This post was written by Jessica Pater, MS, PhD, manager and research scientist, Health Services and Informatics Research, Parkview Mirro Center for Research and Innovation.
Approximately 72% of American adults utilize at least one social media site, many visiting those sites daily. [1] And, given the volume of personal information shared on social media each day, it’s not surprising that researchers are looking at this data to better understand how people use these sites to help draw connections between mental health and well-being. [2]
Interestingly, the ability to predict if a social media user is experiencing a mental health struggle is possible thanks to advancements in computing such as image processing, natural language processing and artificial intelligence. From Pinterest and Facebook to Tik-Tok and Instagram, the images, videos and texts you share tell a story about who you are. Looking at patterns over time provides insights into someone’s overall emotional and mental state. For example, are you experiencing post-partum depression, or have you been more anxious since starting college? Now, while not everyone will share content related to specific mental health challenges, the patterns in social media use and other metadata (a set of data that provides information about other data) can indicate changes in mental well-being like increased depression or anxiety. [3]
Researching health and wellness
Using social media to understand our mental state is nothing new. For the past 10 years, scholars have examined ways to use social media content to predict depression. [4-6] This body of research has grown to assess several mental health conditions including, but not limited to:
- Depression
- Anxiety
- Suicidal risk/ideation
- Eating disorders and other self-harm
- Schizophrenia
- Bipolar
Currently, it’s rare for social media data to be part of the diagnostic or treatment processes employed by mental health care professionals. However, the Division of Digital Psychiatry at Beth Israel Deacon in Boston and our Health Services and Informatics Research lab at Parkview Health are leading the way for new research on how to effectively integrate digital signals (social media content) into the clinical setting. Like other behavior patterns that must change to improve a health condition, social media use could potentially have an adverse effect on recovery. [3] As a result, social media data is another means for a provider to understand their patient and a reflection of a critical health behavior that should be considered when treatment planning. [7]
Accessing social media data
It might surprise you, but there are many ways researchers can access and collect social media data, including:
- Using verified access paths (agreements with social media companies) and submitting requests for permission to download data through a secure connection
- Purchasing large sets of data from third parties
- Asking people to share their information
- Manually collecting publicly available data
Moreover, researchers can also collect data based on a variety of aspects, which could include the groups/forums you belong to within the social media site, hashtags you use and elements of your demographics or user profile.
Plus, if you set your profile or posts to “public,” your data has most likely been used at some point in a social media research study. All social media companies have teams of scientists who regularly research with end-user data.
How social media companies use the information
Many companies utilize end-user data with the best of intentions to help people in moments of distress. One approach provides interventions within the social media site for individuals posting or searching for information related to a mental health issue. For example, on Instagram, if you search for a term commonly associated with eating disorders, you might see a search window popup. While the user can still view all search results, the site will also share related mental health resources and tips like encouragement to reach out to a friend, a link to talk with a helpline volunteer or direct them toward other support paths.
Why it’s important
Social media is a tool that many of us use in our daily lives, but it’s important we understand how the information we share publicly can be used, including:
- Who can see your information
- What the social media company has the right to
- How your data might be used for research (internally by the social media company and externally by the general public)
Granted, you can find most of this information in the Terms of Service and other policy-related documents. Still, you should regularly check your privacy settings because they can switch back to default when site or policy updates occur.
Upcoming Event
What: 4th Annual Tech Talk: Youth, Technology and Mental Health Virtual Conference
When: Wednesday, November 10, 2021, from 6:30 – 8:05 p.m.
This virtual event is open to the community; children, teens and families are encouraged to attend. The conference will focus on educating teens, parents, professionals and educators about healthy technology habits, how to be safe and smart, and the impacts of COVID-19 on children. For more information or to register, please contact Lauren Reining at 260-266-7771.
Sources
[3] Pater, J. A., Farrington, B., Brown, A., Reining, L. E., Toscos, T., & Mynatt, E. D. (2019). Exploring indicators of digital self-harm with eating disorder patients: A case study. Proceedings of the ACM on Human-Computer Interaction, 3 (CSCW), 1-26.
[4] Kotikalapudi, R.; Chellappan, S.; Montgomery, F.; Wunsch, D.; & Lutzen, K. 2012. Associating depressive symptoms in college students with internet usage using real Internet data. IEEE Technology and Society Magazine.
[5] Moreno, M.; Jelenchick, L.; Egan, K.; Cox, E. et al. 2011. Feeling bad on Facebook: depression disclosures by college students on a social networking site. Depression and Anxiety 28(6):447–455.
[6] De Choudhury, M., Gamon, M., Counts, S., & Horvitz, E. (2013, June). Predicting depression via social media. In Seventh International AAAI Conference on weblogs and social media.
[7] Pater, J., Nova, F. F., Coupe, A., Reining, L. E., Kerrigan, C., Toscos, T., & Mynatt, E. D. (2021, May). Charting the Unknown: Challenges in the Clinical Assessment of Patients’ Technology Use Related to Eating Disorders. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (pp. 1-14).
