Drug and alcohol review最新文献

{"title":"A Qualitative Study of How Teens in Washington State Make Sense of Cannabis Edibles Warning Labels and Packaging","authors":"Jessica Fitts Willoughby,&nbsp;Stacey J. T. Hust,&nbsp;Soojung Kang,&nbsp;Leticia Couto,&nbsp;Ron Price,&nbsp;Christina Griselda Nickerson,&nbsp;Opeyemi Johnson,&nbsp;Sarah Ross-Viles","doi":"10.1111/dar.70071","DOIUrl":"10.1111/dar.70071","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Washington state's adult use cannabis market operates under regulations by the Washington State Liquor and Cannabis Board to restrict access and promotion among young people. Cannabis edibles sold in the state are required to contain specific labels that inform consumers that the product contains cannabis and provide contact information for Poison Control. However, it is unclear how teens perceive such labels.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Ten focus groups were conducted with a diverse sample of 28 teens (<i>M =</i> 15.93, SD = 1.25) in Washington state, United States. After viewing images of cannabis edible products available in Washington state, participants shared their thoughts and opinions about the packaging, warning labels and nutrition information.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Through a thematic analysis, we noted that teens may be misinterpreting warning labels, and they think warning labels are hidden or unnoticeable. Most teens paid little attention to nutrition labels and often found serving size information confusing. Teens said if an edible product looked similar to snack products they know, they might perceive them as less risky and more enticing. Knowledge of cannabis products also impacted teens' understanding of edible product packaging.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>Labels alert teens to the fact that products contain cannabis. However, teens often feel such labels apply to younger children and would not keep teens from using a product. Youth might benefit from additional guidance around interpreting cannabis packaging and labels.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12682422/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145476731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Scoping Review of Large Language Model Chatbot Use for Alcohol and Other Drug Health Information","authors":"Natasha Harding,&nbsp;Nataly Bovopoulos,&nbsp;Dotahn Caspi,&nbsp;Craig Martin,&nbsp;Skye McPhie,&nbsp;Mohamed Reda Bouadjenek,&nbsp;Sunil Aryal,&nbsp;Michael Hobbs","doi":"10.1111/dar.70068","DOIUrl":"10.1111/dar.70068","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Issues</h3>\u0000 \u0000 <p>While people prefer to seek alcohol and drug information (AOD) online, there can be quality and accessibility issues with these sources. Large Language Model (LLM) based chatbots are an emerging technology that may present an opportunity to overcome these barriers. We aimed to review the literature on the use of chatbots for seeking AOD health information, particularly the benefits, challenges and recommendations for future use.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Approach</h3>\u0000 \u0000 <p>Scoping review methodology was used to conduct a systematic search of four databases for English language studies relating to the use of chatbots to seek AOD health information in the last 5 years. This resulted in the screening of 243 articles, with five included studies.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Key Findings</h3>\u0000 \u0000 <p>There has been growing interest in the topic, though evidence is still limited. Despite identified benefits of chatbot use such as accuracy, appropriateness, overall experience and the provision of supporting documentation, important challenges in user safety concerns, lack of referral, quality, readability issues, and lack of adherence to current guidelines were noted, with mixed results regarding evidence-based responses. Only three of the five studies recommended chatbots for AOD-information seeking.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Implications/Conclusion</h3>\u0000 \u0000 <p>The current review suggests gaps in knowledge remain in the areas of accuracy, user safety, readability, evidence base and quality of LLM chatbot responses to AOD questions. More research is needed to investigate the applicability of LLMs in obtaining safe, non-stigmatising AOD information.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145457997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effectiveness of Psychosocial Interventions for Adults With Substance Use Disorder That Have a Co-Occurring Common Mental Health Disorder: An Umbrella Review","authors":"Emma L. Simpson,&nbsp;Munira Essat,&nbsp;Ruth Wong,&nbsp;Sarah Stacey,&nbsp;Edward Day","doi":"10.1111/dar.70066","DOIUrl":"10.1111/dar.70066","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Issues</h3>\u0000 \u0000 <p>People with substance use disorders can have co-occurring mental disorders.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Approach</h3>\u0000 \u0000 <p>An umbrella review was conducted to identify evidence of the effectiveness of psychosocial interventions for adults (aged 18+) with substance use disorders and co-occurring common mental health disorders. Systematic reviews were sought of randomised controlled trials of psychosocial interventions compared to each other, treatment as usual or wait-list. Five databases were systematically searched in February 2024. Data, including critical appraisal (Joanna Briggs Institute Checklist), were extracted by one reviewer and checked by another. Data were discussed in a narrative review.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Key Findings</h3>\u0000 \u0000 <p>Of 5420 unique records, 28 systematic reviews were included. The methodological quality of the reviews was good. Most reviews focused on depression, anxiety or post-traumatic stress disorder. There was much heterogeneity between reviews, and randomised controlled trials within reviews. Most of the interventions and many of the treatment-as-usual comparators resulted in significant improvement in substance use and mental health disorders. Results suggested integrated (co-ordinated) treatment for co-occurring diagnosis patients was better than treating one condition alone, and usually better than parallel uncoordinated services. There was limited evidence assessing sequential treatment, but this suggested similar effectiveness to integrated treatment.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Implications</h3>\u0000 \u0000 <p>Implications for current practise could not be recommended due to heterogeneity. Improvement shown by all types of psychosocial intervention including active comparators precluded recommending one type of intervention over another.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>Further research is needed comparing integrated with parallel or sequential treatment, with follow-up of 6 months or longer, and sample size large enough to encompass dropout.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12682421/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145451326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Addressing Hepatitis A and Hepatitis B Risks: Screening and Preventative Care for People Who Inject Drugs","authors":"Wendy Yao,&nbsp;Natali Jokanovic,&nbsp;Hashini Herath,&nbsp;Susan Poole,&nbsp;Kate Mackie,&nbsp;Joseph S. Doyle,&nbsp;Alison Duncan","doi":"10.1111/dar.70064","DOIUrl":"10.1111/dar.70064","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>People who inject drugs (PWID) often present to hospital with complicated infections, providing an opportunity to screen and vaccinate for hepatitis A virus (HAV) and hepatitis B virus (HBV). This study evaluated HAV and HBV screening and vaccination among PWID over 3 years.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A single-centre retrospective study of all infectious diseases inpatient admissions for current PWID with a length of stay ≥ 48 h, Drug and Alcohol team referral and/or opioid substitution therapy was performed (January 2020–December 2022). Serological screening for HAV (immunoglobulin G [IgG]) and HBV (complete screening: surface antigen, core and surface antibodies), and vaccinations were collected per encounter. The prevalence of serological screening and vaccinations against HAV and HBV and differences in characteristics by screening status were explored with descriptive statistics.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Overall, 115 patients from 159 encounters were included (mean age: 41.8 ± 8.4 years; male: <i>n</i> = 73, 64%; median length of stay: 7 (IQR 4–16) days). Anti-HAV screening was performed for 47 patients across 57 (35.8%) encounters, with 12 patients who screened negative vaccinated (<i>n</i> = 12/21, 57.1%). Complete HBV screening was performed for 75 patients across 85 (53.5%) encounters, with 7 patients who screened negative vaccinated (<i>n</i> = 7/12, 58.3%). Males were more likely to receive HAV screening than females (77% vs. 23%, <i>p</i> = 0.015). HBV screening was less likely in encounters requiring intensive care unit admission (12% vs. 88%, <i>p</i> = 0.038).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>While more than half of eligible patients received vaccination, complete screening for both HAV and HBV was low. Initiatives including electronic serology order sets should be considered to improve screening.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145437799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Community Pharmacy-Based Injectable Opioid Agonist Treatment: Findings From a Canadian Pilot Program","authors":"Tamara Mihic,&nbsp;Maria Eugenia Socias,&nbsp;Karen McCrae,&nbsp;Cheyenne Johnson,&nbsp;Seonaid Nolan,&nbsp;Cameron Grant,&nbsp;Christy Sutherland,&nbsp;Nadia Fairbairn","doi":"10.1111/dar.70062","DOIUrl":"10.1111/dar.70062","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Access to evidence-based treatment for opioid use disorder remains limited, particularly for individuals who have not responded to oral opioid agonist treatment (OAT). A community pharmacy-based model of injectable OAT (iOAT) was piloted in Vancouver, Canada from March 2017 to December 2018. This brief report describes the program structure, participant sociodemographics, reported outcomes, and strengths and areas for improvement of the program.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A retrospective review of cross-sectional, interviewer-led questionnaire data from participants who accessed iOAT at the pharmacy site (<i>n</i> = 176) and provided informed consent was conducted. Outcomes include participant-reported changes in symptomatology, function and satisfaction, analysed through descriptive statistics. Open-ended responses were analysed using content analysis to identify strengths and areas for improvement of the program.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Fifty-one participants (29%) completed the questionnaire, and most had multiple previous overdoses and trials of oral OAT. The most commonly reported outcomes were reduction in illicit opioid use (76%), opioid cravings (45%) and illicit substance use (45%). Participants identified key strengths of the program as positive experiences with staff and efficiency of the pharmacy model including flexible dosing time and the ability to pick up other medications at the same time. Suggested improvements focused on medication options (e.g., access to diacetylmorphine, alternate routes of administration), expanded hours and flexibility, additional support services, and increased capacity and space.