MetaLAB-HOI: Template standardization of health outcomes enable massive and accurate detection of adverse drug reactions from electronic health records.

PURPOSE: This study aimed to advance the MetaLAB algorithm and verify its performance with multicenter data to effectively detect major adverse drug reactions (ADRs), including drug-induced liver injury. METHODS: Based on MetaLAB, we created an optimal scenario for detecting ADRs by considering demographic and clinical records. MetaLAB-HOI was developed to identify ADR signals using common model-based multicenter electronic health record (EHR) data from the clinical health outcomes of interest (HOI) template and design for drug-exposed and nonexposed groups. In this study, we calculated the odds ratio of 101 drugs for HOI in Konyang University Hospital, Seoul National University Hospital, Chungbuk National University Hospital, and Seoul National University Bundang Hospital. RESULTS: The overlapping drugs in four medical centers are amlodipine, aspirin, bisoprolol, carvedilol, clopidogrel, clozapine, digoxin, diltiazem, methotrexate, and rosuvastatin. We developed MetaLAB-HOI, an algorithm that can detect ADRs more efficiently using EHR. We compared the detection results of four medical centers, with drug-induced liver injuries as representative ADRs. CONCLUSIONS: MetaLAB-HOI's strength lies in fully utilizing the patient's clinical information, such as prescription, procedure, and laboratory results, to detect ADR signals. Considering changes in the patient's condition over time, we created an algorithm based on a scenario that accounted for each drug exposure and onset period supervised by specialists for HOI. We determined that when a template capable of detecting ADR based on clinical evidence is developed and manualized, it can be applied in medical centers for new drugs with insufficient data.

Cancer depends on fatty acids for ATP production: A possible link between cancer and obesity.

Several metabolic pathways for the supply of adenosine triphosphate (ATP) have been proposed; however, the major source of reducing power for ADP in cancer remains unclear. Although glycolysis is the source of ATP in tumors according to the Warburg effect, ATP levels do not differ between cancer cells grown in the presence and absence of glucose. Several theories have been proposed to explain the supply of ATP in cancer, including metabolic reprograming in the tumor microenvironment. However, these theories are based on the production of ATP by the TCA-OxPhos pathway, which is inconsistent with the Warburg effect. We found that blocking fatty acid oxidation (FAO) in the presence of glucose significantly decreased ATP production in various cancer cells. This suggests that cancer cells depend on fatty acids to produce ATP through FAO instead of glycolysis. We observed that cancer cell growth mainly relies on metabolic nutrients and oxygen systemically supplied through the bloodstream instead of metabolic reprogramming. In a spontaneous mouse tumor model (KrasG12D; Pdx1-cre), tumor growth was 2-fold higher in mice fed a high-fat diet (low-carbo diet) that caused obesity, whereas a calorie-balanced, low-fat diet (high-carbo diet) inhibited tumor growth by 3-fold compared with that in mice fed a control/normal diet. This 5-fold difference in tumor growth between mice fed low-fat and high-fat diets suggests that fat-induced obesity promotes cancer growth, and tumor growth depends on fatty acids as the primary source of energy.

Glucose Deprivation Induces Cancer Cell Death through Failure of ROS Regulation.

In previous work, we showed that cancer cells do not depend on glycolysis for ATP production, but they do on fatty acid oxidation. However, we found some cancer cells induced cell death after glucose deprivation along with a decrease of ATP production. We investigated the different response of glucose deprivation with two types of cancer cells including glucose insensitive cancer cells (GIC) which do not change ATP levels, and glucose sensitive cancer cells (GSC) which decrease ATP production in 24 h. Glucose deprivation-induced cell death in GSC by more than twofold after 12 h and by up to tenfold after 24 h accompanied by decreased ATP production to compare to the control (cultured in glucose). Glucose deprivation decreased the levels of metabolic intermediates of the pentose phosphate pathway (PPP) and the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) in both GSC and GIC. However, glucose deprivation increased reactive oxygen species (ROS) only in GSC, suggesting that GIC have a higher tolerance for decreased NADPH than GSC. The twofold higher ratio of reduced/oxidized glutathione (GSH/GSSG) in GIS than in GSC correlates closely with the twofold lower ROS levels under glucose starvation conditions. Treatment with N-acetylcysteine (NAC) as a precursor to the biologic antioxidant glutathione restored ATP production by 70% and reversed cell death caused by glucose deprivation in GSC. The present findings suggest that glucose deprivation-induced cancer cell death is not caused by decreased ATP levels, but rather triggered by a failure of ROS regulation by the antioxidant system. Conclusion is clear that glucose deprivation-induced cell death is independent from ATP depletion-induced cell death.

