Dataset Information

National substance use patterns on Twitter.

ABSTRACT: We examined openly shared substance-related tweets to estimate prevalent sentiment around substance use and identify popular substance use activities. Additionally, we investigated associations between substance-related tweets and business characteristics and demographics at the zip code level.A total of 79,848,992 tweets were collected from 48 states in the continental United States from April 2015-March 2016 through the Twitter API, of which 688,757 were identified as being related to substance use. We implemented a machine learning algorithm (maximum entropy text classifier) to estimate sentiment score for each tweet. Zip code level summaries of substance use tweets were created and merged with the 2013 Zip Code Business Patterns and 2010 US Census Data.Quality control analyses with a random subset of tweets yielded excellent agreement rates between computer generated and manually generated labels: 97%, 88%, 86%, 75% for underage engagement in substance use, alcohol, drug, and smoking tweets, respectively. Overall, 34.1% of all substance-related tweets were classified as happy. Alcohol was the most frequently tweeted substance, followed by marijuana. Regression results suggested more convenience stores in a zip code were associated with higher percentages of tweets about alcohol. Larger zip code population size and higher percentages of African Americans and Hispanics were associated with fewer tweets about substance use and underage engagement. Zip code economic disadvantage was associated with fewer alcohol tweets but more drug tweets.The patterns in substance use mentions on Twitter differ by zip code economic and demographic characteristics. Online discussions have great potential to glorify and normalize risky behaviors. Health promotion and underage substance prevention efforts may include interactive social media campaigns to counter the social modeling of risky behaviors.

SUBMITTER: Meng HW

PROVIDER: S-EPMC5673183 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

National substance use patterns on Twitter.

Meng Hsien-Wen HW Kath Suraj S Li Dapeng D Nguyen Quynh C QC

PloS one 20171106 11

<h4>Purpose</h4>We examined openly shared substance-related tweets to estimate prevalent sentiment around substance use and identify popular substance use activities. Additionally, we investigated associations between substance-related tweets and business characteristics and demographics at the zip code level.<h4>Methods</h4>A total of 79,848,992 tweets were collected from 48 states in the continental United States from April 2015-March 2016 through the Twitter API, of which 688,757 were identif ...[more]

PMID: 29107961

Similar Datasets

Project description:AIMS:To examine cross-national patterns of 12-month substance use disorder (SUD) treatment and minimally adequate treatment (MAT), and associations with mental disorder comorbidity. DESIGN:Cross-sectional, representative household surveys. SETTING:Twenty-seven surveys from 25 countries of the WHO World Mental Health Survey Initiative. PARTICIPANTS:A total of 2446 people with past-year DSM-IV SUD diagnoses (alcohol or illicit drug abuse and dependence). MEASUREMENTS:Outcomes were SUD treatment, defined as having either received professional treatment or attended a self-help group for substance-related problems in the past 12 months, and MAT, defined as having either four or more SUD treatment visits to a health-care professional, six or more visits to a non-health-care professional or being in ongoing treatment at the time of interview. Covariates were mental disorder comorbidity and several socio-economic characteristics. Pooled estimates reflect country sample sizes rather than population sizes. FINDINGS:Of respondents with past-year SUD, 11.0% [standard error (SE) = 0.8] received past 12-month SUD treatment. SUD treatment was more common among people with comorbid mental disorders than with pure SUDs (18.1%, SE = 1.6 versus 6.8%, SE = 0.7), as was MAT (84.0%, SE = 2.5 versus 68.3%, SE = 3.8) and treatment by health-care professionals (88.9%, SE = 1.9 versus 78.8%, SE = 3.0) among treated SUD cases. Adjusting for socio-economic characteristics, mental disorder comorbidity doubled the odds of SUD treatment [odds ratio (OR) = 2.34; 95% confidence interval (CI) = 1.71-3.20], MAT among SUD cases (OR = 2.75; 95% CI = 1.90-3.97) and MAT among treated cases (OR = 2.48; 95% CI = 1.23-5.02). Patterns were similar within country income groups, although the proportions receiving SUD treatment and MAT were higher in high- than low-/middle-income countries. CONCLUSIONS:Few people with past-year substance use disorders receive adequate 12-month substance use disorder treatment, even when comorbid with a mental disorder. This is largely due to the low proportion of people receiving any substance use disorder treatment, as the proportion of patients whose treatment is at least minimally adequate is high.

Dataset Information

National substance use patterns on Twitter.

Publications

National substance use patterns on Twitter.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets