Retrospective Analysis of Adverse Drug Reactions in Patients with Type 2 Diabetes Mellitus and Development of a Risk Prediction Model.

BACKGROUND: Patients with diabetes are prone to adverse drug reactions (ADRs) during long-term medication, but the related risk factors and predictive models remain to be clarified. OBJECTIVE: To analyze the influencing factors of ADRs in patients with type 2 diabetes mellitus (T2DM) and to develop and validate a risk prediction model. METHODS: Clinical data of 254 randomly selected patients with T2DM were retrospectively analyzed. Patients were divided into a control group (n=198) and an ADR group (n=56) based on the presence of ADRs. Multivariate logistic regression was used to identify independent risk factors associated with the occurrence and establish a prediction model. RESULTS: Among the 56 patients who experienced ADRs, insulin hypoglycemic drugs had the highest proportion of adverse reactions (33.93%). Age ≥ 60 years old, three or more type medication, three or more comorbidities, abnormal BMI (> 24 or < 18 kg/m2), and length of hospitalization > 7 days were independent risk factors associated with the occurrence of ADRs (P < 0.05). A multivariate logistic regression model incorporating age, medication type, number of comorbidities, body mass index, and length of hospitalization demonstrated good predictive performance, with an accuracy of 85.43%, sensitivity of 83.93%, and specificity of 85.86%. CONCLUSION: The above five indicators are independent risk factors associated with the occurrence of ADRs in patients with T2DM. The risk prediction model has certain predictive performance and may help early identification of high-risk patients.