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and <b>Conclusions</b></h3>\u0000 \u0000 <p>Community pharmacy-based iOAT represents a novel strategy to expand access to evidence-based opioid use disorder treatment among individuals who inject opioids and have not responded to or do not prefer oral OAT.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12682424/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145437820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Single-Dose, Long-Acting Injectable Buprenorphine for Opioid Withdrawal Treatment","authors":"Jeffrey S. Kruk,&nbsp;Judith L. Fraser,&nbsp;Prasun Datta,&nbsp;Karen A. Fisher","doi":"10.1111/dar.70070","DOIUrl":"10.1111/dar.70070","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Long-acting, injectable buprenorphine (LAIB) is a recently introduced opioid agonist treatment (OAT) for opioid dependence in Australia. Little research has been performed for its role in opioid withdrawal management.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A retrospective analysis of hospital patient records was conducted for five patients who received a single dose of LAIB as part of an inpatient or outpatient opioid withdrawal management program. Patients analysed had tried and failed conventional opioid withdrawal management and were unwilling to commence or continue an OAT program.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Patient demographics including types of opioid use and inpatient lengths of stay are reported. Patient withdrawal symptoms before treatment and during follow-up periods are also described. Overall, most patients were satisfied with a single dose of LAIB to adequately control withdrawal symptoms in the community.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>Given the stability of OAT patients on LAIB, it was inferred that some patients may do well in opioid withdrawal management using LAIB, should other conventional treatments not succeed. This was based on previous pharmacokinetic data and modelling by others which would allow for a slow, in vivo wean of the drug rather than discrete, daily dose reductions. For patients, this could reduce potential withdrawal symptoms and shorten admission lengths in hospital withdrawal management wards. Further research is needed to verify these results, but these preliminary data are encouraging.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145444207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning Algorithms to Predict Heavy Episodic Drinking in the United States Using Survey Data","authors":"Laura Llamosas-Falcón,&nbsp;Charlotte Probst,&nbsp;Kevin Shield,&nbsp;Erik Spence,&nbsp;Jürgen Rehm","doi":"10.1111/dar.70065","DOIUrl":"10.1111/dar.70065","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Heavy episodic drinking (HED) is a major public health concern but is often missing from surveys or measured unreliably. Predictive models offer a method to estimate HED's likelihood at the individual level in such cases. While logistic regression is commonly used, other machine learning algorithms (MLA) may offer greater accuracy and robustness. This study compares various MLAs to identify the best predictive model of HED.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Data from the 1997–2018 National Health Interview Survey were used. Six MLAs were trained and cross-validated: logistic regression, naïve bayes, k-nearest neighbour, support vector machine, random forest and XGBoost. Model performance was compared, and the SHapley Additive exPlanations (SHAP) method assessed interpretability by ranking features based on their contribution to the model's prediction.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The probability of correctly ranking a randomly selected HED instance higher than a non-HED instance ranged from 0.85 to 0.97 (with values closer to 1 indicating better performance). XGBoost outperformed the other MLAs (sensitivity 0.80, precision 0.83, accuracy 0.92). Amongst the 11 features included in the models, average daily alcohol use and age were the most influential, as determined by SHAP values.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>The strong discriminative ability of our models shows that even a limited number of well-chosen features can yield robust predictions, highlighting the potential of MLAs for modelling health behaviours. Integrating our models into simulation frameworks can help model HED and test scenarios, leading to effective policies. Future studies should incorporate objective sources for external validation and investigate systematic biases to improve predictive accuracy.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12682420/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145444265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Alcohol Purchase Modality: Differences in Demographics, Purchase Patterns and Alcohol-Related Harms Among Australians Who Drink at High-Risk Levels","authors":"Kerri Coomber,&nbsp;Kira Button,&nbsp;Ryan Baldwin,&nbsp;Florentine Martino,&nbsp;Adyya Gupta,&nbsp;Kathryn Backholer,&nbsp;Peter G. Miller,&nbsp;Danica Keric,&nbsp;Julia Stafford","doi":"10.1111/dar.70069","DOIUrl":"10.1111/dar.70069","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Alcohol home delivery increases alcohol availability and is associated with high-risk drinking and increased harms. However, little is known about the specific impacts home delivery has among people who consume alcohol at high-risk levels. The current study examined demographics, purchase patterns and alcohol-related harms among Australian adults classified as drinking at high-risk levels, by purchase modality (in-store vs. online for delivery).