University of California San Francisco
Helen Diller Family Comprehensive Cancer Center

The Growing Role of Social Media in Research and Healthcare

Q&A with Urmimala Sarkar, MD, MPH

October 16, 2018

The Growing Role of Social Media in Research and Healthcare
Casual or even frequent users of social media recognize its power to share ideas and enhance professional networking, but social media’s utility in healthcare research is only just coming to the fore as a powerful tool for understanding, improving, and reinforcing health beliefs and behaviors. Below, Dr. Urmimala Sarkar, Associate Professor of Medicine, talks social media’s value and potential for research, her projects, and the value in attending the PRISM Health Symposium, (Promoting Research in Social Media and Health) December 7th at Mission Bay.

Urmimala SarkarUrmimala Sarkar, MD, PhD  

"Like the internet and mobile technologies more broadly, social media can overcome time and distance constraints.

There is transformative potential to reach patients, especially using audio, video, and translation to address language, literacy, and cognition barriers."

Q. How is social media being used for cancer research and to what extent? Which platforms are the most popular for this?
A: There is a wide range of cancer-relevant social media research across all kinds of platforms, ranging from large, mainstream platforms like Facebook and Twitter to purpose-built, small and specific online communities. I think of three ways that social media can be used for cancer research.

First, cancer researchers can learn a lot from what people post and discuss online. For example, we learned about how people talk about cancer screening like mammography and Pap tests online, and this information can help us craft strategies to encourage people to get the cancer screening they need. Second, social media is a powerful marketing tool for public health campaigns. We can use tools like online videos and messages to encourage positive health behaviors, like exercise, and discourage risky behaviors. UCSF dermatologist Eleni Linos has taken this approach to indoor tanning, creating and posting anti-tanning videos to Facebook. Third, we can deliver important cancer prevention services, like smoking cessation, online, instead of through traditional mechanisms.

Obviously, online doesn’t work for everything, but for peer support, information seeking, and making daily changes in behavior, using social media can extend and enhance existing tools.
Q. How does social media differ from health information technology?
A: Health information technology (HIT) encompasses the electronic systems that healthcare providers and patients use to store and share health information. Examples of HIT include electronic health records where providers document health information, and online patient portals where patients can access their health records. Mobile apps aiming to increase individual healthy behaviors like physical activity or nutrition are also examples of HIT, which mostly facilitates interactions between the care team and patient.

Social media, on the other hand, involves interacting with an online network where there are interactions amongst participants. We know that behaviors are contagious, both in real life and online, and research on social media and health encompasses the concept of networks.
Q. What kinds of research are most compatible with social media? How should basic scientists think about social media for research?
A: I would say that social, behavioral, and public health research are most compatible with social media. Social media has become a popular way for Americans to find and share health information with one another. For example, support groups on social media provide resources for weight loss and managing chronic conditions. With widespread use of social media among adults, social media can be a powerful tool to understand people’s health behaviors and perceptions.

Since social media can be a large data source, potential exists for social media data mining to help us with disaster response and crises like the opioid crisis. Social media “big data” can inform basic science research. Although I am not a basic scientist, my understanding is that using social media to engage large scientific networks has been promising for collaborative problem solving across a range of basic scientific disciplines.
Q. Where is social media research finding the most success or utility so far?
A: Going back to three frameworks I mentioned earlier, social media research has been most successful as a data source and platform for disseminating public health messages and interventions at scale. For example, Nguyen et al. used tweets to predict county-level health outcomes based on geotags (Nguyen, et al. American Journal of Public Health, 2017). Healthy food and physical activity tweets were associated with lower mortality, obesity, and physical inactivity.

