Disease-Specific Hashtags for Online Communication About Cancer Care

Increasingly, patients, caregivers, and health care professionals (HCPs) go online to learn about and discuss cancer care.¹ However, finding other people or organizations with similar interests can be difficult without some structure.

Hashtags are user-generated tags that can organize and aggregate content on social networks. In July 2011, 2 patient advocates started a breast cancer chat on Twitter using the tag #bcsm (breast cancer social media); one of us (D.J.A.) joined as a comoderator. This same model but with hashtag #btsm was used to discuss brain tumors in January 2012. Both tags are now regularly used on Twitter by patients, caregivers, and HCPs.

Dedicated hashtags may make it easier to engage in relevant conversations online for other types of cancer. In this study, we describe a way to use disease-specific hashtags similar to #bcsm and #btsm to organize and increase online discussion of cancer care.

Based on the models using the #bcsm and #btsm hashtags, 2 of us (M.S.K. and P.F.A.) developed a set of 23 new cancer-specific tags that met the following criteria: disease specific, short, unique or minimally used on Twitter, and ending in “sm” for “social media” (as a prompt that online use is public). We selected this design to balance practical use with the ability to organize content. Initially proposed in July 2013, this cancer tag ontology (CTO) was posted on Symplur in November 2013 after public commentary and engagement (Table).²

We analyzed the number of tweets and users of the tags quarterly from April 2011 through June 2015 using Symplur Signals, which accesses Twitter’s application program interface. One of us (M.S.K.) classified the most active 100 users as patients, caregivers and/or advocates, physicians, nonphysician HCPs; individuals not otherwise specified, hospitals, other health care organizations, and spam accounts. Patients were defined by self-identification with the disease associated with each CTO tag. To assess hashtag adoption by well-known institutions, we evaluated the prevalence of CTO use by the National Comprehensive Cancer Network (NCCN)- and National Cancer Institute (NCI)-designated cancer centers as well as the NCI and American Society of Clinical Oncology (ASCO).

During the study period, there were 762 103 tweets by 117 064 user accounts. The hashtags #bcsm and #lcsm had the most use, with 323 720 and 143 089 tweets, respectively.

After mid-2013, the most active tags had organized live chats: #ayacsm, #gyncsm, #lcsm, #mmsm, and #pancsm, accounting for 92.2% of all CTO hashtag activity. Among the top 100 users, 34% were patients, 17% caregivers and/or advocates, 14% physicians, 8% nonphysician HCPs, 7% individuals, 2% hospitals, 14% other organizations, and 4% spam generators.

Quarterly tweet activity for all 25 CTO tags increased from 13 778 tweets in the third quarter of 2011 to 87 319 in second quarter of 2015 (Figure). Among 26 NCCN cancer centers, 80.7% used the CTO tags in tweets (median, 86; range, 18-2555). For 44 NCI-designated comprehensive cancer centers, 63.6% used the CTO. The most active organizations were Dana-Farber Cancer Institute (2555 tweets, 4 accounts) and MD Anderson Cancer Center (1771 tweets, 12 accounts). The NCI (749 tweets, 13 accounts) and ASCO (3238 tweets, 4 accounts) also used the CTO frequently.

In this report, we describe a way to use hashtags to organize disease-specific health information on Twitter. Most hashtag use develops ad hoc rather than in an organized fashion, but some studies suggest that structured tags can be a successful way to share.^3,4 We observed a 13% compound quarterly growth rate in Twitter activity, indicating widespread adoption of disease-specific hashtags by a variety of stakeholders.

The CTO creates a practical way to facilitate patient, clinician, and institutional access to health information and engagement. Preliminary data from the #bcsm hashtag suggest that interaction through Twitter may improve patient knowledge and reduce anxiety.⁵ We have described feasibility and growth of organized cancer information online. Further study is needed to determine whether the CTO can improve health literacy or other meaningful outcomes.

Accepted for Publication: August 21, 2015.

Corresponding Author: Matthew S. Katz, MD, Lowell General Hospital, 295 Varnum Ave, Lowell, MA 01854 (Matthew.Katz@lowellgeneral.org).

Published Online: November 5, 2015. doi:10.1001/jamaoncol.2015.3960.

Author Contributions: Dr Katz had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Katz, Anderson, Thompson, Johnston.

Acquisition, analysis, or interpretation of data: Katz, Utengen, Thompson, Attai, Dizon.

Drafting of the manuscript: Katz, Utengen, Thompson, Dizon.

Critical revision of the manuscript for important intellectual content: Katz, Anderson, Thompson, Attai, Johnston, Dizon.

Statistical analysis: Utengen, Thompson.

Administrative, technical, or material support: Utengen, Attai, Johnston.

Study supervision: Thompson.

Conflict of Interest Disclosures: Mr Utengen is cofounder of Symplur LLC, which promotes the use of social media and Twitter in health care. No other conflicts are reported.

Previous Presentation: This study was presented at the ASCO Annual Meeting; June 2, 2015; Chicago, Illinois.

Additional Contributions: We thank Lee Aase, BS, Mayo Clinic; Michael Fisch, MD, AIM Specialty Health; Thomas Lee, BS NHA, Symplur LLC; and Robert Miller, MD, American Society of Clinical Oncology, for input on the design and presentation of this study at the ASCO Annual Meeting in June 2015.

Correction: This article was corrected on December 23, 2015, to fix an author’s affiliation name.