Studying social networking habits is the latest approach the mental health community is taking to explore new ways to detect early signs of mental health issues and “intervene” with their usual treatment – a psychiatric drug prescription.
One such researcher is Dr. Munmum De Choudhury, a PhD in computer science who is currently an Asst. Professor in the School of Interactive Computing at the Georgia Institute of Technology.
Her earlier work in 2013 involved a study of the Facebook and Twitter activity of expectant mothers in order to attempt to predict whether a particular individual was likely to a victim for the diagnosis of “post-partum depression”.
She claims to have found a model that could predict with 80% accuracy
which moms would end up with depression.
The next step?
“…the design and deployment of new kinds of early warning systems that could bring people timely information and assistance” – assistance being an anti-depressant drug.
She states that her studies will spark discussions about privacy and ethics and that people might be uncomfortable with others making predictions about their mental health and making those predications public. Her justification is that she’s using data “they have shared openly with the public.”
In Dec of 2103 she was interviewed while working at Microsoft Research using Twitter and other social networking to study depression. She had been looking into how often people posted, what types of posts they shared, when they posted, how they were connected to friends and friends of friends and the language style of their posts.
She was surprised to learn, with the help of a psychologist, called in to help on the project that people who never mentioned they were “depressed” still gave off signals in various nuanced ways that showed they were in fact suffering from depression.
“This gives us hope that observing social media use of people over time—something which is increasingly gaining popularity—can be used to build tools, forecasting algorithms, interventions, and prevention strategies for both individuals themselves as well as policymakers to help them deal with and manage this medical condition in a better way.”
Yes, depression is a “medical condition” despite no medical tests for it and it requires “intervention.”
She goes on to state, “In terms of intervention, since our estimates of depression can be made considerably more frequent than conventional surveys such as by the Centers for Disease Control (CDC), the estimates can be utilized time to time to enable early detection and rapid treatment of depression in sufferers.”
Drug companies will love any system that can find more depressed people than the CDC can!
Her latest work was done with another Microsoft Researcher and was a study of two years of Twitter data from 4 cities in Mexico where the violence between the government forces and the drug cartels, between rival drug cartels and between cartels and citizen vigilante groups was so brutal journalists reporting it were in danger and the newspaper and media stopped covering the stories.
Citizen bystanders thus used Twitter and a blog called  the Blog del Narco in order to acquire and share news and warnings about what was going on.
Dr. Choudhury and her researcher partners studied this social networking with algorithms designed to scan tweets for specific words, message frequency and when items were posted and replied to.
They went into the project looking for signs of people getting numb and de-sensitized to violence and for signs of which persons would have post-traumatic stress disorder. They did find some evidence of these mental health issues in the social media posts they studied but acknowledged that the study was purely correlations and they could not draw a cause and effect conclusion that exposure to prolonged violence actually led to desensitization in these people.
They also did not find any simple tips to identify mental health issues in individuals from the study of the social media postings.
Yet despite those acknowledgments, Dr. Choudhury was quoted as saying that what they did find could help “make interventions to make sure that people receive the right kind of attention at the right time”
Recently “TIME” magazine reported on a few more researchers in its Jan 2014 article entitled,  “How Twitter Knows When You’re Depressed.”
They quoted Eric Horvitz, co-director of Microsoft Research Redmond.
“We wondered if we could actually build measures that might be able to detect if someone is severely depressed, just in publicly posted media. What are people telling the world in public spaces? You might imagine tools that could make people aware of a swing in mood, even before they can feel it themselves.”
Microsoft researchers acquired the Twitter records for 476 people users; 171 of them who had  severe depression. They analyzed 2.2 million Tweets going back to a year before depression began and formed a model of what a depressed or soon to be depressed person’s Tweets would look like. They studied the number of Tweets, time of day of Tweeting and words used in the Tweets.
They claim it predicted future depression with 70% accuracy despite being baffled by the study showing that words like she, him, girl, game, men, home, fun, house, favorite, wants, tolerance, cope, amazing, love, care, songs, and movie could be indications of depression as well as ones that they expected such as anxiety, severe, appetite, suicidal, nausea, drowsiness, fatigue, nervousness, addictive, attacks, episodes, and sleep.
Also, they found, if you Tweet after 9:00 PM or stop Tweeting so often, you could soon be a victim of depression.
A different researcher at the Univ. of Washington has identified a group of first-year students at a number of colleges across the country based on their Twitter feeds—hashtags and posts relating to orientation. From this he hopes to isolate the “red flags” that indicate emotional problems.
Another study by The University of California in San Diego has been funded by the federal government’s National Institute of Health. Here Michael Conway is creating models that will eventually track depression in communities and figure out how to apply mental health resources to better assess public health. “The ultimate goal of this work is to provide a cost-effective, real-time means of monitoring the prevalence of depression in the general population,” Conway said in an email.
Prozac in the water supply, anyone?
This “science” has a long way to go and millions of citizens around the world have no interest in being snooped and spied on by social scientists ready to label them “depressed” when they feel just fine, thank you very much.
Researchers like Horvitz admit they will have to deal with those skeptical of their work and concerned with privacy issues
And Conway’s team is looking at the ethics of his work by “investigating public attitudes towards the ethics of using social media for public health monitoring.” In other words, he wants to find out how to spin it so the public will accept more NSA type spying and Big Brother style mental health treatments
This type of research is turning traditional American liberty on its head. This social scientists would label you guilty until proven innocent. The artificial intelligence computers and algorithms will say you are mentally ill and you’d better take your anti-depressants until you can prove you aren’t.
http://www.ihealthbeat.org/articles/2014/3/17/study-social-media-can-locate-identify-mental-health-needs
http://research.microsoft.com/pubs/208580/affect_desensitize-v29.pdf
http://www.nbcnews.com/tech/social-media/can-social-media-help-direct-mental-health-aid-n53801
http://time.com/1915/how-twitter-knows-when-youre-depressed/
http://blog.gnip.com/using-social-data-to-understand-depression/
http://www.munmund.net/pubs/chi_13.pdf