Pub Date : 2025-03-04DOI: 10.1007/s40264-025-01530-z
Ahmed Al-Imam, Riccardo Lora, Marek A Motyka, Erica Marletta, Michele Vezzaro, Jerzy Moczko, Manal Younus, Michal Michalak
Background: Psychedelics are gaining attention for their therapeutic potential in modern and personalized medicine. Online forums such as Erowid provide valuable user insights, but analyses of these experiences using natural language processing (NLP) remain scarce.
Objective: This study aims to utilize NLP, including sentiment and lexicon analysis, to examine user-generated experience reports on psilocybin-containing mushrooms and LSD from the Erowid forum.
Methods: Data from 2188 Erowid users (1161 psilocybin mushrooms and 1027 LSD) was collected via automated web scraping with XPath, CSS selectors, and Selenium WebDriver. The dataset included report titles, substances, and demographics. Sentiment analysis utilized BERT, RoBERTa, and VADER models. Preprocessing involved tokenization, lemmatization, part-of-speech tagging, and stop-word filtering. Lexicon analysis identified themes through recurring n-grams, visualized using Python.
Results: User demographics revealed comparable ages for psilocybin mushrooms (23.8 ± 0.9 years) and LSD users (20.0 ± 0.6 years), with a predominance of male users. The BERT model predominantly labeled experiences as negative (unfavorable), particularly for mushroom users (p = 0.001). VADER indicated more positive experiences for mushroom users (p < 0.001), while RoBERTa mainly classified experiences as negative or neutral. Significant gender differences were found only with VADER, where more male users expressed positive opinions about psilocybin mushrooms (74.09% versus 65.52%, p < 0.021). The VADER model yielded more polarized results, whereas RoBERTa's cautious classifications indicate its suitability for analyzing lengthy and complex psychedelic reports. Further, RoBERTa outperformed other transformer-based models, achieving the highest accuracy. Lexicon analysis revealed emotional, sensory, and temporal themes, with psilocybin reports emphasizing introspection and time dilation phenomenon, while LSD reports highlighted memory issues and cognitive disorientation.
Conclusions: Sentiment analysis showed that VADER produced more polarized results, while RoBERTa offered cautious classifications with the highest accuracy. Lexicon analysis revealed shared themes, with mushroom reports focusing on introspection and time dilation perception, while those of LSD emphasized cognitive disturbances. This study highlights the value of these analyses in understanding psychedelic experiences, informing harm reduction, and guiding policy-making.
{"title":"Opinion Mining of Erowid's Experience Reports on LSD and Psilocybin-Containing Mushrooms.","authors":"Ahmed Al-Imam, Riccardo Lora, Marek A Motyka, Erica Marletta, Michele Vezzaro, Jerzy Moczko, Manal Younus, Michal Michalak","doi":"10.1007/s40264-025-01530-z","DOIUrl":"https://doi.org/10.1007/s40264-025-01530-z","url":null,"abstract":"<p><strong>Background: </strong>Psychedelics are gaining attention for their therapeutic potential in modern and personalized medicine. Online forums such as Erowid provide valuable user insights, but analyses of these experiences using natural language processing (NLP) remain scarce.</p><p><strong>Objective: </strong>This study aims to utilize NLP, including sentiment and lexicon analysis, to examine user-generated experience reports on psilocybin-containing mushrooms and LSD from the Erowid forum.</p><p><strong>Methods: </strong>Data from 2188 Erowid users (1161 psilocybin mushrooms and 1027 LSD) was collected via automated web scraping with XPath, CSS selectors, and Selenium WebDriver. The dataset included report titles, substances, and demographics. Sentiment analysis utilized BERT, RoBERTa, and VADER models. Preprocessing involved tokenization, lemmatization, part-of-speech tagging, and stop-word filtering. Lexicon analysis identified themes through recurring n-grams, visualized using Python.</p><p><strong>Results: </strong>User demographics revealed comparable ages for psilocybin mushrooms (23.8 ± 0.9 years) and LSD users (20.0 ± 0.6 years), with a predominance of male users. The BERT model predominantly labeled experiences as negative (unfavorable), particularly for mushroom users (p = 0.001). VADER indicated more positive experiences for mushroom users (p < 0.001), while RoBERTa mainly classified experiences as negative or neutral. Significant gender differences were found only with VADER, where more male users expressed positive opinions about psilocybin mushrooms (74.09% versus 65.52%, p < 0.021). The VADER model yielded more polarized results, whereas RoBERTa's cautious classifications indicate its suitability for analyzing lengthy and complex psychedelic reports. Further, RoBERTa outperformed other transformer-based models, achieving the highest accuracy. Lexicon analysis revealed emotional, sensory, and temporal themes, with psilocybin reports emphasizing introspection and time dilation phenomenon, while LSD reports highlighted memory issues and cognitive disorientation.</p><p><strong>Conclusions: </strong>Sentiment analysis showed that VADER produced more polarized results, while RoBERTa offered cautious classifications with the highest accuracy. Lexicon analysis revealed shared themes, with mushroom reports focusing on introspection and time dilation perception, while those of LSD emphasized cognitive disturbances. This study highlights the value of these analyses in understanding psychedelic experiences, informing harm reduction, and guiding policy-making.</p>","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143540563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-03DOI: 10.1007/s40264-025-01532-x
Minh-Hoang Tran, Kim-Huong Truong-Nguyen
{"title":"Comment on: \"The Use of Multiple Medications During Pregnancy Among an Ethnically Diverse Population in South-Eastern Melbourne: A Retrospective Analysis to Explore Potential Risks and Complications\".","authors":"Minh-Hoang Tran, Kim-Huong Truong-Nguyen","doi":"10.1007/s40264-025-01532-x","DOIUrl":"https://doi.org/10.1007/s40264-025-01532-x","url":null,"abstract":"","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143540562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-03DOI: 10.1007/s40264-025-01534-9
Yitayeh Belsti, Aya Mousa, Hannah Jackson, Lisa J Moran, Kirsten R Palmer, Raja Ram Dhungana, Emily Callander, Daniel Lorber Rolnik, Helena Teede, Joanne Enticott
{"title":"Authors' response to Tran et al.'s comment on \"The Use of Multiple Medications During Pregnancy Among an Ethnically Diverse Population in South-Eastern Melbourne: A Retrospective Analysis to Explore Potential Risks and Complications\".","authors":"Yitayeh Belsti, Aya Mousa, Hannah Jackson, Lisa J Moran, Kirsten R Palmer, Raja Ram Dhungana, Emily Callander, Daniel Lorber Rolnik, Helena Teede, Joanne Enticott","doi":"10.1007/s40264-025-01534-9","DOIUrl":"https://doi.org/10.1007/s40264-025-01534-9","url":null,"abstract":"","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":""},"PeriodicalIF":4.0,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143540561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2024-11-08DOI: 10.1007/s40264-024-01491-9
Sara Ferraro, Emiliano Cappello, Marco Fornili, Irma Convertino, Marco Bonaso, Ersilia Lucenteforte, Marco Tuccori
<p><strong>Background: </strong>In 2018, the European Medicines Agency issued some risk minimisation measures related to unresolved adverse drug reactions (ADRs) reported for fluoroquinolones, including sensory ADRs. Spontaneous reporting databases frequently report unresolved outcomes for gustatory, olfactory and auditory (GOA) ADRs. However, such a high volume of unresolved GOA ADRs could reflect an under-investigated clinical issue or an intrinsic difficulty in the outcome assessment.</p><p><strong>Objectives: </strong>The objectives of the study were: (1) to investigate whether unresolved outcomes are reported more frequently for GOA ADRs than for other ADRs to systemic antibiotics and (2) to identify possible signals of unresolved GOA ADRs for systemic antibiotics.</p><p><strong>Methods: </strong>We used the EudraVigilance database to extract the number of ADRs to systemic antibiotics of the Anatomical Therapeutic Chemical class J01 up to February 2019. We classified ADRs in "non-GOA ADRs" and "GOA ADRs". Adverse drug reactions were categorised in three groups according to the outcome: defined, persistent/permanent (unresolved) and undetermined ADRs. We performed disproportionality analyses with the case/non-case methodology, by calculating the crude reporting odds ratio (ROR) and 95% confidence interval (CI). Cases were all persistent/permanent ADRs, and non-cases were defined and undetermined ADRs. For the first objective, index groups were gustatory or olfactory or auditory ADRs, while reference group included all non-GOA ADRs. For the second objective, we performed a disproportionality analysis by using the sub-set of GOA ADRs. Index and reference groups varied with subgroups of drugs and drug class, so that each drug and drug class was compared with the others. We conducted two sensitivity analyses for each analysis by varying the case definition.