European Addiction Research最新文献

{"title":"Identify Candidate Intervention and Prevention Targets for Internet Gaming Disorder: Evidence from Network Analysis and Simulation Interventions.","authors":"Chenguang Zhao, Wei Zhang","doi":"10.1159/ear/acrag002","DOIUrl":"10.1159/ear/acrag002","url":null,"abstract":"<p><strong>Introduction: </strong>Internet gaming disorder (IGD) is characterized by persistent and recurrent gaming behavior that can lead to negative psychosocial outcomes. Although the network theory of psychopathology suggests that mental disorders arise from interactions among symptoms, few studies have empirically evaluated how targeting core symptoms affects the broader symptom network.</p><p><strong>Methods: </strong>This study aimed to identify candidate prevention and intervention targets for IGD using network analysis combined with simulation-based interventions.</p><p><strong>Results: </strong>Results found that compulsivity emerged as the candidate intervention target under alleviating simulation, whereas loss of interest in alternative activities represented a key candidate prevention target under aggravating simulation.</p><p><strong>Conclusion: </strong>The present findings may help prioritize symptoms for future longitudinal and experimental studies designed to test whether changes in these symptoms produce broader changes in IGD symptomatology. Moreover, this study demonstrates the methodological utility of simulation-based interventions, offering a framework for exploring targeted symptom manipulation in other behavioral and substance-related addictions.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1"},"PeriodicalIF":1.8,"publicationDate":"2026-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148817903","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":"Erratum.","authors":"","doi":"10.1159/000553055","DOIUrl":"https://doi.org/10.1159/000553055","url":null,"abstract":"<p><p>In the article by Heckel et al. entitled \"Variations in Regulated Cannabis Sales Conditions and Their Effect on Purchase Behaviour: Findings from an Observational Study\" [Eur Addict Res. 2026;32(1):46-55; https://doi.org/10.1159/000551031], an error was introduced during production whereby the affiliation \"Addictive Disorders Research Group\" was mistakenly shown as a co-author. The name of the author Eva Maria Havelka was also mistakenly shown as \"Eva-Maria Havelka.\"The original article has been updated.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1"},"PeriodicalIF":1.8,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148677618","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":"Enhanced and Prolonged Euphoria during Co-Use of Alcohol and Cocaine: A Narrative and Integrative Review.","authors":"Jan van Amsterdam, Wim van den Brink","doi":"10.1159/000553048","DOIUrl":"10.1159/000553048","url":null,"abstract":"<p><strong>Introduction: </strong>Alcohol and cocaine are frequently used simultaneously, with alcohol typically consumed prior to cocaine. Qualitative studies indicate that this combination enhances and prolongs euphoria compared to cocaine alone. The aim of this study was to collect quantitative data that were published several decades ago, but have never been integrated.</p><p><strong>Methods: </strong>A narrative and integrative review was conducted to collect and synthetize research data from quantitative controlled human studies on the mechanisms responsible for enhanced and prolonged euphoric sensations during co-use of alcohol and cocaine compared to cocaine alone. A total of 1,012 unique studies were retrieved via Medline (PubMed), Embase, PsycInfo, and Google Scholar of which 24 studies published between 1987 and 2011 met the eligibility criteria.</p><p><strong>Results: </strong>The current narrative and integrative review shows that the co-use of alcohol and cocaine potentiates and prolongates cocaine's euphoric effects. First, when alcohol is used prior to oral or intranasal cocaine, it inhibits cocaine metabolism, resulting in higher peak plasma levels (+18-52%) and increased bioavailability (+15-40%) of cocaine. Second, alcohol plus cocaine generates cocaethylene (12-34% conversion) which, compared to cocaine, induces a similar or somewhat less intense but more prolonged euphoria. Third, in contrast to cocaine, cocaethylene selectively inhibits dopamine reuptake resulting in higher synaptic dopamine levels, which may explain its greater reinforcement compared to cocaine.</p><p><strong>Conclusion: </strong>Co-use of alcohol and cocaine enhances and prolongs the subjective and reinforcing effects of cocaine. Although this may be perceived as an efficient way of drug use, it also poses serious cardiovascular risks.