Background/Objectives: Diabetes mellitus and hypertension are major chronic conditions that markedly affect patients’ health and quality of life worldwide. With the rapid development of technology, there has been a growing interest in exploring the potential role of artificial intelligence (AI) in the management of such diseases. This study aims to assess the accuracy and reliability of artificial intelligence tools in providing information for diabetes mellitus and hypertension management. Methods: This study assessed the accuracy and reliability of the information provided by major AI tools such as ChatGPT, Gemini, POE, Claude, Consensus, and Perplexity. Twenty questions that are essential for the management of diabetes mellitus and hypertension were constructed based on the chapters of the respective guidelines and were fed to the AI tools. The outcomes were compared with evidence-based treatment guidelines, such as those from the American Diabetes Association (ADA), the American Heart Association (AHA), the European Society of Cardiology (ESC), and the National Institute for Health and Care Excellence (NICE). Answers were classified into “accurate “, “inaccurate”, and “accurate with missing information”. Three rounds of six-week intervals were conducted to assess accuracy and reliability. In addition, they were conducted to evaluate data updates by comparing answers across the rounds. Results: In round one of the evaluations, ChatGPT and Poe showed the highest accuracy, both at 65% (95% CI: 41.0–83.7), followed by Claude at 60% (95% CI: 41.0–83.7). ChatGPT had the lowest inaccuracy rate at 5% (95% CI: 1.75–33.1), while Claude demonstrated the smallest percentage of responses with missing information at only 6%. (95% CI: 12.8–54.3). In round 2, Claude markedly outperformed all other tools, achieving an accuracy rate of 95% (95% CI: 73.0–99.7) and no responses with missing information (0%). In round 3, ChatGPT came second with 70% (95% CI: 45.70–87.2) accuracy and maintained the lowest inaccuracy rate of 5% (95% CI: 0.26–26.9). Consensus had the largest inaccuracy rate at 40% (95% CI: 20.0–63.6) and the lowest accuracy rate at 40% (95% CI: 20.0–63.6). Overall, statistically significant pairwise comparisons showed that Cloud in the second round has the highest accuracy compared to Poe (p = 0.0154), Gemini (p = 0.0421), Consensus (p = 0.0035), and Perplexity (p = 0.0302). In the assessment of performance shift from round 1 to round 2, Claude achieved the greatest improvement in accuracy at 40%. In the assessment of performance shift from round 2 to round 3, Poe improved the most with an accuracy increase of 25%, while ChatGPT followed with 20%. When evaluating the unprompted and guideline-prompted questions for all AI tools using McNemar’s test, it did not reveal a statistically significant distinction in the proportion of accurate responses (p > 0.05). Conclusions: Throughout the three rounds, ChatGPT maintained the best performance, with the fewest missing data. Claude and Poe followed, showing high accuracy with relatively low inaccuracy rates. On the other hand, Perplexity and Gemini performed moderately, while Consensus had the lowest accuracy.

Warfarin is a common anticoagulant drug for thrombo-prophylaxis in stroke and venous thromboembolism, which has many advantages but also some disadvantages including narrow therapeutic window, vast drug interactions (and wide variability with foods/herbs), as well as unpredictability of pharmacodynamics and/or kinetics. Complicating factors can present as challenges for the outpatient clinicians trying to strike that balance due to the potential consequences of over or under dose anticoagulation with associated increased risk of bleeding and/or thromboembolic events, respectively. Because warfarin interactions can drastically affect therapeutic outcomes, patient to healthcare provider communication regarding such potential drug-drug or diet-warfarin interactions is crucial for compliance with the medication and achieving successful treatment. Furthermore, language barriers cause low patient satisfaction scores and poor quality/safety health care. In fact, the advancement and improvements in healthcare technology promise artificial intelligence (AI) as one of ideal options to optimize delivery of health care. The goal of this study is to develop Warfa-Check, a bilingual AI-based web app that matches both speakers of Arabic and English. The application helps users recognize potential warfarin-associated drug interactions with a simple user interface that accepts text, picture or voice commands. Warfa-Check, developed with Python and Flask as well as OpenAI’s GPT-4 API with natural language processing tools trained to correctly interpret outbound warfarin interactions. Multiple validation methods and beta testing have been done to ensure that the app is data-driven, as well color coded alerts for interaction severity provide clear feedback to end-users. This easy-to-use application helps patients identify drug interactions in both English and Arabic. Warfa-Check represents a valuable avenue for improving the safety of our residents, simplifying medication management in high-risk individuals and streamlining workflow. Future development plans are to develop into other anticoagulants, and integrate with Electronic Health Records (EHRs).

