Friday, March 16, 2012

Mid-term Prescription Drug Abuse: Does Social Media Play a Role?

Prescription Drug Abuse:
Does Social Media Play a Role?

Much has been written about illicit drug abuse in the United States. There are Public Health programs geared towards preventing and reducing the number of new users, as well as advising active users on resources to help them stop. In addition extensive law enforcement efforts are in place to address the importation, trafficking, and end-user sales. Despite these efforts and expenditures, elicit drug abuse continues to be a troubling and serious problem for Public Health Departments, the Health care system, Law Enforcement officials, all levels of government, and society at large.

From: Paulozzi, p. 1491

In recent years, a growing problem has emerged involving the recreational or illicit use of prescription drugs. As detailed in a report published last year from the Centers for DiseaseControl and Prevention (CDC), over the 10-year period between 1999-2008, “overdose death rates, sales and substance abuse treatment admissions related to opiate pain relievers all increased substantially.” (Paulozzi, 2011) The same study found that in 2008 in the U.S., opiate pain relievers were involved in 73.8% of the 20,044 reported deaths due to prescription drug overdoses. (see Figure 2, right)

Another study published in 2011 by the CDC showed that 20.2% of high school students reported having taken prescription drugs- including OxyContin®, Percocet®, Vicodin®, Adderall®, Ritalin®, and Xanax®.- that were not specifically prescribed to them one or more times in their lives. (Eaton, 2010, p. 19) Much of the scientific literature examining prescription drug abuse by adolescents and young adults tends to focus on analgesics, specifically opiates, because of the alarming increase in deaths due to overdose among these two groups. 

Figure 3
Stimulant drugs, which usually are prescribed for the treatment of ADD/ADHD, are a growing cause for concern as well, particularly for this demographic. The most common sources of non-prescribed use of stimulant drugs such as Adderall® are either from a person with a legitimate prescription for the drug, or  from individuals who feigned the symptoms of ADD/ADHD in order to obtain the drug. (DuPont, 2010) Swanson et al observed, in a 2011 review of stimulant use, that the total number of prescriptions dispensed for this therapeutic category has risen steadily each year. In fact there has been an over eleven-fold increase between 1990 and 2010 (See Figure 3, left). In this same report the authors estimated that 70% of the prescriptions dispensed were actually used for therapeutic purposes, adding "In 2008 the number of prescriptions was 38 million, so our rough estimate of 30% diversion suggests that about 11.4 million prescription might have been diverted to non-medical use in 2008." (Swanson, p. 744) Another study looking at students who had legitimate prescriptions for either type of drug found that 61.7% diverted their stimulants, while 35.1% diverted their analgesics to another person for non-prescribed use. (Garnier, 2010)

Adderall® tablets & capsules
The principle objective of this monitoring project is to explore whether or not data culled from Social Media can be a useful tool in the phenomenology research of non-medical use of certain prescription drugs- specifically the stimulant Adderall®, and the opiate analgesic OxyContin®. It is difficult to empirically establish and/or quantify correlations between Internet usage and prescription drug abuse for a number of reasons. One study attempted to do this by correlating the rates of increased Internet usage with the number of reported hospital admissions for treatment of prescription drug abuse (PDA). This investigation found “that for every 10 percent increase in high speed Internet use at the state level, associated treatment facility admissions for prescription drug abuse rose by 1 percent.” (Jena, 2011)

OxyContin ®   tablets
While these findings are hardly definitive, this study demonstrates the challenges in applying metrics that would accurately reflect the degree to which this should be considered a problem. In spite of these methodological challenges, there were strong statistical correlations found for both opiates (1.09) and stimulants (1.18), with both having p values of less than 0.001 (p. 1197). The obvious inference that could be made is that wider availability of high speed Internet leads to increased Internet sales of the drugs. In turn this results in increased nefarious use, leading to increased numbers of treatment facility admissions. Logical perhaps, but hardly definitive because in spite of the strong statistical correlations, these data lack the specificity required to reliably draw inferences or conclusions.

With these challenges in mind, the current investigation seeks to understand the role of Social Media in various transactions involving two of the prescription drugs cited in the previous paragraphs- the opiate analgesic  OxyContin® and the stimulant Adderall®. The volume and context of mentions on Facebook, Twitter, and Google Plus of each drug will be tracked daily, over the 2 month period of February 16- April 15, 2012. The primary objectives will be:
  • To explore the types of interaction and frequency of mentions; 
  • To identify any patterns of communication related to these drugs that might emerge; and 
  • To explore some of the methodological, ethical, and practical issues that must be considered in this format. 
The data presented in this interim report will examine patterns of total mentions of each drug. In addition this report will explore whether there are any identifiable correlative factors that can inform our thinking about these phenomena.