Peripheral blood transcriptomic clusters uncovered immune phenotypes of asthma.

BACKGROUND: Transcriptomic analysis has been used to elucidate the complex pathogenesis of heterogeneous disease and may also contribute to identify potential therapeutic targets by delineating the hub genes. This study aimed to investigate whether blood transcriptomic clustering can distinguish clinical and immune phenotypes of asthmatics, and microbiome in asthmatics. METHODS: Transcriptomic expression of peripheral blood mononuclear cells (PBMCs) from 47 asthmatics and 21 non-asthmatics was measured using RNA sequencing. A hierarchical clustering algorithm was used to classify asthmatics. Differentially expressed genes, clinical phenotypes, immune phenotypes, and microbiome of each transcriptomic cluster were assessed. RESULTS: In asthmatics, three distinct transcriptomic clusters with numerously different transcriptomic expressions were identified. The proportion of severe asthmatics was highest in cluster 3 as 73.3%, followed by cluster 2 (45.5%) and cluster 1 (28.6%). While cluster 1 represented clinically non-severe T2 asthma, cluster 3 tended to include severe non-T2 asthma. Cluster 2 had features of both T2 and non-T2 asthmatics characterized by the highest serum IgE level and neutrophil-dominant sputum cell population. Compared to non-asthmatics, cluster 1 showed higher CCL23 and IL1RL1 expression while the expression of TREML4 was suppressed in cluster 3. CTSD and ALDH2 showed a significant positive linear relationship across three clusters in the order of cluster 1 to 3. No significant differences in the diversities of lung and gut microbiomes were observed among transcriptomic clusters of asthmatics and non-asthmatics. However, our study has limitations in that small sample size data were analyzed with unmeasured confounding factors and causal relationships or function pathways were not verified. CONCLUSIONS: Genetic clustering based on the blood transcriptome may provide novel immunological insight, which can be biomarkers of asthma immune phenotypes. Trial registration Retrospectively registered.

Sodium-Glucose Cotransporter-2 Inhibitor-Related Diabetic Ketoacidosis: Accuracy Verification of Operational Definition.

BACKGROUND: The most important aspect of a retrospective cohort study is the operational definition (OP) of the disease. We developed a detailed OP for the detection of sodium-glucose cotransporter-2 inhibitors (SGLT2i) related to diabetic ketoacidosis (DKA). The OP was systemically verified and analyzed. METHODS: All patients prescribed SGLT2i at four university hospitals were enrolled in this experiment. A DKA diagnostic algorithm was created and distributed to each hospital; subsequently, the number of SGLT2i-related DKAs was confirmed. Then, the algorithm functionality was verified through manual chart reviews by an endocrinologist using the same OP. RESULTS: A total of 8,958 patients were initially prescribed SGLT2i. According to the algorithm, 0.18% (16/8,958) were confirmed to have SGLT2i-related DKA. However, based on manual chart reviews of these 16 cases, there was only one case of SGLT2i-related DKA (positive predictive value = 6.3%). Even after repeatedly narrowing the diagnosis range of the algorithm, the effect of a positive predictive value was insignificant (6.3-10.0%, P > 0.999). CONCLUSION: Owing to the nature of electronic medical record data, we could not create an algorithm that clearly differentiates SGLT2i-related DKA despite repeated attempts. In all retrospective studies, a portion of the samples should be randomly selected to confirm the accuracy of the OP through chart review. In retrospective cohort studies in which chart review is not possible, it will be difficult to guarantee the reliability of the results.

Lung virome: New potential biomarkers for asthma severity and exacerbation.

