Social media has forever changed the landscape for our patients. Now clinicians at Zucker Hillside Hospital are working to incorporate social media into clinical care.
"We have the opportunity to transform health care if we can learn how to take advantage of this tool," said Michael Birnbaum, MD, director of Zucker Hillside's Early Treatment Program and assistant investigator at the Center for Psychiatric Neuroscience at Feinstein Institute for Medical Research.
Dr. Birnbaum, who studies social media use and language patterns, and a team of researchers at Zucker Hillside Hospital began by analyzing the linguistic structure of social media posts. They are now adding additional variables, such as the time and frequency of posting, how patients engage others on the platform and the amount of information patients share or what they respond to.
Plugging such data into advanced computational algorithms could give behavioral health clinicians a readily available tool to identify precursors of psychotic episodes, which could facilitate early intervention and significantly enhance long-term outcomes.
Reducing Treatment Wait Times
Recent research conducted by the Zucker Hillside team shows that initial evaluation typically occurs more than one year after the onset of psychosis.
"In a study conducted across 21 states, we found a median duration of untreated psychosis of 74 weeks," said John Kane, MD, Northwell Health's senior vice president for behavioral health and professor and chair of psychiatry at Hofstra Northwell School of Medicine. "This is a huge concern, as psychosis deleteriously affects a person and his or her family. Any tool we can develop to reduce the duration of untreated psychosis is a boon."
Cluing In to Relapse
Changes in language utilization or social media habits could also indicate signs of relapse or exacerbation of symptoms associated with psychosis.
"We're hoping to pinpoint individuals who are in the early stages of relapsing," Dr. Birnbaum said. "If we could identify early on who is relapsing, we could intervene and reduce possible negative outcomes, such as hospitalization. We don't want to wait until patients are in crisis. We want an algorithm that can identify early red flags associated with psychotic symptoms."