A new app tracks breathing to detect an opioid overdose
Called Second Chance, the smartphone application could call for help in an emergency
A new smartphone app may help people who shoot up alone get medical treatment if they accidentally overdose.
The app, dubbed Second Chance, monitors its user for breathing problems that foreshadow an opioid overdose (SN: 3/31/18, p. 18). In an emergency, the app could call 911 or send an SOS to friends or family who could provide opioid-counteracting medication.
“Being able to track an overdose when a person may be by themselves could significantly improve the ability to save lives,” says psychiatrist Nora Volkow, director of the National Institute on Drug Abuse in Bethesda, Md., who was not involved in developing the app. More than 115 people die from an opioid overdose every day in the United States, according to the NIDA, and many victims are alone or with people who are either untrained or too impaired to help.
Second Chance, described online January 9 in Science Translational Medicine, converts a smartphone’s speaker and microphone into a sonar system that works within about a meter of a user’s body. When the app is running, the phone continuously emits sound waves at frequencies too high to hear, which bounce off a user’s chest. Tracking when these echoes reach the phone allows the app to detect two possible signs of an impending overdose: slow breathing or no breathing at all.
Computer scientist Rajalakshmi Nandakumar and colleagues at the University of Washington in Seattle tested Second Chance at a legally sanctioned injection facility in Vancouver, where people self-inject illicit opioids, like heroin and fentanyl, under medical supervision in an effort to prevent overdoses. The researchers used the app to detect dangerous breathing patterns in 94 users that could indicate the beginning of a lethal overdose after drug injection.
Once a user self-injected, Second Chance observed their chest movements for five minutes — the critical window when an overdose is most likely to occur — and raised a red flag if the person stopped breathing for at least 10 seconds or inhaled fewer than eight times per minute. The app caught all but two of 49 cases where the user stopped breathing, and raised a false alarm in one of 45 cases where the user kept breathing.
The system missed six of 47 cases where a patient was breathing too slowly, and misjudged five of the 47 cases where the user was breathing frequently enough.
The researchers also tested the app’s performance in conditions that mimic a life-threatening overdose using anesthesia, where a user stops breathing for an extended period. During surgeries, drugs are often administered to suppress a patient’s breathing, says study coauthor and anesthesiologist Jacob Sunshine. An operating room is “a safe environment to do that, because you’re connected to monitors, you’ve been given lots of oxygen,” he says.
Researchers simulated overdoses with 20 volunteers who received standard anesthetic medications that caused 30 seconds of slow or no breathing. The app detected abnormal respiration in 19 patients.
For real-world use, the researchers envision the app notifying a user if it detects breathing problems and sending for help only if the user doesn’t respond to that notification, says study coauthor and computer scientist Shyam Gollakota. The scientists still need to ensure that this setup could reliably alert emergency contacts or medical personnel in time to resuscitate a person.