Extraction and standardization of patient complaints from electronic medication histories for pharmacovigilance

Natural language processing analysis in Japanese

Misa Usui, Eiji Aramaki, Tomohide Iwao, Shoko Wakamiya, Tohru Sakamoto, Mayumi Mochizuki

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background: Despite the growing number of studies using natural language processing for pharmacovigilance, there are few reports on manipulating free text patient information in Japanese. Objective: This study aimed to establish a method of extracting and standardizing patient complaints from electronic medication histories accumulated in a Japanese community pharmacy for the detection of possible adverse drug event (ADE) signals. Methods: Subjective information included in electronic medication history data provided by a Japanese pharmacy operating in Hiroshima, Japan from September 1, 2015 to August 31, 2016, was used as patients' complaints. We formulated search rules based on morphological analysis and daily (nonmedical) speech and developed a system that automatically executes the search rules and annotates free text data with International Classification of Diseases, Tenth Revision (ICD-10) codes. The performance of the system was evaluated through comparisons with data manually annotated by health care workers for a data set of 5000 complaints. Results: Of 5000 complaints, the system annotated 2236 complaints with ICD-10 codes, whereas health care workers annotated 2348 statements. There was a match in the annotation of 1480 complaints between the system and manual work. System performance was .66 regarding precision, .63 in recall, and .65 for the F-measure. Conclusions: Our results suggest that the system may be helpful in extracting and standardizing patients' speech related to symptoms from massive amounts of free text data, replacing manual work. After improving the extraction accuracy, we expect to utilize this system to detect signals of possible ADEs from patients' complaints in the future.

Original languageEnglish
Article numbere11021
JournalJournal of Medical Internet Research
Volume20
Issue number9
DOIs
Publication statusPublished - 2018 Sep 1

Fingerprint

Natural Language Processing
Pharmacovigilance
International Classification of Diseases
Delivery of Health Care
Pharmacies
Drug-Related Side Effects and Adverse Reactions
Japan

Keywords

  • Adverse drug events
  • Medical informatics
  • Medication history
  • Natural language processing
  • Pharmacovigilance

ASJC Scopus subject areas

  • Health Informatics

Cite this

Extraction and standardization of patient complaints from electronic medication histories for pharmacovigilance : Natural language processing analysis in Japanese. / Usui, Misa; Aramaki, Eiji; Iwao, Tomohide; Wakamiya, Shoko; Sakamoto, Tohru; Mochizuki, Mayumi.

In: Journal of Medical Internet Research, Vol. 20, No. 9, e11021, 01.09.2018.

Research output: Contribution to journalArticle

Usui, Misa ; Aramaki, Eiji ; Iwao, Tomohide ; Wakamiya, Shoko ; Sakamoto, Tohru ; Mochizuki, Mayumi. / Extraction and standardization of patient complaints from electronic medication histories for pharmacovigilance : Natural language processing analysis in Japanese. In: Journal of Medical Internet Research. 2018 ; Vol. 20, No. 9.
@article{0bd6731e26754b0ca0cb26852f136a93,
title = "Extraction and standardization of patient complaints from electronic medication histories for pharmacovigilance: Natural language processing analysis in Japanese",
abstract = "Background: Despite the growing number of studies using natural language processing for pharmacovigilance, there are few reports on manipulating free text patient information in Japanese. Objective: This study aimed to establish a method of extracting and standardizing patient complaints from electronic medication histories accumulated in a Japanese community pharmacy for the detection of possible adverse drug event (ADE) signals. Methods: Subjective information included in electronic medication history data provided by a Japanese pharmacy operating in Hiroshima, Japan from September 1, 2015 to August 31, 2016, was used as patients' complaints. We formulated search rules based on morphological analysis and daily (nonmedical) speech and developed a system that automatically executes the search rules and annotates free text data with International Classification of Diseases, Tenth Revision (ICD-10) codes. The performance of the system was evaluated through comparisons with data manually annotated by health care workers for a data set of 5000 complaints. Results: Of 5000 complaints, the system annotated 2236 complaints with ICD-10 codes, whereas health care workers annotated 2348 statements. There was a match in the annotation of 1480 complaints between the system and manual work. System performance was .66 regarding precision, .63 in recall, and .65 for the F-measure. Conclusions: Our results suggest that the system may be helpful in extracting and standardizing patients' speech related to symptoms from massive amounts of free text data, replacing manual work. After improving the extraction accuracy, we expect to utilize this system to detect signals of possible ADEs from patients' complaints in the future.",
keywords = "Adverse drug events, Medical informatics, Medication history, Natural language processing, Pharmacovigilance",
author = "Misa Usui and Eiji Aramaki and Tomohide Iwao and Shoko Wakamiya and Tohru Sakamoto and Mayumi Mochizuki",
year = "2018",
month = "9",
day = "1",
doi = "10.2196/11021",
language = "English",
volume = "20",
journal = "Journal of Medical Internet Research",
issn = "1439-4456",
publisher = "Journal of medical Internet Research",
number = "9",

