Are lipid-lowering therapies used according to the guidelines?: A real-world retrospective study of 1001 patients with type 2 diabetes

Type 2 diabetes mellitus (T2DM) is frequently complicated by dyslipidemia, which increases the cardiovascular risk. Although guidelines strongly recommend lipid-lowering therapy, real-world treatment thresholds and adherence remain inconsistent in the literature. This retrospective observational study included 1001 patients with T2DM who attended outpatient clinics between March 2024 and March 2025. Demographic characteristics, metabolic parameters, and medication use (statins, fibrates, insulin, and neuropathy drugs) were retrieved from the hospital records. Group comparisons were performed using parametric and nonparametric tests, and multivariate logistic regression was used to identify the independent predictors of drug use. The mean age of the cohort was 60.9 ± 11.4 years, and 57.1% of the patients were female. The mean HbA1c was 7.47 ± 1.56%, low-density lipoprotein (LDL) was 110.7 ± 42.8 mg/dL, high-density lipoprotein was 49.6 ± 14.0 mg/dL, and triglycerides (TG) were 172.7 ± 100.1 mg/dL. Statin users had higher LDL and high-density lipoprotein levels but lower TG levels than nonusers. Fenofibrate users exhibited markedly higher TG levels. Insulin therapy was associated with poorer glycemic control but did not significantly affect the lipid parameters. In logistic regression, statin use was independently predicted by higher HbA1c and LDL levels, whereas fibrate use was predicted by elevated TG levels. Age was the only independent predictor of neuropathic drug use. In this real-world T2DM cohort, the initiation of statins and fibrates did not align with guideline-based thresholds. Statins were prescribed at higher LDL levels, whereas fibrates were reserved for patients with severely elevated TG levels. These findings highlight the persistent gap between guidelines and clinical practice and emphasize the need for strategies to improve lipid management in secondary-care settings.