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-12"},"PeriodicalIF":1.8,"publicationDate":"2026-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13461128/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148619035","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":"Using Machine Learning with the Brief Symptom Inventory to Screen for Comorbid Addictions: A Proof-of-Principle Study.","authors":"Maor Daniel Levitin, Dvora Shmulewitz, Roi Eliashar, Shaul Lev-Ran, Mario Mikulincer","doi":"10.1159/000553511","DOIUrl":"10.1159/000553511","url":null,"abstract":"<p><strong>Introduction: </strong>Mental health disorders are leading causes of morbidity worldwide and often co-occur with substance use and other addictive behaviors. However, many individuals in mental healthcare settings are not screened for addictions. General psychopathology measures, such as the Brief Symptom Inventory (BSI), may offer valuable insights into addiction risk. This study explores whether machine learning models can utilize BSI responses to predict problematic substance use (alcohol, drugs) and addictive behaviors (gambling, gaming, hypersexual behavior, and pornography use).</p><p><strong>Methods: </strong>A population sample of Jewish adults in Israel (N = 2,451) was assessed for mental health, including the BSI, problematic substance use (alcohol, drugs), and potentially addictive behaviors: gambling, gaming, compulsive sexual behavior, and pornography use. Machine learning models - including decision trees, random forest, boosting, LASSO, subset selection, and elastic net regression - were employed to predict addiction outcomes based on BSI responses.</p><p><strong>Results: </strong>The BSI demonstrated varying degrees of predictive ability across different addictions and models. Test set explained variance (R2) on the continuous track ranged from 1% (gambling, decision tree) to 25% (compulsive sexual behavior, random forest); area under the precision-recall curve (AUPRC) ranged from 0.04 (pornography, LASSO) to 0.36 (gaming, decision tree), in each case exceeding the prevalence baseline. Behavioral addictions generally showed descriptively higher predictability than substance use disorders.</p><p><strong>Conclusions: </strong>The BSI contains substantial information on addiction risk, providing preliminary, proof-of-principle evidence that an existing general psychopathology measure carries item-level signal relevant to comorbid addiction screening, with the strongest signal for behavioral addictions. These findings serve as proof of principle for the notion that existing psychopathology measures can be leveraged to inform about potential comorbidity. Being able to efficiently identify which mental health patients must be screened for addictions will help to determine the most appropriate interventions.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-16"},"PeriodicalIF":1.8,"publicationDate":"2026-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13533478/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148419859","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":"There, but Invisible: Warning Messages in Czech Gambling Advertising during Football Broadcasts.","authors":"Filip Kovařík, Steve Sharman, David Fiedor","doi":"10.1159/000553513","DOIUrl":"10.1159/000553513","url":null,"abstract":"<p><strong>Introduction: </strong>Gambling advertising during live sports broadcasts raises significant public health concerns, yet regulatory compliance (particularly regarding warning messages) remains underexplored. This study aims to evaluate how common warning messages are and how much screen space they occupy in gambling adverts during live top-tier Czech football broadcasts.</p><p><strong>Methods: </strong>A total of 1,824 gambling adverts were systematically analysed across 32 top-tier Czech football matches. Variables included the presence of mandatory warnings (about age restrictions and potential addiction) and safe gambling messages (SGMs), along with their delivery methods, gambling types, match segments, and screen coverage. Warning message sizes were measured using pixel-by-pixel calculations. Nonparametric tests (Mann-Whitney U test and Cramér's V) were used for statistical analysis.</p><p><strong>Results: </strong>Overall, 77.7% of all adverts complied with legal requirements. SGM appeared in 58.1% of adverts. Warning messages occupied an average of 1.58% of the screen (SD = 0.47%), with significantly greater coverage in adverts featuring an SGM (p < 0.001). Screen coverage varied significantly across different gambling types (p < 0.001) but not across match segments (p = 0.834).