National substance use patterns on Twitter

Purpose 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. Methods 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. Results 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. Conclusions 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.

[1]  Ben Y. Zhao,et al.  Beyond Social Graphs: User Interactions in Online Social Networks and their Implications , 2012, TWEB.

[2]  J. Jackson,et al.  Neighborhood disadvantage, stress, and drug use among adults. , 2001, Journal of health and social behavior.

[3]  L. Ungar,et al.  Data-Driven Content Analysis of Social Media , 2015 .

[4]  Melissa J. Krauss,et al.  Hookah-Related Twitter Chatter: A Content Analysis , 2015, Preventing chronic disease.

[5]  R. Kosterman,et al.  Neighborhood Typologies Associated with Alcohol Use among Adults in Their 30s: a Finite Mixture Modeling Approach , 2017, Journal of Urban Health.

[6]  S. Galea,et al.  Neighborhood smoking norms modify the relation between collective efficacy and smoking behavior. , 2009, Drug and alcohol dependence.

[7]  M. Moreno,et al.  Influence of Social Media on Alcohol Use in Adolescents and Young Adults , 2014, Alcohol research : current reviews.

[8]  A. Lenhart.  Teens, Social Media & Technology Overview 2015 , 2015 .

[9]  Ming Wen,et al.  Building a National Neighborhood Dataset From Geotagged Twitter Data for Indicators of Happiness, Diet, and Physical Activity , 2016, JMIR public health and surveillance.

[10]  M. J. Mcdermott,et al.  Prevalence and Sociodemographic Correlates of Lifetime Substance Use among a Rural and Diverse Sample of Adolescents , 2013, Substance abuse.

[11]  J. Bauermeister,et al.  Neighborhood variation in adolescent alcohol use: examination of socioecological and social disorganization theories. , 2011, Journal of studies on alcohol and drugs.

[12]  Christopher M. Danforth,et al.  Climate Change Sentiment on Twitter: An Unsolicited Public Opinion Poll , 2015, PloS one.

[13]  Christophe Giraud-Carrier,et al.  An Explor ation of Social Circles and Prescr iption Drug Ab use Through Twitter , 2013 .

[14]  Z. Yin,et al.  Ethnic Identity and Risky Health Behaviors in School-Age Mexican-American Children , 2006, Psychological reports.

[15]  Kevin J. Anstrom,et al.  Using Inverse Probability-Weighted Estimators in Comparative Effectiveness Analyses With Observational Databases , 2007, Medical care.

[16]  Marc A Zimmerman,et al.  Permissive norms and young adults' alcohol and marijuana use: the role of online communities. , 2012, Journal of studies on alcohol and drugs.

[17]  Mark Dredze,et al.  What Are Health-Related Users Tweeting? A Qualitative Content Analysis of Health-Related Users and Their Messages on Twitter , 2014, Journal of medical Internet research.

[18]  L. Kann,et al.  Population Density and Alcohol-related Risk Behaviors among us High School Students , 2005 .

[19]  Greg S. Weaver,et al.  Predictors of Alcohol and Drug Use: A Multi-Ethnic Comparison , 1995 .

[20]  Rui Fan,et al.  Anger Is More Influential than Joy: Sentiment Correlation in Weibo , 2013, PloS one.

[21]  B. Lewis,et al.  Methods of using real-time social media technologies for detection and remote monitoring of HIV outcomes. , 2014, Preventive medicine.

[22]  K. Karriker-Jaffe,et al.  Neighborhood disadvantage and adult alcohol outcomes: differential risk by race and gender. , 2012, Journal of studies on alcohol and drugs.

[23]  J. Miles,et al.  Predictors and consequences of prescription drug misuse during middle school. , 2015, Drug and alcohol dependence.

[24]  K. Theall,et al.  The neighborhood alcohol environment and at-risk drinking among African-Americans. , 2011, Alcoholism, clinical and experimental research.

[25]  G. Eysenbach,et al.  Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak , 2010, PloS one.