</p><p><strong>Results: </strong>We extracted 748,798 ADRs, including 10,770 GOA ADRs. The first analysis showed that GOA ADRs were reported more frequently as unresolved events compared with all other ADRs (ROR: 2.68 95% CI 2.51-2.85; ROR: 5.20 95% CI 4.66-5.81; and ROR: 2.64 (95% CI 2.51-2.79, respectively). Gustatory ADRs were reported more frequently as unresolved for doxycycline (ROR: 1.69, 95% CI 1.18-2.41, p = 0.0038), azithromycin (ROR: 2.07, 95% CI 1.58-2.72, p < 0.0001) and levofloxacin (ROR: 1.59, 95% CI 1.22-2.07, p < 0.001) compared with GOA ADRs of all other antibiotics. Olfactory ADRs were reported more frequently as unresolved for doxycycline (ROR: 2.4, 95% CI 1.26-4.58, p = 0.0078) and levofloxacin (ROR: 1.92, 95% CI 1.28-2.86, p = 0.0014). Auditory ADRs were reported more frequently as unresolved for doxycycline (ROR: 1.52, 95% CI 1.09-2.12, p = 0.013) and clarithromycin (ROR: 1.31, 95% CI 1.09-1.59, p = 0.0049).</p><p><strong>Conclusions: </strong>We tested and used an appropriate expected frequency standard, which allows us to identify possible signals of unresolved GOA ADRs
背景:2018 年,欧洲药品管理局发布了一些与氟喹诺酮类药物(包括感官 ADR)报告的未解决药物不良反应(ADR)相关的风险最小化措施。自发报告数据库经常报告未解决的味觉、嗅觉和听觉(GOA)ADR 结果。然而,大量未解决的味觉、嗅觉和听觉不良反应可能反映出临床问题未得到充分调查或结果评估存在内在困难:本研究的目的是研究目的:(1)调查与全身用抗生素的其他不良反应相比,GOA ADR 的未解决结果报告是否更频繁;(2)确定全身用抗生素的 GOA ADR 未解决的可能信号:我们使用 EudraVigilance 数据库提取了截至 2019 年 2 月的解剖治疗化学类 J01 全身用抗生素的 ADR 数量。我们将ADR分为 "非GOA ADR "和 "GOA ADR"。药物不良反应根据结果分为三类:已确定、持续/永久(未解决)和未确定的 ADR。我们采用病例/非病例方法进行了比例失调分析,计算了粗报告几率比(ROR)和 95% 置信区间(CI)。病例是指所有持续性/永久性 ADR,非病例是指已确定和未确定的 ADR。在第一个目标中,指标组为味觉、嗅觉或听觉 ADR,参照组包括所有非 GOA ADR。对于第二个目标,我们使用全球海洋观测系统 ADR 子集进行了比例失调分析。指数组和参照组随着药物和药物类别的分组而变化,因此每种药物和药物类别都与其他药物和药物类别进行了比较。通过改变病例定义,我们对每项分析进行了两次敏感性分析:我们提取了 748,798 例 ADR,其中包括 10,770 例 GOA ADR。第一项分析表明,与所有其他 ADR 相比,GOA ADR 作为未解决事件报告的频率更高(ROR:2.68 95% CI 2.51-2.85;ROR:5.20 95% CI 4.66-5.81;ROR:2.64 (95% CI 2.51-2.79,分别为 2.68、5.20 和 2.64)。与所有其他抗生素的GOA ADR相比,多西环素(ROR:1.69,95% CI 1.18-2.41,p = 0.0038)、阿奇霉素(ROR:2.07,95% CI 1.58-2.72,p < 0.0001)和左氧氟沙星(ROR:1.59,95% CI 1.22-2.07,p < 0.001)的口腔ADR未解决的报告频率更高。多西环素(ROR:2.4,95% CI 1.26-4.58,p = 0.0078)和左氧氟沙星(ROR:1.92,95% CI 1.28-2.86,p = 0.0014)的嗅觉 ADR 更常被报告为未解决。多西环素(ROR:1.52,95% CI 1.09-2.12,p = 0.013)和克拉霉素(ROR:1.31,95% CI 1.09-1.59,p = 0.0049)的听觉 ADR 更常被报告为未解决:我们测试并使用了一种适当的预期频率标准,该标准使我们能够识别 EudraVigilance 数据库中可能存在的抗生素药物未解决的 GOA ADR 信号。这种方法可用于生成其他药物或不良反应的持续性甚至不可逆事件信号。不过,这些信号必须通过全面的临床评估加以确认。
{"title":"Signals of Possibly Persistent Gustatory, Olfactory and Auditory Adverse Drug Reactions to Antibiotic Drugs: A Disproportionality Analysis Using the EudraVigilance Database.","authors":"Sara Ferraro, Emiliano Cappello, Marco Fornili, Irma Convertino, Marco Bonaso, Ersilia Lucenteforte, Marco Tuccori","doi":"10.1007/s40264-024-01491-9","DOIUrl":"10.1007/s40264-024-01491-9","url":null,"abstract":"<p><strong>Background: </strong>In 2018, the European Medicines Agency issued some risk minimisation measures related to unresolved adverse drug reactions (ADRs) reported for fluoroquinolones, including sensory ADRs. Spontaneous reporting databases frequently report unresolved outcomes for gustatory, olfactory and auditory (GOA) ADRs. However, such a high volume of unresolved GOA ADRs could reflect an under-investigated clinical issue or an intrinsic difficulty in the outcome assessment.</p><p><strong>Objectives: </strong>The objectives of the study were: (1) to investigate whether unresolved outcomes are reported more frequently for GOA ADRs than for other ADRs to systemic antibiotics and (2) to identify possible signals of unresolved GOA ADRs for systemic antibiotics.</p><p><strong>Methods: </strong>We used the EudraVigilance database to extract the number of ADRs to systemic antibiotics of the Anatomical Therapeutic Chemical class J01 up to February 2019. We classified ADRs in \"non-GOA ADRs\" and \"GOA ADRs\". Adverse drug reactions were categorised in three groups according to the outcome: defined, persistent/permanent (unresolved) and undetermined ADRs. We performed disproportionality analyses with the case/non-case methodology, by calculating the crude reporting odds ratio (ROR) and 95% confidence interval (CI). Cases were all persistent/permanent ADRs, and non-cases were defined and undetermined ADRs. For the first objective, index groups were gustatory or olfactory or auditory ADRs, while reference group included all non-GOA ADRs. For the second objective, we performed a disproportionality analysis by using the sub-set of GOA ADRs. Index and reference groups varied with subgroups of drugs and drug class, so that each drug and drug class was compared with the others. We conducted two sensitivity analyses for each analysis by varying the case definition.