate for each question if your control chart is exactly the same as seen in Teach One or other formats.  For each question, you must indicate if the answers you have provided is the same as the answers supplied on the web.  If there are no answers provided, you must indicate that there were no answers available on the web to compare your answers to.

Question 1: Select  your teach one assignments.  We subscribe to the principle of "Learn one, do one, teach one."  The best way to learn is to teach the topic.  At start of the course, you select a topic in the course to teach.  A week prior to the due date, you submit all assignments in the topic to the instructor. On due date, you email to all students in the class a You Tube video showing how to solve assigned homework.  Missing any of the deadlines on this assignment will lead to 10% reduction in your grade for each week of delay.  If you and others have the same topic, you can focus on different parts of the lecture or assignments. Use Blackboard discussion to indicate the topic you want to focus on. If a student has already taken the topic, it is your responsibility to negotiate how your focus will differ.

Question 2: Plot the following data, where the X axis is the time periods, the Y-axis shows the observed value, the upper control limit and the lower control limits.  Distinguish between the first seven data periods as these were collected pre-intervention.  Remaining data points were collected post intervention.  Title the chart. Create a legend that defines the name for various lines.  Make sure that the observation line has markers and the control limits have no markers.  Make portion of Upper and Lower limit lines that are post intervention dashed.  Make portion of the line that is pre-intervention straight line.  Make all Upper and Lower limit lines red.  Data► Slides► Answer►

Question 3: This problem will test your familiarity with Excel. Consider the following population data, P: {0, 4, 10, 15, 47, 100}. Assume that you are sampling WITHOUT replacement i.e. once a population element is selected in a sample, that element cannot be selected again for that sample.

  •  List all possible samples of size 2 which can be drawn from set P. For example (0, 4) is a sample of size 2. Another such sample is (15, 47). Verify that you have listed all 15 possible samples.
  • Calculate the arithmetic mean for each individual sample that you listed earlier. Since there are 15 samples you should end up with 15 means. List all means.
  •  Calculate the arithmetic mean of the 15 means that you calculated earlier. This is the average value of all sample means. Compare this value to average of the initial 6 numbers, the population mean. Are they close? Same?

Question 4: In the following Table, fill in all cell values so that all indicated relationships hold. X and Y are random variables.  X with a bar above it indicates the average of X values in the column X. This problem will test your understanding of several principles related to elementary statistics and algebraic operations. Submitted your completed work as an Excel file. Data► Answer►

Question 5: Use snippets of videos from YouTube, quotes from Yelp (r) reviews, and data from Hospital Compare to define a problem in a healthcare system. Make sure that the definition of the problem follows the rules discussed in the lecture including:  (1) not blaming anyone, (2) numerically describe the gap in the desired and actual state, (3) using patient’s voices and/or quotes to accentuate the problem description. Hospital Compare► Bad hospital experiences► Yelp Hospital Reviews► Example Response► Another Example►


  • Many books cover the same topics as the current course.  One of our favorites is the free book "OpenIntro Statistics" Read►
  • History of patient safety efforts at AHRQ Read►
  • Psychology of change Read► 


Copyright 1996 Farrokh Alemi, Ph.D. Most recent revision 08/28/2019.  This page is part of the course on Statistical Process Improvement, this is the lecture on Introduction to the Course.