[26]  J. Bauermeister,et al.  Online Network Influences on Emerging Adults’ Alcohol and Drug Use , 2013, Journal of youth and adolescence.

[27]  K. Brady,et al.  Perceived cannabis use norms and cannabis use among adolescents in the United States. , 2015, Journal of psychiatric research.

[28]  M. Fowler,et al.  Geographic mobility and children's emotional/behavioral adjustment and school functioning. , 1994, Pediatrics.

[29]  Urmimala Sarkar,et al.  The Canary in the Coal Mine Tweets: Social Media Reveals Public Perceptions of Non-Medical Use of Opioids , 2015, PloS one.

[30]  Kayo Fujimoto,et al.  A comparison of peer influence measures as predictors of smoking among predominately hispanic/latino high school adolescents. , 2013, The Journal of adolescent health : official publication of the Society for Adolescent Medicine.

[31]  I. Ajzen.  The theory of planned behavior , 1991 .

[32]  J. Brook,et al.  Drug Use among African Americans: Ethnic Identity as a Protective Factor , 1998, Psychological reports.

[33]  Melissa J. Krauss,et al.  Twitter chatter about marijuana. , 2015, The Journal of adolescent health : official publication of the Society for Adolescent Medicine.

[34]  E. T. Tchetgen Tchetgen,et al.  The Relationship Between Neighborhood Poverty and Alcohol Use: Estimation by Marginal Structural Models , 2010, Epidemiology.

[35]  Ruth Etzioni,et al.  Building Better Models , 2012, Medical decision making : an international journal of the Society for Medical Decision Making.

[36]  V. Rovnyak,et al.  Mental Disorders and Help Seeking in a Rural Impoverished Population , 1999, International journal of psychiatry in medicine.

[37]  W. Chapman,et al.  Using Twitter to Examine Smoking Behavior and Perceptions of Emerging Tobacco Products , 2013, Journal of medical Internet research.

[38]  Matthew Hindman,et al.  Building Better Models , 2015 .

[39]  Melissa J. Krauss,et al.  "Hey Everyone, I'm Drunk." An Evaluation of Drinking-Related Twitter Chatter. , 2015, Journal of studies on alcohol and drugs.

[40]  C. Perry,et al.  Who needs liquor stores when parents will do? The importance of social sources of alcohol among young urban teens. , 2007, Preventive medicine.

[41]  J. Brownstein,et al.  A Case Study of the New York City 2012-2013 Influenza Season With Daily Geocoded Twitter Data From Temporal and Spatiotemporal Perspectives , 2014, Journal of medical Internet research.

[42]  D. Fiellin,et al.  Prevalence and correlates for nonmedical use of prescription opioids among urban and rural residents. , 2013, Drug and alcohol dependence.

[43]  T. Naimi,et al.  Binge Drinking and Associated Health Risk Behaviors Among High School Students , 2007, Pediatrics.

[44]  A. Mokdad,et al.  Binge drinking among US adults. , 2003, JAMA.

[45]  K. Karriker-Jaffe,et al.  Understanding Associations Between Neighborhood Socioeconomic Status and Negative Consequences of Drinking: a Moderated Mediation Analysis , 2016, Prevention Science.

[46]  N. D. de Vries,et al.  Youth crowds and substance use: the impact of perceived group norm and multiple group identification. , 2007, Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors.

[47]  B. Psaty,et al.  Assessment of community-level influences on individuals' attitudes about cigarette smoking, alcohol use, and consumption of dietary fat. , 1993, American journal of preventive medicine.

[48]  T. Chan,et al.  Availability of convenience stores and adolescent alcohol use in Taiwan: a multi-level analysis of national surveys. , 2013, Addiction.

[49]  Michael J. Paul,et al.  Discovering Health Topics in Social Media Using Topic Models , 2014, PloS one.

[50]  J. Havens,et al.  Nonmedical prescription drug use in a nationally representative sample of adolescents: evidence of greater use among rural adolescents. , 2011, Archives of pediatrics & adolescent medicine.