</p><p><strong>Conclusion: </strong>The limited screen coverage and compliance issues identified highlight notable regulatory gaps, particularly regarding minimum size, font, and colour contrast requirements for warning messages. The simultaneous display of warnings alongside promotional content also reduces their effectiveness. Current regulations should be strengthened to better protect consumers from gambling-related harm.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-12"},"PeriodicalIF":1.8,"publicationDate":"2026-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13472589/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148419813","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":"Preliminary Evidence of Inhibitory and Error-Related Neural Alterations Associated with Early Relapse in Cocaine Use Disorder: A Retrospective Go/No-Go Event-Related Potential Study.","authors":"Anaïs Ingels, Christie Chenut, Alix Pendry, Félix Hever, Clémence Dousset, Hendrik Kajosch, Matthieu Hein, Salvatore Campanella","doi":"10.1159/000553386","DOIUrl":"10.1159/000553386","url":null,"abstract":"<p><strong>Introduction: </strong>Cocaine use disorder (CUD) is associated with impairments in inhibitory control and error-monitoring, two neurocognitive mechanisms strongly linked to relapse. Characterizing how these processes differ at treatment entry between patients who later remain abstinent versus those who relapse may help clarify the neurophysiological factors associated with early recovery.</p><p><strong>Methods: </strong>Thirty-one CUD patients undergoing a 15-day inpatient detoxification program were included and classified as abstainers (n = 16) or relapsers (n = 15) based on drug use reports 1 month after discharge. Participants completed an EEG Go/No-Go task approximately 4 to 7 days after admission, once acute withdrawal symptoms had subsided. Inhibitory (N2d, P3d) and error-related (ERN, PE) ERP components were extracted. Multivariate analyses controlled for psychotropic medication status.</p><p><strong>Results: </strong>No behavioral differences were observed between groups. However, abstainers showed larger P3d and ERN amplitudes, reduced N2d and PE amplitudes, and longer P3d latencies than relapsers. This profile suggests more efficient inhibitory processing and sharper early error monitoring in abstainers, while relapsers exhibited attenuated error detection and greater reliance on later evaluative processes.</p><p><strong>Conclusion: </strong>The findings indicate that CUD patients differ markedly in their inhibitory and error-monitoring neural responses already at treatment entry. ERP measures may therefore constitute clinically relevant indicators of cognitive functioning in early detoxification and could support more individualized therapeutic strategies.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-15"},"PeriodicalIF":1.8,"publicationDate":"2026-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13480973/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148403794","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":"Linking Affective Symptoms and Brain Morphology in Cocaine Use Disorder: An Integrated Network Approach.","authors":"Diego Angeles-Valdez, Said Jiménez, Marie-José van Tol, Eduardo A Garza-Villarreal","doi":"10.1159/000553421","DOIUrl":"10.1159/000553421","url":null,"abstract":"<p><strong>Introduction: </strong>Cocaine use disorder (CUD) often co-occurs with depressive and anxiety symptoms, suggesting shared neurobiological bases. This study aimed to understand the relationship between these heterogeneous individual symptoms and structural brain characteristics in CUD using an integrated network approach.</p><p><strong>Method: </strong>Data from 116 individuals (17 women) with CUD (SUDMEX project) were analyzed. We used item-level depression and anxiety scores from the SCL-90-R, along with measures of cortical thickness (insula, inferior/middle frontal gyrus, anterior cingulate cortex) and hippocampal volume. Four brain-behavior integrated networks were estimated: (1) brain regions with depressive symptoms, (2) brain regions with anxiety symptoms, (3) brain regions with global depression scores, and (4) brain regions with global anxiety scores.</p><p><strong>Results: </strong>The symptom-level networks revealed specific associations. The insula was a central finding, showing negative associations in the depression network with Crying (partial r = -0.078) and Self-blame (partial r = -0.024), and in the anxiety network with Being/Feeling Scared (partial r = -0.058). Furthermore, insular thickness was negatively associated with the global depression score (partial r = -0.068). These associations were not moderated by duration of cocaine use.