AI and Big Data approaches to addressing the opioid crisis: a scoping review protocol

BMJ Open, Journal Year: 2024, Volume and Issue: 14(8), P. e084728 - e084728

Published: Aug. 1, 2024

Introduction This paper outlines the steps necessary to assess latest developments in artificial intelligence (AI) as well Big Data technologies and their relevance opioid crisis. Fatal overdoses have risen over 82 998 annually USA. highlights need for urgent effective data-driven solutions. AI approaches, such machine learning, deep learning natural language processing, been employed analyse patterns trends overdose data facilitate timely interventions. However, a comprehensive scoping review on effectiveness of AI-driven detect, treat, prevent or respond crisis remains absent. Thus, it is important identify recent advancements addressing Methods analysis We will electronically search four scientific databases (PubMed, Web Science, Engineering Village PsycInfo), including finding reference lists grey literature from 2013 2023. Covidence be used screening selecting papers. extract information citation details, study context, used, AI/Big technologies, features, algorithms evaluation metrics. synthesised, analysed summarised draw meaningful conclusions future directions tackle Ethics dissemination approval not required. Results disseminated via conference presentations peer-reviewed publication.

Language: Английский

---

Mapping drug epidemiology futures

International Journal of Drug Policy, Journal Year: 2021, Volume and Issue: 94, P. 103378 - 103378

Published: July 25, 2021

Language: Английский

---

Education as drug policy: A realist synthesis of continuing professional development for opioid agonist therapy

International Journal of Drug Policy, Journal Year: 2022, Volume and Issue: 108, P. 103807 - 103807

Published: Aug. 2, 2022

Continuing professional development (CPD) for opioid agonist therapy (OAT) has been identified as a key health policy strategy to improve care people living with use disorder (OUD) and address rising opioid-related harms. To design deliver effective CPD programs, there is need clarify how they work within complex system contexts. This review synthesizes the literature on OAT programs educational theory which interventions work, whom, in what contexts.A systematic realist synthesis of evaluations focused was conducted. included record identification screening, familiarization, data collection, analysis, expert consultation, iterative context-intervention-mechanism-outcome (CIMO) configuration development.Twenty-four reports comprising 21 evaluation studies from 5 countries 3373 providers were reviewed. Through testing relevant theory, five CIMO configurations developed. The categorized by who drove learning outcomes (i.e., system/policy, instructor, learner) their spheres influence micro, meso, macro). There predominance instructor-driven driving change at micro level, few policy-driven macro-influential inconsistent promotion clear crisis policy-level intervention.OAT challenged mismatches program justifications, objectives, activities, outcomes. Depending these factors interact, can operate barrier or facilitator OUD care. With more deliberate planning consideration directly addressing diverse learner needs may be developed delivered. drug does not isolation; feed into each other intercalate initiatives have macro impacts population

Language: Английский

---

Estimating County-Level Overdose Rates Using Opioid-Related Twitter Data: Interdisciplinary Infodemiology Study

JMIR Formative Research, Journal Year: 2022, Volume and Issue: 7, P. e42162 - e42162

Published: Dec. 22, 2022

There were an estimated 100,306 drug overdose deaths between April 2020 and 2021, a three-quarter increase from the prior 12-month period. is approximate 6-month reporting lag for provisional counts of National Vital Statistics System, highest level geospatial resolution at state level. By contrast, public social media data are available close to real-time often accessible with precise coordinates.The purpose this study assess whether county-level mortality burden could be using opioid-related Twitter data.International Classification Diseases (ICD) codes poisoning or exposure county obtained CDC WONDER. Demographics collected American Community Survey. The Application Programming Interface was used obtain tweets that contained any 36 terms names. An unsupervised classification approach clustering tweets. Population-normalized variables polynomial population-normalized produced. Furthermore, z scores Getis Ord Gi statistic produced, both these their counterparts explored in regression modeling burden. A series linear models predictive explore interpretability analytical output.Modeling normalized demographic alone explained only 7.4% variability mortality, whereas approximately doubled by use specific covariates based on backward selection approach. adjusted R2 lowest AIC (Akaike Info Criterion) model variables, analyses, topic (adjusted R2=0.133, AIC=8546.8). appeared have improved utility over population-normalization alone. In model, median age, female population, about web-based sales positively associated opioid mortality. Asian race Hispanic ethnicity significantly negatively burdens mortality.Social data, when transformed certain statistical approaches, may add goal producing closer estimates Prediction outcomes can advanced inform prevention treatment decisions. This interdisciplinary facilitate evidence-based funding decisions various substance disorder programs.

Language: Английский

---

Drug Use and Artificial Intelligence: Weighing Concerns and Possibilities for Prevention

American Journal of Preventive Medicine, Journal Year: 2023, Volume and Issue: 66(3), P. 568 - 572

Published: Dec. 4, 2023

Language: Английский

---