Journal of Addictive Diseases最新文献

{"title":"Effects of seven traditional Chinese sports on physiological outcomes in individuals receiving treatment for substance use disorder: a network meta-analysis of randomized controlled trials.","authors":"Jianqiang Guan, Xiaofei Zhang, Yuqi Wang, Jing Gao, Xin Gao","doi":"10.1080/10550887.2026.2746585","DOIUrl":"https://doi.org/10.1080/10550887.2026.2746585","url":null,"abstract":"<p><strong>Background: </strong>Substance use disorder is a major public health challenge, with physiological consequences, including dysregulated blood pressure, abnormal heart rate, and altered body composition, which increase the risk of cardiovascular disease and mortality. Physical exercise shows rehabilitation potential, but evidence-based guidance on optimal traditional Chinese exercise regimens is lacking.</p><p><strong>Objective: </strong>This network meta-analysis evaluated and ranked the comparative effects of seven traditional Chinese sports (Taiji Rouliqiu, combined Tai Chi and Qigong, 24-form Tai Chi, Baduanjin, Tai Chi rehabilitation exercise, shuttlecock, and Tai Chi Thirteen Postures Daoyin) on physiological outcomes in individuals with substance use disorder.</p><p><strong>Methods: </strong>Randomized controlled trials were searched from PubMed, Web of Science, CNKI, and VIP databases from inception to February 2026. A Bayesian network meta-analysis was performed, with treatment rankings assessed by surface under the cumulative ranking curve (SUCRA) values and consistency evaluated <i>via</i> node-splitting analysis.</p><p><strong>Results: </strong>Seventeen RCTs involving 1,133 participants were included. Combined Tai Chi and Qigong ranked highest for BMI (SUCRA = 92.4%), diastolic blood pressure (SUCRA = 73.8%), and resting heart rate (SUCRA = 82.9%). Tai Chi rehabilitation exercise ranked highest for systolic blood pressure (SUCRA = 66.3%), and Tai Chi Thirteen Postures Daoyin ranked highest for body fat percentage (SUCRA = 87.9%). No significant inconsistency was detected, and sensitivity analyses confirmed robustness.</p><p><strong>Conclusion: </strong>Combined Tai Chi and Qigong, Tai Chi rehabilitation exercise, and Tai Chi Thirteen Postures Daoyin emerged as the most promising interventions. These findings provide preliminary evidence for exercise-based rehabilitation strategies, though future head-to-head trials are needed to confirm the observed rankings.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"1-14"},"PeriodicalIF":1.7,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Substance use diagnoses among patients who present to federally run mental health facilities with co-morbid ADHD and bipolar disorder between 2013-2023: a national analysis using random forest methods.","authors":"Christine Ramdin, Jamey J Lister, Ellen Kurtzman, Sally Kuo, Stephanie Shiau, Howard Greller, Lewis S Nelson","doi":"10.1080/10550887.2026.2733227","DOIUrl":"https://doi.org/10.1080/10550887.2026.2733227","url":null,"abstract":"<p><strong>Background/objectives: </strong>Although substance use disorder is common in patients with attention deficit hyperactivity disorder or bipolar disorder, limited recent data exist for those with these co-occurring conditions. Our study analyzes recent trends and predictors of these diagnoses.</p><p><strong>Methods: </strong>We conducted a secondary analysis of 2013-2023 Mental Health Client-Level Data, identifying patients with co-morbid ADHD and bipolar disorder and their SUD prevalence. Joinpoint regression was used to analyze trends, and random forest models identified the most important predictors.</p><p><strong>Results: </strong>Between 2013 and 2023, 5,691 patients (0.008%) had comorbid ADHD and bipolar disorder, with rates increasing by 0.01% per year starting in 2016 (<i>p</i> < 0.0001). All patients had a documented SUD. Most patients were White (75.7%) and unemployed (1,284; 22.6%). The most common SUD diagnoses were 1) other unspecified substance dependence (665, 11.7%), 2) alcohol dependence (474, 8.3%), and 3) cannabis dependence (450, 7.9%). Key predictors included age, education, employment, and marital status.</p><p><strong>Conclusions: </strong>Our study identified the most common types of SUDs among those with co-morbid ADHD and bipolar disorder and associated risk factors. This will guide prevention and early intervention to reduce adverse outcomes.