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>An online panel survey of 700 Australian adults who drink alcohol at high-risk levels measured demographics, usual purchasing behaviour, most recent alcohol purchase and alcohol-related harms. Participants were classified according to how they purchase alcohol: (i) only in-store (<i>n</i> = 202); (ii) occasionally online (<i>n</i> = 299); and (iii) frequently online (<i>n</i> = 199). Regression models controlling for demographics and clustering by location examined the association between purchase modality and purchasing behaviours and harms.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Participants who purchased alcohol frequently online had significantly higher AUDIT-10 scores, were younger, less likely to be female, had higher levels of education, and were more likely to live in high socio-economic areas compared to those purchasing in-store or occasionally online. Participants purchasing frequently online were also significantly more likely to use buy-now-pay-later options than the occasionally online participant group. There were no clear differences in alcohol-related harms experienced by participant groups.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>The current study reaffirms the need for effective regulations targeting home delivery of alcohol, such as mandated delivery timeframes between sale and delivery and removal of buy-now-pay-later payment options from online alcohol stores.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145444118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Language Models for Standardising Clinical Notes and Information Extraction in Addiction Psychiatry—An Empirical Study","authors":"Haritha Gireesh,&nbsp;Lekhansh Shukla,&nbsp;Prakrithi Shivaprakash,&nbsp;Animesh Mukherjee,&nbsp;Prabhat Chand,&nbsp;Pratima Murthy","doi":"10.1111/dar.70059","DOIUrl":"10.1111/dar.70059","url":null,"abstract":"&lt;div&gt;\u0000 \u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Introduction&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;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.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Methods&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;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.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Results&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;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 (&lt;i&gt;p&lt;/i&gt; &lt; 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.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Discussion and Conclusions&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;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","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145387909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Demographic Predictors of Medicinal Cannabis Users' Perceived Level of Physician Support for Medicinal Cannabis Prescriptions in New Zealand","authors":"Vinuli Withanarachchie,&nbsp;Chris Wilkins,&nbsp;Karl Parker,&nbsp;Marta Rychert","doi":"10.1111/dar.70063","DOIUrl":"10.1111/dar.70063","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Studies indicate that access to prescribed medicinal cannabis (MC) is influenced by demographic factors, though few have explored the roles of ethnicity and income. Patients' perceptions of physician support for MC may be an important barrier. New Zealand implemented a legal MC scheme in 2020. This paper aimed to identify demographic predictors of accessing an MC prescription and perceptions of physician support for MC.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>An anonymous online survey (New Zealand Drug Trends Survey), promoted via Meta was completed by 10,781 New Zealanders aged 16+ years in 2024. This paper focuses on 1742 respondents who used cannabis primarily for medicinal purposes in the previous 6 months. Logistic regression models predicted obtaining MC prescriptions, perceived physician refusal to prescribe, and perceptions of physician support.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Mental health (73.8%), sleep (71.6%) and pain (59.5%) were most treated with MC. Two-fifths of female MC users (41.5%) were treating women's health conditions. MC prescriptions were less common among females, Māori, Pacific Peoples, lower income-earners (≤ $60,000) and those treating substance dependency. Individuals aged 60+ years, tertiary educated, with a medical diagnosis, earning $60,001+, treating sleep, gastrointestinal, neurological, or mental health were more likely to be MC prescription holders. Younger individuals, not employed, Māori or Pacific were more likely <i>not</i> to request a MC prescription because of anticipated refusal. Older MC prescription holders perceived more support from physicians for MC use.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Discussion and Conclusions</h3>\u0000 \u0000 <p>Ethnic, gender and socio-economic disparities limit access to MC prescriptions. Improving equitable access to legal MC requires greater support for open patient-physician conversations about MC use.</p>\u0000 </section>\u0000 </div>","PeriodicalId":11318,"journal":{"name":"Drug and alcohol review","volume":"45 1","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145387941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}