A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems

Matheus Araújo; Paula Van Dommelen; Jaideep Srivastava; Ekaterina Koledova

doi:10.3233/SHTI210803

A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems

Matheus Araújo, Paula Van Dommelen, Jaideep Srivastava, Ekaterina Koledova

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Recombinant human growth hormone (r-hGH) is an established therapy for growth hormone deficiency (GHD); yet, some patients fail to achieve their full height potential, with poor adherence and persistence with the prescribed regimen often a contributing factor. A data-driven clinical decision support system based on 'traffic light' visualizations for adherence risk management of patients receiving r-hGH treatment was developed. This research was feasible thanks to data-sharing agreements that allowed the creation of these models using real-world data of r-hGH adherence from easypod™ connect; data was retrieved for 11,015 children receiving r-hGH therapy for ≥180 days. Patients' adherence to therapy was represented using four values (mean and standard deviation [SD] of daily adherence and hours to next injection). Cluster analysis was used to categorize adherence patterns using a Gaussian mixture model. Following a traffic lights-inspired visualization approach, the algorithm was set to generate three clusters: green, yellow, or red status, corresponding to high, medium, and low adherence, respectively. The area under the receiver operating characteristic curve (AUC-ROC) was used to find optimum thresholds for independent traffic lights according to each metric. The most appropriate traffic light used the SD of the hours to the next injection, with an AUC-ROC value of 0.85 when compared to the complex clustering algorithm. For the daily adherence-based traffic lights, optimum thresholds were >0.82 (SD, <0.37), 0.53-0.82 (SD, 0.37-0.61), and <0.53 (SD, >0.61) for high, medium, and low adherence, respectively. For hours to next injection, the corresponding optimum thresholds were <27.18 (SD, <10.06), 27.18-34.01 (SD, 10.06-29.63), and >34.01 (SD, >29.63). Our research indicates that implementation of a practical data-driven alert system based on recognised traffic-light coding would enable healthcare practitioners to monitor sub-optimally-adherent patients to r-hGH treatment for early intervention to improve treatment outcomes.

Original language	English (US)
Title of host publication	Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference
Editors	Jaime Delgado, Arriel Benis, Paula de Toledo, Parisis Gallos, Mauro Giacomini, Alicia Martinez-Garcia, Dario Salvi
Publisher	IOS Press BV
Pages	23-27
Number of pages	5
ISBN (Electronic)	9781643682365
DOIs	https://doi.org/10.3233/SHTI210803
State	Published - Nov 18 2021
Event	2021 European Federation for Medical Informatics (EFMI) Special Topic Conference, STC 2021 - Virtual, Online, Spain Duration: Nov 22 2021 → Nov 24 2021

Publication series

Name	Studies in Health Technology and Informatics
Volume	287
ISSN (Print)	0926-9630
ISSN (Electronic)	1879-8365

Conference

Conference	2021 European Federation for Medical Informatics (EFMI) Special Topic Conference, STC 2021
Country/Territory	Spain
City	Virtual, Online
Period	11/22/21 → 11/24/21

Bibliographical note

Funding Information:
This research was made possible thanks to a collaboration between The Netherlands Organization for Applied Scientific Research TNO, Leiden, The Netherlands & Merck Healthcare KGaA, Darmstadt, Germany. MA, JS, and PvD have consultancy agreements with Merck. EK is an employee of Merck Healthcare KGaA, Darmstadt, Germany and holds shares in the company. Editorial assistance was provided by Sinéad Mutton, inScience Communications, Springer Healthcare Ltd, UK, funded by Merck Healthcare KGaA, Darmstadt, Germany.

Publisher Copyright:
© 2021 The European Federation for Medical Informatics (EFMI) and IOS Press. All rights reserved.

Keywords

Adherence
cluster modeling
growth hormone deficiency
pediatrics
recombinant human growth hormone

PubMed: MeSH publication types

Journal Article

Access

10.3233/SHTI210803

OpenUrl availability

Full text

Cite this

Araújo, M., Van Dommelen, P., Srivastava, J., & Koledova, E. (2021). A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems. In J. Delgado, A. Benis, P. de Toledo, P. Gallos, M. Giacomini, A. Martinez-Garcia, & D. Salvi (Eds.), Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference (pp. 23-27). (Studies in Health Technology and Informatics; Vol. 287). IOS Press BV. https://doi.org/10.3233/SHTI210803

A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems. / Araújo, Matheus; Van Dommelen, Paula; Srivastava, Jaideep et al.
Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference. ed. / Jaime Delgado; Arriel Benis; Paula de Toledo; Parisis Gallos; Mauro Giacomini; Alicia Martinez-Garcia; Dario Salvi. IOS Press BV, 2021. p. 23-27 (Studies in Health Technology and Informatics; Vol. 287).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Araújo, M, Van Dommelen, P, Srivastava, J & Koledova, E 2021, A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems. in J Delgado, A Benis, P de Toledo, P Gallos, M Giacomini, A Martinez-Garcia & D Salvi (eds), Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference. Studies in Health Technology and Informatics, vol. 287, IOS Press BV, pp. 23-27, 2021 European Federation for Medical Informatics (EFMI) Special Topic Conference, STC 2021, Virtual, Online, Spain, 11/22/21. https://doi.org/10.3233/SHTI210803