BACKGROUND: Although some respiratory virus infections are known to contribute to the development and exacerbation of asthma, commensal viromes in airway have not been extensively studied due to technical challenges. OBJECTIVES: This study investigated the characteristics of the virome in asthmatic airways. METHODS: Both the bacteriome and virome profiles in sputum from 12 healthy individuals, 15 patients with nonsevere asthma, and 15 patients with severe asthma were analyzed and assessed for the association with clinical characteristics such as severity, exacerbation, Asthma Control Test (ACT), and lung function. RESULTS: While analysis of the 16S ribosomal RNA bacteriome in the airway showed no differences, clear contrasts in the diversity and composition of airway viromes were observed between healthy controls and patients with asthma. Herpesviruses were the most abundant type of virus in the asthma group (44.6 ± 4.6%), mainly with cytomegalovirus (CMV) and EBV accounting for 24.5 ± 3.3% and 16.9 ± 3.5%, respectively, in contrast to those in the healthy controls (5.4 ± 2.5% and 7.1 ± 3.0%, respectively). CMV and EBV were more abundant in patients with asthma who experienced exacerbation, and their abundance showed correlation with more severe asthma, lower ACT score, and lower lung function. On the contrary, bacteriophage that is abundant in healthy controls was severely reduced in patients with asthma in the order of nonsevere and severe asthma and presented significant positive correlation with ACT and FEV1/forced vital capacity. CONCLUSIONS: Lung viromes, especially, CMV, EBV, and bacteriophage may be potential biomarkers of asthma severity and exacerbation.

Fubp1 supports the lactate-Akt-mTOR axis through the upregulation of Hk1 and Hk2.

Cells require energy for homeostatic activities, growth and division. By utilizing glucose as the main energy source, cells generate ATP and metabolic precursors through glycolysis and citric acid cycle. Although the oxidative phosphorylation can produce more ATP molecules from one molecule of glucose than glycolysis, rapidly growing cells primarily metabolize glucose via aerobic glycolysis. This aerobic glycolysis makes cells to uptake glucose at a higher rate and to efficiently convert glucose into the macromolecules required for new daughter cells. Recent evidence suggests that Fubp1 promotes cell proliferation and survival, and it is overexpressed in a variety of cancers. However, the role of Fubp1 in cellular metabolism remains unclear. In the present study, we demonstrated that Fubp1 upregulates the mRNA levels of two hexokinase genes, Hk1 and Hk2. We also found the positive correlation in mRNA expression between Fubp1 and both of hexokinase genes in several types of cancers. We suggest that Fubp1 contributes to cell survival through supporting lactate-Akt-mTOR axis.

Inferring causal relationship in coordinated flight of pigeon flocks.

Collective phenomenon of natural animal groups will be attributed to individual intelligence and interagent interactions, where a long-standing challenge is to reveal the causal relationship among individuals. In this study, we propose a causal inference method based on information theory. More precisely, we calculate mutual information by using a data mining algorithm named "k-nearest neighbor" and subsequently induce the transfer entropy to obtain the causality entropy quantifying the causal dependence of one individual on another subject to a condition set consisting of other neighboring ones. Accordingly, we analyze the high-resolution GPS data of three pigeon flocks to extract the hidden interaction mechanism governing the coordinated free flight. The comparison of spatial distribution between causal neighbors and all other remainders validates that no bias exists for the causal inference. We identify the causal relationships to establish the interaction network and observe that the revealed causal relationship follows a local interaction mode. Interestingly, the individuals closer to the mass center and the average velocity direction are more influential than others.

Age-related prevalence of chronic rhinosinusitis and nasal polyps and their relationships with asthma onset.

BACKGROUND: Chronic rhinosinusitis (CRS) is a major disease condition with high morbidity and can influence lower airway disease status in adults. However, its associations with adult asthma onset and activity have not been examined in detail in a general adult population. OBJECTIVE: To investigate relationships between CRS with nasal polyps (CRSwNP) and asthma characteristics. METHODS: A cross-sectional data set of 17,506 adult participants (≥18 years old) in the Korean National Health and Nutrition Examination Survey from 2010 through 2012 was analyzed. CRS was defined using structured questionnaires according to the international guideline, and presence of nasal polyps was objectively assessed using nasal endoscopy. Presence of asthma and its onset and current activity were assessed using structured questionnaires. RESULTS: CRS was significantly related to asthma, but the relationships were distinct by CRS and asthma status. CRSwNP was significantly associated with adult-onset asthma (onset after 18 years of age) or late-onset asthma (onset after 40 years of age), whereas CRS without nasal polyps was related to childhood-onset asthma (onset before 18 years) or early-onset asthma (onset before 40 years) in adults. The 2 CRS subgroups showed significant associations with current asthma but not with past asthma. However, the comorbid asthma rate was lower than 10% among subjects with CRS. CONCLUSION: This study found distinct age-related patterns of CRSwNP and asthma and demonstrated their significant associations in a general population. However, the low prevalence of asthma in CRSwNP is in sharp contrast to findings in Western populations, which warrants further investigation for ethnic or regional differences in relationships between CRSwNP and asthma.

Large-scale machine learning of media outlets for understanding public reactions to nation-wide viral infection outbreaks.