</p><p><strong>Results: </strong>We extracted 748,798 ADRs, including 10,770 GOA ADRs. The first analysis showed that GOA ADRs were reported more frequently as unresolved events compared with all other ADRs (ROR: 2.68 95% CI 2.51-2.85; ROR: 5.20 95% CI 4.66-5.81; and ROR: 2.64 (95% CI 2.51-2.79, respectively). Gustatory ADRs were reported more frequently as unresolved for doxycycline (ROR: 1.69, 95% CI 1.18-2.41, p = 0.0038), azithromycin (ROR: 2.07, 95% CI 1.58-2.72, p < 0.0001) and levofloxacin (ROR: 1.59, 95% CI 1.22-2.07, p < 0.001) compared with GOA ADRs of all other antibiotics. Olfactory ADRs were reported more frequently as unresolved for doxycycline (ROR: 2.4, 95% CI 1.26-4.58, p = 0.0078) and levofloxacin (ROR: 1.92, 95% CI 1.28-2.86, p = 0.0014). Auditory ADRs were reported more frequently as unresolved for doxycycline (ROR: 1.52, 95% CI 1.09-2.12, p = 0.013) and clarithromycin (ROR: 1.31, 95% CI 1.09-1.59, p = 0.0049).</p><p><strong>Conclusions: </strong>We tested and used an appropriate expected frequency standard, which allows us to identify possible signals of unresolved GOA ADRs","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"217-231"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142603784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2024-12-18DOI: 10.1007/s40264-024-01507-4
Omar Aimer, Maribel Salas, Tarek A Hammad, Rania Mouchantaf, Katarina Ilic, Maxine Gossell-Williams, Ushma Mehta, Robert W Platt, Bruce Carleton
{"title":"Advances in Drug Safety: Highlights from the 23rd Annual Meeting of the International Society of Pharmacovigilance.","authors":"Omar Aimer, Maribel Salas, Tarek A Hammad, Rania Mouchantaf, Katarina Ilic, Maxine Gossell-Williams, Ushma Mehta, Robert W Platt, Bruce Carleton","doi":"10.1007/s40264-024-01507-4","DOIUrl":"10.1007/s40264-024-01507-4","url":null,"abstract":"","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"305-309"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142845957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2025-01-08DOI: 10.1007/s40264-024-01513-6
Mohamed A Elhawary, Rebecca Noss, Loubna Alj, Manal Younus, Mayada Alkhakany, Hadir Rostom, Angela Caro-Rojas, Thamir M Alshammari
{"title":"Pharmacovigilance in the Community: A Special-Interest Group of the International Society of Pharmacovigilance.","authors":"Mohamed A Elhawary, Rebecca Noss, Loubna Alj, Manal Younus, Mayada Alkhakany, Hadir Rostom, Angela Caro-Rojas, Thamir M Alshammari","doi":"10.1007/s40264-024-01513-6","DOIUrl":"10.1007/s40264-024-01513-6","url":null,"abstract":"","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"203-207"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142946484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2024-12-11DOI: 10.1007/s40264-024-01499-1
Jürgen Dietrich, André Hollstein
Introduction: Recent artificial intelligence (AI) advances can generate human-like responses to a wide range of queries, making them a useful tool for healthcare applications. Therefore, the potential use of large language models (LLMs) in controlled environments regarding efficacy, reproducibility, and operability will be of paramount interest.
Objective: We investigated if and how GPT 3.5 and GPT 4 models can be directly used as a part of a GxP validated system and compared the performance of externally hosted GPT 3.5 and GPT 4 against LLMs, which can be hosted internally. We explored zero-shot LLM performance for named entity recognition (NER) and relation extraction tasks, investigated which LLM has the best zero-shot performance to be used potentially for generating training data proposals, evaluated the LLM performance of seven entities for medical NER in zero-shot experiments, selected one model for further performance improvement (few-shot and fine-tuning: Zephyr-7b-beta), and investigated how smaller open-source LLMs perform in contrast to GPT models and to a small fine-tuned T5 Base.
