Addiction最新文献

Background and aims: Tobacco smoking is a major risk factor for cardiovascular and lung diseases. A better understanding of its neurobiological underpinnings will benefit the prevention of smoking-related illnesses and mortality. Previous studies link smoking to increased iron concentration in the striatum, a central component of the brain's reward system, and to reduced cognitive performance. This study aimed to investigate whether smoking and striatal iron share common biological pathways and to assess potential causal relationships between the two.

Methods: Using data from the UK Biobank, we investigated phenotypic and genetic correlations, and causal relationships between smoking initiation and magnetic resonance imaging (MRI)-derived markers of iron content (T2* and quantitative susceptibility mapping) in the bilateral putamen, caudate and accumbens nuclei.

Results: We found positive correlations between smoking and striatal iron (β ∈ [0.03, 0.40], P < 0.001), particularly when comparing current smokers with never smokers. Striatal iron was positively associated with pack-years (β ∈ [0.11, 0.13], P < 0.001) and inversely related to years since smoking cessation (β ∈ [0.06, 0.10], P < 0.001), suggesting iron levels may decrease after quitting. Genetic analysis confirmed phenotypic correlations, with shared genetic associations (P < 2.73 × 10^-6, or 0.01 for candidate genes) in genes related to dopaminergic, glutamatergic and synaptic systems (DRD2, PPP1R1B, NCAM1, DLX5, GGACT, NAT16, PLEKHM1). Causality analysis revealed a relationship from smoking to striatal iron via genes involved in synaptogenesis and plasticity (BAI3, SEMA6D, TENM2), with evidence of reverse causality from iron to smoking through inflammatory and immune system-related genes (ING5, NLRP7).

Conclusions: There appear to be links between smoking and striatal iron with complex causal mechanisms involving synaptic transmission and inflammatory circuits. Striatal iron content could serve as a biomarker for smoking-related neurobiological changes and a potential target for interventions aimed at mitigating cognitive decline related to striatal iron accumulation.

Background and aims: Thanks to smart devices, social media and streaming platforms, watching videos, like movies or short social media clips, has become extremely popular. Alcohol portrayals are frequent in videos, yet their prevalence is difficult to quantify using traditional methods such as manual coding. Artificial intelligence (AI) offers a scalable solution to analyse large volumes of video images. This study aimed to compare the accuracy of three AI models in detecting alcohol presence in video images.

Method: Experimental evaluation of three models: one supervised deep learning model (ABIDLA2) and two zero-shot learning models (ZSL-CLIP and ZSL-LLaVA). The models were tested on datasets of video frames that had been annotated by researchers for whether they included alcohol or not. Three datasets of increasing complexity were used: (1) a Google/Bing image set of clearly visible alcohol and non-alcohol images; (2) a set of movie frames manually annotated as containing or not containing alcohol; and (3) a contextually challenging set of movie frames from alcohol-related settings (e.g. bars, parties) that may or may not include visible alcohol. Model performance was assessed using accuracy, unweighted average recall (UAR) and F1 score, representing the balance between precision and recall. Execution time per frame was also measured to evaluate computational efficiency.

Results: Across the three datasets, ABIDLA2, ZSL-CLIP and ZSL-LLaVA achieved percentage accuracies of 90%, 91% and 92% on the Google/Bing images; 70%, 65% and 95% on the diverse movie-scene dataset; and 67%, 63% and 94% on the most complex alcohol-related dataset, respectively. In terms of execution time, ABIDLA2 processed a single frame the fastest (0.21 seconds), followed by ZSL-LLaVA (0.45 seconds), while ZSL-CLIP was the slowest (0.58 seconds).

Conclusion: Automated artificial intelligence (AI) models appear to be able to detect alcohol imagery in videos at large scale with high accuracy and in near real time. Of the three AI models tested, ZSL-LLaVA achieved the best balance between accuracy and speed. Offering a cost- and time-efficient alternative to labour-intensive manual coding, ZSL-LLaVA could be used to monitor alcohol-related visual content in videos across diverse media platforms.