</p><p><strong>Conclusion: </strong>The findings identify the insula as a potential neural hub linking affective symptoms with brain morphology in CUD. This integrated approach offers a more nuanced understanding of the complex relationships between psychiatric symptoms and brain structure in CUD.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-13"},"PeriodicalIF":1.8,"publicationDate":"2026-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13577654/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148403791","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":"Benchmarking Motivational Interviewing Competence of Large Language Models.","authors":"Aishwariya Jha, Prakrithi Shivaprakash, Lekhansh Shukla, Animesh Mukherjee, Prabhat Chand, Pratima Murthy","doi":"10.1159/000553455","DOIUrl":"10.1159/000553455","url":null,"abstract":"<p><strong>Introduction: </strong>Motivational interviewing (MI) promotes behavioural change in substance use disorders. Its fidelity is measured using the Motivational Interviewing Treatment Integrity (MITI) framework. While large language models (LLMs) can potentially generate MI-consistent therapist responses, their competence using MITI is not well-researched, especially in real-world clinical transcripts. We aim to benchmark MI competence of proprietary and open-source models compared to human therapists in real-world transcripts and assess distinguishability from human therapists.</p><p><strong>Methods: </strong>We shortlisted 3 proprietary and 7 open-source LLMs from LMArena, evaluated performance using MITI 4.2 framework on two datasets (96 handcrafted model transcripts and 34 real-world clinical transcripts). We generated parallel LLM-therapist utterances iteratively for each transcript while keeping client responses static and ranked performance using a composite ranking system with MITI components and verbosity. We conducted a distinguishability experiment with two independent psychiatrists to identify human-versus-LLM responses.</p><p><strong>Results: </strong>All 10 tested LLMs had fair (MITI global scores of >3.5) to good (MITI global scores of >4) competence across MITI measures, and the three best-performing models (gemma-3-27b-it, gemini-2.5-pro, and grok-3) were tested on real-world transcripts. All showed good competence, with LLMs outperforming human-expert in complex reflection percentage (39% vs. 96%) and reflection-question ratio (1.2 vs. >2.8). In the distinguishability experiment, psychiatrists identified LLM responses with only 56% accuracy, with d-prime of 0.17 and 0.25 for gemini-2.5-pro and gemma-3-27b-it, respectively.</p><p><strong>Conclusion: </strong>LLMs can achieve good MI proficiency in real-world clinical transcripts using the MITI framework. However, a high complex reflection percentage may result in a technically correct but unnatural conversation style. These findings suggest that even open-source LLMs are viable candidates for expanding MI counselling sessions in low-resource settings, warranting further rigorous clinical validation.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-12"},"PeriodicalIF":1.8,"publicationDate":"2026-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13472593/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148387478","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":"Validation of the English Social Attunement Questionnaire: Associations with Substance Use, Impulsivity, and Sensation Seeking across the Lifespan.","authors":"Emese Kroon, Karis Colyer-Patel, Christophe Romein, Janna Cousijn","doi":"10.1159/000553273","DOIUrl":"10.1159/000553273","url":null,"abstract":"<p><strong>Introduction: </strong>Our social environment strongly influences behaviors like substance use and sensation seeking. While most research focuses on group pressure and conformity, the role of social attunement - the tendency to align with one's social environment for social reward without explicit pressure - has been largely overlooked.</p><p><strong>Methods: </strong>This study evaluated the English version of the Social Attunement Questionnaire (SAQ) and examined its links to age, alcohol and cannabis use, impulsivity, and sensation seeking (N = 447; 54.5% female).</p><p><strong>Results: </strong>Confirmatory factor analysis showed acceptable to good fit of the originally proposed two factor structure, with the one factor solution showing a substantially worse fit. Internal consistency was good across the full scale (α = 0.82) and acceptable for the Cognition subscale (α = 0.75) and the Behavior subscale (α = 0.78). The SAQ also showed similar scale and subscale scores and measurement invariance across sex. Higher social attunement was observed in younger individuals. SAQ scores - particularly the behavior subscale - were positively associated with alcohol and cannabis use and with impulsivity. Age moderated these associations: among younger individuals, higher social attunement was linked to greater alcohol use and sensation seeking, while these associations reversed with age.