From May to July 2015, there was a nation-wide outbreak of Middle East respiratory syndrome (MERS) in Korea. MERS is caused by MERS-CoV, an enveloped, positive-sense, single-stranded RNA virus belonging to the family Coronaviridae. Despite expert opinions that the danger of MERS might be exaggerated, there was an overreaction by the public according to the Korean mass media, which led to a noticeable reduction in social and economic activities during the outbreak. To explain this phenomenon, we presumed that machine learning-based analysis of media outlets would be helpful and collected a number of Korean mass media articles and short-text comments produced during the 10-week outbreak. To process and analyze the collected data (over 86 million words in total) effectively, we created a methodology composed of machine-learning and information-theoretic approaches. Our proposal included techniques for extracting emotions from emoticons and Internet slang, which allowed us to significantly (approximately 73%) increase the number of emotion-bearing texts needed for robust sentiment analysis of social media. As a result, we discovered a plausible explanation for the public overreaction to MERS in terms of the interplay between the disease, mass media, and public emotions.

Capturing fine-scale travel behaviors: a comparative analysis between personal activity location measurement system (PALMS) and travel diary.

BACKGROUND: Device-collected data from GPS and accelerometers for identifying active travel behaviors have dramatically changed research methods in transportation planning and public health. Automated algorithms have helped researchers to process large datasets with likely fewer errors than found in other collection methods (e.g., self-report travel diary). In this study, we compared travel modes identified by a commonly used automated algorithm (PALMS) that integrates GPS and accelerometer data with those obtained from travel diary estimates. METHODS: Sixty participants, who made 2100 trips during seven consecutive days of data collection, were selected from among the baseline sample of a project examining the travel behavior impact of a new light rail system in the greater Seattle, WA (USA) area. GPS point level analyses were first conducted to compare trip/place and travel mode detection results using contingency tables. Trip level analyses were then performed to investigate the effect of proportions of time overlap between travel logs and device-collected data on agreement rates. Global performance (with all subjects' data combined) and subject-level performance of the algorithm were compared at the trip level. RESULTS: At the GPS point level, the overall agreement rate of travel mode detection was 77.4% between PALMS and the travel diary. The agreement rate for vehicular trip detection (84.5%) was higher than for bicycling (53.5%) and walking (58.2%). At the trip level, the global performance and subject-level performance of the PALMS algorithm were 46.4% and 42.4%, respectively. Vehicular trip detection showed highest agreement rates in all analyses. Study participants' primary travel mode and car ownership were significantly related to the subject-level mode agreement rates. CONCLUSIONS: The PALMS algorithm showed moderate identification power at the GPS point level. However, trip level analyses found lower agreement rates between PALMS and travel diary data, especially for active transportation. Testing different PALMS parameter settings may serve to improve the detection of active travel and help expand PALMS's applicability in geographically different urbanized areas with a variety of travel modes.

The Antitumor Natural Compound Falcarindiol Disrupts Neural Stem Cell Homeostasis by Suppressing Notch Pathway.

Neural stem cells (NSCs) are undifferentiated, multi-potent cells that can give rise to functional neurons and glial cells. The disruption in NSC homeostasis and/or the impaired neurogenesis lead to diverse neurological diseases, including depression, dementia, and neurodegenerative disorders. Falcarindiol (FAD) is a polyacetylene found in many plants, and FAD shows the cytotoxicity against breast cancers and colon cancers. However, there is no research on the consequence of FAD treatment in normal stem cells. Here, we suggest that FAD has anticancer roles against glioblastoma cells by inducing the differentiation of glioblastoma stem-like cells, as well as activating apoptosis pathway in glioblastoma cells. On the other hand, we also show that FAD has detrimental effects by disrupting the maintenance of normal NSCs and altering the balance between self-renewal and differentiation of NSCs.

Epidemiology of drug-induced anaphylaxis in a tertiary hospital in Korea.

