FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890842500> ?p ?o ?g. }

• W2890842500 endingPage "696" @default.
• W2890842500 startingPage "693" @default.
• W2890842500 abstract "Since patterns of drug use evolve constantly 1-4, they require constant monitoring. Two recent developments illustrate this need. First, the development of New Psychoactive Substances (NPS) necessitates careful monitoring—in December 2017, 111 countries reported 779 different NPSs to the United Nations Office on Drugs and Crime 5, with about 100 NPSs appearing annually 6. Second, prescription medication misuse (e.g. opioids, stimulants, sedatives and tranquilizers) 1, 7, 8 has risen sharply—exemplified by some academics labelling opioid misuse as a growing ‘epidemic’ 9-12 (tellingly, Google Scholar returned approximately 3500 exact phrase matches to ‘opioid epidemic’ from 2017 to April 2018 alone). However, change is not limited to drug use. Globally, societies en bloc face radical change. Two of the biggest drivers of recent societal change are internet access (especially, in the last decade, through smartphones) and social media. Despite having 20th century roots, smartphones and social media are properly 21st century phenomena. Modern societies have near-ubiquitous smartphone and internet access. Australia is no exception, with 88% smartphone penetration 13, and social media usage (especially Facebook) at approximately 80% 14, 15, mostly on smartphones 14. Internet searching and social media can no longer responsibly be excluded as research data sources, especially in behavioural medicine, although web surveillance cannot supplant, but should complement, traditional surveillance 16. Accordingly, drug surveillance has taken to the internet. Early attempts included the US National Institute of Health and National Institute on Drug Abuse (NIH/NIDA)-funded Real Drugs in a Virtual World studies 17, the European-funded Psychonaut web mapping project 18, 19 and the more recent I-Trend 20 project. These studies involved mapping drug-discussion web forums and employing manual thematic analysis. The NIH/NIDA more recently funded the Center for Interventions, Treatment and Addictions Research to monitor publicly accessible Twitter feeds and web forums: The National Early Warning System Network 21, aims to combine these data with data from key sentinel toxicologists to identify emerging psychoactive drug use, and the eDrugTrends study 22 aims to establish a platform for the semi-automated monitoring of these data for information on cannabis usage (building on a similar study monitoring buprenorphine 23). The European Monitoring Centre for Drugs and Drug Addiction has also more recently funded the Computer Assisted Solutions for Studying the Availability aNd DistRibution of novel psychoActive substances (CASSANDRA) project, combining manual mapping with automated Twitter/Facebook data scraping to understand the supply and diffusion of NPS 24-26. Both eDrugTrends and CASSANDRA make use of machine learning techniques to semi-automate surveillance efforts. When discussing the internet in the drug research context, one must distinguish between the ‘surface web’ and the ‘darknet’. The former consists of all content available through a standard search engine, whereas darknet content and markets must be accessed through a specialised encrypted browser 27. The NIH/NIDA funded eDarkTrends project monitors the emergence of NPS through darknet markets, using machine learning and automated web scraping 28. The National Drug and Alcohol Research Centre's Drugs and New Technologies programme monitors the darknet for substance availability 29. The CASSANDRA project also monitors darknet markets to track the availability of NPS. Apart from these projects aiming to monitor new drug trends through the tracking and analysis of internet data, several one-off studies have used social media data to describe illicit drug trends. A recent systematic review identified 14 such studies which typically utilised Twitter, YouTube, web forums and Facebook data 30, and another identified 22 studies using social media for pharmacovigilance 31. Are these early forays into the world of drug research and monitoring on the internet sufficient, or are there as yet untouched avenues? We believe that internet-related drug use and surveillance research needs to embrace four important concepts if it is to establish itself as an effective methodology capable of delivering translational impacts. First, the obvious importance of the darknet for drug-related research does not obviate the need to continually monitor surface web information—findings from the 2017 Ecstasy and Related Drugs Reporting System 32 suggested that only 16% of respondents had purchased drugs from the darknet, while 20% had purchased from the surface web. It is not unusual for darknet retailers to maintain surface web front ends or entry portals to their darknet sites, or to have surface web shops that sell psychoactive substances under the guise of selling nootropics, herbal supplements or ‘incense’. Since these web shops are on the surface web, users are searching for information on drugs using conventional search engines, and since people use their smartphones to access the internet when they are on the go (typically not via Tor browsers), researchers could use the Application Programming Interfaces made available by various search portals (e.g. Google Trends) to