Establishing cut-off points for consistency in reporting hypoglycemia symptoms among diabetes patients

• Main Authors: Sharmin, Afsana Al, Zulkafli, Hani Syahida, Mohamed Ali, Nazihah
• Format: Article
• Published: Pushpa Publishing House 2023
• Online Access: http://psasir.upm.edu.my/id/eprint/107746/; https://pphmjopenaccess.com/index.php/jpjb/article/view/1169

Recognizing the onset of hypoglycemia is crucial for managing its effects in diabetes patients, but the variability of symptoms makes accurate reporting challenging. This study is aimed to establish cut-off points for differentiating consistent and inconsistent symptom reporting during hypoglycemic episodes among diabetes patients. The study utilized data collected by the UK Hypoglycemia Study Group, comprising 26 symptoms reported by 66 diabetes patients during 2791 episodes of hypoglycemia. A Bayesian consistency model was developed to assess individual symptom consistency during episodes. The estimated consistency values were assumed to follow a beta distribution between 0 and 1, and the regularized incomplete beta function was used to determine cut-off points. The analysis revealed significant variability in symptom reporting both within episodes and between patients. Using the assumed distribution and the method of moments estimation, cut-off points were established to categorize patients into low (0 to 0.44), medium (0.44 to 0.60), and high (0.60 to 1) consistency levels. Approximately 35 of the patients demonstrated a high level of consistency, while the majority exhibited lower consistency in reporting symptoms. The classification of consistencies from an assumed distribution provides meaningful and relevant cut-off points, enabling a more accurate differentiation between reporting of consistent and inconsistent symptoms. The results contribute to improving the understanding of symptom variability and can aid in personalized treatment approaches for diabetes patients.