ABSTRACT
Social media sites have struggled with the presence of emotional and physical self-injury content. Individuals who share such content are often challenged with severe mental illnesses like eating disorders. We present the first study quantifying levels of mental illness severity (MIS) in social media. We examine a set of users on Instagram who post content on pro-eating disorder tags (26M posts from 100K users). Our novel statistical methodology combines topic modeling and novice/clinician annotations to infer MIS in a user's content. Alarmingly, we find that proportion of users whose content expresses high MIS have been on the rise since 2012 (13%/year increase). Previous MIS in a user's content over seven months can predict future risk with 81% accuracy. Our model can also forecast MIS levels up to eight months in the future with performance better than baseline. We discuss the health outcomes and design implications as well as ethical considerations of this line of research.
- Sofiane Abbar, Yelena Mejova, and Ingmar Weber. 2015. You Tweet What You Eat: Studying Food Consumption Through Twitter. In Proc. CHI. Google ScholarDigital Library
- Nazanin Andalibi, Pinar Ozturk, and Andrea Forte. 2015. Depression-related Imagery on Instagram. In Proc. CSCW'15 Companion. 231–234. Google ScholarDigital Library
- Jon Arcelus, Alex J Mitchell, Jackie Wales, and Sren Nielsen. 2011. Mortality rates in patients with anorexia nervosa and other eating disorders: a meta-analysis of 36 studies. Archv. Gen. Psychiatry 68, 7 (2011), 724–731.Google ScholarCross Ref
- American Psychiatric Association and others. 2013. Diagnostic and statistical manual of mental disorders, (DSM-5). American Psychiatric Pub.Google Scholar
- Jin Yeong Bak, Suin Kim, and Alice Oh. 2012. Self-disclosure and relationship strength in twitter conversations. In Proc. ACL. 60–64. Google ScholarDigital Library
- Anna M Bardone-Cone and Kamila M Cass. 2006. Investigating the impact of pro-anorexia websites: A pilot study. European Eating Disorders Review 14, 4 (2006), 256–262.Google ScholarCross Ref
- David M Blei, Andrew Y Ng, and Michael I Jordan. 2003. Latent dirichlet allocation. Journal of Machine Learning Research (JMLR) 3 (2003), 993–1022. Google ScholarDigital Library
- Dina LG Borzekowski, Summer Schenk, Jenny L Wilson, and Rebecka Peebles. 2010. e-Ana and e-Mia: A content analysis of pro–eating disorder web sites. American journal of public health 100, 8 (2010), 1526.Google Scholar
- Leo Breiman. 2001. Random forests. Machine learning 45, 1 (2001), 5–32. Google ScholarDigital Library
- Jay Callahan. 1996. A specific therapeutic approach to suicide risk in borderline clients. Clinical Social Work Journal 24, 4 (1996), 443–459.Google ScholarCross Ref
- Richard F Catalano and J David Hawkins. 1996. A Theory of Antisocial Behavior. Delinquency and crime: Current theories (1996), 149.Google Scholar
- Laurence Claes, Koen Luyckx, Patricia Bijttebier, Brianna Turner, Amarendra Ghandi, Jos Smets, Jan Norre, Leen Van Assche, Els Verheyen, Yvienne Goris, and others. 2015. Non-Suicidal Self-Injury in Patients with Eating Disorder: Associations with Identity Formation Above and Beyond Anxiety and Depression. European Eating Disorders Review 23, 2 (2015), 119–125.Google ScholarCross Ref
- Michael Collins, Robert E Schapire, and Yoram Singer. 2002. Logistic regression, AdaBoost and Bregman distances. Machine Learning 48, 1-3 (2002), 253–285. Google ScholarDigital Library
- Karen Conterio, W Lader, and JK Bloom. 1998. Bodily harm: The breakthrough treatmentprogramfor self-injurers. (1998).Google Scholar
- Glen Coppersmith, Mark Dredze, and Craig Harman. 2014. Quantifying mental health signals in twitter. In Proc. ACL CLCP Workshop.Google ScholarCross Ref
- Munmun De Choudhury. 2015. Anorexia on Tumblr: A Characterization Study. In Proc. Digital Health. Google ScholarDigital Library
- Munmun De Choudhury, Scott Counts, Eric Horvitz, and Aaron Hoff. 2014. Characterizing and Predicting Postpartum Depression from Facebook Data. In Proc. CSCW. Google ScholarDigital Library
- Munmun De Choudhury, Michael Gamon, Scott Counts, and Eric Horvitz. 2013. Predicting depression via social media. In Proc. ICWSM.Google Scholar
- Sharon K Farber, Craig C Jackson, Johanna K Tabin, and Eytan Bachar. 2007. Death and annihilation anxieties in anorexia nervosa, bulimia, and self-mutilation. Psychoanalytic Psych 24, 2 (2007), 289.Google ScholarCross Ref
- Angela Favaro and Paolo Santonastaso. 1998. Impulsive and compulsive self-injurious behavior in bulimia nervosa: Prevalence and psychological correlates. The Journal of Nervous and Mental Disease 186, 3 (1998), 157–165.Google ScholarCross Ref
