Drug and alcohol review最新文献

Introduction: Electronic health records contain both structured and unstructured data, with unstructured clinical notes widely used in addiction psychiatry. Clinical notes have numerous errors and require proofreading to ensure accuracy and readability. This study evaluates natural language processing methods and adapts a Large Language Model (LLM) for proofreading clinical notes and extracting substance-related information.

Methods: We analysed clinical notes from a 5-year addiction medicine electronic health record dataset (2018-2023), selecting 6500 notes. The proofreading task involved correcting spelling and expanding abbreviations, while information extraction identified the presence of substance use and quantified the time since last use. Annotations by a team of doctors and nurses provided the gold standard. Against this, we compared the performance of existing solutions, including LLMs, and adapted an LLM for these tasks. The final model (fine-tuned LLAMA-3.2-3b) is also compared against a state-of-the-art commercial model (Generative Pretrained Transformer-4-o), and a human-preference experiment is done with masked raters choosing between model-generated and human-generated proofread versions.

Results: Proofreading improved readability and decreased out-of-vocabulary words. LLM-based solutions outperformed simpler approaches. The fine-tuned model outperformed the Generative Pretrained Transformer-4-o on both tasks. Masked human evaluators chose model-corrected clinical notes over the human-corrected version in 62% of trials (p < 0.001). On the information extraction task, while the overall performance is satisfactory (Mean F1 0.99), it is poor on rarer substance classes like hallucinogens.

Discussion and conclusions: Fine-tuned LLMs effectively standardised clinical notes and extracted structured information from addiction psychiatry records. Both these functionalities have important applications. Standardising improves the readability of clinical documentation and facilitates communication within and between interdisciplinary teams. Automated information extraction can decrease the burden on clinical staff, allow the creation of research cohorts from existing records and improve treatment outcomes by extracting critical information, such as 'time since last drink', which can be used to raise alerts. Even with limited computational resources, it is possible to adapt open-source LLMs for bespoke tasks in the field of addiction psychiatry. Our proposed solution is a model that can be deployed on consumer-grade servers, thus ensuring data privacy and security.

Introduction: Public health planning in jurisdictions considering cannabis policy shifts may benefit from projections concerning post-policy changes in cannabis use and user profiles. Multiple scenarios concerning decriminalised cannabis use among Norwegian high school students were projected using a simple but under-utilised tool: their self-reported willingness to do so.

Methods: A nationally representative sample of Norwegian high school students (N = 3490) reported on their characteristics, cannabis use, and whether they would use cannabis if doing so were non-punishable. These estimates were scaled up to census data and used to project the population sizes of students who may initiate/increase decriminalised cannabis use under various hypothetical willingness-to-behaviour transition scenarios. Multinomial regression models were used to identify risk profiles associated with willingness to use decriminalised cannabis.

Results: One in five participants (20.3%) reported lifetime cannabis use, corresponding to 38,200 student users nationally. Among never-users, 8.0% reported willingness to initiate decriminalised cannabis use, while 22.3% were unsure. Among users, 40.3% reported willingness to increase decriminalised cannabis use, while 30.3% were unsure. Multiple projection scenarios indicated that both the prevalence (e.g., 12,000 new users nationally if all non-users willing to initiate use would do so) and frequency of cannabis use (e.g., 15,400 more frequent users nationally if all users willing to increase use would do so) would increase under decriminalisation, especially among youth with certain socio-demographic characteristics and recent histories of frequent use.

Discussion and conclusions: Asking young people about their intended cannabis use under different policies provides an easily understandable and highly relevant tool in public health research and planning.