Notably, one of our colleagues here at the Cancer Center, Eleni Linos, MD, DrPH, MPH, used social media to spread an evidence-based skin cancer prevention message to high-risk audiences using Google Ads (Serrano et al. JAMA Dermatology, 2016). Her research has shown that it is feasible and relatively inexpensive to expose target populations to public health messages using Google Ad Words. Also, UCSF faculty member Danielle Ramo, PhD, created a Facebook intervention for tobacco cession that prompted young adults to undertake a large number of quit attempts. Participants tried to quit without additional assistance after participating in Facebook intervention (Thrul and Ramo. Subst. Use Misuse 2017).
Q. How did you become involved in social media related research?
A: My interest in social media related research really came from my patients. I am a primary care provider at the Richard Fine People’s Clinic at Zuckerberg San Francisco General. Many of my patients have difficulty finding health education resources that are in their language or reading level. One of my patients, a monolingual Spanish speaker, was caring for her aging mother with Alzheimer’s disease. She found it very difficult and isolating to be a 24/7 caregiver, and because she could not leave her mother alone, she could not participate in caregiving support groups. Her refuge and main source of social support was a Facebook support group for Alzheimer’s caregivers in Spanish.

Support groups like these improve access to peer support and can serve as a place to scale public health interventions for a wide audience. Like the internet and mobile technologies more broadly, social media can overcome time and distance constraints. There is transformative potential to reach patients, especially using audio, video, and translation to address language, literacy, and cognition barriers.
Q. Describe some of your work, the questions you have been pursuing, and the results?
A: My study, “Drowning in #Codeine,” characterizes the extent of codeine misuse on Instagram. We identified hashtags and searchable text phrases associated with codeine misuse and performed a content analysis on these posts. We found common themes arising in the images identifying a culture around codeine misuse, the glamorization of ingestion with soda and alcohol, and integration with mainstream popular culture (Cherian et al., JMIR Public Health Surveill, 2018).

“SAVE DR MoM: Second Adverse Event Victim Experiences: Decisions and Repercussions for Mothers in Medicine” characterizes and measures burnout in physician mothers. We surveyed The Physician Mom’s Group (PMG), an active online Facebook community of physician mothers. Because of the online group, we were able to reach over 1,500 women physicians across practice settings and specialties around the U.S. in a way that is not possible in offline settings. We conducted a mixed-methods analysis to understand the extent and types of mistakes physicians make, the impact of those mistakes on patients and physicians, and how making mistakes related to feelings of burnout. (Gupta et al. Under Review, 2018).

Social Networks Online Working As a Behavioral and Learning Laboratory (SNOWBALL) is a three-year R01 grant with my co-PI Damon Centola. SNOWBALL uses social media to influence cervical cancer prevention and detection online. We disseminated test Twitter messages on cervical cancer prevention in an online experimental forum. We found that facts were shared more than narratives, which runs counter to the evidence that narratives change behavior. Organizations’ tweets shared more than individual tweets (Zhang et al. Under Review, 2018)
Q. What is the next frontier for social media research in terms of how it could be used in new ways? Is it just Facebook and Twitter?
A: Since social media has become a popular medium of communication, it is incumbent upon public health researchers to harness social media for health promotion. Facebook and Twitter are just the beginning and serve as platforms for public health interventions. I’d like to harness artificial intelligence to effectively target public health promotion content to people the way that advertisers seem to know exactly what to offer to me when I am on social media. I’d also like to use tools like bots, which have been used for nefarious purposes, to support health and cancer prevention.

In the future, I hope to see more science influencers. According to Science, “Katy Perry has more than 100 million followers on Twitter. Meanwhile, celebrity scientist Neil deGrasse Tyson has only 8.7 million followers.” Finding public health advocates to share evidence-based messages will be the next frontier.
Q. The third annual PRISM Conference is scheduled for Friday, December 7th in Mission Bay. Who could benefit from attending?
A: The PRISM Health Symposium is a unique opportunity for stakeholders involved in all aspects of social media and health research to gather and advance their work. Academic researchers, industry partners, patients, advocates, and others are invited to join us for an exciting day of innovation and collaboration. PRISM provides a one-of-a-kind opportunity for participants to meet, collaborate, and advance their work. We highly encourage early career professionals, students, trainees, and groups traditionally under-represented in medicine to attend.
PRISM PRISM Health Symposium, Dec 7th at Mission Bay.
Deadline to register is October 21st.
Please visit for more information.