- Armando R Favazza, Lori DeRosear, and K Conterio. 1989. Self-Mutilation and Eating Disorders. Suicide & Life-Threatening Behavior 19, 4 (1989), 352–361.Google Scholar
- Rachel A Fleming-May and Laura E Miller. 2010. I'm scared to look. But I'm dying to know: Information seeking and sharing on Pro-Ana weblogs. Proc. ASIST 47, 1 (2010), 1–9. Google ScholarDigital Library
- Yoav Freund, Robert E Schapire, and others. 1996. Experiments with a new boosting algorithm. In ICML, Vol. 96. 148–156.Google ScholarDigital Library
- W Gaebel, H Zäske, and AE Baumann. 2006. The relationship between mental illness severity and stigma. Acta Psychiatrica Scandinavica 113, s429 (2006), 41–45.Google ScholarCross Ref
- Sarah A St Germain and Jill M Hooley. 2012. Direct and indirect forms of non-suicidal self-injury: Evidence for a distinction. Psychiatry research 197, 1 (2012), 78–84.Google Scholar
- Erving Goffman and others. 1959. The presentation of self in everyday life. (1959).Google Scholar
- Kim L Gratz, Sheree Dukes Conrad, and Lizabeth Roemer. 2002. Risk factors for deliberate self-harm among college students. American Journal of Orthopsychiatry 72, 1 (2002), 128.Google ScholarCross Ref
- P Hasking, R Momeni, S Swannell, and S Chia. 2008. The nature and extent of non-suicidal self-injury in a non-clinical sample of young adults. Archives of Suicide Research 12, 3 (2008), 208–218.Google ScholarCross Ref
- Matthew Hoffman, Francis R Bach, and David M Blei. 2010. Online learning for latent dirichlet allocation. In Proc. NIPS.Google Scholar
- Christopher M Homan, Ravdeep Johar, Tong Liu, Megan Lytle, Vincent Silenzio, and Cecilia Alm. 2014a. Toward Macro-Insights for Suicide Prevention: Analyzing Fine-Grained Distress at Scale. ACL 2014 (2014), 107.Google ScholarCross Ref
- Christopher M Homan, Naiji Lu, Xin Tu, Megan C Lytle, and Vincent Silenzio. 2014b. Social structure and depression in TrevorSpace. In Computer-Supported Cooperative Work and Social Computing (CSCW). Google ScholarDigital Library
- James I Hudson, Eva Hiripi, Harrison G Pope, and Ronald C Kessler. 2007. The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication. Biological psychiatry 61, 3 (2007), 348–358.Google Scholar
- Grace J Johnson and Paul J Ambrose. 2006. Neo-tribes: The power and potential of online communities in health care. Commun. ACM 49, 1 (2006), 107–113. Google ScholarDigital Library
- Adrienne S Juarascio, Amber Shoaib, and C Alix Timko. 2010. Pro-eating disorder communities on social networking sites: a content analysis. Eating disorders 18, 5 (2010), 393–407.Google Scholar
- Kelly L Klump, Cynthia M Bulik, Walter H Kaye, Janet Treasure, and Edward Tyson. 2009. Academy for eating disorders position paper: eating disorders are serious mental illnesses. Int J Eat Disord 42, 2 (2009), 97–103.Google ScholarCross Ref
- Katrina Kostro, Jessica B Lerman, and Evelyn Attia. 2014. The current status of suicide and self-injury in eating disorders: a narrative review. Journal of Eating Disorders 2, 1 (2014), 19.Google ScholarCross Ref
- Reeva Lederman, Greg Wadley, John Gleeson, Sarah Bendall, and Mario Álvarez-Jiménez. 2014. Moderated online social therapy: Designing and evaluating technology for mental health. ACM Transactions on Computer-Human Interaction (TOCHI) 21, 1 (2014), 5. Google ScholarDigital Library
- Stephen P Lewis and Alexis E Arbuthnott. 2012. Searching for thinspiration: the nature of internet searches for pro-eating disorder websites. Cyberpsychology, Behavior, and Social Networking 15, 4 (2012), 200–204.Google ScholarCross Ref
- Rebecca MacKinnon. 2008. Flatter world and thicker walls? Blogs, censorship and civic discourse in China. Public Choice 134, 1-2 (2008), 31–46.Google Scholar
- Diana MacLean, Sonal Gupta, Anna Lembke, Christopher Manning, and Jeffrey Heer. 2015. Forum77: An Analysis of an Online Health Forum Dedicated to Addiction Recovery. In Proc. CSCW. Google ScholarDigital Library
- Jeanne B Martin. 2010. The development of ideal body image perceptions in the United States. Nutrition Today 45, 3 (2010), 98–110.Google ScholarCross Ref
- Terence C Mills. 1991. Time series techniques for economists. Cambridge University Press.Google Scholar
- Paul Moran, Carolyn Coffey, Helena Romaniuk, Craig Olsson, Rohan Borschmann, John B Carlin, andGoogle Scholar
- ge C Patton. 2012. The natural history of self-harm from adolescence to young adulthood: a population-based cohort study. The Lancet 379, 9812 (2012), 236–243.Google Scholar