}

TY - JOUR

T1 - Extraction and standardization of patient complaints from electronic medication histories for pharmacovigilance

T2 - Natural language processing analysis in Japanese

AU - Usui, Misa

AU - Aramaki, Eiji

AU - Iwao, Tomohide

AU - Wakamiya, Shoko

AU - Sakamoto, Tohru

AU - Mochizuki, Mayumi

PY - 2018/9/1

Y1 - 2018/9/1

N2 - Background: Despite the growing number of studies using natural language processing for pharmacovigilance, there are few reports on manipulating free text patient information in Japanese. Objective: This study aimed to establish a method of extracting and standardizing patient complaints from electronic medication histories accumulated in a Japanese community pharmacy for the detection of possible adverse drug event (ADE) signals. Methods: Subjective information included in electronic medication history data provided by a Japanese pharmacy operating in Hiroshima, Japan from September 1, 2015 to August 31, 2016, was used as patients' complaints. We formulated search rules based on morphological analysis and daily (nonmedical) speech and developed a system that automatically executes the search rules and annotates free text data with International Classification of Diseases, Tenth Revision (ICD-10) codes. The performance of the system was evaluated through comparisons with data manually annotated by health care workers for a data set of 5000 complaints. Results: Of 5000 complaints, the system annotated 2236 complaints with ICD-10 codes, whereas health care workers annotated 2348 statements. There was a match in the annotation of 1480 complaints between the system and manual work. System performance was .66 regarding precision, .63 in recall, and .65 for the F-measure. Conclusions: Our results suggest that the system may be helpful in extracting and standardizing patients' speech related to symptoms from massive amounts of free text data, replacing manual work. After improving the extraction accuracy, we expect to utilize this system to detect signals of possible ADEs from patients' complaints in the future.

AB - Background: Despite the growing number of studies using natural language processing for pharmacovigilance, there are few reports on manipulating free text patient information in Japanese. Objective: This study aimed to establish a method of extracting and standardizing patient complaints from electronic medication histories accumulated in a Japanese community pharmacy for the detection of possible adverse drug event (ADE) signals. Methods: Subjective information included in electronic medication history data provided by a Japanese pharmacy operating in Hiroshima, Japan from September 1, 2015 to August 31, 2016, was used as patients' complaints. We formulated search rules based on morphological analysis and daily (nonmedical) speech and developed a system that automatically executes the search rules and annotates free text data with International Classification of Diseases, Tenth Revision (ICD-10) codes. The performance of the system was evaluated through comparisons with data manually annotated by health care workers for a data set of 5000 complaints. Results: Of 5000 complaints, the system annotated 2236 complaints with ICD-10 codes, whereas health care workers annotated 2348 statements. There was a match in the annotation of 1480 complaints between the system and manual work. System performance was .66 regarding precision, .63 in recall, and .65 for the F-measure. Conclusions: Our results suggest that the system may be helpful in extracting and standardizing patients' speech related to symptoms from massive amounts of free text data, replacing manual work. After improving the extraction accuracy, we expect to utilize this system to detect signals of possible ADEs from patients' complaints in the future.

KW - Adverse drug events

KW - Medical informatics

KW - Medication history

KW - Natural language processing

KW - Pharmacovigilance

UR - http://www.scopus.com/inward/record.url?scp=85054281228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054281228&partnerID=8YFLogxK

U2 - 10.2196/11021

DO - 10.2196/11021

M3 - Article

VL - 20

JO - Journal of Medical Internet Research

JF - Journal of Medical Internet Research

SN - 1439-4456

IS - 9

M1 - e11021

ER -