Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model

PURPOSEThis study aimed to differentiate nonscheduled visits (NSVs) in an outpatient palliative care setting that are driven by or accompanied by uncontrolled symptoms from those that are administrative or routine, such as prescription refills and examination readings. A small language model (SLM) w...

Full description

Saved in:
Bibliographic Details
Main Authors: Javier Retamales, Juan Pablo Retamales, Ana Maria Demarchi, Marcela Gonzalez, Caroll Lopez, Nina Ramirez, Tamara Retamal, Virginia Sun
Format: Article
Language:English
Published: American Society of Clinical Oncology 2025-04-01
Series:JCO Global Oncology
Online Access:https://ascopubs.org/doi/10.1200/GO-24-00432
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849761058502213632
author Javier Retamales
Juan Pablo Retamales
Ana Maria Demarchi
Marcela Gonzalez
Caroll Lopez
Nina Ramirez
Tamara Retamal
Virginia Sun
author_facet Javier Retamales
Juan Pablo Retamales
Ana Maria Demarchi
Marcela Gonzalez
Caroll Lopez
Nina Ramirez
Tamara Retamal
Virginia Sun
author_sort Javier Retamales
collection DOAJ
description PURPOSEThis study aimed to differentiate nonscheduled visits (NSVs) in an outpatient palliative care setting that are driven by or accompanied by uncontrolled symptoms from those that are administrative or routine, such as prescription refills and examination readings. A small language model (SLM) was used to enhance the detection and management of symptoms, thus improving health care resource allocation.METHODSA retrospective analysis was performed on 25,867 patient visits to an outpatient palliative care unit, including 7,036 NSVs. A stratified random sample of 384 NSVs was reviewed to determine the presence of symptoms, using physician audits as the gold standard. A Phi-3–based SLM was validated against these audits to assess its accuracy in detecting the symptoms. The validated SLM was then applied to the entire NSV data set to identify symptom patterns. Multivariate linear regression was used to analyze the association of age, cancer type, and insurance category with the presence of symptoms.RESULTSSLM demonstrated high sensitivity (99.4%) and accuracy (95.3%) in identifying symptom-driven NSVs. The analysis revealed that 85.7% of the NSVs were driven by symptoms, indicating a significant hidden burden of unmanaged symptoms. The study found that certain demographic and clinical factors, including younger age groups and specific cancer types, were significantly associated with an increased symptom burden.CONCLUSIONThis study highlights the substantial burden of symptom-driven NSVs in palliative care and demonstrates the effectiveness of using a SLM to identify and manage symptoms. Implementing such models in clinical practice can improve patient care by optimizing the allocation of health care resources and tailoring interventions to the needs of patients with advanced illnesses.
format Article
id doaj-art-b35d1d46ae9e44238c4b95b30d01568d
institution DOAJ
issn 2687-8941
language English
publishDate 2025-04-01
publisher American Society of Clinical Oncology
record_format Article
series JCO Global Oncology
spelling doaj-art-b35d1d46ae9e44238c4b95b30d01568d2025-08-20T03:06:09ZengAmerican Society of Clinical OncologyJCO Global Oncology2687-89412025-04-011110.1200/GO-24-00432Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language ModelJavier Retamales0Juan Pablo Retamales1Ana Maria Demarchi2Marcela Gonzalez3Caroll Lopez4Nina Ramirez5Tamara Retamal6Virginia Sun7Hospital Sotero del Río, Santiago, ChileQJV, Santiago, ChileHospital Sotero del Río, Santiago, ChileHospital Sotero del Río, Santiago, ChileHospital Sotero del Río, Santiago, ChileHospital Sotero del Río, Santiago, ChileHospital Sotero del Río, Santiago, ChileCity of Hope Comprehensive Cancer Center, Duarte, CAPURPOSEThis study aimed to differentiate nonscheduled visits (NSVs) in an outpatient palliative care setting that are driven by or accompanied by uncontrolled symptoms from those that are administrative or routine, such as prescription refills and examination readings. A small language model (SLM) was used to enhance the detection and management of symptoms, thus improving health care resource allocation.METHODSA retrospective analysis was performed on 25,867 patient visits to an outpatient palliative care unit, including 7,036 NSVs. A stratified random sample of 384 NSVs was reviewed to determine the presence of symptoms, using physician audits as the gold standard. A Phi-3–based SLM was validated against these audits to assess its accuracy in detecting the symptoms. The validated SLM was then applied to the entire NSV data set to identify symptom patterns. Multivariate linear regression was used to analyze the association of age, cancer type, and insurance category with the presence of symptoms.RESULTSSLM demonstrated high sensitivity (99.4%) and accuracy (95.3%) in identifying symptom-driven NSVs. The analysis revealed that 85.7% of the NSVs were driven by symptoms, indicating a significant hidden burden of unmanaged symptoms. The study found that certain demographic and clinical factors, including younger age groups and specific cancer types, were significantly associated with an increased symptom burden.CONCLUSIONThis study highlights the substantial burden of symptom-driven NSVs in palliative care and demonstrates the effectiveness of using a SLM to identify and manage symptoms. Implementing such models in clinical practice can improve patient care by optimizing the allocation of health care resources and tailoring interventions to the needs of patients with advanced illnesses.https://ascopubs.org/doi/10.1200/GO-24-00432
spellingShingle Javier Retamales
Juan Pablo Retamales
Ana Maria Demarchi
Marcela Gonzalez
Caroll Lopez
Nina Ramirez
Tamara Retamal
Virginia Sun
Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
JCO Global Oncology
title Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
title_full Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
title_fullStr Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
title_full_unstemmed Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
title_short Leveraging Artificial Intelligence to Uncover Symptom Burden in Palliative Care: Analysis of Nonscheduled Visits Using a Phi-3 Small Language Model
title_sort leveraging artificial intelligence to uncover symptom burden in palliative care analysis of nonscheduled visits using a phi 3 small language model
url https://ascopubs.org/doi/10.1200/GO-24-00432
work_keys_str_mv AT javierretamales leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT juanpabloretamales leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT anamariademarchi leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT marcelagonzalez leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT carolllopez leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT ninaramirez leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT tamararetamal leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel
AT virginiasun leveragingartificialintelligencetouncoversymptomburdeninpalliativecareanalysisofnonscheduledvisitsusingaphi3smalllanguagemodel