AI text analysis uncovers hidden mental health signs in emergency dispatchers and 911 dispatchers

Emergency dispatchers and emergency dispatchers act as an invisible front line of public safety, overcoming life-or-death crises through telephone lines and radio without ever seeing the results of their efforts. Now, a new study from San Antonio Polytechnic Institute has revealed that the specific words these workers use to describe their most stressful calls may provide early hidden warning signs of depression and anxiety.

Published in a peer-reviewed journal Pro SwanThis study demonstrates how natural language processing, a branch of artificial intelligence, can assess the psychological well-being of an important but historically understudied workforce.

A research team led by Dr. Vivian Ta Johnson and Dr. Sandra B. Morissette, professors of psychology in the College of Health, Community and Policy, analyzed written accounts from 106 emergency communicators. This investigation was conducted in conjunction with the San Antonio Police Department and with assistance from Bexar Metro 911.

“This collaboration is critical to the advancement of science and the well-being of 911 responders,” said Morissette, who led the establishment of San Antonio’s community partnership.

Persistent occupational gap

Emergency dispatchers and 911 dispatchers are exposed to indirect and ongoing trauma on a daily basis by handling a high volume of calls involving violence, injury, and death. However, they face serious occupational disparities. Unlike on-the-ground first responders such as police officers and firefighters, command center personnel in many jurisdictions are denied classification as official first responders, limiting their access to dedicated health-related resources and funding streams.

“They are beginning However, they often do not receive the same support,” Ta-Johnson explained.

They also face unique psychological burdens due to their lack of closure.

“The common theme is the uncertainty that lingers after an intense and devastating phone call,” Ta-Johnson said, reflecting on the stories she read. “It’s not uncommon for emergency dispatchers and 911 dispatchers to wonder what happened and whether the person they were talking to got the help they needed. They usually don’t have that information directly because the information always leads to the next call.”

Measuring emotional tone and intensity

To understand how this chronic stress shapes mental health, researchers asked participants to complete standard self-report assessments of depression, anxiety, and stress before entering detailed descriptions of high-stress workplace events.

Ta Johnson’s team analyzed these testimonies using natural language processing tools extracted from a validated database of approximately 14,000 English words, each rated based on several emotional factors. Using these ratings, the researchers quantified two key aspects of emotional reactivity as reflected in participants’ language. They are valence, which measures the emotional tone from positive to negative, and arousal, which measures the intensity of the emotion from calm to excitement or panic.

The results showed that stories containing higher proportions of negatively valenced words, such as “blood,” “danger,” “hate,” and “shooting,” reliably predicted more acute symptoms of depression and anxiety. Follow-up tests confirmed that people who met clinical thresholds for moderate or higher levels of depression and anxiety used significantly more negative language than those below the threshold.

Although all emergency dispatchers and dispatchers experience objectively negative scenarios through calls, the words they choose to describe a call can reveal their subjective, emotional interpretation of the call, explained Ta Johnson.

“What was noticeable was the use of more negative language, which seems to reflect not only what they have experienced, but also how they are processing that experience psychologically,” she added.

Surprisingly, language intensity or arousal level did not predict symptoms of anxiety, depression, or stress.

All of the experiences they described were quite emotionally intense, including suicides and medical emergencies. Language that reflects intensity is to some extent an inevitable part of the profession. ”

Dr. Vivian Ta Johnson, Professor of Psychology, College of Health, Community, and Policy, UT San Antonio

This finding also suggests that the selection of positive and negative words operates on independent tracks of emotional expression rather than direct opposition. Highly negative language indicates a mental health risk, but a lack of positive language doesn’t necessarily mean call takers or dispatchers are struggling. In chronically stressful occupations, positive expression is often inhibited or constrained by the nature of the job, rather than by poor mental health.

Providing active health support

The researchers emphasized that these language-based AI tools are not designed to diagnose mental health conditions, monitor employees in a punitive manner, or replace human therapists. Instead, they hope the findings will lead to simple wellness tools that can be integrated into normal workplace protocols and used to support employees who opt in.

For example, call takers and dispatchers can complete periodic, short written reflections that language software can screen to flag individuals who may benefit from additional mental health resources. This discreet strategy can be critical for first responders, who are often hesitant to seek mental health support due to structural workload demands or workplace stigma.

Looking forward, Ta-Johnson and Morissette plan to investigate whether changing the way telecommunicators talk about their trauma (such as writing in a more positive or past tense frame) affects their responses to trauma over time, including during a narrative writing intervention. They are also in the early stages of establishing partnerships with other police departments and communications departments to replicate and expand this research.

Co-authors of the paper include San Antonio College psychology doctoral students Isabella M. Swafford and Paige Lindsey, psychology postdoctoral researcher Lillian D. Effinger, Ph.D., and supervising assistant professor Janelle Kohler, Ph.D., as well as Michael Fernandez of the San Antonio Police Department, and researchers from the Warriors Institute, Baylor Scott & White Health, Texas A&M University, and Baylor University. medicine.