- Elizabeth L Murnane and Scott Counts. 2014. Unraveling abstinence and relapse: smoking cessation reflected in social media. In Proc. CHI. Google ScholarDigital Library
- Matthew K Nock and Mitchell J Prinstein. 2004. A functional approach to the assessment of self-mutilative behavior. Journal of consulting and clinical psychology 72, 5 (2004), 885.Google ScholarCross Ref
- Mark L Norris, Katherine M Boydell, Leora Pinhas, and Debra K Katzman. 2006. Ana and the Internet: A review of pro-anorexia websites. International Journal of Eating Disorders 39, 6 (2006), 443–447.Google ScholarCross Ref
- Minsu Park, David W McDonald, and Meeyoung Cha. 2013. Perception Differences between the Depressed and Non-depressed Users in Twitter. In Proc. ICWSM.Google Scholar
- Michael J Paul and Mark Dredze. 2011. You Are What You Tweet: Analyzing Twitter for Public Health. In Proc. ICWSM.Google Scholar
- Rebecka Peebles, Jenny L Wilson, and James D Lock. 2011. Self-injury in adolescents with eating disorders:Google Scholar
- elates and provider bias. Journal of Adolescent Health 48, 3 (2011), 310–313.Google Scholar
- Mirella Ruggeri, Morven Leese, Graham Thornicroft, Giulia Bisoffi, and Michele Tansella. 2000. Definition and prevalence of severe and persistent mental illness. The British Journal of Psychiatry 177, 2 (2000), 149–155.Google ScholarCross Ref
- Hansen Andrew Schwartz, Johannes C Eichstaedt, Margaret L Kern, and Lukasz et al. Dziurzynski. 2013. Characterizing Geographic Variation in Well-Being Using Tweets. In ICWSM.Google Scholar
- Yukari Seko. 2013. Picturesque Wounds: A Multimodal Analysis of Self-Injury Photographs on Flickr. In Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, Vol. 14.Google Scholar
- Leslie Regan Shade. 2003. Weborexics: The Ethical Issues Surrounding Pro-Ana Websites. SIGCAS Comput. Soc. 33, 4 (Dec. 2003). Google ScholarDigital Library
- Frédérique RE Smink, Daphne van Hoeken, and Hans W Hoek. 2012. Epidemiology of eating disorders: incidence, prevalence and mortality rates. Current psychiatry reports 14, 4 (2012), 406–414.Google Scholar
- John R Suler and Wende L Phillips. 1998. The bad boys of cyberspace: Deviant behavior in a multimedia chat community. CyberPsychology & Behavior 1, 3 (1998), 275–294.Google ScholarCross Ref
- Karen L Suyemoto. 1998. The functions of self-mutilation. Clinical psych rev 18, 5 (1998), 531–554.Google Scholar
- Brianna J Turner, Angelina Yiu, Brianne K Layden, Laurence Claes, Shannon Zaitsoff, and Alexander L. Chapman. 2015. Temporal associations between disordered eating and nonsuicidal self-injury: examining symptom overlap over 1 year. Behavior Therapy 46, 1 (2015), 125–138.Google ScholarCross Ref
- Robert F Valois, Keith J Zullig, and Amy A Hunter. 2013. Association between adolescent suicide ideation, suicide attempts and emotional self-efficacy. Journal of Child and Family Studies 24, 2 (2013), 237–248.Google ScholarCross Ref
- Markus Wolf, Florian Theis, and Hans Kordy. 2013. Language Use in Eating Disorder Blogs Psychological Implications of Social Online Activity. Journal of Language and Social Psychology 32, 2 (2013), 212–226.Google ScholarCross Ref
- Elad Yom-Tov, Luis Fernandez-Luque, Ingmar Weber, and P Steven Crain. 2012. Pro-Anorexia and Pro-Recovery Photo Sharing: A Tale of Two Warring Tribes. J Med Internet Res (2012).Google Scholar
Recommendations
Detecting and Characterizing Eating-Disorder Communities on Social Media
WSDM '17: Proceedings of the Tenth ACM International Conference on Web Search and Data MiningEating disorders are complex mental disorders and responsible for the highest mortality rate among mental illnesses. Recent studies reveal that user-generated content on social media provides useful information in understanding these disorders. Most ...
Sensitive Self-disclosures, Responses, and Social Support on Instagram: The Case of #Depression
CSCW '17: Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social ComputingPeople can benefit from disclosing negative emotions or stigmatized facets of their identities, and psychologists have noted that imagery can be an effective medium for expressing difficult emotions. Social network sites like Instagram offer ...
Beyond the Coded Gaze: Analyzing Expression of Mental Health and Illness on Instagram
In CSCW and HCI, work examining expression of mental health and illness on social media frequently aims to classify content, quantify visual trends, and predict user states. This approach to analysis is a form of the coded gaze, a type of algorithmic '...
Comments