The charts above compare the absolute volume of mentions (top) and the percent of messages containing the mention, out of the total number of blog posts (bottom). If one were to rely on the bottom chart alone, it would appear that neither of these terms receives significant mention relative to the total "chatter" on the Web. However, paired with the pattern of absolute increased number of mentions, some possible patterns emerge. 

Adderall Total Buzz, Feb. 16- Mar. 15, 2012, Source:
In this chart one can see the daily volume of "Adderall" mentions on the three social networking sites being tracked, as well as sub-categories of "positive" and "negative" tweets, and the category total (burnt orange). Although "positive" and "negative" Tweets have been separated out by, the data provider, closer analysis of the individual Tweets demonstrates that these are not reliable characterizations, for the purposes of this investigation. Rather, these characterizations seem arbitrary, if not pejorative. It is not clear what qualifies a comment as "negative" or "positive". Furthermore, an overwhelming majority of comments were characterized as simply "neutral". [The same is true for the OxyContin® data to be presented in the next section.] The majority of mentions came from Tweets, which may indicate that they are being generated by a younger audience, perhaps college-aged or young adults. The most logical correlation to the peak increase, if one assumes that the use is related to attempts to improve academic performance, is that they coincide with periods that universities are entering the mid-term. In addition, 7 out of 10 of the highest volume "Tweeters" were online pharmacies advertising sale of Adderall® and other drugs without a prescription.

OxyContin Total Buzz, Feb. 16- Mar. 15, 2012, Source:
This chart presents data similar to the Adderall® chart, except in this case monitoring mentions of OxyContin® during the same time period. While there is more variability in the origin of the mentions, the vast majority of mentions come from Twitter. Closer examination of the individual Tweets, as well as temporal peak patterns can be correlated with issues in the news at the time. The first peak occurred in late February and early March. This coincides with intense media coverage of OxyContin® being removed from the Canadian market due to increasing problems with illicit use of the drug. The second peak occurred in the second week of March, following the unfortunate remarks radio talk-show host Rush Limbaugh made about Georgetown Law School student Sandra Fluke. Just as in the case of Adderall®, there were Tweets from online pharmacies advertising sale of the drug without a prescription. However the numbers of these Tweets were dwarfed by the Canadian market and Limbaugh-Fluke related activity.

The final report will provide a more detailed data analysis and include in depth discussion of the findings in the context of the study questions previously outlined.
The author would like to thank for their generous support in providing daily tracking and data computation services free of charge.

Works Cited

DuPont, R. (2010, June). Prescription Drug Abuse: An Epidemic Dilemma. Journal of Psychoactive Drugs , pp. 127-132.
Eaton, D. K. (2010, June 4). Youth Risk Behavior Surveillance — United States, 2009. MMWR, Special Report No. 5 , pp. 1-38.
Garnier, L., Arria, A., Caldeira, K., Vincent, K., & O’Grady, K. (2010, March). Sharing and selling of prescription medications in a college student sample. Journal of Clinical Psychiatry , pp. 262-269.
Jena, A. a. (2011, June). Growing Internet Use May Help Explain the Rise in Prescription Drug Abuse in the United States. Health Affairs , pp. 1192-1199.
Paulozzi, L. J. (2011, November 4). Vital Signs: Overdoses of Prescription Opioid Pain Relievers -- United States, 1999-2008. Morbidity & Mortality Weekly Report , 60 (43), pp. 1487-1492.
Swanson, J., Wigal, T., & Volkow, N. (2011, September). Contrast of Medical and Nonmedical Use of Stimulant Drugs, Basis for the Distinction, and Risk of Addiction: Comment on Smith and Farah (2011). Psychological Bulletin , 137 (5), pp. 742-748.


  1. The following paper will most likely be interesting to you. It found what you are finding about Adderall:
    Hanson CL, Burton SH, Giraud-Carrier C, West JH, Barnes MD, Hansen B. Tweaking and tweeting: exploring Twitter for nonmedical use of a psychostimulant drug (Adderall) among college students. J Med Internet Res 2013;15(4):e62

    1. Thanks for the suggestion Christophe. I've downloaded the article and will read it. I think I saw your email address on it as the corresponding author. I'll follow up with you by email, OK?

  2. I really like the dear information you offer in your articles. I'm able to bookmark your site and show the kids check out up here generally. Im fairly positive theyre likely to be informed a great deal of new stuff here than anyone else!
    Drug Treatment Center