Background and Objectives: Investigating the adherence to antihypertensive medication and identifying patients with low adherence allows targeted interventions to improve therapeutic outcomes. Artificial intelligence (AI) offers advanced tools for analyzing medication adherence data. This study aimed to develop and validate several predictive models for non-adherence, using patient-reported data collected via a structured questionnaire. Materials and Methods: A cross-sectional, multi-center study was conducted on 3095 hypertensive patients from community pharmacies. A structured questionnaire was administered, collecting data on sociodemographic factors, medical history, self-monitoring behaviors, and informational exposure, alongside medication adherence measured using the Romanian-translated and validated ARMS (Adherence to Refills and Medications Scale). Five machine learning models were developed to predict non-adherence, defined by ARMS quartile-based thresholds. The models included Logistic Regression, Random Forest, and boosting algorithms (CatBoost, LightGBM, and XGBoost). Models were evaluated based on their ability to stratify patients according to adherence risk. Results: A total of 79.13% of respondents had an ARMS Score ≥ 15, indicating a high prevalence of suboptimal adherence. Better adherence was statistically associated (adjusted for age and sex) with more frequent blood pressure self-monitoring, a reduced salt intake, fewer daily supplements, more frequent reading of medication leaflets, and the receipt of specific information from pharmacists. Among the ML models, CatBoost achieved the highest ROC AUC Scores across the non-adherence classifications, although none exceeded 0.75. Conclusions: Several machine learning models were developed and validated to estimate levels of medication non-adherence. While the performance was moderate, the results demonstrate the potential of AI in identifying and stratifying patients by adherence profiles. Notably, to our knowledge, this study represents the first application of permutation and SHapley Additive exPlanations feature importance in combination with probability-based adherence stratification, offering a novel framework for predictive adherence modelling.

Background: The pharmacological management of type 2 diabetes mellitus has become increasingly complex due to expanding therapeutic options and the high prevalence of multimorbidity in affected patients. As a result, the risk of drug–drug and drug–disease interactions has grown significantly, with potential implications for glycemic control, safety, and treatment outcomes. Objective: This narrative review provides a comprehensive, class-based overview of clinically relevant interactions associated with antidiabetic medications, highlighting their mechanisms, clinical consequences, and practical management strategies. Methods: A targeted literature search was conducted using major medical databases to identify clinical studies, meta-analyses, pharmacovigilance reports, and evidence-based guidelines concerning interactions related to key antidiabetic drug classes. Interactions were categorized as pharmacokinetic, pharmacodynamic, or disease-related. Results: Significant variability exists across antidiabetic drug classes in terms of interaction profile and clinical relevance. Metformin presents interaction risks mainly through renal impairment or co-administration with drugs affecting lactate metabolism. Sulfonylureas and glinides are strongly associated with hypoglycemia-enhancing interactions. DPP-4 inhibitors generally exhibit a low interaction burden, whereas GLP-1 receptor agonists may interact through delayed gastric emptying. SGLT2 inhibitors require caution in patients with diuretics or conditions predisposing them to dehydration or ketoacidosis. Insulin remains highly sensitive to pharmacodynamic interactions with a broad spectrum of therapies. Underlying renal, hepatic, and cardiovascular conditions further modify the interaction risk. Conclusions: Understanding class-specific interaction profiles is essential for personalized and safe diabetes management. Careful medication review, close metabolic monitoring, and individualized dose adjustments can mitigate the risk of harmful interactions. Further research is needed to elucidate interactions in populations with advanced multimorbidity and polypharmacy.

In the context of an aging population and the growing prevalence of polypharmacy, optimizing pharmacological treatment in older adults has become essential for minimizing risks and for improving the quality of life. The STOPP/START criteria, now in their third version as of 2023, provide a clear and practical clinical guide for identifying potentially inappropriate medications (STOPP) and potential prescribing omissions (START) in older individuals. Validated through the Delphi consensus method and tailored to European clinical practice, these criteria offer a more nuanced approach than other tools, such as the Beers Criteria, which primarily reflect the realities of the U.S. healthcare system. This paper reviews the STOPP/START v3 recommendations on avoiding harmful prescriptions, offering concrete examples from the STOPP list (e.g., inappropriate use of opioids, benzodiazepines, antipsychotics, estrogens, or antidiabetic agents), with pharmacological justifications based on the mechanisms of action, adverse effects, and considerations specific to older patients. A follow-up article will focus on identifying missing yet necessary treatments, a critical issue in geriatric care. Overall, STOPP/START v3 represents a modern, balanced, and more Europe-oriented tool that contributes to safer, more personalized, and effective pharmacotherapy in older populations. Integrating these criteria into everyday medical decision-making is a key step toward delivering evidence-based, patient-centered care in geriatrics.

Probleme de terapie medicamentoasă la vârstnic. Criteriile STOPP/START (I) Polimorfismul genetic și farmacocinetica preparatelor antihipertensive

Outside end-of-life situations, where treatment is predominantly palliative, the START criteria (Screening Tool to Alert to Right Treatment) highlight therapies that should be considered in older patients if they are missing without a valid clinical justification, with careful respect for contraindications and prescribing guidelines. These include cardiovascular drugs (statins, ACE inhibitors/ARBs, beta-blockers, antiplatelets, anticoagulants, heart failure therapies), respiratory drugs (LAMA, LABA, inhaled corticosteroids, home oxygen), bone metabolism therapies (vitamin D, calcium, antiresorptive or anabolic agents), gastrointestinal therapies (proton pump inhibitors, fiber, laxatives, probiotics, Helicobacter pylori eradication), neurological and psychiatric drugs (L-DOPA, non-TCA antidepressants, acetylcholinesterase inhibitors, SSRIs/SNRIs, propranolol), musculoskeletal and rheumatological drugs (DMARDs, bisphosphonates), genitourinary therapies (alpha-blockers, 5-alpha-reductase inhibitors, vaginal estrogens, PDE5 inhibitors), renal therapies (alfacalcidol/calcitriol, phosphate binders, erythropoietin, RAAS blockers), and vaccines (influenza, pneumococcal, varicella-zoster, SARS-CoV-2).