Health timeline: An insight-based study of a timeline visualization of clinical data

Andres Ledesma; Niranjan Bidargaddi; Jörg Strobel; Geoffrey Schrader; Hannu Nieminen; Ilkka Korhonen; Miikka Ermes

doi:10.1186/s12911-019-0885-x

Health timeline: An insight-based study of a timeline visualization of clinical data

Andres Ledesma^* (Corresponding Author), Niranjan Bidargaddi, Jörg Strobel, Geoffrey Schrader, Hannu Nieminen, Ilkka Korhonen, Miikka Ermes

^*Corresponding author for this work

Research output: Contribution to journal › Article › Scientific › peer-review

18 Citations (Scopus)

Abstract

BACKGROUND: The increasing complexity and volume of clinical data poses a challenge in the decision-making process. Data visualizations can assist in this process by speeding up the time required to analyze and understand clinical data. Even though empirical experiments show that visualizations facilitate clinical data understanding, a consistent method to assess their effectiveness is still missing.

METHODS: The insight-based methodology determines the quality of insights a user acquires from the visualization. Insights receive a value from one to five points based on a domain-specific criteria. Five professional psychiatrists took part in the study using real de-identified clinical data spanning 4 years of medical history.

RESULTS: A total of 50 assessments were transcribed and analyzed. Comparing a total of 558 insights using Health Timeline and 576 without, the mean value using the Timeline (1.7) was higher than without (1.26; p<0.01), similarly the cumulative value with the Timeline (11.87) was higher than without (10.96: p<0.01). The average time required to formulate the first insight with the Timeline was higher (13.16 s) than without (7 s; p<0.01). Seven insights achieved the highest possible value using Health Timeline while none were obtained without it.

CONCLUSIONS: The Health Timeline effectively improved understanding of clinical data and helped participants recognize complex patterns from the data. By applying the insight-based methodology, the effectiveness of the Health Timeline was quantified, documented and demonstrated. As an outcome of this exercise, we propose the use of such methodologies to measure the effectiveness of visualizations that assist the clinical decision-making process.

Original language	English
Article number	170
Number of pages	14
Journal	BMC Medical Informatics and Decision Making
Volume	19
Issue number	1
DOIs	https://doi.org/10.1186/s12911-019-0885-x
Publication status	Published - 22 Aug 2019
MoE publication type	A1 Journal article-refereed

Funding

VTT Technical Research Centre of Finland has funded the design of the study, collection, analysis and interpretation of data. The writing of the manuscript was funded by the EU program for Research and Innovation Horizon 2020 under the Grant Agreement number 689260.

Keywords

Clinical data
Data visualization
Electronic health record
Health informatics
Insight-based methodology

Access to Document

10.1186/s12911-019-0885-x

Cite this

@article{2fc7c7a245eb4cf8b2db3855ea4e3f5a,

title = "Health timeline: An insight-based study of a timeline visualization of clinical data",

abstract = "BACKGROUND: The increasing complexity and volume of clinical data poses a challenge in the decision-making process. Data visualizations can assist in this process by speeding up the time required to analyze and understand clinical data. Even though empirical experiments show that visualizations facilitate clinical data understanding, a consistent method to assess their effectiveness is still missing.METHODS: The insight-based methodology determines the quality of insights a user acquires from the visualization. Insights receive a value from one to five points based on a domain-specific criteria. Five professional psychiatrists took part in the study using real de-identified clinical data spanning 4 years of medical history.RESULTS: A total of 50 assessments were transcribed and analyzed. Comparing a total of 558 insights using Health Timeline and 576 without, the mean value using the Timeline (1.7) was higher than without (1.26; p<0.01), similarly the cumulative value with the Timeline (11.87) was higher than without (10.96: p<0.01). The average time required to formulate the first insight with the Timeline was higher (13.16 s) than without (7 s; p<0.01). Seven insights achieved the highest possible value using Health Timeline while none were obtained without it.CONCLUSIONS: The Health Timeline effectively improved understanding of clinical data and helped participants recognize complex patterns from the data. By applying the insight-based methodology, the effectiveness of the Health Timeline was quantified, documented and demonstrated. As an outcome of this exercise, we propose the use of such methodologies to measure the effectiveness of visualizations that assist the clinical decision-making process.",

keywords = "Clinical data, Data visualization, Electronic health record, Health informatics, Insight-based methodology",

