Activity in empathy-related brain regions predicts clinicians’ pain treatment decisions

826 

Abstract Submission 

Nikta Khalilkhani¹, Theoni Varoudaki², Morgan Gianola³, Elizabeth Losin⁴

¹Pennsylvania state University, state college, PA, ²Penn State Univeristy, State College, PA, ³University of Miami, Coral Gables, FL, ⁴Penn State University, State College, PA

Nikta Khalilkhani  
Pennsylvania state University
state college, PA

Theoni Varoudaki  
Penn State Univeristy
State College, PA

Morgan Gianola  
University of Miami
Coral Gables, FL

ELIZABETH LOSIN, Ph.D.  
Penn State University
State College, PA

Pain is a global public health problem, disproportionately affecting individuals from marginalized groups, women, and older adults. Both undertreatment and overtreatment of pain contribute to adverse downstream effects, such as ineffective pain management, unnecessary exposure to opioids, and health disparities. Understanding underlying mechanisms of pain assessment and treatment decisions is essential to developing intervention aimed at improving pain management. Despite evidence that pain management decisions are influenced by many factors beyond patient reported pain, the neurobehavioural process underlying these decisions are poorly understood. Here we focused on the role of the clinician's cognition and brain activity during the clinical assessment and pain management decisions. We hypothesize that the greater a clinician's pain-related empathic brain activity is when seeing their patient in pain the more accurate they will be at assessing that patient's pain and need for treatment.

To test this hypothesis we recruited N=67 (34 f) medical students and had them complete a virtual pain management task while undergoing fMRI. Each clinician interacted with a diverse group of 36 mock patients. Each simulated clinical interaction consisted of 4 sections: 1) a medical vignette with mock patient injury information, 2) pain behavior videos meant to simulate the clinical exam: 3 x 4-second clips of previous research participants responding to evoked pain, and 3-4) pain and treatment rating, where clinicians rated how much pain they thought the patient was in and the likelihood they would prescribe any analgesic.
To investigate clinician's brain activity related to pain empathy we chose an a priori brain mask consisting of regions related to empathy from a Neurosynth automated meta-analysis. To test which (if any) aspects of empathy these brain regions were related to in our sample of clinicians, we ran an exploratory analysis using Stochastic Search Variable Selection. We predicted average brain activity within the Neurosynth empathy mask using all 28 of the questions in the Interpersonal Reactivity Index (IRI), a well-validated empathy questionnaire. We then included IRI questions that had a maximum inclusion probability above 0.5 in a multiple regression to test their relationship with average activity within the empathy mask.

We found that clinicians who reported having higher empathic concern for others' misfortune, had significantly higher activity within the empathy brain mask while observing patients in pain, suggesting brain activity within this Neurosynth mask was indeed related to trait empathy in our sample of clinicians. We found that average activity within the Neurosynth empathy mask predicted the accuracy of both clinicians' assessment of patients' pain (defined as the difference between the clinician and patients' pain intensity rating) and their assessment of the patient's need for treatment (defined as the difference between the clinicians' likelihood of prescribing an analgesic and the patients' ratings of their pain unpleasantness).

Our findings suggest that clinicians with higher neural responses associated with pain empathy when viewing their patients in pain may be able to more accurately perceive the pain of their patient, and, more importantly, may be able to more effectively assess the patient's need for treatment. More broadly, our findings suggest that medical training that teaches clinicians to empathize more with their patient's pain, rather than detach from it as is often found to occur across the course of medical training and practice, could improve the efficacy of pain management.

Social Cognition

Social Interaction ¹

Activation (eg. BOLD task-fMRI) ²

Univariate Modeling

Perception: Pain and Visceral

Cognition

FUNCTIONAL MRI

Pain

Social Interactions