Can ‘Linguistics’ Determine Patient’s Mental Health?

The words we choose. Our tone of voice. Can those aspects of daily speech offer any insights into the mental health of patients or each clinician’s ability to convey empathy? Or could the linguistic tones used by patients be associated with depression and health-related anxiety?

A new study tackled these intriguing questions and formally tested the hypothesis that there is an association between both patient and clinician linguistic tones and the patient’s perception of the provider’s empathy. The study, titled “Are Patient Linguistic Tones Associated with Mental Health and Perceived Clinician Empathy?” appears in the December 1, 2021, edition of The Journal of Bone and Joint Surgery.

Giving background on the study was co-author David Ring, M.D., Ph.D., associate dean for Comprehensive Care in the Department of Surgery and Perioperative Care at the University of Texas at Austin Dell Medical School. He told OTW, “We have a set of transcripts from three prior studies that we are using to address a new hypothesis. We have also made increasing use of natural language processing tools. The transcripts offered a great opportunity to look at tones, word count, and other aspects of clinician and patient language to see if it was associated with patient mental health. An association would open the possibility of using computer analysis of patient and clinician language to identify mental health opportunities.”

The authors wrote, “Trained and attentive musculoskeletal specialists can use aspects of patient communication to distinguish between:

1. symptoms that correspond well with observed pathophysiology as taught in the traditional biomedical paradigm and
2. disproportionate or incongruent symptoms that suggest mental and social health opportunities that are increasingly recognized in comprehensive biopsychosocial approaches.”

The team examined video and audio recordings of 109 adult patients who sought out musculoskeletal specialty care between September 2019 and January 2020. This study includes secondary data from studies where permission from new and returning patients was obtained to record the visit; the patients completed a series of questionnaires afterward. After dividing the transcripts into clinician and patient dialogue, the researchers used Linguistic Inquiry and Word Count to analyze the text and detect “the relative strength of various emotional tones, cognitive processes, and core drives and needs.”

“The percentage of words in each of the dictionary categories is calculated and compared with the total number of words in the text. For example, if we counted 15 words (using a linguistic inquiry and word count identifier program) in the category of “family” (e.g., aunt, uncle, nephew, niece) in a text with 100 words in total, the relative strength of this category would be 15%.”

“For patient dialogue, we were interested in the following Linguistic Inquiry and Word Count domains: the number of words, questions, and adjectives as well as the relative scores indicating the prevalence of words representing the following tone categories: positive emotion, negative emotion, anxiety, sadness, anger, social, and family.”

“For clinician dialogue, we were interested in the number of words spoken and score indicating the prevalence of words representing the following tone categories: analytical, clout, authentic, positive emotion, negative emotion, social, family, insight, causation, tentativeness, and risk.”

Dr. Ring, also a courtesy professor in the departments of Health Social Work and Psychiatry and Behavioral Sciences, told OTW, “Patient words and tones were not associated with mental health. If anything, speaking less was a sign of symptoms of depression. What this means is that clinicians and computers cannot rely on patient language to discern mental health opportunities.”

What, we wondered, were the clinical implications of this work. Dr. Ring explained, “Over the years, I’ve often heard surgeons say with confidence that a given patient they are caring for had no mental health opportunities. Based on this study, I would say that in many cases, those surgeons were being over-confident in their assertions. We all need to be attuned for mental health opportunities because the data are clear and consistent: mental and social health opportunities are common in musculoskeletal specialty care.”