Dictionary construction and identification of possible adverse drug events in Danish clinical narrative text

Robert Eriksson, Peter Bjødstrup Jensen, Sune Frankild, Lars Juhl Jensen, Søren Brunak

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

OBJECTIVE: Drugs have tremendous potential to cure and relieve disease, but the risk of unintended effects is always present. Healthcare providers increasingly record data in electronic patient records (EPRs), in which we aim to identify possible adverse events (AEs) and, specifically, possible adverse drug events (ADEs).

MATERIALS AND METHODS: Based on the undesirable effects section from the summary of product characteristics (SPC) of 7446 drugs, we have built a Danish ADE dictionary. Starting from this dictionary we have developed a pipeline for identifying possible ADEs in unstructured clinical narrative text. We use a named entity recognition (NER) tagger to identify dictionary matches in the text and post-coordination rules to construct ADE compound terms. Finally, we apply post-processing rules and filters to handle, for example, negations and sentences about subjects other than the patient. Moreover, this method allows synonyms to be identified and anatomical location descriptions can be merged to allow appropriate grouping of effects in the same location.

RESULTS: The method identified 1 970 731 (35 477 unique) possible ADEs in a large corpus of 6011 psychiatric hospital patient records. Validation was performed through manual inspection of possible ADEs, resulting in precision of 89% and recall of 75%.

DISCUSSION: The presented dictionary-building method could be used to construct other ADE dictionaries. The complication of compound words in Germanic languages was addressed. Additionally, the synonym and anatomical location collapse improve the method.

CONCLUSIONS: The developed dictionary and method can be used to identify possible ADEs in Danish clinical narratives.

OriginalsprogEngelsk
TidsskriftJournal of the American Medical Informatics Association : JAMIA
Vol/bind20
Udgave nummer5
Sider (fra-til)947-53
Antal sider7
DOI
StatusUdgivet - 25. maj 2013
Udgivet eksterntJa

Fingeraftryk

Drug-Related Side Effects and Adverse Reactions
Hospital Records
Psychiatric Hospitals
Pharmaceutical Preparations
Health Personnel
Language

Citer dette

@article{1fc17ef8ee38484dab7932f31ccf19d7,
title = "Dictionary construction and identification of possible adverse drug events in Danish clinical narrative text",
abstract = "OBJECTIVE: Drugs have tremendous potential to cure and relieve disease, but the risk of unintended effects is always present. Healthcare providers increasingly record data in electronic patient records (EPRs), in which we aim to identify possible adverse events (AEs) and, specifically, possible adverse drug events (ADEs).MATERIALS AND METHODS: Based on the undesirable effects section from the summary of product characteristics (SPC) of 7446 drugs, we have built a Danish ADE dictionary. Starting from this dictionary we have developed a pipeline for identifying possible ADEs in unstructured clinical narrative text. We use a named entity recognition (NER) tagger to identify dictionary matches in the text and post-coordination rules to construct ADE compound terms. Finally, we apply post-processing rules and filters to handle, for example, negations and sentences about subjects other than the patient. Moreover, this method allows synonyms to be identified and anatomical location descriptions can be merged to allow appropriate grouping of effects in the same location.RESULTS: The method identified 1 970 731 (35 477 unique) possible ADEs in a large corpus of 6011 psychiatric hospital patient records. Validation was performed through manual inspection of possible ADEs, resulting in precision of 89{\%} and recall of 75{\%}.DISCUSSION: The presented dictionary-building method could be used to construct other ADE dictionaries. The complication of compound words in Germanic languages was addressed. Additionally, the synonym and anatomical location collapse improve the method.CONCLUSIONS: The developed dictionary and method can be used to identify possible ADEs in Danish clinical narratives.",
keywords = "Data Mining, Denmark, Dictionaries, Medical, Drug-Related Side Effects and Adverse Reactions, Electronic Health Records, Humans, Narration",
author = "Robert Eriksson and Jensen, {Peter Bj{\o}dstrup} and Sune Frankild and Jensen, {Lars Juhl} and S{\o}ren Brunak",
year = "2013",
month = "5",
day = "25",
doi = "10.1136/amiajnl-2013-001708",
language = "English",
volume = "20",
pages = "947--53",
journal = "Journal of the American Medical Informatics Association",
issn = "1067-5027",
publisher = "Heinemann",
number = "5",

}

Dictionary construction and identification of possible adverse drug events in Danish clinical narrative text. / Eriksson, Robert; Jensen, Peter Bjødstrup ; Frankild, Sune; Jensen, Lars Juhl; Brunak, Søren.