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"1-12"},"PeriodicalIF":1.7,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review and meta-analysis exploring the recruitment hurdles and nonparticipation rate in noninvasive brain stimulation for addictive disorders.","authors":"Pinki Sevda, Priyanka Saha, Siddharth Sarkar, Rohit Verma, Naresh Nebhinani","doi":"10.1080/10550887.2025.2564446","DOIUrl":"10.1080/10550887.2025.2564446","url":null,"abstract":"<p><strong>Background: </strong>Substance use disorders (SUDs) are a growing public health concern with high relapse rates. Noninvasive brain stimulation techniques are emerging as potential adjunctive therapies; however, recruitment challenges persist in related clinical trials. To use systematic review and meta-analysis to determine the rate and reasons for nonparticipation in randomized controlled trials (RCTs) involving noninvasive brain stimulation for substance-related and addictive disorders.</p><p><strong>Materials and methods: </strong>Major databases (PubMed, CENTRAL, and Google Scholar) were searched from inception to January 2025 for RCTs of noninvasive brain stimulation and SUD. Of 3589 abstracts searched, 55 RCTs met the inclusion criteria. Eligible studies were peer-reviewed, published in English and examined any form of noninvasive brain stimulation in individuals with substance-related or behavioral addictions. Studies were required to report the number of individuals screened and ultimately randomized. Non-randomized studies, those involving invasive brain stimulation, were excluded.Following PRISMA guidelines, data were extracted independently by two reviewers, with discrepancies resolved by a third. Risk of bias (RoB) was assessed using the Cochrane RoB tool. Random-effects model was used for meta-analysis.</p><p><strong>Results: </strong>A total of 55 randomized trials were included which involved diverse SUDs and behavioral addictions. The pooled nonparticipation rate was 55.9% (95% CI: 45.9-65.8%; <i>I</i>² = 99.56%, <i>p</i> < 0.001). Majority of the studies reported craving as an outcome measure. Key reasons included failure to meet inclusion criteria (reported in over 25 studies), refusal to participate due to anxiety or lack of motivation (≥15 studies), and logistical barriers such as travel or session frequency (≥10 studies).</p><p><strong>Conclusion: </strong>Nonparticipation in noninvasive brain stimulation trials for SUD are high, largely due to stringent eligibility criteria, procedural apprehension, and practical burdens faced by participants. Addressing these barriers through broader inclusion criteria, participant education, incentives, and flexible scheduling is essential to enhance recruitment, trial generalizability, and future clinical applicability of NIBS in addiction treatment.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"459-478"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145253297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Theta burst stimulation (TBS) in methamphetamine use disorder: a meta-analysis and systematic review.","authors":"Gopalkumar Rakesh, Thomas G Adams, Dylan H Ballard, Christopher J McLouth, Craig R Rush","doi":"10.1080/10550887.2025.2541165","DOIUrl":"10.1080/10550887.2025.2541165","url":null,"abstract":"<p><p>Methamphetamine use disorder (MUD) currently lacks FDA-approved treatments. Previous studies involving transcranial magnetic stimulation (TMS) over the left dorsolateral prefrontal cortex (L.dlPFC) have shown promise in TMS's effectiveness at decreasing craving for methamphetamine. Theta burst stimulation (TBS), which includes intermittent (iTBS) and continuous (cTBS) protocols, is increasingly being used for substance use disorders, including MUD. Previous reviews of TMS in MUD performed subgroup meta-analyses of studies that delivered TBS in MUD. However, these meta-analyses included studies with overlapping participant cohorts, limiting their validity. To address this limitation, we reviewed randomized controlled trials (RCTs) using unique patient cohorts from PubMed/Medline, EMBASE, and Google Scholar until September 1, 2024, comparing the impact of TBS versus sham on cue-induced methamphetamine craving in patients with MUD. We performed a meta-analysis with four eligible RCTs that delivered iTBS. Results suggest iTBS was more effective in reducing cue-induced methamphetamine craving than sham iTBS (standardized mean difference [SMD] in change = 1.04; 95% CI [0.16, 1.92]). Two additional RCTs without sham control arms were reviewed, and one demonstrated a significant reduction in craving following accelerated iTBS. Future studies should examine whether neuroimaging-targeted iTBS can impact outcome measures other than craving, such as methamphetamine use, by measuring return to use. Exploring accelerated iTBS and cTBS for MUD, targeting alternative cortical sites such as the frontal pole, and studying their effects on relevant MUD biomarkers is also pertinent. This review demonstrates the effectiveness of TBS for MUD, emphasizing its potential to advance treatment options for MUD.