This project is a requirement of the GOA Abnormal Psychology Course. Using the process of design thinking, a challenge in the world of mental health was identified, interviews and research were undertaken, and a solution prototype was developed. Below you will find information about the identified area of concern and my proposed solution. Please feel free to provide feedback on this idea. For more information on the process of Design Thinking, click here.
For many industries, professions, and lives big data has become a big deal. As it becomes possible to collect more data and develop increasingly complex algorithms, we have transformed the way many systems work, for the better. Yet this transformation is just starting to reaching the treatment and prevention of mental illness. There are hundreds of ways data can and will be used to aid mental health professionals in caring for patients, and in this project, we will explore just a few of the ways data is starting to be used, it’s potential, and the challenges that still remain in the way of using more data-driven systems. While data may seem like a very impersonal way to look at mental health treatment, given that data can be the difference between noticing when someone is deteriorating and getting them the support they need, and that same person, at the fault of no one person, falling through the cracks, data, in fact, can be very human.
Preventing suicide among teens and young adults is something many people work tirelessly to do, myself included. But, until this year I had never thought that data fit into that equation, however, with the recent rise in social media use among teens and young adults it does. Dr. Molly C. Adrian a psychologist from Seattle Children’s Hospital is developing a program that uses teen’s social media profiles to assess and predict suicide risk (Matlick).
The program is modeled off of the Durkheim Project, a program that uses keywords from analysis of psychiatrist notes about veteran patients to predict up to 70% of suicide attempts in the population (Matlick). The ability to use data to predict 70% of suicide attempts is incredible, adapting this to teens and social media has the potential to save lives is huge. This system could also be expanded past just mental health professionals, schools, parents, or individual teams could begin using the program, setting it to alert a chosen adult if there is a risk of suicide detected (Adrian). This study, however, is still looking for help collecting data, if you want to learn more about donating your data please follow the guidelines in the action section below.
A similar idea was also investigated by a recent study that creates models to predict when people were likely to develop depression or Post-Traumatic Stress Disorder from their Twitter accounts. In the study data from Twitter as well as personal mental health, details were collected from 204 individuals (105 depressed, 99 healthy), to develop an algorithm to predict users at risk (Reece).
Feedback Informed Treatment
Using social media is just one new innovative approach to data usage, another comes in the form of assisting therapists. It is often much more difficult then one might imagine for therapists to get an accurate sense of whether or not a patient is deteriorating or will deteriorate (deterioration is when a patient begins to get worse, deterioration can lead to not showing up for sessions, returning to negative habits, and even suicide). One study found that of several hundred clients being seen by 48 therapists at a single clinic, only one of the therapists actually was able to identify a client at risk, when in fact 40 clients did deteriorate (“What Your Therapist…”).
But the use of feedback informed treatment or FIT for short has begun to address this issue.
In an article in The Atlantic titled What Your Therapist Doesn’t Know?, therapist and member of the clinical faculty at the University of Washington Tony Rousmaniere talks about his experience using FIT in his practice. Rousmaniere says “The system aids therapy in two primary ways. First, it provides an element of blunt performance feedback that therapists too often lack. Many clients are more willing to report worsening symptoms to a computer—even if they know that their therapist will see the results—than disappoint their therapist face-to-face.The second benefit comes from the metrics: Risk alerts allow therapists to adjust treatment and can help them compensate for natural overconfidence and clinical blind spots.” (“What Your Therapist…”).
While FIT is fairly new we are already starting to see how effective it can be. In a study of the Calgary Counselling Centre (CCC) where the CEO Robbie Babins Wagner began requiring therapists to use FIT, a small but consistent improvement was found in clinical effectiveness every year for seven years, this may sound simple but this is actually a very big deal. In his article Rousmaniere says that this is only the “second-time year-over-year improvement in therapist effectiveness -measured by improved client outcomes – has been empirically demonstrated” (“What Your Therapist…”), meaning the use of FIT systems is offering consistent therapists improvement, something that is very hard to achieve. Yet when at the CCC Babins-Wagner started mandating the use of data 40% of the therapists who worked for her reigned within a few months (“What Your Therapist…”).
Despite the evidence that FIT works many therapists still feel skeptical in using data in a profession that honors a long tradition of intuition and personal interaction. However as Rousinamiere pointed out when I spoke to him over the phone, FIT systems only aim to help alert therapists when they need to be concerned, at which point they would speak with the patient or seek outside counsel (Rousmaniere). FIT doesn’t replace the therapist personal touch or knowledge, similar to how a thermometer would help a doctor, it is just a metric to help therapists catch clients who are at risk. Rousmaniere even points out that when the thermometer was initially introduced, many doctors pushed back against the use of it saying it could not possibly take into account everything they knew and would lead to negative patient outcomes.
What’s Next? Take Action!
Where else data within the prevention and treatment of mental illness can go from here is yet to be seen, but you can help in supporting these data-driven ideas by donating data to the ASSES suicide project here and read the rest of Tony’s article here. While in the world today marketing and commercial interests are some of the biggest users of data, the potential to use big data to significantly better the lives of people all over the world should not be forgotten.
Adrian, Molly C. Interview. 15 Mar. 2018.
Matlick, Justin. “Researchers Use Social Media to Gauge Suicide Risk.” Seattle Children’s, Seattle Children’s Hospital, 24 Mar. 2015, pulse.seattlechildrens.org/researchers-use-social-media-to-gauge-suicide-risk/. Accessed 15 Apr. 2018.
Reece, Andrew G. “Forecasting the onset and course of mental illness with Twitter data.” Nature.nature.com, Nature, doi:10.1038/s41598-017-12961-9. Accessed 15 Apr. 2018. Editorial.
Rousmaniere, Tony. Interview. 4 Feb. 2018.
“What Your Therapist Doesn’t Know.” The Atlantic, Apr. 2017, www.theatlantic.com/magazine/archive/2017/04/what-your-therapist-doesnt-know/517797/. Accessed 15 Apr. 2018.