I: Journal of the American Medical Informatics Association : JAMIA, Bind 20, Nr. 5, 25.05.2013, s. 947-53.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Dictionary construction and identification of possible adverse drug events in Danish clinical narrative text

AU - Eriksson, Robert

AU - Jensen, Peter Bjødstrup

AU - Frankild, Sune

AU - Jensen, Lars Juhl

AU - Brunak, Søren

PY - 2013/5/25

Y1 - 2013/5/25

N2 - OBJECTIVE: Drugs have tremendous potential to cure and relieve disease, but the risk of unintended effects is always present. Healthcare providers increasingly record data in electronic patient records (EPRs), in which we aim to identify possible adverse events (AEs) and, specifically, possible adverse drug events (ADEs).MATERIALS AND METHODS: Based on the undesirable effects section from the summary of product characteristics (SPC) of 7446 drugs, we have built a Danish ADE dictionary. Starting from this dictionary we have developed a pipeline for identifying possible ADEs in unstructured clinical narrative text. We use a named entity recognition (NER) tagger to identify dictionary matches in the text and post-coordination rules to construct ADE compound terms. Finally, we apply post-processing rules and filters to handle, for example, negations and sentences about subjects other than the patient. Moreover, this method allows synonyms to be identified and anatomical location descriptions can be merged to allow appropriate grouping of effects in the same location.RESULTS: The method identified 1 970 731 (35 477 unique) possible ADEs in a large corpus of 6011 psychiatric hospital patient records. Validation was performed through manual inspection of possible ADEs, resulting in precision of 89% and recall of 75%.DISCUSSION: The presented dictionary-building method could be used to construct other ADE dictionaries. The complication of compound words in Germanic languages was addressed. Additionally, the synonym and anatomical location collapse improve the method.CONCLUSIONS: The developed dictionary and method can be used to identify possible ADEs in Danish clinical narratives.

AB - OBJECTIVE: Drugs have tremendous potential to cure and relieve disease, but the risk of unintended effects is always present. Healthcare providers increasingly record data in electronic patient records (EPRs), in which we aim to identify possible adverse events (AEs) and, specifically, possible adverse drug events (ADEs).MATERIALS AND METHODS: Based on the undesirable effects section from the summary of product characteristics (SPC) of 7446 drugs, we have built a Danish ADE dictionary. Starting from this dictionary we have developed a pipeline for identifying possible ADEs in unstructured clinical narrative text. We use a named entity recognition (NER) tagger to identify dictionary matches in the text and post-coordination rules to construct ADE compound terms. Finally, we apply post-processing rules and filters to handle, for example, negations and sentences about subjects other than the patient. Moreover, this method allows synonyms to be identified and anatomical location descriptions can be merged to allow appropriate grouping of effects in the same location.RESULTS: The method identified 1 970 731 (35 477 unique) possible ADEs in a large corpus of 6011 psychiatric hospital patient records. Validation was performed through manual inspection of possible ADEs, resulting in precision of 89% and recall of 75%.DISCUSSION: The presented dictionary-building method could be used to construct other ADE dictionaries. The complication of compound words in Germanic languages was addressed. Additionally, the synonym and anatomical location collapse improve the method.CONCLUSIONS: The developed dictionary and method can be used to identify possible ADEs in Danish clinical narratives.

KW - Data Mining

KW - Denmark

KW - Dictionaries, Medical

KW - Drug-Related Side Effects and Adverse Reactions

KW - Electronic Health Records

KW - Humans

KW - Narration

U2 - 10.1136/amiajnl-2013-001708

DO - 10.1136/amiajnl-2013-001708

M3 - Journal article

C2 - 23703825

VL - 20

SP - 947

EP - 953

JO - Journal of the American Medical Informatics Association

JF - Journal of the American Medical Informatics Association

SN - 1067-5027

IS - 5

ER -