gauge trends in interest for a number of substances. Prescription medication misuse can also be tracked using social media 33. This illustrates the potential of surface web data for drug monitoring. The same argument concerning cryptomarkets must be applied to drug trading via smartphone apps—since the data are inaccessible, this is best studied through surveys, but it still does not negate the necessity of studying surface web data. Second, we may need to rethink our methodologies and toolsets with replicability in mind. All the aforementioned studies monitoring social media or the darknet use their own proprietary systems, developed as part of their work. If these tools are not shared, they are time-consuming and costly to replicate. Even if they are shared, they may be so custom-designed that adapting them for a new context may match redeveloping new tools in difficulty. Replicability is thus a big concern with current research 31. To address this challenge, researchers could consider using commercial, ‘off-the-shelf’ solutions. There may often be costs associated with these tools, but these costs can be negotiated and may compare very favourably with the resource-intensive process of developing custom tools. Several tools exist for commercial brand management on social media (so-called ‘social listening’), and these can easily be used for other topics such as drug use. The publicly available search portal Application Programming Interfaces (mentioned above) that provide access to internet search data may also be useful. Another solution to increasing replicability is the sharing of curated datasets. This approach was taken by Carnegie Mellon University to support their research into darknet markets 34. A variety of ethical and legal issues must be worked through to better support data sharing of this kind 35, 36. Third, researchers need to rethink their research questions when it comes to the internet. Researchers in the drugs field have turned to the internet for data, but we argue that many researchers are not thinking deeply enough about the internet as a source of data 37. For many, the internet is seen (not incorrectly) as a tool enabling access to larger numbers of respondents for conventional surveys or to hidden populations not otherwise accessible 38. The real value of internet data, however, may not be that it gives access to respondents, nor that it allows researchers to overcome the small-sample limitations of their studies, but rather that it provides access to new modes of human interaction, new data that cannot be obtained by surveys 36, 39. We feel that researchers in the drugs field are only just beginning to explore these avenues in the depth required. This effort should also extend to the question of how the analysis of internet data can meaningfully supplement (not supplant) survey data 40, which is, in terms of implementation, far from a trivial methodological question. This question may be fruitfully explored using internet data for the contemporaneous forecasting of new drugs and drug use patterns which are confirmed and validated in due course through conventional survey and other surveillance methods (e.g. wastewater analysis). Finally, while cross-sectional or one-off studies are useful for demonstrating possibilities, to make full use of internet data for drug trend monitoring, studies should be designed to detect trends over long time-periods and funded accordingly. It is regrettable that there is currently no Australian drug trend monitoring system that systematically includes monitoring of both surface and darknet internet data. Ideally, given the global nature of most social media platforms, global multi-disciplinary consortiums would share in systematic extraction methodologies, creating replicability datasets for research and monitoring purposes 30. In closing, it should be remembered that, even apart from the question of suitability to context, all research is imperfect. Drug surveillance using internet data has advantages compared to conventional methodologies, such as lower cost and the immediacy with which data are obtained and updated. But research using internet data also has liabilities; it can address some conventional forms of bias but introduces biases of its own. The acquisition and analysis of internet data also requires additional skillsets beyond the conventional training of researchers in the drugs field, with the additional requirement of continual retraining to keep on top of developments in internet data. Despite these challenges, we argue that studying internet data to monitor new drug trends is no longer an optional choice for researchers. The question, instead, is how best to approach these data to derive the maximum impact. JR is supported by the National Health and Medical Research Council through the Translational Australian Clinical Toxicology Program (grant ID1055176). MB is supported by an Early Career Research fellowship from the National Health and Medical Research Council (APP1070140). The National Drug and Alcohol Research Centre and the National Drug Research Institute are supported by funding from the Australian Government under the Drug and Alcohol Program Fund. We also acknowledge the contribution of the Victorian Operational Infrastructure Support Program received by the Burnet Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders listed. The authors have no conflicts of interest." @default.
