It’s a Process

DMAIC helps manage diabetes

by John W. Jennings III

More than half of Americans will have diabetes or be pre-diabetic by 2020.1 I have been a Type 2 diabetic for 40 years—about as long as I’ve been in the quality field. Currently, there is no cure for diabetes. But it can be managed. I found that statistical thinking and the define, measure, analyze, improve and control (DMAIC) method can reduce variation in diabetes management and create control.


Insulin converts glucose into cell nutrients. Type 1 diabetics produce little to no insulin and need insulin injections. Type 2 diabetics either produce too little insulin or have become insulin resistant. They control their blood sugar levels (BSL) by modifying their diet, exercising and, in some cases, taking medication.2

Type 1 and 2 diabetics use a device to measure their BSLs. Most diabetics aim to maintain a BSL between 80 and 150 milligrams per deciliter.3 Monitoring provides input data for statistical tools to evaluate the level of control the diabetic has on his or her body’s ability to regulate its BSL.

Statistical thinking supports BSL control with three basic assumptions:

  • Work occurs in a system of interconnected processes.
  • Variation exists in all processes.
  • Understanding and reducing variation are the keys to success.


An Excel spreadsheet is useful for recording BSL readings, medication use and comments. I use Excel functions to calculate averages and standard deviations on a daily, weekly and monthly basis. Figure 1 shows an example of a truncated data collection matrix.

Figure 1


Landmark studies published in the 1990s showed the benefit of tightly controlled BSLs in the management of both types of diabetes.4, 5 Key factors that influence BSL—exercise, weight and medication—are identified by doctors. Modifications to these factors help establish a zone of tight control. 

Design of experiments can determine factor levels for minimizing glucose level variation after exercise. This approach also can help determine calorie limits, meal times and what to eat.

Analysis of variance (ANOVA) can analyze the effects of medication changes. In August, I began a new weight loss medication, Victoza. Using ANOVA single-factor test, I evaluated my BSL averages, which are shown in Figure 2. The null hypothesis represents June and July daily averages without Victoza. The test shows u1 and u2 equals August’s daily averages with Victoza (u3). Because the null hypothesis is accepted, I can determine the medication has not affected my BSL.

Figure 2


The hardest part about managing diabetes is changing your lifestyle into a disciplined process. That usually means participation in an exercise program, controlling carbohydrate intake, monitoring BSLs, taking medication and sometimes abstaining from alcohol and nicotine.

Doctors set BSL goals and diabetics achieve them with lifestyle changes. A dietician helps plan a proper diet and identify foods to avoid. Medication may be a part of the management plan. Support from your family, friends and medical team is crucial.


Statistical process control (SPC) is used to monitor and analyze variation in manufacturing processes. Processes are in control when there are no special causes of variation present. A diabetic’s BSL is similar to a process with special cause variation because BSL regulation does not occur automatically.

Most diabetics track their BSL daily and may need to modify their food intake, exercise and medication accordingly. Figure 3 is an X-bar chart of my daily BSL averages in September. There were several days with daily averages on the high end due to special cause variation. My BSL was managed with more insulin, less food and more exercise.

Figure 3

It is possible for diabetics to experience a high quality of life. The key to achieving treatment goals is a combination of lifestyle changes, close monitoring and heeding doctors’ advice. DMAIC and statistical thinking provides a disciplined and data driven approach for making decisions about your health.


  1. Bill Berkrot, "Half of Americans facing diabetes by 2020: report," Reuters, Nov. 23 2010, www.reuters.com/article/2010/11/23/us-unitedhealth-diabetes-idusTRE6AM0NH20101123.
  2. American Diabetes Association, www.diabetes.org.
  3. "Hyperglycemia: When Your Blood Sugar is Too High," Diabetic Gourmet, http://diabeticgourmet.com/articles/206.html.
  4. National Institute of Diabetes and Digestive and Kidney Diseases, "The Diabetes Control and Complications Trial," Health Trends, 1994, http://diabetes.niddk.nih.gov/dm/pubs/control.
  5. B.M. Margetts, United Kingdom Prospective Diabetes Study (UKPDS) 13, BMJ, 1995, www.ncbi.nlm.nih.gov/pubmed/7833731.

John W. Jennings III is a compliance manager for Vertex Business Services in Richardson, TX. An ASQ fellow, Jennings is an ASQ-certified Six Sigma Black Belt, quality engineer, reliability engineer, software quality engineer, quality auditor, quality technician, biomedical auditor and a manager of quality/organizational excellence. He is an RABQSA-certified ISO 9001:2008 lead auditor, is certified by the National Association of Radio and Telecommunications Engineers as an electrostatic discharge control engineer, and is a senior examiner for the Texas Quality Award program.

Average Rating


Out of 0 Ratings
Rate this article

Add Comments

View comments
Comments FAQ

Featured advertisers