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"446-458"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144776561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sequential dual-target tDCS for tobacco use disorder: A longitudinal pilot case series targeting craving and inhibitory control.","authors":"Thiago P Fernandes, Givago S Souza, Natanael A Santos","doi":"10.1080/10550887.2026.2676591","DOIUrl":"10.1080/10550887.2026.2676591","url":null,"abstract":"<p><strong>Background: </strong>Transcranial direct current stimulation (tDCS) is increasingly explored as an adjunctive intervention for Tobacco Use Disorder (TUD). Studies typically employ single-site stimulation protocols, and reported effects on craving and relapse have varied.</p><p><strong>Objective: </strong>To examine whether sequential dual-target tDCS could influence smoking behavior and relapse-relevant mechanisms in TUD.</p><p><strong>Methods: </strong>In this case series, three adults with TUD underwent a two-week sequential tDCS protocol targeting the left dorsolateral prefrontal cortex and pre-supplementary motor area. Assessments were conducted at baseline, post-intervention, and at 1- and 6-month follow-up characterize clinical, biological, and cognitive trajectories over time.</p><p><strong>Results: </strong>All cases showed reductions in cigarette consumption across the six-month follow-up, accompanied by decreases in biochemical exposure markers. Trajectories were heterogeneous across participants. Two individuals achieved abstinence at follow-up, while one maintained sustained reduction. Craving and withdrawal symptoms declined, and inhibitory control improved alongside reductions in exposure markers. The nicotine metabolite ratio remained stable across assessments, consistent with its role as an index of metabolic phenotype.</p><p><strong>Conclusions: </strong>Sequential dual-target tDCS was associated with longitudinal improvements in smoking behavior and relapse-related clinical and cognitive markers. These findings support further investigation of dual-target neuromodulation as a potential adjunctive strategy for smoking cessation.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"568-578"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148273342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predictive value of the Clinical Institute Withdrawal Assessment for Alcohol Scale, Revised (CIWA-Ar) score on admission for alcohol withdrawal delirium (AWD) in patients with alcohol use disorder (AUD): A prospective observational study.","authors":"Masahiro Uzawa, Shin Inuzuka, Masayo Adachi, Yosuke Yoshizaki, Hitoshi Mutai, Tokiji Hanihara","doi":"10.1080/10550887.2026.2616724","DOIUrl":"10.1080/10550887.2026.2616724","url":null,"abstract":"<p><strong>Background: </strong>Alcohol withdrawal delirium (AWD) is the most severe and potentially fatal manifestation of alcohol withdrawal syndrome (AWS). Early identification of patients at risk is critical for preventing medical complications and optimizing withdrawal management. Although the Clinical Institute Withdrawal Assessment for Alcohol Scale, Revised (CIWA-Ar), is widely used to assess AWS severity, its predictive value for AWD remains unclear, particularly in Asian inpatient settings.</p><p><strong>Objective: </strong>To examine the incidence of AWS and AWD among patients with alcohol use disorder (AUD) in a Japanese rehabilitation center and to identify admission factors associated with AWD onset.</p><p><strong>Methods: </strong>Among 295 consecutively admitted patients, 273 were analyzed; 22 were excluded for ongoing AWD at admission. AWS was evaluated using CIWA-Ar throughout hospitalization. Associations between admission variables and AWD onset were analyzed using logistic regression and receiver operating characteristic (ROC) analysis.</p><p><strong>Results: </strong>Of 273 patients, 152 (55.7%) exhibited AWS and 24 (8.8%) developed AWD during hospitalization. Stepwise logistic regression identified only the CIWA-Ar score at admission as a significant predictor (odds ratio = 1.50; 95% confidence interval, 1.28-1.74; <i>p</i> < 0.001). ROC analysis yielded an area under the curve (AUC) of 0.906 with a CIWA-Ar cutoff of 4 (sensitivity 91.7%, specificity 84.3%).</p><p><strong>Conclusions: </strong>The admission CIWA-Ar score is a strong and practical predictor of AWD in AUD inpatients. Even mild withdrawal symptoms may indicate early autonomic hyperactivity and high AWD risk, underscoring the need for vigilant monitoring and timely intervention.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"549-557"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146041718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the clinical competence of large language models for tobacco use disorder: A multi-domain expert evaluation.","authors":"Thiago P Fernandes, Linnea Dahlgren, Natanael A Santos, Zeke Degraff","doi":"10.1080/10550887.2026.2636620","DOIUrl":"10.1080/10550887.2026.2636620","url":null,"abstract":"<p><strong>Background: </strong>Tobacco use disorder (TUD) remains the leading preventable cause of death globally, yet fewer than one-third of users receive guideline-concordant care due to workforce shortages and training gaps. Emerging artificial intelligence (AI) systems, particularly large language models (LLMs), may help expand access to evidence-based cessation support, but their clinical competence remains insufficiently characterized.