Customize your JAMA Network experience by selecting one or more topics from the list below.
Shive M, Bhatt M, Cantino A, Kvedar J, Jethwani K. Perspectives on Acne: What Twitter Can Teach Health Care Providers. JAMA Dermatol. 2013;149(5):621–622. doi:10.1001/jamadermatol.2013.248
Author Affiliations: University of California, San Francisco School of Medicine, San Francisco (Ms Shive); Department of Health Policy and Management, Harvard School of Public Health, Boston, Massachusetts (Ms Shive); Harvard Business School, Boston (Mr Bhatt); Iteration Labs LLC, San Francisco (Mr Cantino); Department of Dermatology (Drs Kvedar and Jethwani), Harvard Medical School (Mr Bhatt), Boston; Center for Connected Health, Partners Healthcare, Boston (Drs Kvedar and Jethwani); and Department of Dermatology, Massachusetts General Hospital, Boston (Drs Kvedar and Jethwani).
Acne is one of the most common skin diseases, with an estimated prevalence of 50 million people in the United States alone, and has a significant impact on quality of life. The high prevalence and seriousness of acne makes crafting innovative avenues for patient education about this disease very important. Twitter has become a popular social networking phenomenon with a user base of over 140 million active users and 340 million tweets per day.1 Its popularity makes it a potentially powerful source of information and route of communication for acne, especially since the Internet can be an adolescent's primary sources of health information.2
Twitter is an online social networking service that allows users to post 140-character messages called “tweets” and to subsequently repost or “retweet” these messages from their own account. Using a form of real-time data capture through the use of the Twitter Streaming Application Programming Interface (API), we obtained an institutional review board–exempt status and collected all tweets that contained 1 or more of 5 keywords: pimple, pimples, zit, zits, and acne for a 2-week period from June 10 through June 23, 2012, with additional data monitoring in order to calculate a 1-week retweet count for each tweet. We applied an English filter that we determined had a 93% sensitivity and 97% specificity.3 These data were then exported as a comma-separated values (CSV) file via the TwitterToCSV software library that one of us (A.C.) developed for this research and that we have made available as an open-source package.4High-impact tweets, defined as tweets with one or more retweets, were the only tweets examined in this study. These high-impact tweets were frequency weighted by retweet count and categorized by content into 4 main categories: personal, celebrity, education, and irrelevant/excluded. The education category was further subdivided into: disease question, disease information, treatment question, treatment information (branded), treatment information (non-branded), and treatment information (ambiguous). The language and content of these high-impact tweets were then compared to the acne patient information website published by the American Academy of Dermatology (AAD).5
There were a total of 8192 English high-impact tweets of a total of 392 617 tweets collected. Personal tweets about acne were the most common type of high-impact tweets (43.1%), followed by tweets about celebrities (20.4%), and then education-related tweets (27.1%); 9.4% of tweets were excluded. By education subcategory, 16.9% and 8.9% of all high-impact tweets were about disease information and treatment information, respectively.
Results of a more detailed analysis of disease and treatment related tweets are summarized in Table. A total of 67.3% of disease question tweets related to some variant of “what is acne” or “why does acne exist.” Two-thirds of disease question tweets asserted in some way that stress causes pimples, and 9% of retweets commented that makeup causes pimples. The most common treatment question was “how do I get rid of my pimples?” There were a large variety of acne home remedies suggested, including topical food-based remedies ranging from eggs to herbs to fruit. A total of 3.3% recommended using over-the-counter products topically, including aloe vera gel, crushed aspirin (as a source of salicylic acid), and baking soda.
Word frequency comparisons between the AAD website (http://www.aad.org/) and Twitter showed a strong discrepancy in both the topics and level of language used to talk about acne. People on Twitter more often use the nonmedical terms of “pimple” or “pimples” in their discussions and referred more frequently to “toothpaste” and “Proactive” [sic] as treatments (Proactiv; Guthy-Renker LLC), while the AAD website used words like “skin,” “pore,” “cells,” “dermatologist,” and “treatment.” In addition, the AAD website did not address topics that are commonly discussed on Twitter, like makeup, stress, and the efficacy of diet, toothpaste, or other home remedies on acne.
Twitter is emerging as a popular forum where people exchange health information. Health providers can not only learn about the perceptions and misperceptions of diseases like acne, but they might also communicate reliable medical information. There is a significant amount of negative commentary about acne on Twitter, in addition to myths, incorrect information, and unconventional home remedies. The dermatology community should be aware of these popular beliefs to effectively address them and to deliver the best patient education and care possible, both online and in the clinic.
Correspondence: Dr Jethwani, Center for Connected Health, 25 New Chardon St, Boston, MA 02114 (email@example.com).
Accepted for Publication: November 29, 2012.
Author Contributions: Ms Shive, Mr Bhatt, and Dr Jethwani had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Shive, Bhatt, Kvedar, and Jethwani. Acquisition of data: Shive, Bhatt, and Cantino. Analysis and interpretation of data: Shive, Bhatt, Cantino, and Jethwani. Drafting of the manuscript: Shive and Jethwani. Critical revision of the manuscript for important intellectual content: Shive, Bhatt, Cantino, Kvedar, and Jethwani. Statistical analysis: Shive, Cantino, and Jethwani. Administrative, technical, and material support: Cantino and Jethwani. Study supervision: Kvedar and Jethwani.
Conflict of Interest Disclosures: Dr Kvedar has served as consultant for and holds stock in Healthrageous.
Funding/Support: None reported.