[LUM#11] Prevention 2.0
Preventing suicide risk… with a cell phone. This is the project led by psychiatrist Philippe Courtet, who has assembled a team of computer scientists from Lirmm to develop the Emma app, designed to detect suicide risk.
With more than 9,000 deaths each year, France has one of the highest suicide rates in Europe. It is even the leading cause of death among people aged 25–34. Worldwide, someone takes their own life every 40 seconds. This is a major public health issue and a challenge for psychiatrists grappling with this thorny problem: how can we prevent suicide?
“We are powerless when it comes to predicting suicide risk,” explains Philippe Courtet. “A patient never commits suicide in front of their psychiatrist during a consultation. Until now, all we could do was diagnose the risk of suicide after the fact, when we see a patient and discuss with them the events leading up to their act,” emphasizes the head of the psychiatric emergency department at Montpellier University Hospital.
"Recall Bias"
There is one major obstacle, however: during these consultations, patients have little or no recollection of those moments. To avoid this “recall bias” and better understand the context in which the suicide occurred, specialists lacked a “tool for continuous monitoring to determine the risk of suicide in real time, ” emphasizes Philippe Courtet.
It was to overcome this problem that the specialist developed the Emma app—which stands for “Ecological Mental Momentary Assessment.” It is an essential tool “for monitoring the patient in real time, in their everyday life and daily environment, and not just during a monthly consultation.”
In practice, the app uses two types of assessments: passive and active. “Passive assessment analyzes how the patient uses their phone: calls, texts, social media…,” explain Jérôme Azé and Sandra Bringay of the Montpellier Laboratory of Computer Science, Robotics, and Microelectronics. While the app knows nothing about the content of these interactions, simply knowing how a person uses their phone provides valuable insight. “If we notice that a patient starts using their phone all night long, or that they’re no longer contacting their friends, or generally that their habits have changed, it’s an indicator that perhaps they’re not doing well, ” the psychiatrist points out.
Valuable clues
For the interactive portion of the assessment, Emma will directly engage the user through “electronic interviews.” The app asks the patient to answer questions such as “How are you doing?”, “Are you feeling anxious?”, or “Are you having dark thoughts?”. If the answers suggest distress, the app triggers what’s known as the intervention module. “It’s a personalized safety plan that was developed in advance with the patient and their psychiatrist,” explains Philippe Courtet. “The app then offers the patient access to a relaxation and stress management module developed by a doctor on the team. They can also listen to music or view photos or videos they’ve selected in advance.”
As a last resort, the app may suggest that the patient contact pre-selected loved ones. “With this system, we’re tapping into the protective power of social connections to prevent suicide, ” explains the specialist. Emma may also suggest that the patient call 15 or the VigilanS center, a “re-engagement program ” for people who have attempted suicide. “The goal of this safety plan is to delay the very brief window of time that can lead to a suicide attempt, ” explains the psychiatrist.
Reach out
Initially, the app is being tested with 100 patients, who are monitored for 6 months. This phase allows researchers to assess how patients use Emma, but it also aims to collect as much data as possible, which Jérôme Azé and Sandra Bringay’s team will use to create an algorithm to refine the app’s functionality. “We’re using artificial intelligence—and more specifically, what’s known as deep learning—to try to build the most accurate algorithm possible for predicting suicide risk, which will be used in a second version of the app, ” the computer scientists explain.
Supported by the FondaMental Foundation, this project has already proven its value. “The initial feedback has been positive,” explains Philippe Courtet. “Patients report that the app’s suggestion to call a loved one has helped them realize that people care about them. With Emma, we’re reaching out to people rather than waiting for them to reach out to us.”
Check out theUM podcasts, now available on your favorite platform (Spotify, Deezer, Apple Podcasts, Amazon Music, etc.).