Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Bioinformatics, Computational and Systems Biology

Bioinformatics, Computational and Systems Biology - Poster Session A

Poster A9 - Capturing Lived Experiences of Chronic Pain through a Listening Experience Model

Thursday, October 24, 2024
10:00 AM - 11:00 AM EST
Location: Exhibit Hall E, F & G

Introduction:  Approximately 51 million Americans (more than 1 out of 5 Americans) suffer from chronic pain [1]. Chronic pain patients often have symptoms that lack a clear cause of identifiable physical pathology, and traditional pain assessment methods, which focus on clinical history and physical examinations, tend to overlook the unique lived experiences of these patients. These experiences encompass the daily challenges, routines, personal needs, and aspirations disrupted by chronic pain. Current documentation practices, including clinician notes and patient-completed medical intake forms, offer a limited perspective and fail to capture the patient’s voice fully. Similarly, patient-reported outcomes may not genuinely reflect the patient’s experience, potentially omitting vital information relevant to their clinical status and overall well-being. Although clinicians desire to capture this patient information, it has remained impractical to collect, analyze, and summarize all the salient information of a patient’s lived experiences. However, emerging machine learning technologies, such as Natural Language Processing (NLP), show promise for examining the language and expression used to describe their pain and allow for efficient analysis over large datasets. This method could provide insights often overlooked by traditional documentation, promoting a holistic and patient-centered approach. The objective of this ongoing study is to collect and analyze first-hand narratives of patients’ lived experiences with chronic pain using NLP. This research aspires to foster better communication between patients and healthcare providers, ultimately leading to more personalized and effective pain management strategies.

Materials and

Methods:  We interviewed twelve adults who self-reported as having chronic pain (age = 29.5±11.52 years). Each interview was 30-60 minutes and audio recorded. Two researchers facilitated the interview: one to engage and ask participants questions and another to record interviewer-interviewee interactions, body language, tonal inflections, and facial expressions. To explore the multifactorial influencers of pain expression, interview questions were devised using factors from the National Institute on Minority Health and Health Disparities Research Framework which included parameters in physical environment, behavioral patterns, cultural identity, family, and peer networks [2]. The interview format was based on a Motivational Interviewing (MI) scheme [3], such that we did not intend to alter the behaviors of our subjects, but rather to gain a comprehensive understanding of their lived experiences (i.e., using the engaging, focusing, empathizing, and reflective listening principles of MI but not evoking and planning). To fully capture participants’ voices, the interviewer asked open-ended questions, acknowledged participants’ feelings to honor autonomy, gave participants the freedom to proceed in their desired direction during dialogue, reflected and summarized their experience at the end of the interview (as detailed in the OARS technique [3]). We then uploaded the audio recordings to a secure server, transcribing and de-identifying the audio using multiple AI models: Whisper from OpenAI [4], Segmentation-3.0 [5], and a PII removal model [6]. The output of these scripts used a combination of a text file with the labeled speakers, timestamps of each segment, and personal information replaced with tags (I.E. John replaced with [NAME]).

Results, Conclusions, and Discussions: The results from this study are considered preliminary as the study remains on going. From the interviews conducted so far, subjects have expressed the following:

Community Challenges:

"I've never really talked to people [in my community] about this because I'm afraid to hear people try to diagnose me"

"I wonder if I would have done more schooling because sitting is so painful for me"

Emotional Impact:

"They'll say you're totally normal. It… am I making this up?"

“Just wishing for relief and knowing the only way you're going to get it is if you're not wrapped because your disease is not curable..."

Clinician-Patient Interaction:

"So, it’s kind of impossible to fully treat me… in 15 mins and … you’re only looking at one part, because it is an interaction with everything”

Interview Impact:

"I like how you ask because nobody really asks about remedies your cultural remedies"



The narratives collected so far provide a rich and nuanced understanding of the lived experiences of individuals with chronic pain, highlighting aspects often missed by traditional documenting methods. The themes captured underscore the multifaceted nature of chronic pain, encompassing physical, emotional, and social dimensions. NLP techniques, using Whisper, Segmentation-3.0, and a PII removal model, have effectively transcribed and analyzed the interviews while preserving participant privacy. The results suggest that chronic pain significantly impacts daily activities, emotional wellbeing, and relationships. Participants’ detailed descriptions of everyday struggles and emotional distress emphasize the value of first-hand narratives to thoroughly understand chronic pain. Actively listening to patients’ voices and using analytical methods could enhance our understanding and management of chronic pain, thus advancing towards more empathetic and effective care. Preliminary findings indicate that traditional pain assessments may not fully capture patient experiences. Moreover, NLP techniques provide insights for a patient-centered approach, capturing the depth and complexity of experiences to optimize personalized therapeutic strategies and improve clinical outcomes. Future steps will include examining patterns in speech, sentiment, and linguistic cues during interviews to help clinicians in better understanding patients’ cognitive processes and attention dynamics. Furthermore, a larger sample is needed to validate these findings and explore additional themes.

Acknowledgements (Optional):  This project was supported by the Department of Bioengineering at George Mason University
The Office of Student Creative Activities and Research at George Mason University
NIH HEAL #1R61AT012286