Background: Levetiracetam (LEV) is among the most frequently prescribed antiseizure medications, yet its tolerability is limited in some patients by psychiatric and behavioral adverse events. Because psychiatric comorbidity and psychotropic co-medication are common in epilepsy, attributing such events to LEV rather than to the underlying disease remains difficult. Thus, we sought to characterize LEV-associated psychiatric reporting and to determine whether it differs by age, sex, and concurrent psychotropic use. Methods: Different safety reports submitted to the FDA Adverse Event Reporting System (FAERS) from 2013 through 2022 were extracted and joined by case identifier, restricted to LEV as the primary suspect drug, with co-administered antidepressants and antipsychotics flagged. Reports lacking age or sex were excluded. Groups were compared using the t-test and the chi-square or Fisher exact test, with odds ratios referenced to LEV without a recorded antidepressant or antipsychotic. Results: Among 928 reports (mean age 48.0 ± 20.1 years; 60.9% female), dizziness (30.1%), anxiety (16.5%), agitation (10.3%), insomnia (9.8%), depression (9.6%), and amnesia (8.8%) accounted for more than 85% of adverse events. The profile was age dependent. Agitation was reported more often for adolescents than for adults (adjusted OR 3.22, 95% CI 1.67–6.20). Psychotropic co-medication was recorded in 230 reports (24.8%). Furthermore, anxiety (OR 2.07, 95% CI 1.43–2.99) and depression (OR 1.82, 95% CI 1.15–2.89) were reported more frequently in co-medicated reports, whereas dizziness (OR 0.59, 95% CI 0.42–0.84) and agitation (OR 0.44, 95% CI 0.25–0.83) were reported less frequently. The association with agitation was attenuated and was no longer statistically significant after adjustment (aOR 0.61, 95% CI 0.33–1.14). Conclusions: These findings characterize age- and co-medication-related differences in psychiatric adverse-event reporting with levetiracetam. After correction for multiple comparisons, agitation remained more frequently reported among adolescents, while other apparent age-related differences were not statistically significant. These findings provide descriptive pharmacovigilance evidence regarding psychiatric adverse-event reporting patterns associated with levetiracetam.

Background: Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used for pain, inflammation, and fever. Although many NSAIDs are available without a prescription, inappropriate use may increase the risk of renal, gastrointestinal, and cardiovascular complications. This study evaluated NSAID dispensing patterns and potential risk practices in community pharmacies in the Eastern Province of Saudi Arabia. Methods: A cross-sectional study was conducted in 90 community pharmacies across Al-Ahsa, Al-Dammam, and Jubail Industrial City between February and March 2025. Data on demographics, NSAID type, indication, dosage, frequency, duration, and concomitant medication use were collected and analyzed descriptively. Results: Among the 171 documented NSAID dispensing cases, 98 (57.3%) occurred without a prescription, whereas 73 (42.7%) involved prescription-based dispensing. Ibuprofen (62%) and diclofenac (36.3%) were the most commonly dispensed agents, with oral tablets being the preferred formulation (76.6%). Most documented treatment durations were short (3–7 days) with once- or twice-daily dosing. Common documented indications included dental conditions, pain, and inflammation, although the indication was not specified in 57.3% of cases. At least one potential interaction was identified in 11(7.8%, 95% CI 4.4–13.4%) of the 141 cases with concomitant medication use, corresponding to 13 (5.9%, 95% CI 3.5–9.9%) potential interaction pairs. Conclusions: Frequent non-prescription NSAID dispensing, incomplete documentation, and potentially clinically relevant drug interactions highlight the need to strengthen medication review, documentation, public awareness, regulatory oversight, and pharmacist counseling in community pharmacy practice.

BACKGROUND Narrow therapeutic index medications require careful dose individualization because small changes in drug exposure may result in treatment failure or toxicity. Their management often requires multiple pharmacokinetic calculations, therapeutic drug monitoring, and interpretation of clinical guidelines, which can make the dosing process complex and time-consuming. OBJECTIVE This study aimed to develop KFU Dose AI, an integrated clinical decision-support application that combines validated pharmacokinetic calculations with artificial intelligence assisted interpretation to support individualized dosing and therapeutic drug monitoring of NTI medications. METHODS KFU Dose AI was developed using R for deterministic pharmacokinetic calculations and incorporated the OpenAI GPT-4o model with a local Retrieval-Augmented Generation system. Validated pharmacokinetic equations and established clinical guidelines were integrated into medication-specific modules. The AI component was designed as a clinical auditing and interpretation layer rather than as the primary calculation engine. Patient-specific clinical and laboratory parameters are entered into the application, after which the system performs pharmacokinetic calculations and generates clinical assessments, dosing recommendations, and safety alerts. The interface also incorporates a color-coded system to facilitate rapid interpretation of therapeutic, subtherapeutic, and supratherapeutic results. RESULTS The developed platform incorporated modules for several NTI medications, including vancomycin, gentamicin, phenytoin, valproate, lithium, and digoxin. The system successfully integrated patient-specific pharmacokinetic calculations with guideline-based AI-assisted interpretation within a single interface. Depending on the medication, the application provides outputs such as estimated drug exposure, corrected drug concentrations, predicted doses or trough concentrations, clinical interpretation, dosing recommendations, and relevant safety-monitoring alerts. CONCLUSIONS KFU Dose AI provides an integrated approach to pharmacokinetic dosing and therapeutic drug monitoring by combining deterministic pharmacokinetic calculations with guideline-grounded AI-assisted clinical interpretation. The platform may simplify complex dosing workflows and support clinical decision-making for NTI medications. Further prospective clinical validation is required to evaluate its accuracy, usability, and impact on clinical outcomes.