Model-Based Tool for Personalized Adjustment of Basal Insulin Supply in Patients With Intensified Conventional Insulin Therapy

Abstract

Background: The decisive factor in successful intensive insulin therapy is the ability to deliver need-based-adjusted nutrition-independent insulin dosages at the closest possible approximation to the physiological insulin level. Because this basal insulin requirement is strongly influenced by the patient’s lifestyle, its subtlety is of great importance. This challenge is very different between patients with type 1 diabetes and those with insulin-dependent type 2 diabetes. Furthermore, it is more difficult to finetune a basal insulin dosage with intensified conventional insulin therapy (ICT), due to delayed insulin delivery, compared to insulin pump therapy, which provides continuous delivery of small doses of exclusively short-acting insulin. In all cases, the goal is to achieve an optimal basal delivery rate. Method: We hypothesized that this goal could be achieved with a modeling tool that determined the optimal basal insulin supply based on the patient’s anamnestic data and monitored glucose values. This type of modeling tool has been used in health insurance programs in Germany to improve insulin control in patients that receive ICT. Results: Our retrospective data analysis showed that this modeling tool provided a significant improvement in metabolic control, significant reductions in HbA1c and Q scores, and improved time-in-range values, with reduced daily insulin levels. Conclusion: The model-based basal rate test could provide additional data of the actual effect of the basal insulin adjustment in intensified insulin treated diabetes to the physician or treatment team.

Efficacy of a basal insulin dose management smartphone application for controlling fasting blood glucose in patients with type‐2 diabetes mellitus: A single‐centre, randomised clinical study

“…We believe that publishing the formula can improve the traceability of insulin doses, enable timely detection of sources of dose errors, and enhance the safety of the app. Compared with other apps, 33,34 our app serves as a clinical decision‐making tool for clinicians, with low cost of use and easy‐to‐use operation, improving the learning curve for operating the app and reducing the chance of patient errors in dose adjustment during treatment. Compared with the conventional regular expert consultation to titrate the BI dose regimen, our app shortens the FBG‐TRT, thus demonstrating the effectiveness and safety of the app.…”

“…sources of dose errors, and enhance the safety of the app. Compared with other apps,33,34 our app serves as a clinical decision-making tool for clinicians, with low cost of use and easy-to-use operation, improving the learning curve for operating the app and reducing the chance of patient errors in dose adjustment during treatment.Compared with the conventional regular expert consultation to titrate the BI dose regimen, our app shortens the FBG-TRT, thus demonstrating the effectiveness and safety of the app.A limitation of this study is that it is a single-centre, open-label study and these factors may lead to study bias. The study subject population did not include patients who were combined with oral hypoglycaemia agents or once-weekly glucagon-like peptide-1 receptor agonists, leading to a limited population for application of the study results.…”

The Effectiveness of an App (Insulia) in Recommending Basal Insulin Doses for French Patients With Type 2 Diabetes Mellitus: Longitudinal Observational Study

“…Franc et al [1] reported on the TeleDiab-2 trial that, at month 4, twice as many patients using such a device compared to the control group achieved an HbA 1c level <7% (29.8% vs 12.5%). Other similar devices have also shown positive results [5,6]. However, the translation of clinical trial results into real life often raises a series of questions that lead to an interest in conducting postmarketing observational studies of products.…”

Modelos Personalizados de Metabolismo de Glucosa para Pacientes Diabéticos Tipo 2 Mediante Algoritmos Evolutivos

“…Para poder conocer de mejor manera cómo cambia el metabolismo de cada paciente se recurre a los modelos personalizados. Dichos modelos están generalmente basados en los datos de los pacientes utilizando alguna técnica de ajuste paramétrico [7]- [10]. Por lo anterior, en el presente artículo se presenta la metodología para ajustar un modelo matemático del metabolismo de glucosa mediante una técnica de algoritmos evolutivos.…”