Treatment-dependent and treatment-independent risk factors associated with the risk of diabetes-related events: a retrospective analysis based on 229,042 patients with type 2 diabetes mellitus

Background: The aim of this study was to analyse which factors predict the real world macro/microvascular event, hospitalisation and death risk in patients with type 2 diabetes mellitus. Furthermore, we aimed to investigate whether there exists both an under- and over-treatment risk of these patients. Methods: We used a German claims/clinical data set covering the years 2010-12. Diabetes-related events were defined as (1) macro-, (2) microvascular events leading to inpatient hospitalisation, (3) other hospitalisations with type 2 diabetes mellitus as main diagnosis, (4) all-cause death and (5) a composite outcome including all event categories 1-4. Factors associated with event risk were analysed by a Kaplan Meier curve analysis and by multivariable Cox regression models. Results: 229,042 patients with type 2 diabetes mellitus (mean age 70.2 years; mean CCI 6.03) were included. Among factors that increased the event risk were patients' age, male gender, the adapted Charlson Comorbidity Index, the adapted Diabetes Complication Severity Index, previous events, and number of prescribed chronic medications. For systolic blood pressure/HbA(1C), a double-J/U-curve pattern was detected: HbA(1C) of 6-6.5% (42-48 mmol/mol) and systolic blood pressure of 130-140 mmHg (17.3-18.7kPa) were associated with the lowest event risk, values below/above that range were associated with higher risk. However, this pattern was mainly driven by the death risk and was much less clearly observed for the macrovascular/microvascular/hospitalization risk and for young/less comorbid patients. Conclusions: Both blood pressure and HbA(1C) seem to be very important treatment targets, especially in comorbid old patients. It is of particular clinical importance that both over- and under-treatment pose a threat to patients with type 2 diabetes mellitus.