BACKGROUND: Epidemiology and risk factors of drug-induced anaphylaxis are difficult to estimate due to lack of confirmative diagnosis and under reporting. Here we report the current state of drug-induced anaphylaxis in Korea based on an in-hospital pharmacovigilance database in a tertiary hospital. METHODS: This study is a retrospective analysis of drug-induced anaphylaxis, reported to an in-hospital pharmacovigilance center in Seoul National University Hospital from June 2009 to May 2013. Anaphylaxis occurred in patients under 18 years of age or developed by medications administered from outside pharmacies or hospitals were excluded. We assessed causative drug, incidence per use of each drug and risk factors of fatal anaphylactic shock. RESULTS: A total of 152 in-hospital drug-induced anaphylaxis cases were reported during the study period. The single most frequently reported drug was platinum compound and the incidence of anaphylaxis and anaphylactic shock in platinum compounds users was 2.84 and 1.39 per 1000 patients use. Risk factors of anaphylactic shock among total anaphylaxis cases were identified as older age ≥70 years [Odd's ratio (OR), 5.86; 95% confidence interval (CI), 1.70-20.14]. The use of iodinated contrast media (OR, 6.19; 95% CI, 1.87-20.53) and aminosteroid neuromuscular blocking agent (NMBA) (OR, 12.82; 95% CI, 1.50-109.92) were also a risk factor for the development of anaphylactic shock. CONCLUSIONS: Platinum compounds are the most commonly reported causative agents of in-hospital drug-induced anaphylaxis. Older age ≥70 years and drugs such as iodinated contrast media and aminosteroid NMBA are related with high risk of anaphylactic shock.

Urban-rural differences in the prevalence of allergen sensitization and self-reported rhinitis in the elderly population.

BACKGROUND: Urbanization is frequently associated with allergic conditions during childhood; however, the literature lacks studies on the association between allergies and degree of urbanization in the elderly population. OBJECTIVE: To determine how the degree of urbanization affects the prevalence of allergic sensitization and self-reported rhinitis symptoms in elderly community populations. METHODS: The study population consisted of 1,311 elderly subjects identified from 2 community population cohort datasets who were divided into 3 groups according to the degree of urbanization (urban, semirural, and rural) where they resided. Current rhinitis symptoms were assessed using a questionnaire. Sensitization to inhalant allergen was measured using skin prick tests for 9 common allergens. RESULTS: Sensitization to inhalant allergen showed a positive correlation with degree of urbanization (urban 17.2%, semirural 9.8%, rural 6.0%; P for trend <.001), with a significant correlation observed between house dust mite allergens and degree of urbanization. Self-reported rhinitis symptoms were mostly nonallergic, but showed a positive correlation with degree of urbanization (urban 26.8%, semirural 18.2%, rural 11.5%; P for trend <.001). Self-reported rhinoconjunctivitis also correlated with urbanization. Correlations between self-reported allergic conditions and urbanization remained statistically significant in multivariate logistic regression tests. CONCLUSION: The present analyses found significant correlations between degree of urbanization with self-reported rhinitis symptoms and sensitization to inhalant allergen in the elderly population. These findings warrant further investigation of the roles that urban factors play in the development of elderly rhinitis and allergen sensitization.

A ferroelectric photocatalyst for enhancing hydrogen evolution: polarized particulate suspension.

A particle-based photocatalyst with a permanent internal field prepared by a corona poling method is presented as a novel approach to enhance the hydrogen evolution reaction in a particulate-suspension system. Photocatalytic activity of K0.5Na0.5NbO3 was significantly improved by 7.4 times after the polarization.

CM-GAN: A Cross-Modal Generative Adversarial Network for Imputing Completely Missing Data in Digital Industry.

Multimodal data fusion analysis is essential to model the uncertainty of environment awareness in digital industry. However, due to communication failure and cyberattack, the sampled time-series data often have the issue of data missing. In some extreme cases, part of units are unobservable for a long time, which results in complete data missing (CDM). To impute missing data, many models have been proposed. However, they cannot address the CDM issue, because no observation data of the unobservable units can be obtained in this case. Thus, to address the CDM issue, a novel cross-modal generative adversarial network (CM-GAN) is proposed in this article. It combines the cross-modal data fusion technique and the deep adversarial generation technique to construct a cross-modal data generator. This generator can generate long-term time-series data from widely existing spatio-temporal modal data in modern industrial system, and then impute missing value by replacing them with generated data. To test the performance of CM-GAN, extensive experiments are conducted on photovoltaic (PV) power output dataset. Compared with other baseline models, the performance of CM-GAN is generally better and reaches the state-of-the-art level. Moreover, sufficient ablation studies are conducted to present the contribution of the cross-modal data fusion technique and show the reasonability of parameter settings of CM-GAN. Apart from this, some prediction experiments are also conducted. The results show that the PV data recovered by CM-GAN can provide more predictability information for improving the prediction accuracy of deep learning model.

Spatio-Temporal Influence on the Distribution of Forensically Relevant Blowflies (Diptera: Calliphoridae) in Gyeongsangnam-do, South Korea.

