Antimicrobials associated adverse drug reaction profiling: a four years retrospective study (Pharmacovigilance study)

Epidemiology of reported serious adverse drug reactions due to anti-infectives using nationwide database of Thailand

Serious Adverse Drug Reactions (ADRs) can cause a longer stay, which can result in fatal outcomes. Understanding the prognostic factors for the serious ADRs play a vital role in developing appropriate serious ADR prevention strategies. This study aimed to analyze nationwide database in Thailand to identify predisposing factors associated with the serious ADRs, explore drug exposure, distribution of serious ADRs, types of ADRs, and classify the determinants of serious ADR due to anti-infective in Thailand. The national database of anti-infective-induced ADRs from January 2012 to December 2021 in Thailand's 77 provinces, Thai Vigibase at the Health Product Vigilance Center (HPVC), was considered. After pre-processing, frequencies and percentages were used to investigate the distribution of ADR seriousness. To determine the significance of the independent variables on the seriousness of anti-infective-induced ADRs, logistic regression and the Classification and Regression Tree (CART) model were performed. A p-value < 0.05 was considered statistically significant. A total of 82,333 ADR cases, of which 20,692 were serious ADRs (25.13%). Serious ADRs is statistically associated with region, gender, ethnicity, age, type of patient, history of drug allergy, chronic disease and dose frequency (p-value < 0.001). The most commonly reported serious ADRs were in the South region of Thailand (OR = 1.92, 95% CI = 1.88-1.97), followed by the North region (OR = 1.68, 95% CI = 1.64-1.71) of Thailand. Gender and history of drug allergy were also statistically associated with the seriousness of ADRs (p-value = 0.001). Reported ADRs revealed that patients were males (OR =  1.11, 95% CI = 1.11-1.13) and those with a prior history of drug allergy (OR = 1.22, 95% CI = 1.20-1.24) were more likely to experience serious ADRs. The risk of having an ADR reported as serious was significantly higher in patients aged 60 and over (OR = 1.42, 95% CI = 1.39-1.46) and patients aged 40-59 years (OR = 1.34, 95% CI = 1.31-1.37) compared to patients aged 0-19 years. IPD patients most commonly associated with serious ADRs. The results of this study will enable healthcare professionals to use caution when prescribing to those groups. Furthermore, developing a reporting system to reduce serious ADR evidence, such as software with electronic prescribing databases or applications that enable efficient detection of ADRs in high-risk groups, was critical in order to closely monitor and improve patient safety.

A Knowledge Graph Embedding Based Approach to Predict the Adverse Drug Reactions Using a Deep Neural Network

Recently Artificial Intelligence(AI) has not only been used to diagnose the disease but also to cure the disease. Researchers started using AI for drug discovery. Predicting the Adverse Drug Reactions(ADRs) caused by the drug in the manufacturing stage or in the clinical trial stage is a very important problem in drug discovery. ADRs have become a major concern resulting in injuries and also becoming fatal sometimes. Drug safety has gained much importance over the years propelling to the forefront investigation of predicting the ADRs. Although prior studies have queried diverse approaches to predict ADRs, very few were found to be effective. Also, the problem of having fewer reports makes the prediction of ADRs more difficult. To tackle this problem effectively, a novel method has been proposed in this paper. The proposed method is based on Knowledge Graph(KG) embedding. Using the KG embedding, we designed and trained a custom-made Deep Neural Network(DNN) called KGDNN(Knowledge Graph DNN) for predicting the ADRs. A KG has been constructed with 6 types of entities: drugs, ADRs, target proteins, indications, pathways, and genes. Using the Node2Vec algorithm, each node has been embedded into a feature space. Using those embeddings, the ADRs are classified by the KGDNN model. The proposed method has obtained an AUROC score of 0.917 and significantly outperformed the existing methods. Two case studies on drugs causing liver injury and COVID-19 recommended drugs have been performed to illustrate the model efficacy.