</p><p><strong>Conclusion: </strong>These findings suggest that social attunement may act as both a risk and protective factor for risk-taking behaviors, depending on developmental stage. The English SAQ demonstrated solid psychometric properties across its two subscales - Cognition and Behavior - while also proving suitable for use as a single composite scale, supporting its flexibility as a reliable measure of social attunement.</p>","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-11"},"PeriodicalIF":1.8,"publicationDate":"2026-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13450979/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148359988","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":"Integrating Psychosocial Factors into Artificial Intelligence Models for Predicting Addiction Treatment Outcomes: A Systematic Review.","authors":"Guillermo Francisco Martínez-Montás, Marina Iniesta-Sepúlveda, Orquiria Rafaela Colón-Santana","doi":"10.1159/000553186","DOIUrl":"10.1159/000553186","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Introduction: &lt;/strong&gt;The integration of artificial intelligence (AI) into addiction research has expanded rapidly, yet it remains unclear how psychosocial, behavioral, and social-structural determinants are incorporated into predictive models of addiction treatment outcomes. Because recovery is strongly shaped by psychological, social, and environmental context, assessing how AI approaches operationalize these dimensions is essential for developing clinically meaningful and equitable tools.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We conducted a systematic review of peer-reviewed studies published from January 1, 2020 to October 30, 2025 across PubMed, Scopus, and Web of Science. Eligible studies applied AI or machine learning (ML) models to addiction treatment outcomes and explicitly included psychosocial, behavioral, or social-structural predictors. Two reviewers independently screened studies, extracted data, and evaluated methodological quality using Joanna Briggs Institute and Cochrane Risk of Bias 2 domain structures. The protocol was prospectively registered on the Open Science Framework and in PROSPERO.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Fifteen studies met inclusion criteria, including electronic health record (EHR), administrative-, claims-based, program-level clinical models and psychosocial assessment datasets, digital phenotyping/ecological momentary assessment (EMA) studies, natural language processing/large language model approaches, and one causal ML analysis of randomized controlled trial data. Across modalities, models consistently identified housing instability, psychiatric comorbidity, employment status, craving, stress, legal involvement, prior overdose, treatment history, medication adherence, and neighborhood disadvantage as influential predictors of treatment dropout, discontinuation, overdose risk, relapse, or poor engagement - often adding prognostic value beyond medication-related, diagnostic, and routinely available clinical variables. EMA and digital phenotyping showed the highest short-term predictive accuracy for near-term risk prediction, whereas structured EHR-, administrative-, claims-based, and program-level clinical models achieved moderate but clinically actionable performance. Methodological quality was moderate overall, with limited external validation and infrequent assessment of calibration, fairness, transportability, or reproducibility practices.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;Current evidence indicates that psychosocial, behavioral, and social-structural determinants are central to AI-based prediction of addiction treatment outcomes. Although findings are promising, existing models remain preliminary and should not yet guide clinical decisions without external validation and implementation evaluation. Future work should prioritize multisite validation, transparent reporting, fairness evaluation, and co-development with clinicians and individuals with lived experience to ensure that AI tools strengt","PeriodicalId":11902,"journal":{"name":"European Addiction Research","volume":" ","pages":"1-18"},"PeriodicalIF":1.8,"publicationDate":"2026-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13441268/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148283476","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}