Methods: We performed reproducibility experiments to evaluate if LLMs can be used in controlled environments and utilized guided generation to use the same prompt across multiple models. Few-shot learning and quantized low rank adapter (QLoRA) fine-tuning were applied to further improve LLM performance.
Results and conclusion: We demonstrated that zero-shot GPT 4 performance is comparable with a fine-tuned T5, and Zephyr performed better than zero-shot GPT 3.5, but the recognition of product combinations such as product event combination was significantly better by using a fine-tuned T5. Although Open AI launched recently GPT versions to improve the generation of consistent output, both GPT variants failed to demonstrate reproducible results. The lack of reproducibility together with limitations of external hosted systems to keep validated systems in a state of control may affect the use of closed and proprietary models in regulated environments. However, due to the good NER performance, we recommend using GPT for creating annotation proposals for training data as a basis for fine-tuning.
{"title":"Performance and Reproducibility of Large Language Models in Named Entity Recognition: Considerations for the Use in Controlled Environments.","authors":"Jürgen Dietrich, André Hollstein","doi":"10.1007/s40264-024-01499-1","DOIUrl":"10.1007/s40264-024-01499-1","url":null,"abstract":"<p><strong>Introduction: </strong>Recent artificial intelligence (AI) advances can generate human-like responses to a wide range of queries, making them a useful tool for healthcare applications. Therefore, the potential use of large language models (LLMs) in controlled environments regarding efficacy, reproducibility, and operability will be of paramount interest.</p><p><strong>Objective: </strong>We investigated if and how GPT 3.5 and GPT 4 models can be directly used as a part of a GxP validated system and compared the performance of externally hosted GPT 3.5 and GPT 4 against LLMs, which can be hosted internally. We explored zero-shot LLM performance for named entity recognition (NER) and relation extraction tasks, investigated which LLM has the best zero-shot performance to be used potentially for generating training data proposals, evaluated the LLM performance of seven entities for medical NER in zero-shot experiments, selected one model for further performance improvement (few-shot and fine-tuning: Zephyr-7b-beta), and investigated how smaller open-source LLMs perform in contrast to GPT models and to a small fine-tuned T5 Base.</p><p><strong>Methods: </strong>We performed reproducibility experiments to evaluate if LLMs can be used in controlled environments and utilized guided generation to use the same prompt across multiple models. Few-shot learning and quantized low rank adapter (QLoRA) fine-tuning were applied to further improve LLM performance.</p><p><strong>Results and conclusion: </strong>We demonstrated that zero-shot GPT 4 performance is comparable with a fine-tuned T5, and Zephyr performed better than zero-shot GPT 3.5, but the recognition of product combinations such as product event combination was significantly better by using a fine-tuned T5. Although Open AI launched recently GPT versions to improve the generation of consistent output, both GPT variants failed to demonstrate reproducible results. The lack of reproducibility together with limitations of external hosted systems to keep validated systems in a state of control may affect the use of closed and proprietary models in regulated environments. However, due to the good NER performance, we recommend using GPT for creating annotation proposals for training data as a basis for fine-tuning.</p>","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"287-303"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11829833/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142806427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2025-01-22DOI: 10.1007/s40264-024-01493-7
Victoria Prudence Nambasa, Hannah May Gunter, Modupe Bamidele Adeyemo, Neetesh Yanish Bhawaneedin, Marc Blockman, George Tsey Sabblah, John Owusu Gyapong, Eric Muriithi Guantai, Tamrat Abebe, Workeabeba Abebe, Henry Jeremy Lawson, Mercedes Chawada Leburu, Abdullahi Mohammed, Kwame Amponsa-Achiano, Mafora Florah Matlala, Uchenna Geraldine Elemuwa, Hudu Mogtari, Alexander Kwadwo Nyarko, Marione Schönfeldt, Mercy Kamupira, Kerrigan McCarthy, Yohannes Lakew Tefera, Asnakech Alemu, Kabir Mawashi Yusuf, Obi Emelife, Ladji Sidibe, Kudakwashe Dandajena, Kenneth Onu, Mojisola Christianah Adeyeye, Delese Mimi Darko, Heran Gerba, Boitumelo Semete, Fred Siyoi, Aggrey Ambali, Johanna Catharina Meyer
Introduction: The COVID-19 pandemic accelerated new vaccine development. Limited safety data necessitated robust global safety surveillance to accurately identify and promptly communicate potential safety issues. The African Union Smart Safety Surveillance (AU-3S) program established the Joint Signal Management (JSM) group to support identification of potential vaccine safety concerns in five pilot countries (Ethiopia, Ghana, Kenya, Nigeria, South Africa), accounting for approximately 35% of the African population.