• W2890842500 created "2018-09-27" @default.
• W2890842500 creator A5055738358 @default.
• W2890842500 creator A5061940307 @default.
• W2890842500 date "2018-09-01" @default.
• W2890842500 modified "2023-09-26" @default.
• W2890842500 title "Digital era drug surveillance: Quo vadis, Australia?" @default.
• W2890842500 cites W1714785915 @default.
• W2890842500 cites W2020047442 @default.
• W2890842500 cites W2020562792 @default.
• W2890842500 cites W2061427388 @default.
• W2890842500 cites W2076045136 @default.
• W2890842500 cites W2088481526 @default.
• W2890842500 cites W2102742655 @default.
• W2890842500 cites W2117021707 @default.
• W2890842500 cites W2141588998 @default.
• W2890842500 cites W2224826979 @default.
• W2890842500 cites W2300421333 @default.
• W2890842500 cites W2353893441 @default.
• W2890842500 cites W2414519424 @default.
• W2890842500 cites W2495418488 @default.
• W2890842500 cites W2587886959 @default.
• W2890842500 cites W2621112056 @default.
• W2890842500 cites W2626210685 @default.
• W2890842500 cites W2720850745 @default.
• W2890842500 cites W2793361321 @default.
• W2890842500 doi "https://doi.org/10.1111/dar.12853" @default.
• W2890842500 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30175478" @default.
• W2890842500 hasPublicationYear "2018" @default.
• W2890842500 type Work @default.
• W2890842500 sameAs 2890842500 @default.
• W2890842500 citedByCount "4" @default.
• W2890842500 countsByYear W28908425002019 @default.
• W2890842500 countsByYear W28908425002020 @default.
• W2890842500 countsByYear W28908425002021 @default.
• W2890842500 crossrefType "journal-article" @default.
• W2890842500 hasAuthorship W2890842500A5055738358 @default.
• W2890842500 hasAuthorship W2890842500A5061940307 @default.
• W2890842500 hasBestOaLocation W28908425001 @default.
• W2890842500 hasConcept C118552586 @default.
• W2890842500 hasConcept C144133560 @default.
• W2890842500 hasConcept C17744445 @default.
• W2890842500 hasConcept C199539241 @default.
• W2890842500 hasConcept C2776748549 @default.
• W2890842500 hasConcept C2780035454 @default.
• W2890842500 hasConcept C71924100 @default.
• W2890842500 hasConceptScore W2890842500C118552586 @default.
• W2890842500 hasConceptScore W2890842500C144133560 @default.
• W2890842500 hasConceptScore W2890842500C17744445 @default.
• W2890842500 hasConceptScore W2890842500C199539241 @default.
• W2890842500 hasConceptScore W2890842500C2776748549 @default.
• W2890842500 hasConceptScore W2890842500C2780035454 @default.
• W2890842500 hasConceptScore W2890842500C71924100 @default.
• W2890842500 hasFunder F4320334705 @default.
• W2890842500 hasIssue "6" @default.
• W2890842500 hasLocation W28908425001 @default.
• W2890842500 hasLocation W28908425002 @default.
• W2890842500 hasOpenAccess W2890842500 @default.
• W2890842500 hasPrimaryLocation W28908425001 @default.
• W2890842500 hasRelatedWork W2349190201 @default.
• W2890842500 hasRelatedWork W2351709931 @default.
• W2890842500 hasRelatedWork W2354677115 @default.
• W2890842500 hasRelatedWork W2356066271 @default.
• W2890842500 hasRelatedWork W2377965897 @default.
• W2890842500 hasRelatedWork W2386054794 @default.
• W2890842500 hasRelatedWork W2387758490 @default.
• W2890842500 hasRelatedWork W2393626223 @default.
• W2890842500 hasRelatedWork W2748952813 @default.
• W2890842500 hasRelatedWork W2899084033 @default.
• W2890842500 hasVolume "37" @default.
• W2890842500 isParatext "false" @default.
• W2890842500 isRetracted "false" @default.
• W2890842500 magId "2890842500" @default.
• W2890842500 workType "article" @default.