Vertical Compliance: A novel method of reporting patient specific ERAS compliance for real-time risk assessment.
Compliance
EIAS
ERAS
hepatobiliary
outcomes
Journal
International journal of medical informatics
ISSN: 1872-8243
Titre abrégé: Int J Med Inform
Pays: Ireland
ID NLM: 9711057
Informations de publication
Date de publication:
09 2020
09 2020
Historique:
received:
09
03
2020
revised:
17
04
2020
accepted:
21
05
2020
pubmed:
7
6
2020
medline:
22
12
2020
entrez:
7
6
2020
Statut:
ppublish
Résumé
ERAS protocol compliance is traditionally measured and reported as 'items compliance' which is retrospective longitudinal adherence of all patients to an index element. Reporting in this manner is restrictive and cannot impact care in real-time. In order to modify behavior effectively and instantaneously, we seek to introduce the novel concept of "vertical compliance". Vertical compliance is defined as assessment of an individual's compliance along his/her own surgical pathway as ERAS index items are encountered. Prospectively entered data from the ERAS Interactive Audit System (EIAS) database was queried for all patients who underwent pancreatic or hepatic resections. Individual linear regression models were generated using compliance data from 46 ERAS index items against LOS. Significant items with p value < 0.05 were compiled into multivariable linear regression models; each with a unique coefficient that could be used to predict effect on LOS as well as control for the effect of the other items in the model. Compliance data from 483 patients who underwent pancreatic resections and 292 patients who underwent hepatic resections was compiled. Seven ERAS items for pancreatic resections and six ERAS items for hepatic resections were found to significantly impact LOS. Regression models were created for each of the items in an additive fashion. Calculations to determine predicted LOS as a patient progressed through a pathway was able to be determined. Vertical compliance is a novel metric, described in this study, that can provide significant and accurate patient-specific risk prediction to impact care in real-time. This can allow for creation of a variable echelon such that pathway items can be ranked by importance on clinical outcome effect and patient progress can be monitored and altered.
Sections du résumé
BACKGROUND
ERAS protocol compliance is traditionally measured and reported as 'items compliance' which is retrospective longitudinal adherence of all patients to an index element. Reporting in this manner is restrictive and cannot impact care in real-time. In order to modify behavior effectively and instantaneously, we seek to introduce the novel concept of "vertical compliance".
METHODS
Vertical compliance is defined as assessment of an individual's compliance along his/her own surgical pathway as ERAS index items are encountered. Prospectively entered data from the ERAS Interactive Audit System (EIAS) database was queried for all patients who underwent pancreatic or hepatic resections. Individual linear regression models were generated using compliance data from 46 ERAS index items against LOS. Significant items with p value < 0.05 were compiled into multivariable linear regression models; each with a unique coefficient that could be used to predict effect on LOS as well as control for the effect of the other items in the model.
RESULTS
Compliance data from 483 patients who underwent pancreatic resections and 292 patients who underwent hepatic resections was compiled. Seven ERAS items for pancreatic resections and six ERAS items for hepatic resections were found to significantly impact LOS. Regression models were created for each of the items in an additive fashion. Calculations to determine predicted LOS as a patient progressed through a pathway was able to be determined.
CONCLUSION
Vertical compliance is a novel metric, described in this study, that can provide significant and accurate patient-specific risk prediction to impact care in real-time. This can allow for creation of a variable echelon such that pathway items can be ranked by importance on clinical outcome effect and patient progress can be monitored and altered.
Identifiants
pubmed: 32504910
pii: S1386-5056(20)30341-5
doi: 10.1016/j.ijmedinf.2020.104194
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
104194Informations de copyright
Copyright © 2020 Elsevier B.V. All rights reserved.