The study of blowfly (Diptera: Calliphoridae) biodiversity and distribution is crucial for forensic investigations. Abiotic and biotic factors, such as season and habitat type, have a significant impact on blowfly populations. However, only a few forensic entomology studies have been conducted in South Korea, particularly in the Gyeongsangnam-do region. To address this, an extensive year-long survey was conducted to analyze the compositions, habitat preferences, distribution, and seasonal abundance of forensically relevant blowflies in urban and forested habitats of Gyeongsangnam-do, with sampling conducted twice a month using mouse carcass-baited traps set for 48 h each time. A total of 3470 adult blowflies were recorded, encompassing five genera and 13 species, with a noted absence of specimens during the winter months. The predominant species was Lucilia porphyrina, accounting for 37.2% of the total sample, followed by Chrysomya pinguis (27.6%), Lucilia sericata (7.6%), and Lucilia illustris (7.1%). The species composition was consistent across all surveyed regions; however, seasonal variation in species diversity was evident, with a peak in spring and a decline in summer. Notably, certain species exhibited clear preferences for either urban (Calliphora calliphoroides and L. sericata) or forested habitats (L. porphyrina and Ch. pinguis). This pioneering study elucidates the diverse blowfly communities in Gyeongsangnam-do, highlighting significant seasonal and habitat-dependent variations. These findings enrich our understanding of blowfly ecology in this region, offering valuable insights for forensic applications and underscoring the necessity for ongoing entomological surveillance and research.

Seeking a quantum advantage with trapped-ion quantum simulations of condensed-phase chemical dynamics.

Simulating the quantum dynamics of molecules in the condensed phase represents a longstanding challenge in chemistry. Trapped-ion quantum systems may serve as a platform for the analog-quantum simulation of chemical dynamics that is beyond the reach of current classical-digital simulation. To identify a 'quantum advantage' for these simulations, performance analysis of both analog-quantum simulation on noisy hardware and classical-digital algorithms is needed. In this Review, we make a comparison between a noisy analog trapped-ion simulator and a few choice classical-digital methods on simulating the dynamics of a model molecular Hamiltonian with linear vibronic coupling. We describe several simple Hamiltonians that are commonly used to model molecular systems, which can be simulated with existing or emerging trapped-ion hardware. These Hamiltonians may serve as stepping stones towards the use of trapped-ion simulators for systems beyond the reach of classical-digital methods. Finally, we identify dynamical regimes in which classical-digital simulations seem to have the weakest performance with respect to analog-quantum simulations. These regimes may provide the lowest hanging fruit to make the most of potential quantum advantages.

Angiotensin Receptor Blockers and the Risk of Suspected Drug-Induced Liver Injury: A Retrospective Cohort Study Using Electronic Health Record-Based Common Data Model in South Korea.

INTRODUCTION: Angiotensin receptor blockers are widely used antihypertensive drugs in South Korea. In 2021, the Korea Ministry of Food and Drug Safety acknowledged the need for national compensation for a drug-induced liver injury (DILI) after azilsartan use. However, little is known regarding the association between angiotensin receptor blockers and DILI. OBJECTIVE: We conducted a retrospective cohort study in incident users of angiotensin receptor blockers from a common data model database (1 January, 2017-31 December, 2021) to compare the risk of DILI among specific angiotensin receptor blockers against valsartan. METHODS: Patients were assigned to treatment groups at cohort entry based on prescribed angiotensin receptor blockers. Drug-induced liver injury was operationally defined using the International DILI Expert Working Group criteria. Cox regression analyses were conducted to derive hazard ratios and the inverse probability of treatment weighting method was applied. All analyses were performed using R. RESULTS: In total, 229,881 angiotensin receptor blocker users from 20 university hospitals were included. Crude DILI incidence ranged from 15.6 to 82.8 per 1000 person-years in treatment groups, most were cholestatic and of mild severity. Overall, the risk of DILI was significantly lower in olmesartan users than in valsartan users (hazard ratio: 0.73 [95% confidence interval 0.55-0.96]). In monotherapy patients, the risk was significantly higher in azilsartan users than in valsartan users (hazard ratio: 6.55 [95% confidence interval 5.28-8.12]). CONCLUSIONS: We found a significantly higher risk of suspected DILI in patients receiving azilsartan monotherapy compared with valsartan monotherapy. Our findings emphasize the utility of real-world evidence in advancing our understanding of adverse drug reactions in clinical practice.