Objective: Our objective was to provide an overview of the JSM group's role in supporting signal management activities for the AU-3S program during the COVID-19 pandemic.
Methods: Spontaneous, electronically reported COVID-19 vaccine adverse events following immunization (AEFI) from each country's safety data were integrated into the interim Data Integration and Signal Detection system. Statistical disproportionality methods were used to identify and review vaccine-event combinations (VECs) for potential safety concerns. The JSM group-which comprised pharmacovigilance and subject matter experts from National Medicine Regulatory Authorities, Expanded Programs on Immunization, and vaccine safety committees-conducted signal detection activities on cross-country safety data and provided recommendations.
Results: From April 2021 to December 2023, a total of 48,294 spontaneously reported AEFI were analyzed for six COVID-19 vaccines (NRVV Ad [ChAdOx1 nCoV-19]; Ad26.COV2.S; Elasomeran; Tozinameran; Covid-19 vaccine [Vero Cell], Inactivated; NRVV Ad26 [Gam-Covid-Vac]) administered in Ethiopia (34.6%), Nigeria (30.3%), South Africa (16.9%), Ghana (13.5%), and Kenya (4.7%). Overall, 2,742 VECs were validated. A causal association between the COVID-19 vaccines and the reported AEFI cannot be inferred, as data were reported spontaneously. JSM group recommendations included monitoring for further evidence, no immediate action required, engaging marketing authorization holder(s) for additional information, or sensitizing healthcare providers and/or the public about events. Although no new safety signals were identified, nine safety-related recommendations were issued, including patient and healthcare provider education.
Conclusions: The JSM group established a scalable and replicable model for future signal management of other priority health products in low- and middle-income countries, fostering ongoing collaboration and capacity building. Knowledge and experience gained from this pilot initiative will guide stakeholders in future safety surveillance initiatives within the African continent.
{"title":"Empowering African Expertise: Enhancing Safety Data Integration and Signal Detection for COVID-19 Vaccines Through the African Union Smart Safety Surveillance Joint Signal Management Group.","authors":"Victoria Prudence Nambasa, Hannah May Gunter, Modupe Bamidele Adeyemo, Neetesh Yanish Bhawaneedin, Marc Blockman, George Tsey Sabblah, John Owusu Gyapong, Eric Muriithi Guantai, Tamrat Abebe, Workeabeba Abebe, Henry Jeremy Lawson, Mercedes Chawada Leburu, Abdullahi Mohammed, Kwame Amponsa-Achiano, Mafora Florah Matlala, Uchenna Geraldine Elemuwa, Hudu Mogtari, Alexander Kwadwo Nyarko, Marione Schönfeldt, Mercy Kamupira, Kerrigan McCarthy, Yohannes Lakew Tefera, Asnakech Alemu, Kabir Mawashi Yusuf, Obi Emelife, Ladji Sidibe, Kudakwashe Dandajena, Kenneth Onu, Mojisola Christianah Adeyeye, Delese Mimi Darko, Heran Gerba, Boitumelo Semete, Fred Siyoi, Aggrey Ambali, Johanna Catharina Meyer","doi":"10.1007/s40264-024-01493-7","DOIUrl":"10.1007/s40264-024-01493-7","url":null,"abstract":"<p><strong>Introduction: </strong>The COVID-19 pandemic accelerated new vaccine development. Limited safety data necessitated robust global safety surveillance to accurately identify and promptly communicate potential safety issues. The African Union Smart Safety Surveillance (AU-3S) program established the Joint Signal Management (JSM) group to support identification of potential vaccine safety concerns in five pilot countries (Ethiopia, Ghana, Kenya, Nigeria, South Africa), accounting for approximately 35% of the African population.</p><p><strong>Objective: </strong>Our objective was to provide an overview of the JSM group's role in supporting signal management activities for the AU-3S program during the COVID-19 pandemic.</p><p><strong>Methods: </strong>Spontaneous, electronically reported COVID-19 vaccine adverse events following immunization (AEFI) from each country's safety data were integrated into the interim Data Integration and Signal Detection system. Statistical disproportionality methods were used to identify and review vaccine-event combinations (VECs) for potential safety concerns. The JSM group-which comprised pharmacovigilance and subject matter experts from National Medicine Regulatory Authorities, Expanded Programs on Immunization, and vaccine safety committees-conducted signal detection activities on cross-country safety data and provided recommendations.</p><p><strong>Results: </strong>From April 2021 to December 2023, a total of 48,294 spontaneously reported AEFI were analyzed for six COVID-19 vaccines (NRVV Ad [ChAdOx1 nCoV-19]; Ad26.COV2.S; Elasomeran; Tozinameran; Covid-19 vaccine [Vero Cell], Inactivated; NRVV Ad26 [Gam-Covid-Vac]) administered in Ethiopia (34.6%), Nigeria (30.3%), South Africa (16.9%), Ghana (13.5%), and Kenya (4.7%). Overall, 2,742 VECs were validated. A causal association between the COVID-19 vaccines and the reported AEFI cannot be inferred, as data were reported spontaneously. JSM group recommendations included monitoring for further evidence, no immediate action required, engaging marketing authorization holder(s) for additional information, or sensitizing healthcare providers and/or the public about events. Although no new safety signals were identified, nine safety-related recommendations were issued, including patient and healthcare provider education.</p><p><strong>Conclusions: </strong>The JSM group established a scalable and replicable model for future signal management of other priority health products in low- and middle-income countries, fostering ongoing collaboration and capacity building. Knowledge and experience gained from this pilot initiative will guide stakeholders in future safety surveillance initiatives within the African continent.</p>","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"233-249"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11829835/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143022752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-01Epub Date: 2024-12-12DOI: 10.1007/s40264-024-01498-2
Mayra Oliveira, Paige Marquez, Carol Ennulat, Phillip Blanc, Kerry Welsh, Narayan Nair, Monica Taminato, Pedro L Moro
Background: On June 8, 2021, a new 20-valent pneumococcal conjugate vaccine (PCV20, PREVNAR 20®, Pfizer, Inc.) was licensed for use in adults aged ≥ 18 years by the US Food and Drug Administration (FDA).
