TY - JOUR
T1 - Natural Language Processing Accurately Measures Adherence to Best Practice Guidelines for Palliative Care in Trauma
AU - Lee, Katherine C.
AU - Udelsman, Brooks V.
AU - Streid, Jocelyn
AU - Chang, David C.
AU - Salim, Ali
AU - Livingston, David H.
AU - Lindvall, Charlotta
AU - Cooper, Zara
N1 - Funding Information:
Portions of this study were presented at the American Geriatrics Society Annual Meeting, May 2019, in Portland, Oregon. Dr. Udelsman was supported by the Society of University Surgeons—Karl Storz Resident Research Award (2017–2018). Dr. Cooper is supported by the Paul B. Beeson Emerging Leaders Career Development Award in Aging ( 1K76AG054859-01 ), the American Federation for Aging Research , and the Cambia Health Foundation . Dr. Lindvall is supported by the Cambia Health Foundation Sojourns Scholar Leadership Program . The authors declare no conflicts of interest.
Funding Information:
Portions of this study were presented at the American Geriatrics Society Annual Meeting, May 2019, in Portland, Oregon. Dr. Udelsman was supported by the Society of University Surgeons?Karl Storz Resident Research Award (2017?2018). Dr. Cooper is supported by the Paul B. Beeson Emerging Leaders Career Development Award in Aging (1K76AG054859-01), the American Federation for Aging Research, and the Cambia Health Foundation. Dr. Lindvall is supported by the Cambia Health Foundation Sojourns Scholar Leadership Program. The authors declare no conflicts of interest.
Publisher Copyright:
© 2019
PY - 2020/2
Y1 - 2020/2
N2 - Context: The Trauma Quality Improvement Program Best Practice Guidelines recommend palliative care (PC) concurrent with restorative treatment for patients with life-threatening injuries. Measuring PC delivery is challenging: administrative data are nonspecific, and manual review is time intensive. Objectives: To identify PC delivery to patients with life-threatening trauma and compare the performance of natural language processing (NLP), a form of computer-assisted data abstraction, to administrative coding and gold standard manual review. Methods: Patients 18 years and older admitted with life-threatening trauma were identified from two Level I trauma centers (July 2016—June 2017). Four PC process measures were examined during the trauma admission: code status clarification, goals-of-care discussion, PC consult, and hospice assessment. The performance of NLP and administrative coding were compared with manual review. Multivariable regression was used to determine patient and admission factors associated with PC delivery. Results: There were 76,791 notes associated with 2093 admissions. NLP identified PC delivery in 33% of admissions compared with 8% using administrative coding. Using NLP, code status clarification was most commonly documented (27%), followed by goals-of-care discussion (18%), PC consult (4%), and hospice assessment (4%). Compared with manual review, NLP performed more than 50 times faster and had a sensitivity of 93%, a specificity of 96%, and an accuracy of 95%. Administrative coding had a sensitivity of 21%, a specificity of 92%, and an accuracy of 68%. Factors associated with PC delivery included older age, increased comorbidities, and longer intensive care unit stay. Conclusion: NLP performs with similar accuracy with manual review but with improved efficiency. NLP has the potential to accurately identify PC delivery and benchmark performance of best practice guidelines.
AB - Context: The Trauma Quality Improvement Program Best Practice Guidelines recommend palliative care (PC) concurrent with restorative treatment for patients with life-threatening injuries. Measuring PC delivery is challenging: administrative data are nonspecific, and manual review is time intensive. Objectives: To identify PC delivery to patients with life-threatening trauma and compare the performance of natural language processing (NLP), a form of computer-assisted data abstraction, to administrative coding and gold standard manual review. Methods: Patients 18 years and older admitted with life-threatening trauma were identified from two Level I trauma centers (July 2016—June 2017). Four PC process measures were examined during the trauma admission: code status clarification, goals-of-care discussion, PC consult, and hospice assessment. The performance of NLP and administrative coding were compared with manual review. Multivariable regression was used to determine patient and admission factors associated with PC delivery. Results: There were 76,791 notes associated with 2093 admissions. NLP identified PC delivery in 33% of admissions compared with 8% using administrative coding. Using NLP, code status clarification was most commonly documented (27%), followed by goals-of-care discussion (18%), PC consult (4%), and hospice assessment (4%). Compared with manual review, NLP performed more than 50 times faster and had a sensitivity of 93%, a specificity of 96%, and an accuracy of 95%. Administrative coding had a sensitivity of 21%, a specificity of 92%, and an accuracy of 68%. Factors associated with PC delivery included older age, increased comorbidities, and longer intensive care unit stay. Conclusion: NLP performs with similar accuracy with manual review but with improved efficiency. NLP has the potential to accurately identify PC delivery and benchmark performance of best practice guidelines.
KW - Palliative care
KW - natural language processing
KW - quality improvement
KW - surgical palliative care
KW - trauma
UR - http://www.scopus.com/inward/record.url?scp=85075428131&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075428131&partnerID=8YFLogxK
U2 - 10.1016/j.jpainsymman.2019.09.017
DO - 10.1016/j.jpainsymman.2019.09.017
M3 - Article
C2 - 31562891
AN - SCOPUS:85075428131
SN - 0885-3924
VL - 59
SP - 225-232.e2
JO - Journal of Pain and Symptom Management
JF - Journal of Pain and Symptom Management
IS - 2
ER -