December 15th, 2016 16.30 - 18.00
Mining Internet Conversations for Evidence of Adverse Events Related to Herbal Supplements or Prescription Drugs
John H. Holmes, PhD
University of Pennsylvania Perelman School of Medicine Philadelphia, Pennsylvania, USA
Often, users of online resources such as discussion boards seek advice about topics they are hesitant to discuss with providers. One such topic is the use of herbal supplements. Relatively little is known about the use and effects of supplements, primarily due to patient and provider reticence, but also owing to the lack of a formal reporting mechanism that could be used for monitoring such events. As a result, adverse events that are possibly related to herbal supplement use are difficult to evaluate in a population-based investigation. Another discussion topic is the use of prescription drugs. Many people experience some degree of uncertainty about a drug that has been prescribed to themselves or a family member, and some express their concerns on social media platforms with the convenience and anonymity these venues offer. Even though prescription drugs rarely carry the stigma that herbal supplements often do, participants on social media can reveal their concerns more freely on social media than they might in the context of a visit to a health care practitioner. This is especially true of an adverse reaction that does not seem life-threatening but is nevertheless disconcerting, and even more so when the reaction is noted as rare on the drug’s package insert- and therefore not expected. Prescription drugs are marketed only after extensive testing that includes randomized clinical trials. However, these trials are relatively small in size, typically with fewer than 800 subjects. Even though such trials are considered to be the gold standard for testing a drug, some adverse events may be so rare that they are not identified in these studies. Social media provides a potentially rich data source for identifying such rare, yet potentially important events. The goal of our project was to develop, apply, and evaluate computational intelligence tools for mining conversational text for evidence of adverse events and side effects of herbals and prescription drugs reported by users of online message boards relating to breast cancer. In this talk, we will examine some of these tools and review some findings for herbal supplements as well as conversations about a drug class (aromatase inhibitors) that is commonly used in breast cancer.
John H. Holmes is Professor of Medical Informatics in Epidemiology at the University of Pennsylvania Perelman School of Medicine. He is the Associate Director of the Penn Institute for Biomedical Informatics and is Chair of the Graduate Group in Epidemiology and Biostatistics. Dr. Holmes’ research interests are focused on several areas in medical informatics, including evolutionary computation and machine learning approaches to knowledge discovery in clinical databases (data mining), interoperable information systems infrastructures for epidemiologic surveillance, regulatory science as it applies to health information and information systems, clinical decision support systems, semantic analysis, shared decision making and patient-physician communication, and information systems user behavior. Dr. Holmes is a principal or co-investigator on projects funded by the National Cancer Institute, the Patient-Centered Outcomes Research Institute, the National Library of Medicine, and the Agency for Healthcare Research and Quality, and he is the principal investigator of the NIH-funded Penn Center of Excellence in Prostate Cancer Disparities. Dr. Holmes is engaged with the Botswana-UPenn Partnership, assisting in building informatics education and clinical research capacity in Botswana. He leads the evaluation of the National Obesity Observational Studies of the Patient-Centered Clinical Research Network. Dr. Holmes is an elected Fellow of the American College of Medical Informatics and the American College of Epidemiology
How to get to the nearest parking lot