Lay people’s terms of health concepts are widely found in social medial data. Understanding their terms can help learn what they have experienced. The existing consumer health vocabularies do not include many health concept phrases found in social media. Recent advancement of pretrained large language models has demonstrated their state-of-the-art performance in a number of natural language processing tasks. In this work, we investigated 3 commonly used pretrained large language models on their ability of understanding consumer health concepts. The results on a corpus of 156 phrases related to 3 common COVID-19 symptoms showed that these pretrained general domain language models generated encouraging outputs, indicating that they may be a promising alternative to the current consumer health vocabularies.

Gathering the information pertaining to health evidence occurring on particular days can help understand the progression of a disease over time, and such evidence as symptoms was widely shared by the social media users during the COVID-19 pandemic. Identifying this type of evidence is challenging. In this work, we investigated pretrained large language models on their ability to identify the day-based COVID-19 symptom mentions in Twitter posts. Our results on a corpus of 635 tweets show that without any supervision and optimization both GPT-3.5 and GPT 4 models achieved impressive performance, much better than Web-based ChatGPT and Google Bard. In addition, we explored and utilized GPT-4’s ability to determine the number of matches between a list of predicted symptom expressions and a list of annotated symptom expressions, reducing the effort of designing a sophisticated algorithm for finding the matches.

How to deliver best care in various clinical settings remains a vexing problem. All pertinent healthcare-related questions have not, cannot, and will not be addressable with costly time- and resource-consuming controlled clinical trials. At present, evidence-based guidelines can address only a small fraction of the types of care that clinicians deliver. Furthermore, underserved areas rarely can access state-of-the-art evidence-based guidelines in real-time, and often lack the wherewithal to implement advanced guidelines. Care providers in such settings frequently do not have sufficient training to undertake advanced guideline implementation. Nevertheless, in advanced modern healthcare delivery environments, use of eActions (validated clinical decision support systems) could help overcome the cognitive limitations of overburdened clinicians. Widespread use of eActions will require surmounting current healthcare technical and cultural barriers and installing clinical evidence/data curation systems. The authors expect that increased numbers of evidence-based guidelines will result from future comparative effectiveness clinical research carried out during routine healthcare delivery within learning healthcare systems.

Clinical decision-making is based on knowledge, expertise, and authority, with clinicians approving almost every intervention-the starting point for delivery of "All the right care, but only the right care," an unachieved healthcare quality improvement goal. Unaided clinicians suffer from human cognitive limitations and biases when decisions are based only on their training, expertise, and experience. Electronic health records (EHRs) could improve healthcare with robust decision-support tools that reduce unwarranted variation of clinician decisions and actions. Current EHRs, focused on results review, documentation, and accounting, are awkward, time-consuming, and contribute to clinician stress and burnout. Decision-support tools could reduce clinician burden and enable replicable clinician decisions and actions that personalize patient care. Most current clinical decision-support tools or aids lack detail and neither reduce burden nor enable replicable actions. Clinicians must provide subjective interpretation and missing logic, thus introducing personal biases and mindless, unwarranted, variation from evidence-based practice. Replicability occurs when different clinicians, with the same patient information and context, come to the same decision and action. We propose a feasible subset of therapeutic decision-support tools based on credible clinical outcome evidence: computer protocols leading to replicable clinician actions (eActions). eActions enable different clinicians to make consistent decisions and actions when faced with the same patient input data. eActions embrace good everyday decision-making informed by evidence, experience, EHR data, and individual patient status. eActions can reduce unwarranted variation, increase quality of clinical care and research, reduce EHR noise, and could enable a learning healthcare system.

Growing uses of opioids for pain management have led to a crisis of addiction and thousands of deaths in the United States. Although many of the opioid effects have been observed and reported, experience directly from the opioid users may help provide additional information in identifying any potentially unreported effects. In this study, we developed a neural embedding-based method to discover potential opioid-effect relations through similar relations of known medication effects. Using a corpus of 3.6 million clean unannotated tweets, a vector space model was learned with word2vec, and the model was used to identify potential opioid effects. The inferred results were further verified against 5 authoritative sources of medication effects. Seven of inferred effects were identified as potentially unreported, demonstrating the power and utility of our method. It is conceivable that our approach can be applied to discovery of potentially unreported effects of other medications.

BACKGROUND: As Twitter has become an active data source for health surveillance research, it is important that efficient and effective methods are developed to identify tweets related to personal health experience. Conventional classification algorithms rely on features engineered by human domain experts, and engineering such features is a challenging task and requires much human intelligence. The resultant features may not be optimal for the classification problem, and can make it challenging for conventional classifiers to correctly predict personal experience tweets (PETs) due to the various ways to express and/or describe personal experience in tweets. In this study, we developed a method that combines word embedding and long short-term memory (LSTM) model without the need to engineer any specific features. Through word embedding, tweet texts were represented as dense vectors which in turn were fed to the LSTM neural network as sequences. RESULTS: Statistical analyses of the results of 10-fold cross-validations of our method and conventional methods indicate that there exist significant differences (p < 0.01) in performance measures of accuracy, precision, recall, F1-score, and ROC/AUC, demonstrating that our approach outperforms the conventional methods in identifying PETs. CONCLUSION: We presented an efficient and effective method of identifying health-related personal experience tweets by combining word embedding and an LSTM neural network. It is conceivable that our method can help accelerate and scale up analyzing textual data of social media for health surveillance purposes, because of no need for the laborious and costly process of engineering features.