Araújo M, Van Dommelen P, Srivastava J, Koledova E. A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems. In Delgado J, Benis A, de Toledo P, Gallos P, Giacomini M, Martinez-Garcia A, Salvi D, editors, Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference. IOS Press BV. 2021. p. 23-27. (Studies in Health Technology and Informatics). doi: 10.3233/SHTI210803

Araújo, Matheus ; Van Dommelen, Paula ; Srivastava, Jaideep et al. / A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems. Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference. editor / Jaime Delgado ; Arriel Benis ; Paula de Toledo ; Parisis Gallos ; Mauro Giacomini ; Alicia Martinez-Garcia ; Dario Salvi. IOS Press BV, 2021. pp. 23-27 (Studies in Health Technology and Informatics).

@inproceedings{783c29a2df8545f5a322cbcfc2f5c34c,

title = "A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems",

abstract = "Recombinant human growth hormone (r-hGH) is an established therapy for growth hormone deficiency (GHD); yet, some patients fail to achieve their full height potential, with poor adherence and persistence with the prescribed regimen often a contributing factor. A data-driven clinical decision support system based on 'traffic light' visualizations for adherence risk management of patients receiving r-hGH treatment was developed. This research was feasible thanks to data-sharing agreements that allowed the creation of these models using real-world data of r-hGH adherence from easypod{\texttrademark} connect; data was retrieved for 11,015 children receiving r-hGH therapy for ≥180 days. Patients' adherence to therapy was represented using four values (mean and standard deviation [SD] of daily adherence and hours to next injection). Cluster analysis was used to categorize adherence patterns using a Gaussian mixture model. Following a traffic lights-inspired visualization approach, the algorithm was set to generate three clusters: green, yellow, or red status, corresponding to high, medium, and low adherence, respectively. The area under the receiver operating characteristic curve (AUC-ROC) was used to find optimum thresholds for independent traffic lights according to each metric. The most appropriate traffic light used the SD of the hours to the next injection, with an AUC-ROC value of 0.85 when compared to the complex clustering algorithm. For the daily adherence-based traffic lights, optimum thresholds were >0.82 (SD, <0.37), 0.53-0.82 (SD, 0.37-0.61), and <0.53 (SD, >0.61) for high, medium, and low adherence, respectively. For hours to next injection, the corresponding optimum thresholds were <27.18 (SD, <10.06), 27.18-34.01 (SD, 10.06-29.63), and >34.01 (SD, >29.63). Our research indicates that implementation of a practical data-driven alert system based on recognised traffic-light coding would enable healthcare practitioners to monitor sub-optimally-adherent patients to r-hGH treatment for early intervention to improve treatment outcomes. ",

keywords = "Adherence, cluster modeling, growth hormone deficiency, pediatrics, recombinant human growth hormone",

author = "Matheus Ara{\'u}jo and {Van Dommelen}, Paula and Jaideep Srivastava and Ekaterina Koledova",

note = "Funding Information: This research was made possible thanks to a collaboration between The Netherlands Organization for Applied Scientific Research TNO, Leiden, The Netherlands & Merck Healthcare KGaA, Darmstadt, Germany. MA, JS, and PvD have consultancy agreements with Merck. EK is an employee of Merck Healthcare KGaA, Darmstadt, Germany and holds shares in the company. Editorial assistance was provided by Sin{\'e}ad Mutton, inScience Communications, Springer Healthcare Ltd, UK, funded by Merck Healthcare KGaA, Darmstadt, Germany. Publisher Copyright: {\textcopyright} 2021 The European Federation for Medical Informatics (EFMI) and IOS Press. All rights reserved.; 2021 European Federation for Medical Informatics (EFMI) Special Topic Conference, STC 2021 ; Conference date: 22-11-2021 Through 24-11-2021",

year = "2021",

month = nov,

day = "18",

doi = "10.3233/SHTI210803",

language = "English (US)",

series = "Studies in Health Technology and Informatics",

publisher = "IOS Press BV",

pages = "23--27",

editor = "Jaime Delgado and Arriel Benis and {de Toledo}, Paula and Parisis Gallos and Mauro Giacomini and Alicia Martinez-Garcia and Dario Salvi",

booktitle = "Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference",

address = "Netherlands",

}

TY - GEN

T1 - A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems

AU - Araújo, Matheus

AU - Van Dommelen, Paula

AU - Srivastava, Jaideep

AU - Koledova, Ekaterina

N1 - Funding Information: This research was made possible thanks to a collaboration between The Netherlands Organization for Applied Scientific Research TNO, Leiden, The Netherlands & Merck Healthcare KGaA, Darmstadt, Germany. MA, JS, and PvD have consultancy agreements with Merck. EK is an employee of Merck Healthcare KGaA, Darmstadt, Germany and holds shares in the company. Editorial assistance was provided by Sinéad Mutton, inScience Communications, Springer Healthcare Ltd, UK, funded by Merck Healthcare KGaA, Darmstadt, Germany. Publisher Copyright: © 2021 The European Federation for Medical Informatics (EFMI) and IOS Press. All rights reserved.

