A Time Series Control Chart for Monitoring Abnormal Blood Glucose Levels.

Global diabetes statistics indicate a continuous rise in prevalence and complications, highlighting the need for more effective monitoring and management strategies. Selecting techniques for monitoring blood glucose levels is essential in detecting abnormalities, identifying root causes, and facilitating behavioral adjustments. This study proposes a control chart constructed by using a robust estimator concept, which is suitable for monitoring the autocorrelated blood glucose data as a time-series control chart based on . Its performance is evaluated by using a Monte Carlo simulation under varying parameters and compared with existing charts based on the average run length. Results will show that the proposed chart is the quickest in detecting abnormalities when the data are highly correlated and performs comparably in medium-to-low correlations. It is also applied to real patient self-monitoring data and interpreted with treatment guidelines to support behavioral adjustment. A case study will confirm its capability, particularly when used with physician guidance. The proposed chart provides timely behavior-linked insights, enhancing diabetes management.