</p><p><strong>Objective: </strong>To systematically evaluate the clinical accuracy, safety, guideline adherence, and communication quality of five leading LLMs across standardized tobacco-cessation scenarios.</p><p><strong>Methods: </strong>We developed 84 clinical vignettes, covering five core competency domains (screening, diagnosis, pharmacotherapy, behavioral counseling, and harm reduction) based on DSM-5-TR, U.S. Public Health Service Guidelines, NICE NG209, ASAM guidelines, SAMHSA protocols, and the WHO MPOWER framework. Five LLMs, GPT-4.5, Claude 3.5 Sonnet, Gemini 2.5 Pro, Llama 3.1-70B, and DeepSeek-V3, were independently evaluated by four addiction-medicine experts across clinical accuracy, guideline adherence, safety, and clinical utility.</p><p><strong>Results: </strong>GPT-4.5 and Claude 3.5 Sonnet achieved the highest composite scores (<i>M</i> = 4.25, SD = 0.68; <i>M</i> = 4.18, SD = 0.71), with 74-78% of responses rated ≥4.0 and superior safety performance (88% ≥4.0). Gemini 2.5 Pro showed moderate performance (<i>M</i> = 3.72; 52% ≥4.0). Open-weight models (Llama 3.1-70B, M = 3.48; DeepSeek-V3, M = 3.35) lagged behind overall, with only 34-38% achieving the ≥4.0 benchmark.</p><p><strong>Conclusion: </strong>All evaluated AI systems demonstrated competence in tobacco-cessation counseling, but GPT-4.5 and Claude 3.5 Sonnet reached performance levels consistent with supervised clinical use and safety-critical scenarios. Clinician oversight remains essential for all medication-based interventions, and open-weight models warrant further validation before consideration for clinical implementation.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"413-422"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147366799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Balancing transition and readmission: a comparative study of low-dose vs standard buprenorphine initiation for medically managed opioid withdrawal.","authors":"Tara Nichols, Sean T Lynch, Emily Groenendaal, Eldene Towey, Stephen Ferrando","doi":"10.1080/10550887.2026.2625026","DOIUrl":"10.1080/10550887.2026.2625026","url":null,"abstract":"<p><strong>Background: </strong>The rising prevalence of fentanyl has complicated buprenorphine (BUP) initiation due to increased risks for precipitated withdrawal. Low-dose initiation (LDI), or \"micro-dosing\", has been proposed as an alternative to standard-dose initiation (SDI) to improve tolerability and engagement with care. This study aimed to compare inpatient and post-discharge outcomes of LDI versus SDI among patients with opioid use disorder (OUD) undergoing medically managed withdrawal.</p><p><strong>Methods: </strong>This retrospective cohort study included adults admitted for medically managed OUD withdrawal during two time periods: SDI (February 2020-April 2021) and LDI (May 2021-August 2022). Sociodemographic, clinical, and treatment data were extracted from electronic medical records. Outcomes included inpatient length of stay (LOS), BUP acceptance and discharge dose, transition to rehabilitation, readmissions within one year, BUP refill rates, and outpatient treatment engagement. Analyses included logistic regression and negative binomial regression with adjustment for baseline urine drug screen results.</p><p><strong>Results: </strong>The sample included 184 patients (127 SDI, 57 LDI) with a mean age of 37 years, predominately White N=140 (76.1%) and female. Groups were demographically similar. Inpatient outcomes did not differ in LOS, BUP acceptance, or discharge dose. The LDI group was less likely to transition to rehabilitation (26% vs 48%). LDI was associated with significantly lower odds of readmission within 1 year, while 30-day readmission rates showed a similar directional trend that did not reach statistical significance. Post-discharge BUP refill rates and outpatient attendance rates were similar between groups.</p><p><strong>Conclusions: </strong>LDI and SDI produced comparable outcomes across most measures, with LDI associated with fewer readmissions over one year. Both approaches appear to be viable strategies in the fentanyl era.