Survey of Potential Drug Interactions, Use of Non-Medical Health Products, and Immunization Status among Patients Receiving Targeted Therapies.

In recent years, several changes have occurred in the management of chronic immunological conditions with the emerging use of targeted therapies. This two-phase cross-sectional study was conducted through structured in-person interviews in 2018-2019 and 2022. Additional data sources included ambulatory medical records and the itemized reimbursement reporting interface of the National Health Insurance Fund. Drug interactions were analyzed using the UpToDate Lexicomp, Medscape drug interaction checker, and Drugs.com databases. The chi-square test was used, and odds ratios (ORs) were calculated. In total, 185 patients participated. In 53% of patients (n = 53), a serious drug-drug interaction (DDI) was identified (mean number: 1.07 ± 1.43, 0-7), whereas this value was 38% (n = 38) for potential drug-supplement interactions (mean number: 0.58 ± 0.85, 0-3) and 47% (n = 47) for potential targeted drug interactions (0.72 ± 0.97, 0-5) in 2018. In 2022, 78% of patients (n = 66) were identified as having a serious DDI (mean number: 2.27 ± 2.69, 0-19), 66% (n = 56) had a potential drug-supplement interaction (mean number: 2.33 ± 2.69, 0-13), and 79% (n = 67) had a potential targeted drug interactions (1.35 ± 1.04, 0-5). Older age (>60 years; OR: 2.062), female sex (OR: 3.387), and polypharmacy (OR: 5.276) were identified as the main risk factors. Screening methods and drug interaction databases do not keep pace with the emergence of new therapeutics.

Leveraging code-free deep learning for pill recognition in clinical settings: A multicenter, real-world study of performance across multiple platforms.

BACKGROUND: Preventable patient harm, particularly medication errors, represent significant challenges in healthcare settings. Dispensing the wrong medication is often associated with mix-up of lookalike and soundalike drugs in high workload environments. Replacing manual dispensing with automated unit dose and medication dispensing systems to reduce medication errors is not always feasible in clinical facilities experiencing high patient turn-around or frequent dose changes. Artificial intelligence (AI) based pill recognition tools and smartphone applications could potentially aid healthcare workers in identifying pills in situations where more advanced dispensing systems are not implemented. OBJECTIVE: Most of the published research on pill recognition focuses on theoretical aspects of model development using traditional coding and deep learning methods. The use of code-free deep learning (CFDL) as a practical alternative for accessible model development, and implementation of such models in tools intended to aid decision making in clinical settings, remains largely unexplored. In this study, we sought to address this gap in existing literature by investigating whether CFDL is a viable approach for developing pill recognition models using a custom dataset, followed by a thorough evaluation of the model across various deployment scenarios, and in multicenter clinical settings. Furthermore, we aimed to highlight challenges and propose solutions to achieve optimal performance and real-world applicability of pill recognition models, including when deployed on smartphone applications. METHODS: A pill recognition model was developed utilizing Microsoft Azure Custom Vision platform and a large custom training dataset of 26,880 images captured from the top 30 most dispensed solid oral dosage forms (SODFs) at the three participating hospitals. A comprehensive internal and external testing strategy was devised, model's performance was investigated through the online API, and offline using exported TensorFlow Lite model running on a Windows PC and on Android, using a tailor-made testing smartphone application. Additionally, model's calibration, degree of reliance on color features and device dependency was thoroughly evaluated. Real-world performance was assessed using images captured by hospital pharmacists at three participating clinical centers. RESULTS: The pill recognition model showed high performance in Microsoft Azure Custom Vision platform with 98.7 % precision, 95.1 % recall, and 98.2 % mean average precision (mAP), with thresholds set to 50 %. During internal testing utilizing the online API, the model reached 93.7 % precision, 88.96 % recall, 90.81 % F1-score and 87.35 % mAP. Testing the offline TensorFlow Lite model on Windows PC showed a slight performance reduction, with 91.16 % precision, 83.82 % recall, 86.18 % F1-score and 82.55 % mAP. Performance of the model running offline on the Android application was further reduced to 86.50 % precision, 75.00 % recall, 77.83 % F1-score and 69.24 % mAP. During external clinical testing through the online API an overall precision of 83.10 %, recall of 71.39 %, and F1-score of 75.76 % was achieved. CONCLUSION: Our study demonstrates that using a CFDL approach is a feasible and cost-effective method for developing AI-based pill recognition systems. Despite the limitations encountered, our model performed well, particularly when accessed through the online API. The use of CFDL facilitates interdisciplinary collaboration, resulting in human-centered AI models with enhanced real-world applicability. We suggest that rather than striving to build a universally applicable pill recognition system, models should be tailored to the medications in a regional formulary or needs of a specific clinic, which can in turn lead to improved performance in real-world deployment in these locations. Parallel to focusing on model development, it is crucial to employ a human centered approach by training the end users on how to properly interact with the AI based system to maximize benefits. Future research is needed on refining pill recognition models for broader adaptability. This includes investigating image pre-processing and optimization techniques to enhance offline performance and operation on handheld devices. Moreover, future studies should explore methods to overcome limitations of CFDL development to enhance the robustness of models and reduce overfitting. Collaborative efforts between researchers in this domain and sharing of best practices are vital to improve pill recognition systems, ultimately enhancing patient safety and healthcare outcomes.