Objective: To describe reports to the Vaccine Adverse Event Reporting System (VAERS) after administration of the 20-valent pneumococcal conjugate vaccine in adults.
Methods: We searched the VAERS for reports of adverse events involving persons aged ≥ 19 years who received PCV20 during October 20, 2021, through December 31, 2023. Our evaluation included automated analysis of reports, clinical review of serious reports and pre-specified events of special interest, empirical Bayesian data mining to assess for disproportionate reporting, and estimation of reporting rates for reports of Guillain-Barré syndrome (GBS).
Results: The VAERS received 1976 reports after PCV20 administration in persons aged ≥ 19 years (6% of reports involved serious events). The most common adverse events among persons aged 19-64 years (n = 798) were injection-site reactions (231, 29%), pain (134, 17%), erythema (118, 15%), and fever (117, 15%). For persons aged ≥ 65 years (n = 1178), the most common adverse events were injection-site reactions (417, 35%), pain (180, 15%), pain in extremity (162, 14%), and erythema (158, 13%). A data mining alert (EB05 = 3.812) for the MedDRA Preferred Term "Guillain-Barre syndrome" was observed for serious reports. Clinical review verified 11 of 20 GBS reports; 7/11 vaccine recipients were aged ≥ 65 years. Among the 11 verified cases, the median time from vaccination to symptom onset was 14 days. Five persons received another vaccine on the same visit. The reporting rate of GBS after PCV20 receipt was 0.5 cases per million doses distributed. No other safety concern was identified.
Conclusions: During the period of this post-licensure review of PCV20, we found most reports were non-serious and comprised mostly local and systemic (e.g., fever) reactions consistent with prelicensure studies. In serious reports, we also identified a data mining alert for GBS after receipt of PCV20, which Centers for Disease Control and Prevention and the FDA are investigating further. No other new or unexpected safety concern was identified.
{"title":"Post-licensure Safety Surveillance of 20-Valent Pneumococcal Conjugate Vaccine (PCV20) Among US Adults in the Vaccine Adverse Event Reporting System (VAERS).","authors":"Mayra Oliveira, Paige Marquez, Carol Ennulat, Phillip Blanc, Kerry Welsh, Narayan Nair, Monica Taminato, Pedro L Moro","doi":"10.1007/s40264-024-01498-2","DOIUrl":"10.1007/s40264-024-01498-2","url":null,"abstract":"<p><strong>Background: </strong>On June 8, 2021, a new 20-valent pneumococcal conjugate vaccine (PCV20, PREVNAR 20<sup>®</sup>, Pfizer, Inc.) was licensed for use in adults aged ≥ 18 years by the US Food and Drug Administration (FDA).</p><p><strong>Objective: </strong>To describe reports to the Vaccine Adverse Event Reporting System (VAERS) after administration of the 20-valent pneumococcal conjugate vaccine in adults.</p><p><strong>Methods: </strong>We searched the VAERS for reports of adverse events involving persons aged ≥ 19 years who received PCV20 during October 20, 2021, through December 31, 2023. Our evaluation included automated analysis of reports, clinical review of serious reports and pre-specified events of special interest, empirical Bayesian data mining to assess for disproportionate reporting, and estimation of reporting rates for reports of Guillain-Barré syndrome (GBS).</p><p><strong>Results: </strong>The VAERS received 1976 reports after PCV20 administration in persons aged ≥ 19 years (6% of reports involved serious events). The most common adverse events among persons aged 19-64 years (n = 798) were injection-site reactions (231, 29%), pain (134, 17%), erythema (118, 15%), and fever (117, 15%). For persons aged ≥ 65 years (n = 1178), the most common adverse events were injection-site reactions (417, 35%), pain (180, 15%), pain in extremity (162, 14%), and erythema (158, 13%). A data mining alert (EB05 = 3.812) for the MedDRA Preferred Term \"Guillain-Barre syndrome\" was observed for serious reports. Clinical review verified 11 of 20 GBS reports; 7/11 vaccine recipients were aged ≥ 65 years. Among the 11 verified cases, the median time from vaccination to symptom onset was 14 days. Five persons received another vaccine on the same visit. The reporting rate of GBS after PCV20 receipt was 0.5 cases per million doses distributed. No other safety concern was identified.</p><p><strong>Conclusions: </strong>During the period of this post-licensure review of PCV20, we found most reports were non-serious and comprised mostly local and systemic (e.g., fever) reactions consistent with prelicensure studies. In serious reports, we also identified a data mining alert for GBS after receipt of PCV20, which Centers for Disease Control and Prevention and the FDA are investigating further. No other new or unexpected safety concern was identified.</p>","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"279-286"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142812553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Introduction and objective: Proton pump inhibitor (PPI) use in children increases the risk of infections, prompting inquiry into the impact of prenatal PPIs exposure on serious infections in offspring. As a research gap in this area exists, this study aimed to address it by assessing the association between prenatal PPIs exposure and serious infections in infants during their first year of life.