PY - 2021/11/18

Y1 - 2021/11/18

N2 - Recombinant human growth hormone (r-hGH) is an established therapy for growth hormone deficiency (GHD); yet, some patients fail to achieve their full height potential, with poor adherence and persistence with the prescribed regimen often a contributing factor. A data-driven clinical decision support system based on 'traffic light' visualizations for adherence risk management of patients receiving r-hGH treatment was developed. This research was feasible thanks to data-sharing agreements that allowed the creation of these models using real-world data of r-hGH adherence from easypod™ connect; data was retrieved for 11,015 children receiving r-hGH therapy for ≥180 days. Patients' adherence to therapy was represented using four values (mean and standard deviation [SD] of daily adherence and hours to next injection). Cluster analysis was used to categorize adherence patterns using a Gaussian mixture model. Following a traffic lights-inspired visualization approach, the algorithm was set to generate three clusters: green, yellow, or red status, corresponding to high, medium, and low adherence, respectively. The area under the receiver operating characteristic curve (AUC-ROC) was used to find optimum thresholds for independent traffic lights according to each metric. The most appropriate traffic light used the SD of the hours to the next injection, with an AUC-ROC value of 0.85 when compared to the complex clustering algorithm. For the daily adherence-based traffic lights, optimum thresholds were >0.82 (SD, <0.37), 0.53-0.82 (SD, 0.37-0.61), and <0.53 (SD, >0.61) for high, medium, and low adherence, respectively. For hours to next injection, the corresponding optimum thresholds were <27.18 (SD, <10.06), 27.18-34.01 (SD, 10.06-29.63), and >34.01 (SD, >29.63). Our research indicates that implementation of a practical data-driven alert system based on recognised traffic-light coding would enable healthcare practitioners to monitor sub-optimally-adherent patients to r-hGH treatment for early intervention to improve treatment outcomes.

AB - Recombinant human growth hormone (r-hGH) is an established therapy for growth hormone deficiency (GHD); yet, some patients fail to achieve their full height potential, with poor adherence and persistence with the prescribed regimen often a contributing factor. A data-driven clinical decision support system based on 'traffic light' visualizations for adherence risk management of patients receiving r-hGH treatment was developed. This research was feasible thanks to data-sharing agreements that allowed the creation of these models using real-world data of r-hGH adherence from easypod™ connect; data was retrieved for 11,015 children receiving r-hGH therapy for ≥180 days. Patients' adherence to therapy was represented using four values (mean and standard deviation [SD] of daily adherence and hours to next injection). Cluster analysis was used to categorize adherence patterns using a Gaussian mixture model. Following a traffic lights-inspired visualization approach, the algorithm was set to generate three clusters: green, yellow, or red status, corresponding to high, medium, and low adherence, respectively. The area under the receiver operating characteristic curve (AUC-ROC) was used to find optimum thresholds for independent traffic lights according to each metric. The most appropriate traffic light used the SD of the hours to the next injection, with an AUC-ROC value of 0.85 when compared to the complex clustering algorithm. For the daily adherence-based traffic lights, optimum thresholds were >0.82 (SD, <0.37), 0.53-0.82 (SD, 0.37-0.61), and <0.53 (SD, >0.61) for high, medium, and low adherence, respectively. For hours to next injection, the corresponding optimum thresholds were <27.18 (SD, <10.06), 27.18-34.01 (SD, 10.06-29.63), and >34.01 (SD, >29.63). Our research indicates that implementation of a practical data-driven alert system based on recognised traffic-light coding would enable healthcare practitioners to monitor sub-optimally-adherent patients to r-hGH treatment for early intervention to improve treatment outcomes.

KW - Adherence

KW - cluster modeling

KW - growth hormone deficiency

KW - pediatrics

KW - recombinant human growth hormone

UR - http://www.scopus.com/inward/record.url?scp=85120523639&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85120523639&partnerID=8YFLogxK

U2 - 10.3233/SHTI210803

DO - 10.3233/SHTI210803

M3 - Conference contribution

C2 - 34795072

AN - SCOPUS:85120523639

T3 - Studies in Health Technology and Informatics

SP - 23

EP - 27

BT - Applying the FAIR Principles to Accelerate Health Research in Europe in the Post COVID-19 Era - Proceedings of the 2021 EFMI Special Topic Conference

A2 - Delgado, Jaime

A2 - Benis, Arriel

A2 - de Toledo, Paula

A2 - Gallos, Parisis

A2 - Giacomini, Mauro

A2 - Martinez-Garcia, Alicia

A2 - Salvi, Dario

PB - IOS Press BV

T2 - 2021 European Federation for Medical Informatics (EFMI) Special Topic Conference, STC 2021

Y2 - 22 November 2021 through 24 November 2021

ER -

A Data-iven Intervention Framework for Improving Adherence to Growth Hormone Therapy ased on Clustering Analysis and Traffic Light Alerting Systems

Abstract

Publication series

Conference

Bibliographical note

Keywords

PubMed: MeSH publication types

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this