Brooks Udelsman1, Isabel Chien2,3, Kei Ouchi4, Kate Brizzi5,6, James A Tulsky2,7, Charlotta Lindvall2,7. 1. 1 Department of Surgery, Massachusetts General Hospital, Boston, Massachusetts. 2. 2 Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, Massachusetts. 3. 3 Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, Boston, Massachusetts. 4. 4 Department of Emergency Medicine, Brigham and Women's Hospital, Boston, Massachusetts. 5. 5 Division of Neurology, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts. 6. 6 Division of Palliative Care, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts. 7. 7 Division of Palliative Medicine, Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts.
Abstract
BACKGROUND: Alone, administrative data poorly identifies patients with palliative care needs. OBJECTIVE: To identify patients with uncommon, yet devastating, illnesses using a combination of administrative data and natural language processing (NLP). DESIGN/ SETTING: Retrospective cohort study using the electronic medical records of a healthcare network totaling over 2500 hospital beds. We sought to identify patient populations with two unique disease processes associated with a poor prognosis: pneumoperitoneum and leptomeningeal metastases from breast cancer. MEASUREMENTS: Patients with pneumoperitoneum or leptomeningeal metastasis from breast cancer were identified through administrative codes and NLP. RESULTS: Administrative codes alone resulted in identification of 6438 patients with possible pneumoperitoneum and 557 patients with possible leptomeningeal metastasis. Adding NLP to this analysis reduced the number of patients to 869 with pneumoperitoneum and 187 with leptomeningeal metastasis secondary to breast cancer. Administrative codes alone yielded a 13% positive predictive value (PPV) for pneumoperitoneum and 25% PPV for leptomeningeal metastasis. The combination of administrative codes and NLP achieved a PPV of 100%. The entire process was completed within hours. CONCLUSIONS: Adding NLP to the use of administrative codes allows for rapid identification of seriously ill patients with otherwise difficult to detect disease processes and eliminates costly, tedious, and time-intensive manual chart review. This method enables studies to evaluate the effectiveness of treatment, including palliative interventions, for unique populations of seriously ill patients who cannot be identified by administrative codes alone.
BACKGROUND: Alone, administrative data poorly identifies patients with palliative care needs. OBJECTIVE: To identify patients with uncommon, yet devastating, illnesses using a combination of administrative data and natural language processing (NLP). DESIGN/ SETTING: Retrospective cohort study using the electronic medical records of a healthcare network totaling over 2500 hospital beds. We sought to identify patient populations with two unique disease processes associated with a poor prognosis: pneumoperitoneum and leptomeningeal metastases from breast cancer. MEASUREMENTS: Patients with pneumoperitoneum or leptomeningeal metastasis from breast cancer were identified through administrative codes and NLP. RESULTS: Administrative codes alone resulted in identification of 6438 patients with possible pneumoperitoneum and 557 patients with possible leptomeningeal metastasis. Adding NLP to this analysis reduced the number of patients to 869 with pneumoperitoneum and 187 with leptomeningeal metastasis secondary to breast cancer. Administrative codes alone yielded a 13% positive predictive value (PPV) for pneumoperitoneum and 25% PPV for leptomeningeal metastasis. The combination of administrative codes and NLP achieved a PPV of 100%. The entire process was completed within hours. CONCLUSIONS: Adding NLP to the use of administrative codes allows for rapid identification of seriously ill patients with otherwise difficult to detect disease processes and eliminates costly, tedious, and time-intensive manual chart review. This method enables studies to evaluate the effectiveness of treatment, including palliative interventions, for unique populations of seriously ill patients who cannot be identified by administrative codes alone.
Entities:
Keywords:
critical illness; natural language processing; patient identification
Authors: Brian Badgwell; Robert Krouse; Janice Cormier; Caesar Guevara; V Suzanne Klimberg; Betty Ferrell Journal: Ann Surg Oncol Date: 2012-06-06 Impact factor: 5.344
Authors: Joan M Teno; Pedro L Gozalo; Julie P W Bynum; Natalie E Leland; Susan C Miller; Nancy E Morden; Thomas Scupp; David C Goodman; Vincent Mor Journal: JAMA Date: 2013-02-06 Impact factor: 56.272
Authors: Brian Badgwell; Barry W Feig; Merrick I Ross; Paul F Mansfield; Sijin Wen; George J Chang Journal: Ann Surg Oncol Date: 2007-08-07 Impact factor: 5.344
Authors: Katherine C Lee; Anne M Walling; Steven S Senglaub; Amy S Kelley; Zara Cooper Journal: J Pain Symptom Manage Date: 2019-08-09 Impact factor: 3.612
Authors: Tamryn F Gray; Anne Kwok; Khuyen M Do; Sandra Zeng; Edward T Moseley; Yasser M Dbeis; Renato Umeton; James A Tulsky; Areej El-Jawahri; Charlotta Lindvall Journal: JMIR Med Inform Date: 2022-06-15
Authors: Nicola Fawcett; Bernadette Young; Leon Peto; T Phuong Quan; Richard Gillott; Jianhua Wu; Chris Middlemass; Sheila Weston; Derrick W Crook; Tim E A Peto; Berit Muller-Pebody; Alan P Johnson; A Sarah Walker; Jonathan A T Sandoe Journal: BMC Med Date: 2019-09-04 Impact factor: 8.775
Authors: Ali Pourmand; Keith S Boniface; Katherine Douglass; Colton Hood; Sarah E Frasure; Jeremy Barnett; Kunj Bhatt; Neal Sikka Journal: Cureus Date: 2021-11-13