Novel psychoactive substances: An investigation of temporal trends in social media and electronic health records

Kolliakou, Anna; Ball, Michael; Derczynski, Leon; Chandran, David; Gkotsis, George; Deluca, Paolo; Jackson, Roy; Shetty, Hitesh; Stewart, Robert

doi:10.1016/j.eurpsy.2016.05.006

Cited by 20 publications

(10 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The separate codes for each search and extraction can be found at github.com/ leondz/medtermfilter. Another investigation under PHEME using a similar methodology and examining novel psychoactive substances in the EHR and online data sources has also been described elsewhere 35 .…”

Section: Discussionmentioning

confidence: 99%

Mental health-related conversations on social media and crisis episodes: a time-series regression analysis

Kolliakou

Bakolis

Chandran

et al. 2020

Sci Rep

View full text Add to dashboard Cite

We aimed to investigate whether daily fluctuations in mental health-relevant Twitter posts are associated with daily fluctuations in mental health crisis episodes. We conducted a primary and replicated time-series analysis of retrospectively collected data from twitter and two London mental healthcare providers. Daily numbers of 'crisis episodes' were defined as incident inpatient, home treatment team and crisis house referrals between 2010 and 2014. Higher volumes of depression and schizophrenia tweets were associated with higher numbers of same-day crisis episodes for both sites. After adjusting for temporal trends, seven-day lagged analyses showed significant positive associations on day 1, changing to negative associations by day 4 and reverting to positive associations by day 7. There was a 15% increase in crisis episodes on days with above-median schizophrenia-related Twitter posts. A temporal association was thus found between twitter-wide mental health-related social media content and crisis episodes in mental healthcare replicated across two services. Seven-day associations are consistent with both precipitating and longer-term risk associations. Sizes of effects were large enough to have potential local and national relevance and further research is needed to evaluate how services might better anticipate times of higher risk and identify the most vulnerable groups. It has long been recognised that material circulated via news media can have important mental health-relevant outcomes. Research in this area has particularly focused on suicide or non-fatal self-harm carried out by well-known people or fictional portrayals of these issues in widely viewed sources, addressing concerns about imitative behaviour 1. In parallel, there has also been concern around potentially stigmatising material and its longer-term impacts on the wellbeing of people with mental disorders and those who care for them 2. Depictions of mental illness on television and in film 3 and newspaper coverage of high-profile suicides 4 can have profound implications not only for public health and opinion but also directly on those experiencing the mental health issues portrayed 5. The explosion of social media channels and record levels of public use 6 , round-the-clock access to breaking news, stories and discussions as well as the sheer volume of decreasingly controlled information have transformed and amplified exposure to mental health dialogues and representations of mental illness. The use of data-rich, novel sources such as results from internet searches, social media communications and environmental records is becoming a potentially fast and cost-effective way to identify population needs and predict or prevent healthcare emergencies, such as pharmacovigilance 7. More recently, information from Twitter has been utilised for analysing behaviours, attitudes and experiences related to mental health 8 , particularly depression 9,10. In parallel, the expansion of electronic health records (EHRs) has offered extensive, longitudinal clinical

show abstract

Section: Discussionmentioning

confidence: 99%

Mental health-related conversations on social media and crisis episodes: a time-series regression analysis

Kolliakou

Bakolis

Chandran

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…All the users, indeed, can be part of a #highsociety, allowing them to share their #proudstoners daily states of mind. Very recently, epidemiologists and linguistic scientists used Twitter to test the feasibility of producing a fully-automated “drug term discovery” system capable of tracking emerging NPS terms in real time [ 49 ], which confirms that data collected on Twitter may be used to explore trends in NPS selling and use [ 50 ]. Along with other cyber drug communities (e.g., blogs, drug fora, Facebook), Twitter allowed the identification and characterization of a new generation of NPS users, the so-called “e-psychonauts”, that considered themselves as psychedelic researchers, mind navigators, or chemicals experimenters [ 51 ].…”

Section: Social Network and Smartphone Appsmentioning

confidence: 92%

Sales and Advertising Channels of New Psychoactive Substances (NPS): Internet, Social Networks, and Smartphone Apps

et al. 2018

View full text Add to dashboard Cite

In the last decade, the trend of drug consumption has completely changed, and several new psychoactive substances (NPS) have appeared on the drug market as legal alternatives to common drugs of abuse. Designed to reproduce the effects of illegal substances like cannabis, ecstasy, cocaine, or ketamine, NPS are only in part controlled by UN conventions and represent an emerging threat to global public health. The effects of NPS greatly differ from drug to drug and relatively scarce information is available at present about their pharmacology and potential toxic effects. Yet, compared to more traditional drugs, more dangerous short- and long-term effects have been associated with their use, and hospitalizations and fatal intoxications have also been reported after NPS use. In the era of cyberculture, the Internet acts as an ideal platform to promote and market these compounds, leading to a global phenomenon. Hidden by several aliases, these substances are sold across the web, and information about consumption is shared by online communities through drug fora, YouTube channels, social networks, and smartphone applications (apps). This review intends to provide an overview and analysis of social media that contribute to the popularity of NPS especially among young people. The possibility of using the same channels responsible for their growing diffusion to make users aware of the risks associated with NPS use is proposed.

show abstract

“…Finally, we reported training and testing on the in-domain data only and showed that the results are substantially lower compared to the case when the training data is augmented with the out-domain data. Our rumour stance classifier applied to new, obscure domains with shortage of training data has numerous applications that we aim to explore in the near future, such as rumours around bullying and suicide [10], or disputed perceptions around psychoactive substances [11]. Further improving our classifier, in future work we also aim to perform a more detailed analysis underpinning the reasons for the performance drop.…”

Section: Discussionmentioning

confidence: 99%

Stance Classification in Out-of-Domain Rumours: A Case Study Around Mental Health Disorders

Aker

Zubiaga

Bontcheva

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Social media being a prolific source of rumours, stance classification of individual posts towards rumours has gained attention in the past few years. Classification of stance in individual posts can then be useful to determine the veracity of a rumour. Research in this direction has looked at rumours in different domains, such as politics, natural disasters or terrorist attacks. However, work has been limited to in-domain experiments, i.e. training and testing data belong to the same domain. This presents the caveat that when one wants to deal with rumours in domains that are more obscure, training data tends to be scarce. This is the case of mental health disorders, which we explore here. Having annotated collections of tweets around rumours emerged in the context of breaking news, we study the performance stability when switching to the new domain of mental health disorders. Our study confirms that performance drops when we apply our trained model on a new domain, emphasising the differences in rumours across domains. We overcome this issue by using a little portion of the target domain data for training, which leads to a substantial boost in performance. We also release the new dataset with mental health rumours annotated for stance.

show abstract

Novel psychoactive substances: An investigation of temporal trends in social media and electronic health records

Cited by 20 publications

References 35 publications

Mental health-related conversations on social media and crisis episodes: a time-series regression analysis

Mental health-related conversations on social media and crisis episodes: a time-series regression analysis

Sales and Advertising Channels of New Psychoactive Substances (NPS): Internet, Social Networks, and Smartphone Apps

Stance Classification in Out-of-Domain Rumours: A Case Study Around Mental Health Disorders

Contact Info

Product

Resources

About