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"558-567"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13052367/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147345415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An unsupervised machine learning analysis of biopsychosocial characteristics and treatment outcome of alcohol use disorder.","authors":"Aravind Karunakaran, Aparna Sasikumar, Jaimon Plathottathil Michael, Akhil Das, Varghese Punnoose","doi":"10.1080/10550887.2026.2696426","DOIUrl":"10.1080/10550887.2026.2696426","url":null,"abstract":"<p><p><b>Background:</b> Alcohol Use Disorder is a heterogeneous condition where standard severity measures often fail to predict individual treatment responses. Precision medicine requires identifying distinct biopsychosocial profiles to guide targeted interventions.<b>Objectives:</b> To identify clinically meaningful Alcohol Use Disorder profiles using k-means clustering based on eight baseline biopsychosocial variables and validate their prognostic utility by comparing treatment outcomes.<b>Methods:</b> A retrospective observational study included 102 patients at a tertiary care center in India. K-means clustering was applied to baseline variables: age, Alcohol use duration, alcohol severity, craving, co-occurring psychiatric conditions, self-efficacy, education, and occupation. Cluster stability was assessed <i>via</i> bootstrap resampling. The primary outcome was 3-month abstinence, corroborated by GGT levels.<b>Results:</b> Three distinct profiles emerged: (1) Late-Onset (<i>n</i> = 38), with older age and lowest craving, achieving 65.8% abstinence; (2) High-Functioning (<i>n</i> = 34), defined by high socioeconomic status and zero co-occurring disorders; and (3) Severe (<i>n</i> = 30), characterized by early onset, high co-occurring psychiatric conditions (73.3%), and intense craving. Despite similar alcohol severity scores (<i>p</i> = 0.215), the Severe profile group had poorer outcomes (36.7% abstinence; <i>p</i> = 0.004) and a steep return to use trajectory.<b>Conclusion:</b> Identifying three distinct Alcohol Use Disorder profiles through k-means cluster analysis advances precision medicine in substance use treatment. Return to use risk was driven by craving and co-occurring conditions rather than AUD severity. These findings support a stratified approach, where Severe profile group patients require immediate, intensive \"front-loaded\" intervention.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"432-443"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148686153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-world deployment of machine learning models for opioid overdose and opioid use disorder: a systematic review of clinical and operational lessons for addiction medicine.","authors":"Kirolos Eskandar","doi":"10.1080/10550887.2025.2607456","DOIUrl":"10.1080/10550887.2025.2607456","url":null,"abstract":"<p><strong>Background: </strong>Machine learning (ML) is increasingly explored for opioid overdose and opioid use disorder (OUD) detection and prevention. Regional burden is uneven: the United States currently has among the highest rates of drug-overdose deaths worldwide, underscoring the urgent need for operationalized ML tools in US clinical and public-health settings. While many models exist, few have been tested in live settings. Understanding how deployed systems perform and are governed is essential for safe, equitable use in addiction medicine.</p><p><strong>Methods: </strong>We systematically searched PubMed, Embase, IEEE Xplore, and Scopus from January 2019 to August 2025 for studies describing real-world ML deployments for overdose or OUD. Eligible studies required live integration into clinical, public health, or consumer workflows with prospective or concurrent evaluation. Risk of bias and operational robustness were assessed using the PROBAST-R framework (a PROBAST extension for deployed ML; 'risk of bias' here denotes systematic error in reported performance due to study design, data, or reporting).</p><p><strong>Results: </strong>Fifteen studies were included, spanning health systems, public health surveillance, emergency services, and wearable detection. Most achieved good discrimination (AUC > 0.80) or high precision, though tradeoffs between sensitivity and specificity were common. Governance structures were more consistently reported in large deployments, yet no study described automated monitoring, and only two examined subgroup performance. Fairness audits and human-factors evaluations were rare.</p><p><strong>Conclusion: </strong>Deployed ML for opioid outcomes is feasible but uneven in maturity. Clinical implications: addiction medicine clinicians should (1) request subgroup performance results before adoption; (2) confirm post-deployment monitoring plans for drift and calibration; and (3) use human-in-the-loop safeguards (clinician or human verification before automated actions) to reduce harms from false positives/negatives.</p>","PeriodicalId":47493,"journal":{"name":"Journal of Addictive Diseases","volume":" ","pages":"380-398"},"PeriodicalIF":1.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145960572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}