Liqin Wang1,2, Suzanne V Blackley3, Kimberly G Blumenthal2,4, Sharmitha Yerneni1, Foster R Goss5, Ying-Chih Lo1,2,6,7, Sonam N Shah1,2,8, Carlos A Ortega1, Zfania Tom Korach1,2, Diane L Seger1,3, Li Zhou1,2. 1. Division of General Internal Medicine and Primary Care, Brigham and Women's Hospital, Boston, Massachusetts, USA. 2. Harvard Medical School, Harvard University, Boston, Massachusetts, USA. 3. Partners HealthCare, Boston, Massachusetts, USA. 4. Division of Rheumatology, Allergy, and Immunology, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts, USA. 5. Department of Emergency Medicine, School of Medicine, University of Colorado, Aurora, Colorado, USA. 6. Division of Nephrology, Department of Medicine, Taichung Veterans General Hospital, Taichung, Taiwan. 7. Department of Data Science and Big Data Analytics, Providence University, Taichung, Taiwan. 8. Massachusetts College of Pharmacy and Health Sciences, Boston, Massachusetts, USA.
Abstract
OBJECTIVE: Incomplete and static reaction picklists in the allergy module led to free-text and missing entries that inhibit the clinical decision support intended to prevent adverse drug reactions. We developed a novel, data-driven, "dynamic" reaction picklist to improve allergy documentation in the electronic health record (EHR). MATERIALS AND METHODS: We split 3 decades of allergy entries in the EHR of a large Massachusetts healthcare system into development and validation datasets. We consolidated duplicate allergens and those with the same ingredients or allergen groups. We created a reaction value set via expert review of a previously developed value set and then applied natural language processing to reconcile reactions from structured and free-text entries. Three association rule-mining measures were used to develop a comprehensive reaction picklist dynamically ranked by allergen. The dynamic picklist was assessed using recall at top k suggested reactions, comparing performance to the static picklist. RESULTS: The modified reaction value set contained 490 reaction concepts. Among 4 234 327 allergy entries collected, 7463 unique consolidated allergens and 469 unique reactions were identified. Of the 3 dynamic reaction picklists developed, the 1 with the optimal ranking achieved recalls of 0.632, 0.763, and 0.822 at the top 5, 10, and 15, respectively, significantly outperforming the static reaction picklist ranked by reaction frequency. CONCLUSION: The dynamic reaction picklist developed using EHR data and a statistical measure was superior to the static picklist and suggested proper reactions for allergy documentation. Further studies might evaluate the usability and impact on allergy documentation in the EHR.
OBJECTIVE: Incomplete and static reaction picklists in the allergy module led to free-text and missing entries that inhibit the clinical decision support intended to prevent adverse drug reactions. We developed a novel, data-driven, "dynamic" reaction picklist to improve allergy documentation in the electronic health record (EHR). MATERIALS AND METHODS: We split 3 decades of allergy entries in the EHR of a large Massachusetts healthcare system into development and validation datasets. We consolidated duplicate allergens and those with the same ingredients or allergen groups. We created a reaction value set via expert review of a previously developed value set and then applied natural language processing to reconcile reactions from structured and free-text entries. Three association rule-mining measures were used to develop a comprehensive reaction picklist dynamically ranked by allergen. The dynamic picklist was assessed using recall at top k suggested reactions, comparing performance to the static picklist. RESULTS: The modified reaction value set contained 490 reaction concepts. Among 4 234 327 allergy entries collected, 7463 unique consolidated allergens and 469 unique reactions were identified. Of the 3 dynamic reaction picklists developed, the 1 with the optimal ranking achieved recalls of 0.632, 0.763, and 0.822 at the top 5, 10, and 15, respectively, significantly outperforming the static reaction picklist ranked by reaction frequency. CONCLUSION: The dynamic reaction picklist developed using EHR data and a statistical measure was superior to the static picklist and suggested proper reactions for allergy documentation. Further studies might evaluate the usability and impact on allergy documentation in the EHR.
Authors: Thomas Isaac; Joel S Weissman; Roger B Davis; Michael Massagli; Adrienne Cyrulik; Daniel Z Sands; Saul N Weingart Journal: Arch Intern Med Date: 2009-02-09
Authors: Tejal K Gandhi; Saul N Weingart; Joshua Borus; Andrew C Seger; Josh Peterson; Elisabeth Burdick; Diane L Seger; Kirstin Shu; Frank Federico; Lucian L Leape; David W Bates Journal: N Engl J Med Date: 2003-04-17 Impact factor: 91.245
Authors: Maxim Topaz; Diane L Seger; Kenneth Lai; Paige G Wickner; Foster Goss; Neil Dhopeshwarkar; Frank Chang; David W Bates; Li Zhou Journal: Stud Health Technol Inform Date: 2015
Authors: Sharmitha Yerneni; Sonam N Shah; Suzanne V Blackley; Carlos A Ortega; Kimberly G Blumenthal; Foster Goss; Diane L Seger; Paige G Wickner; Christian M Mancini; David W Bates; Li Zhou Journal: Appl Clin Inform Date: 2022-05-26 Impact factor: 2.762
Authors: Neelam A Phadke; Paige Wickner; Liqin Wang; Li Zhou; Elizabeth Mort; David W Bates; Claire Seguin; Xiaoqing Fu; Kimberly G Blumenthal Journal: J Allergy Clin Immunol Pract Date: 2022-04-08
Authors: Patrick Wu; Scott D Nelson; Juan Zhao; Cosby A Stone; QiPing Feng; Qingxia Chen; Eric A Larson; Bingshan Li; Nancy J Cox; C Michael Stein; Elizabeth J Phillips; Dan M Roden; Joshua C Denny; Wei-Qi Wei Journal: J Am Med Inform Assoc Date: 2021-07-14 Impact factor: 7.942