author = "Andres Ledesma and Niranjan Bidargaddi and J{\"o}rg Strobel and Geoffrey Schrader and Hannu Nieminen and Ilkka Korhonen and Miikka Ermes",

note = "Publisher Copyright: {\textcopyright} 2019 The Author(s).",

year = "2019",

month = aug,

day = "22",

doi = "10.1186/s12911-019-0885-x",

language = "English",

volume = "19",

journal = "BMC Medical Informatics and Decision Making",

issn = "1472-6947",

publisher = "BioMed Central",

number = "1",

}

TY - JOUR

T1 - Health timeline

T2 - An insight-based study of a timeline visualization of clinical data

AU - Ledesma, Andres

AU - Bidargaddi, Niranjan

AU - Strobel, Jörg

AU - Schrader, Geoffrey

AU - Nieminen, Hannu

AU - Korhonen, Ilkka

AU - Ermes, Miikka

PY - 2019/8/22

Y1 - 2019/8/22

N2 - BACKGROUND: The increasing complexity and volume of clinical data poses a challenge in the decision-making process. Data visualizations can assist in this process by speeding up the time required to analyze and understand clinical data. Even though empirical experiments show that visualizations facilitate clinical data understanding, a consistent method to assess their effectiveness is still missing.METHODS: The insight-based methodology determines the quality of insights a user acquires from the visualization. Insights receive a value from one to five points based on a domain-specific criteria. Five professional psychiatrists took part in the study using real de-identified clinical data spanning 4 years of medical history.RESULTS: A total of 50 assessments were transcribed and analyzed. Comparing a total of 558 insights using Health Timeline and 576 without, the mean value using the Timeline (1.7) was higher than without (1.26; p<0.01), similarly the cumulative value with the Timeline (11.87) was higher than without (10.96: p<0.01). The average time required to formulate the first insight with the Timeline was higher (13.16 s) than without (7 s; p<0.01). Seven insights achieved the highest possible value using Health Timeline while none were obtained without it.CONCLUSIONS: The Health Timeline effectively improved understanding of clinical data and helped participants recognize complex patterns from the data. By applying the insight-based methodology, the effectiveness of the Health Timeline was quantified, documented and demonstrated. As an outcome of this exercise, we propose the use of such methodologies to measure the effectiveness of visualizations that assist the clinical decision-making process.

AB - BACKGROUND: The increasing complexity and volume of clinical data poses a challenge in the decision-making process. Data visualizations can assist in this process by speeding up the time required to analyze and understand clinical data. Even though empirical experiments show that visualizations facilitate clinical data understanding, a consistent method to assess their effectiveness is still missing.METHODS: The insight-based methodology determines the quality of insights a user acquires from the visualization. Insights receive a value from one to five points based on a domain-specific criteria. Five professional psychiatrists took part in the study using real de-identified clinical data spanning 4 years of medical history.RESULTS: A total of 50 assessments were transcribed and analyzed. Comparing a total of 558 insights using Health Timeline and 576 without, the mean value using the Timeline (1.7) was higher than without (1.26; p<0.01), similarly the cumulative value with the Timeline (11.87) was higher than without (10.96: p<0.01). The average time required to formulate the first insight with the Timeline was higher (13.16 s) than without (7 s; p<0.01). Seven insights achieved the highest possible value using Health Timeline while none were obtained without it.CONCLUSIONS: The Health Timeline effectively improved understanding of clinical data and helped participants recognize complex patterns from the data. By applying the insight-based methodology, the effectiveness of the Health Timeline was quantified, documented and demonstrated. As an outcome of this exercise, we propose the use of such methodologies to measure the effectiveness of visualizations that assist the clinical decision-making process.

KW - Clinical data

KW - Data visualization

KW - Electronic health record

KW - Health informatics

KW - Insight-based methodology

UR - http://www.scopus.com/inward/record.url?scp=85071301071&partnerID=8YFLogxK

U2 - 10.1186/s12911-019-0885-x

DO - 10.1186/s12911-019-0885-x

M3 - Article

C2 - 31438942

AN - SCOPUS:85071301071

SN - 1472-6947

VL - 19

JO - BMC Medical Informatics and Decision Making

JF - BMC Medical Informatics and Decision Making

IS - 1

M1 - 170

ER -

Health timeline: An insight-based study of a timeline visualization of clinical data

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this