Background and aims: Substance-related emergency department (ED) visits represent a critical opportunity to link individuals with sustained treatment for substance use disorders (SUD), yet few transition to post-acute treatment. Among emerging initiatives to integrate specialized SUD care into ED services, the Stockholm SUD-ED is an example of a fully integrated model that remains unmapped. This study aimed to systematically map care pathways and identify predictors of post-acute SUD treatment engagement among SUD-ED patients in Stockholm.

Design: Retrospective cohort study. Electronic health records data were linked with five registries covering, among other aspects, clinical history and healthcare consumption.

Setting: The Stockholm SUD-ED, Sweden.

Participants: n = 9771 SUD-ED patients during a sixteen-month period (2018-2019).

Measurements: For care flow mapping, post-acute SUD treatment engagement was defined as ≥2 outpatient visits during a six-month follow-up. Four engagement levels (none, low, moderate, high) were derived using a quantile regression approach and regressed on 19 candidate predictors in multinomial logistic regression models.

Findings: Of all SUD-ED patients, one-third (33.2%; n = 3248) primarily engaged in post-acute outpatient SUD treatment, 16.9% (n = 1651) primarily engaged in non-SUD psychiatric services and 49.9% (n = 4872) did not engage in either. Police-initiated admissions had lower odds of moderate or high post-acute treatment engagement than ambulance-initiated admissions [high engagement: adjusted odds ratio (aOR) = 0.71, 95% confidence interval (CI) = 0.53-0.95]. Prior needle and syringe program visits increased the odds of post-acute treatment engagement (aOR = 2.09, 95% CI = 1.59-2.75). Lack of prior SUD outpatient treatment was consistently associated with lower odds of post-acute engagement across all levels (e.g. moderate engagement: aOR = 0.3, 95% CI = 0.25-0.35).

Conclusions: The Stockholm model for fully integrating specialized substance use disorder care into emergency departments appears to generate high post-acute treatment engagement and highlights the need for (1) targeted interventions for police-initiated admissions and (2) broad aftercare options to better attract treatment-naïve patients.

Background and aims: There has been limited evidence synthesis examining the treatment of buprenorphine precipitated opioid withdrawal (BPOW). We aimed to conduct the first systematic review to assess the clinical utility of any pharmacological intervention in the management of BPOW.

Methods: Systematic review searching Medline, Embase, PsychINFO and CENTRAL from the date of database inception to 26 August 2025 for studies of any design reporting any pharmacological intervention in the management of BPOW compared with any or no other interventions, using adult participants aged 18 or over receiving buprenorphine and experiencing BPOW. We planned to combine outcomes using random-effects meta-analysis; where this was not possible results were reported narratively. We considered two outcomes and extracted (1) any reported measure or description of the change in opioid withdrawal symptoms (OWS) and (2) the number of individuals retained in buprenorphine treatment. Outcome quality was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.

Results: Forty-three studies met inclusion criteria reporting on 137 participants. These comprised one pilot randomised controlled trial and 42 uncontrolled observational case series or case reports. Meta-analysis was not possible, and all evidence was of low or very low quality. The currently available randomised evidence suggests that use of intravenous magnesium sulphate, in addition to symptomatic treatment with clonidine, paracetamol and diazepam, may statistically significantly reduce OWS when compared with symptomatic treatment alone. The currently available observational evidence suggests that treatment strategies which include additional doses of transmucosal buprenorphine may demonstrate higher rates of symptom control and treatment retention than strategies which do not include additional doses of transmucosal buprenorphine.

Conclusions: There is a paucity of research into pharmacological management of buprenorphine precipitated opioid withdrawal. The limited very low to low quality evidence suggests treatment regimens that include magnesium sulphate and additional doses of transmucosal buprenorphine are potentially the most salient current options and avenues for future research in the management of buprenorphine precipitated opioid withdrawal.