Blood glucose bought back into a state of control

Mr Christopher Ude Mr Christopher Ude

Diabetes is a serious health issue of epidemic proportions, affecting millions of people every year. According to the American Diabetes Association, in 2015, diabetes remained the 7th leading cause of death in the United States. You cannot reverse diabetes, only learn to effectively manage it.

So how does one do so? Medication/insulin, lifestyle changes, blood glucose monitoring, etc springs to mind. Whilst I will not go into medication or pass on lifestyle recommendations, the remainder of this article will focus on self-monitoring in diabetes management, using an SPC approach.

No matter the application, when there is data, statistical analysis is required to uncover meaning. Diabetes management is no exception!

Let's consider the following analogy!

In manufacturing, SPC is used to derive statistical knowledge from data, required to adequately control processes to a target. Upper and lower limits for all process variables are defined, indicating that processes must not exceed these, otherwise quality will be compromised.

Similarly, SPC can too be applied to diabetes management. Afterall, a typical blood glucose reading is ‘data’, which generally is recorded against a date and time. Like in manufacturing, blood glucose also has a target range which diabetics need to manage to. Should blood glucose exceed these limits, health could potentially be compromised.

The best way to describe the effectiveness of SPC in diabetes management is through a case study.


The following is about a 55-year-old diabetic, Patient X, who successfully brought her blood glucose back to a state-of-control through lifestyle changes, using change analysis as her tool of insight.

When Patient X was first diagnosed with diabetes type 2, she was devastated. She had no idea as to what caused her diabetes, and more so, how to effectively manage it.

Patient X did not know where to start. It was assumed that because she was now taking medication, this would keep her diabetes at bay.

Each morning, Patient X took her blood glucose readings. It was high one day, low the next. Rather than jot it down, she discarded it, only to try again the next morning, and so on. Why was her blood glucose all over the place? Medication not working? Was she more worse off than initially thought?

Patient X visited the doctor, who advised she needs to keep an account of all her measurements plus may have to make lifestyle changes. Convinced her lifestyle was not causing her harm, Patient X proceeded like normal, but no change. Patient X’s daughter, who used change analysis for weight loss during her 12-week challenge, suggested her mother use change analysis. This required Patient X to record her blood glucose readings and use change analysis to interpret the data.

For Patient X, change analysis would analyse her data (blood glucose readings) to uncover underlying trends, detect changes in blood glucose, help identify causes to changes, and more. It would also serve as a statistical-based profile on her blood glucose which Patient X could then share with her GP.

Figure 1 is a change analysis chart representing Patient X’s blood glucose readings

Shewhart chart demonstrating the issues with spc theory Figure 1: Patient X's fasting blood glucose readings plotted on a change analysis

As you can see, there is an underlying, statistically-based trend in the form of a step, taking variability of the results into consideration. Patient X was clearly not in control. Results are very variable. Yes, she already knew something was wrong, but was unable to understand the magnitude, detect changes and even know the causes.

If you move along the chart in Figure 1 you will see an upward step change. It was identified that the step change correlated to patient x having overseas visitors during that period therefore were ‘eating-out’ every night. It was at this moment that Patient X realised her lifestyle may indeed be effecting her health and therefore decided to ‘eat-in’ the following week. The step change decreased as shown in Figure 2. So clearly there was a lifestyle factor.

Shewhart chart demonstrating the issues with spc theory Figure 2: Decrease in fasting blood glucose and cause identified

Pleased with the insights, Patient X drilled deeper into her lifestyle and made more positive changes. Yet, another step change was detected also shown in Figure2.

More and more changes were made, resulting in Patient X’s blood glucose drop further to within the acceptable levels.

Through change analysis, Patient X was able to fully understand the effect’s her lifestyle was having on her diabetes, forcing her to take control and make important changes. Reacting to a single measurement was impossible due to the variability. The change analysis depicted an underlying trend of all readings, taking variability into account, thereby making it possible to react and drive improvement which is what change analyis is all about.

Today Patient X is living a healthy life, is a lot more energetic, and uses change analysis everyday to keep track of her blood glucose. Her doctor is extremely pleased with her progress.

200 Analytics Tools For Quality

Download 200 statistical analytics tools for quality improvement - Mobile Version

Improve product quality through smart data insights using new-age, machine-powered driven analytics, available to download through a suite of Apps!

  • Apps for SPC, MSA, Process Performance, Inferences, Visualization, and much more
  • No licencing or subscriptions! Pay ONLY per analysis billed monthly - don't use, don't pay!
  • Always up-to-date with the latest features