Exploring core and sentinel symptoms in elderly patients with type 2 diabetes through network analysis and the Apriori algorithm

Objective To construct the symptom network of elderly patients with type 2 diabetes mellitus (T2DM) and perform association rule analysis for each symptom cluster, identify core symptoms, and explore the sentinel symptoms, to provide references for accurate and efficient symptom management. Methods Convenience sampling was employed. Eligible elderly patients with T2DM were enrolled from the department of endocrinology, affiliated hospital of North Sichuan Medical College, Nanchong City, from March 2024 to July 2024, with data collected using the Chinese Version of Diabetes Symptom Checklist-Revised (DSC-R). The correlation was analyzed using the Apriori algorithm model to identify the sentinel symptoms of each symptom cluster, and the core symptom was identified based on the network analysis method. Results The most prevalent symptoms were thirst and dry mouth (71.69%), general fatigue (70.78%), and frothy urine (69.58%). Regarding symptom burden severity, general fatigue (2.48 ± 1.89), asthenia (2.42 ± 1.94), and insomnia (2.39 ± 1.98) had the highest scores. Sentinel symptoms, including excessive sleepiness, limb numbness, limb swelling, and polydipsia, are characteristic manifestations of distinct symptom clusters: psychological cognition symptom cluster, peripheral nerve symptom cluster, nephropathy symptom cluster, and hyperglycemia symptom cluster. At the global level, general fatigue (r s = 1.207) emerged as the core symptom. Conclusion This study identifies general fatigue as the core symptom, with excessive sleepiness, limb numbness, limb swelling, and polydipsia emerging as sentinel symptoms, within their respective symptom clusters. From the perspective of symptom network analysis and sentinel symptoms, clinical staff can develop accurate and efficient intervention measures to improve the efficiency of symptom management and reduce the burden of symptoms in elderly patients with T2DM.