15 January 2022 in Diabetology & Metabolic Syndrome
The objective of this paper was to review the adherence and persistence rates of major antidiabetic medication classes (i.e., metformin, sulfonylureas, sodium glucose cotransporter-2 inhibitors, dipeptidyl peptidase-4 inhibitors, insulin, glucagon-like peptide-1 receptor agonists, and thiazolidinediones) by summarizing the major findings of the studies published since 2017. In addition, we reported the potential causes for low adherence and persistence of antidiabetic medications. Based on the literature, the highest rate of adherence and persistence was consistently observed in metformin users. Second to metformin were sodium glucose cotransporter-2 inhibitors. Injectable therapies such as insulin and glucagon-like peptide-1 receptor agonists trailed low on the adherence and persistence rates. To the best of our knowledge, no studies published since the year 2017 analyzed the adherence and persistence of thiazolidinediones independently. The most frequently cited cause for low adherence and persistence was the severity of adverse events. Baseline characteristics (e.g., baseline HbA1c level), demographic information (e.g., age, gender, or ethnicity), and comorbidity profiles also had significant impacts on adherence and persistence in patients with type 2 diabetes mellitus.
1 June 2023 in Journal of Diabetes Research
Aims. This study is aimed at identifying clinical characteristics associated with adherence and persistence in patients with type 2 diabetes mellitus (T2DM) treated with dulaglutide. Materials and Methods. This retrospective observational cohort study used the Common Data Model at Seoul National University Hospital, Seoul, South Korea. Eligible subjects were followed for one year. Multivariate logistic and linear regressions were used to identify the factors associated with categorical (i.e., adherence status and continuation status) and continuous (i.e., proportion of days covered, or PDC, and treatment duration) outcome measures, respectively. Subgroup analysis was conducted involving patients at high cardiovascular disease (CVD) risk (i.e., having 2 identifiable risk factors). Results. A total of 236 patients were included. Increase in age and estimated glomerular filtration rate significantly increased the likelihood of adherence and treatment continuation. In contrast, baseline obesity and baseline use of sulfonylurea and insulin significantly reduced the likelihood of continuing dulaglutide. Similarly, increase in age, switching dulaglutide dose, and baseline neuropathy significantly increased PDC and treatment duration. None of the adherence or persistence outcome measures were significantly different between patients at high CVD risk and their matched controls. Baseline hypertension and the higher baseline LDL-C level significantly increased the likelihood of adherence in patients at high CVD risk. Conclusion. Clinical characteristics of dulaglutide users that could have affected their adherence and persistence were identified. Physicians treating T2DM patients with dulaglutide can refer to those clinical characteristics identified in this study to optimize the adherence and persistence to dulaglutide.
28 May 2025 in Scientific Reports
Our study aims to improve the prediction performance of machine learning (ML) models by addressing false records (i.e., false positive, false negative, or missingness) in binary categorical variables in electronic medical records (EMRs) using propensity score (PS). This study used the EMRs of patients with type 2 diabetes mellitus (T2DM) treated with basal insulin at a tertiary university hospital in South Korea. We expanded the definition of PS into the probability of having a record for a binary variable given covariates. We calculated PS for the binary categorical variables in their EMRs and developed PS datasets. By utilizing various ML algorithms, we developed and validated ACS prediction models on 80% and 20% of the dataset, respectively. We evaluated model performance using accuracy, recall, precision, F1 score, and AUROC. Additionally, the Shapley Additive Explanation (SHAP) method was used to identify important clinical predictors of ACS. The study included 9,338 patients (with an average age of 60.2 years and 56.6% of whom were male) over 10,184 treatment periods. The most prevalent comorbidities were hypertension (31.5%) and dyslipidemia (28.9%). Notably, 6.9% experienced ACS during their insulin treatment. The ML models trained on PS datasets generally outperformed the models trained on raw datasets. The results of SHAP analysis showed that older age, higher baseline weight, higher baseline glucose, history of antithrombotic therapy, history of chest pain, and indicators of T2DM progression (e.g., senile cataract) were important ACS risk factors. We have developed an ACS prediction model with an improved performance and higher reliance on clinical predictors that are in alignment with current medical understanding.
30 March 2022 in Korean Journal of Health Promotion
Background: This study describes the development of a multi-dimensional questionnaire for the personalized recommendation of health functional foods.Methods: The questionnaire was designed to include two parts: common and detailed. Each question was formulated based on clinical evidence and physician-administered diagnostic tools provided by reputable sources such as government agencies and medical institutions. The common questions were intended to assess overall health status. The detailed questions were organized into 28 categories based on the classification of health functional foods’ health claims by the Ministry of Food and Drug Safety. Each question’s response was evaluated on a scale from 0 to 1. The final score of each category was calculated as the sum of the scores normalized by the number of questions in each category. Two expert physicians in health functional food assessed the content validity of the questionnaire, and the questionnaire was revised accordingly.Results: A total of 31 common questions evaluated overall health conditions based on demographic information, medical history, allergies, prescription medicines, dietary habits, and lifestyle. A question that surveys interest areas was also included to let the responders choose up to three categories for the detailed question part. A total of 173 detailed questions assessed risk factors and symptoms relevant to each health claim.Conclusions: This questionnaire recommends health functional foods by measuring health risks and symptoms, following the classification of health functional foods by the Ministry of Food and Drug Safety. This questionnaire may serve as an evidence-based information collecting tool for a personalized health functional foods recommendation system.