Methods: Using the French health insurance data warehouse (SNDS) (2013-2018), we conducted a retrospective cohort study on singleton, full-term liveborn non-immunocompromised infants, stratified by PPI use during the first three months of life (early-life use). Proton pump inhibitor dispensing in ambulatory care settings during pregnancy defined exposure. Outcomes concerned any serious infections in offspring aged between 3 and 12 months. Adjusted odds ratios (aORs) were estimated using logistic regression with multivariable models to control for potential confounders.
Results: Of the 2,485,545 infants included, 497,060 (23.3%) were prenatally exposed to PPIs and 97,767 (4.6%) had PPI use during the first three months of life. Prenatal PPI exposure was associated with serious infections in offspring (aOR, 1.09 [95% CI, 1.07-1.10]) in infants without early-life PPIs use. No association was found for infants with early-life PPI use (aOR, 1.05 [95% CI, 1.00-1.11]). Gastrointestinal infections were the sole site with persistent significance.
Conclusion: Prenatal PPI exposure is common and is not associated with a major risk of serious infections in infants during their first year. However, even after adjusting for several confounding factors, a weak association remains, especially in infants without early-life PPI use. While offering reassurance, adherence to clinical guidelines is still crucial.
{"title":"Prenatal Exposure to Proton Pump Inhibitors and Risk of Serious Infections in Offspring During the First Year of Life: A Nationwide Cohort Study.","authors":"Mylène Tisseyre, Mathis Collier, Nathanaël Beeker, Florentia Kaguelidou, Jean-Marc Treluyer, Laurent Chouchana","doi":"10.1007/s40264-024-01496-4","DOIUrl":"10.1007/s40264-024-01496-4","url":null,"abstract":"<p><strong>Introduction and objective: </strong>Proton pump inhibitor (PPI) use in children increases the risk of infections, prompting inquiry into the impact of prenatal PPIs exposure on serious infections in offspring. As a research gap in this area exists, this study aimed to address it by assessing the association between prenatal PPIs exposure and serious infections in infants during their first year of life.</p><p><strong>Methods: </strong>Using the French health insurance data warehouse (SNDS) (2013-2018), we conducted a retrospective cohort study on singleton, full-term liveborn non-immunocompromised infants, stratified by PPI use during the first three months of life (early-life use). Proton pump inhibitor dispensing in ambulatory care settings during pregnancy defined exposure. Outcomes concerned any serious infections in offspring aged between 3 and 12 months. Adjusted odds ratios (aORs) were estimated using logistic regression with multivariable models to control for potential confounders.</p><p><strong>Results: </strong>Of the 2,485,545 infants included, 497,060 (23.3%) were prenatally exposed to PPIs and 97,767 (4.6%) had PPI use during the first three months of life. Prenatal PPI exposure was associated with serious infections in offspring (aOR, 1.09 [95% CI, 1.07-1.10]) in infants without early-life PPIs use. No association was found for infants with early-life PPI use (aOR, 1.05 [95% CI, 1.00-1.11]). Gastrointestinal infections were the sole site with persistent significance.</p><p><strong>Conclusion: </strong>Prenatal PPI exposure is common and is not associated with a major risk of serious infections in infants during their first year. However, even after adjusting for several confounding factors, a weak association remains, especially in infants without early-life PPI use. While offering reassurance, adherence to clinical guidelines is still crucial.</p>","PeriodicalId":11382,"journal":{"name":"Drug Safety","volume":" ","pages":"265-277"},"PeriodicalIF":4.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142767382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}