E-cigarettes are being widely viewed as a disruptive technology with the potential to reshape the tobacco industry and perhaps even pharmaceutical companies that market nicotine replacement products. E-cigarette vendors, consumers, and other related stakeholders have been engaged in voluminous discussions in social media, which has become a critical source for understanding e- cigarette use behavior, health effects, and marketing practice. Yet, research in the intersection of e- cigarettes and social media analytics is almost non-existent, let alone automated systems that could potentially significantly improve the e-cigarette researchers and regulatory bodies' information gathering and analysis capabilities to facilitate evidence-based scientific inquiries and policy-making. This proposed interdisciplinary e-cigarette research is primarily motivated to provide social media data- driven scientific findings and technology-driven informatics tools to address the FDA CTP research priorities concerning understanding of reasons for e-cigarette use, e-cigarettes' health effects and efficacy as a smoking reduction and cessation aid, and vendors' marketing practice and related stakeholders' response.
The specific aims of the proposed efforts are as follows.
Aim 1 is focused on developing a dataset covering a significant portion of e-cigarette-related discussions in the English-based social media, as well as related social networking activities.
Aim 2 is concerned with identifying e-cigarette consumers from the developed data collection, extracting reasons for e-cigarette use among social media users, and understanding e-cigarettes' health effects and efficacy for smoking reduction and cessation.
Aim 3 focuses on understanding e-cigarette marketing in social media and corresponding responses of the related stakeholders.
Aim 4 is concerned with the development of an integrated informatics platform for social media-based e-cigarette research. The proposed work aims to implement a social media analytics plan that can increase our knowledge of e-cigarette use reasons, health effects, efficacy for smoking reduction and cessation, marketing practice, related stakeholders' response in the U.S. Our data could deliver significant benefit to regulatory agencies like the FDA so that they can develop a better understanding of marketing and media that could come under their regulatory review. From the point of view of data sources, our studies will tap into th wealth of social media for analytical and regulatory purposes. We will also develop novel methods and tools to enable automatic analysis of large-scale social media datasets relevant to e-cigarettes. The proposed empirical studies using these tools are expected to generate fresh insights that complement existing research, much of which relies on survey data and anecdotes only. An integrated Web-based social media informatics system, along with a live data collection dedicated to e-cigarette research, will provide much-needed information and situational awareness support to benefit CTP missions and regulatory work.

Public Health Relevance

This project aims to advance the understanding of reasons for e-cigarette use, e-cigarettes' health effects and efficacy as a smoking reduction and cessation aid, as well as e-cigarette vendors' marketing and promotional efforts through analyzing social media content and online social networking activities, complementing traditional survey-based and clinical studies. A Web-based social media informatics platform will be developed to provide near real-time e-cigarette data feeds and analysis capabilities. The proposed efforts could help the FDA CTP and tobacco control researchers with their e- cigarette adoption and consumer behavioral analysis, and related regulatory work.

Agency
National Institute of Health (NIH)
Institute
National Institute on Drug Abuse (NIDA)
Type
Research Project (R01)
Project #
5R01DA037378-04
Application #
9273497
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Kimmel, Heather L
Project Start
2014-06-15
Project End
2019-05-31
Budget Start
2017-06-01
Budget End
2018-05-31
Support Year
4
Fiscal Year
2017
Total Cost
Indirect Cost
Name
University of Arizona
Department
Administration
Type
Sch of Business/Public Admin
DUNS #
806345617
City
Tucson
State
AZ
Country
United States
Zip Code
85721
Li, Qiudan; Wang, Can; Liu, Ruoran et al. (2018) Understanding Users' Vaping Experiences from Social Media: Initial Study Using Sentiment Opinion Summarization Techniques. J Med Internet Res 20:e252
Zhou, Jiaqi; Zhang, Qingpeng; Zeng, Daniel Dajun et al. (2018) Influence of Flavors on the Propagation of E-Cigarette-Related Information: Social Media Study. JMIR Public Health Surveill 4:e27
Xie, Jiaheng; Liu, Xiao; Dajun Zeng, Daniel (2018) Mining e-cigarette adverse events in social media using Bi-LSTM recurrent neural network with word embedding representation. J Am Med Inform Assoc 25:72-80
Chen, Zhipeng; Zeng, Daniel D (2017) Mining online e-liquid reviews for opinion polarities about e-liquid features. BMC Public Health 17:633
Zhan, Yongcheng; Liu, Ruoran; Li, Qiudan et al. (2017) Identifying Topics for E-Cigarette User-Generated Contents: A Case Study From Multiple Social Media Platforms. J Med Internet Res 19:e24
Xie, Jiaheng; Zeng, Daniel Dajun; Marcum, Zachary A (2017) Using deep learning to improve medication safety: the untapped potential of social media. Ther Adv Drug Saf 8:375-377
He, Saike; Zheng, Xiaolong; Zeng, Daniel et al. (2016) Exploring Entrainment Patterns of Human Emotion in Social Media. PLoS One 11:e0150630
Li, Qiudan; Zhan, Yongcheng; Wang, Lei et al. (2016) Analysis of symptoms and their potential associations with e-liquids' components: a social media study. BMC Public Health 16:674
Zhang, Zhu; Zheng, Xiaolong; Zeng, Daniel Dajun (2016) A framework for diversifying recommendation lists by user interest expansion. Knowl Based Syst 105:83-95
Zhang, Zhu; Zheng, Xiaolong; Zeng, Daniel Dajun et al. (2016) Tracking Dabbing Using Search Query Surveillance: A Case Study in the United States. J Med Internet Res 18:e252

Showing the most recent 10 out of 16 publications