Dietary quercetin supplements: Assessment of online product informations and quantitation of quercetin in the products by high-performance liquid chromatography.

Administration of the increasingly popular dietary supplements containing quercetin may interfere with drug therapy. We intended to evaluate the online availability and quercetin content of the high-dose mono-component quercetin products and to review the potential use of quercetin products and their interactions with drugs. We monitored the online access to quercetin-containing dietary supplements, collected the relevant information from the websites, procured selected products from the vendors, and subjected them to substance analysis. The quercetin content was quantified by an HPLC-UV method. Twenty-five websites offered mono-component quercetin products, and nine products were procured. The quercetin content of eight products differed only ±10% from the nominal dose, whereas one product contained almost 30% more quercetin. Misleading indications such as antitumor and cardiovascular effects were often found on the sellers' websites. Quercetin-containing dietary supplements are available online with misleading indications. The recommended daily doses are often high (occasionally over 1,000 mg), which may induce clinically relevant interactions with medications. Because high-quercetin content of dietary supplements was confirmed, health care professionals should be aware of the unregulated internet market of dietary supplements and should consider the interactions of these substances with drugs.

Countrywide prevalence of critical drug interactions in Hungarian outpatients: a retrospective analysis of pharmacy dispensing data.

BACKGROUND: Drug-drug interactions (DDIs) present a significant source of adverse drug reactions. Despite being one of the commonly cited risks to patient safety, prevention of DDIs still poses a challenge to healthcare systems. The prevalence of DDIs can be used as a quality indicator for the safety of prescribing. With the analysis of drug utilization databases, real-world data on critical DDIs can be obtained. The aim of this study was to establish a list of critical DDIs and estimate their prevalence in the Hungarian outpatient population. METHODS: Since there is no conclusive and generally accepted repository of high-risk DDIs, a systematic search of the literature for consensus-based lists was performed. Based on these results and their analysis with 5 interaction compendia, we propose a simple methodology to identify critical combinations. Present study focused on DDIs which are (1) of high clinical importance thus being most likely to cause significant harm if not detected, (2) well-supported by available evidence and (3) affect drugs which are routinely dispensed in the community pharmacy setting. A retrospective analysis of prescriptions filled between 2013 and 2016 was performed. The source of drug utilization data was the IQVIA's national prescription fill database. The number of interacting drug pairs dispensed at the same time to the same patient was established. RESULTS: After excluding drugs with low dispensing rates, the analysis covered 39 DDIs. The distribution of risk categories of the analysed DDIs was inconsistent among different drug interaction compendia. The total number of prescriptions filled varied between 173924449 and 176368468 per year. The prevalence of the selected potential DDIs ranged from 0.00 to 355.89 per 100000 prescriptions per year. There was significant variation between how the number of cases had changed for each DDI throughout the study period, no general tendency could have been described. CONCLUSIONS: There were 1.8 million cases of co-dispensing each year, where prescribers' and community pharmacists' role in recognizing and managing potentially serious interactions was or would have been critical. The method presented to identify high-risk DDIs can serve as a starting point for the much-needed improvement of routine interaction screening.

[Comprehensive evaluation of drug interaction screening programs: discrepancies and concordances]. / Gyógyszerkölcsönhatások kiszurésére szolgáló adatbázisok értékelése: ellentmondások és egyezoségek.

INTRODUCTION: Recognition of potentially harmful drug interactions is one of the duties of healthcare. However, solutions involving databases are fraught with contradictions due to the lack of standardized principles and data. AIM: The aims of the authors were to perform a comparative evaluation of Hungarian and international databases and to explore ambiguities and contradictions in order to develop more standardized criteria for screening interactions. METHOD: Four Hungarian and two English-language websites and software, and the summaries of product characteristics were compared. The authors analyzed 40 drug-drug and 8 drug-supplement interactions and looked at 8 cases, which represent 28 pairs of interacting substances. RESULTS: The databases warn about most interactions, but these warnings were rarely helpful in preventing undesired consequences. The authors found discrepancies between the databases in 70% of interactions. When looking at different products with the same active ingredients, discrepancies cropped up in 0-66.7% of the cases. Up to 80% of searches for supplementary product interactions did not produce satisfactory results. CONCLUSIONS: In the present situation mapping these ambiguities and creating a standardized classification system would be advantageous.