Assess and validate predictive performance of models for in-hospital mortality in COVID-19 patients: A retrospective cohort study in the Netherlands comparing the value of registry data with high-granular electronic health records

Iacopo Vagliano; Martijn C Schut; Ameen Abu-Hanna; Dave A Dongelmans; Dylan W de Lange; Diederik Gommers; Olaf L Cremer; Rob J Bosman; Sander Rigter; Evert-Jan Wils; Tim Frenzel; Remko de Jong; Marco A A Peters; Marlijn J A Kamps; Dharmanand Ramnarain; Ralph Nowitzky; Fleur G C A Nooteboom; Wouter de Ruijter; Louise C Urlings-Strop; Ellen G M Smit; D Jannet Mehagnoul-Schipper; Tom Dormans; Cornelis P C de Jager; Stefaan H A Hendriks; Sefanja Achterberg; Evelien Oostdijk; Auke C Reidinga; Barbara Festen-Spanjer; Gert B Brunnekreef; Alexander D Cornet; Walter van den Tempel; Age D Boelens; Peter Koetsier; Judith Lens; Harald J Faber; A Karakus; Robert Entjes; Paul de Jong; Thijs C D Rettig; M C Reuland; Sesmu Arbous; Lucas M Fleuren; Tariq A Dam; Patrick J Thoral; Robbert C A Lalisang; Michele Tonutti; Daan P de Bruin; Paul W G Elbers; Nicolette F de Keizer; Dutch COVID-19 Research Consortium, the Dutch ICU Data Sharing Against COVID-19 Collaborators

doi:10.1016/j.ijmedinf.2022.104863

Assess and validate predictive performance of models for in-hospital mortality in COVID-19 patients: A retrospective cohort study in the Netherlands comparing the value of registry data with high-granular electronic health records

Int J Med Inform. 2022 Nov:167:104863. doi: 10.1016/j.ijmedinf.2022.104863. Epub 2022 Sep 22.

Authors

Iacopo Vagliano¹, Martijn C Schut², Ameen Abu-Hanna², Dave A Dongelmans³, Dylan W de Lange⁴, Diederik Gommers⁵, Olaf L Cremer⁶, Rob J Bosman⁷, Sander Rigter⁸, Evert-Jan Wils⁹, Tim Frenzel¹⁰, Remko de Jong¹¹, Marco A A Peters¹², Marlijn J A Kamps¹³, Dharmanand Ramnarain¹⁴, Ralph Nowitzky¹⁵, Fleur G C A Nooteboom¹⁶, Wouter de Ruijter¹⁷, Louise C Urlings-Strop¹⁸, Ellen G M Smit¹⁹, D Jannet Mehagnoul-Schipper²⁰, Tom Dormans²¹, Cornelis P C de Jager²², Stefaan H A Hendriks²³, Sefanja Achterberg²⁴, Evelien Oostdijk²⁵, Auke C Reidinga²⁶, Barbara Festen-Spanjer²⁷, Gert B Brunnekreef²⁸, Alexander D Cornet²⁹, Walter van den Tempel³⁰, Age D Boelens³¹, Peter Koetsier³², Judith Lens³³, Harald J Faber³⁴, A Karakus³⁵, Robert Entjes³⁶, Paul de Jong³⁷, Thijs C D Rettig³⁸, M C Reuland², Sesmu Arbous³⁹, Lucas M Fleuren⁴⁰, Tariq A Dam⁴⁰, Patrick J Thoral⁴⁰, Robbert C A Lalisang⁴¹, Michele Tonutti⁴¹, Daan P de Bruin⁴¹, Paul W G Elbers⁴⁰, Nicolette F de Keizer⁴²; Dutch COVID-19 Research Consortium, the Dutch ICU Data Sharing Against COVID-19 Collaborators

Affiliations

¹ Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands. Electronic address: [email protected].
² Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands.
³ National Intensive Care Evaluation (NICE) foundation, Amsterdam, The Netherlands; Department of Intensive Care Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.
⁴ National Intensive Care Evaluation (NICE) foundation, Amsterdam, The Netherlands; Department of Intensive Care Medicine, University Medical Center Utrecht, University Utrecht, Utrecht, The Netherlands.
⁵ Department of Intensive Care, Erasmus Medical Center, Rotterdam, The Netherlands.
⁶ Intensive Care, UMC Utrecht, Utrecht, The Netherlands.
⁷ ICU, OLVG, Amsterdam, The Netherlands.
⁸ Department of Anesthesiology and Intensive Care, St. Antonius Hospital, Nieuwegein, The Netherlands.
⁹ Department of Intensive Care, Franciscus Gasthuis & Vlietland, Rotterdam, The Netherlands.
¹⁰ Department of Intensive Care Medicine, Radboud University Medical Center, Nijmegen, The Netherlands.
¹¹ Intensive Care, Bovenij Ziekenhuis, Amsterdam, The Netherlands.
¹² Intensive Care, Canisius Wilhelmina Ziekenhuis, Nijmegen, The Netherlands.
¹³ Intensive Care, Catharina Ziekenhuis Eindhoven, Eindhoven, The Netherlands.
¹⁴ Department of Intensive Care, ETZ Tilburg, Tilburg, The Netherlands.
¹⁵ Intensive Care, Haga Ziekenhuis, Den Haag, The Netherlands.
¹⁶ Intensive Care, Laurentius Ziekenhuis, Roermond, The Netherlands.
¹⁷ Department of Intensive Care Medicine, Northwest Clinics, Alkmaar, The Netherlands.
¹⁸ Intensive Care, Reinier de Graaf Gasthuis, Delft, The Netherlands.
¹⁹ Intensive Care, Spaarne Gasthuis, Haarlem en Hoofddorp, The Netherlands.
²⁰ Intensive Care, VieCuri Medisch Centrum, Venlo, The Netherlands.
²¹ Intensive care, Zuyderland MC, Heerlen, The Netherlands.
²² Department of Intensive Care, Jeroen Bosch Ziekenhuis, Den Bosch, The Netherlands.
²³ Intensive Care, Albert Schweitzerziekenhuis, Dordrecht, The Netherlands.
²⁴ ICU, Haaglanden Medisch Centrum, Den Haag, The Netherlands.
²⁵ ICU, Maasstad Ziekenhuis Rotterdam, Rotterdam, The Netherlands.
²⁶ ICU, SEH, BWC, Martiniziekenhuis, Groningen, The Netherlands.
²⁷ Intensive Care, Ziekenhuis Gelderse Vallei, Ede, The Netherlands.
²⁸ Department of Intensive Care, Ziekenhuisgroep Twente, Almelo, The Netherlands.
²⁹ Department of Intensive Care, Medisch Spectrum Twente, Enschede, The Netherlands.
³⁰ Department of Intensive Care, Ikazia Ziekenhuis Rotterdam, Rotterdam, The Netherlands.
³¹ Anesthesiology, Antonius Ziekenhuis Sneek, Sneek, The Netherlands.
³² Intensive Care, Medisch Centrum Leeuwarden, Leeuwarden, The Netherlands.
³³ ICU, IJsselland Ziekenhuis, Capelle aan den IJssel, The Netherlands.
³⁴ ICU, WZA, Assen, The Netherlands.
³⁵ Department of Intensive Care, Diakonessenhuis Hospital, Utrecht, The Netherlands.
³⁶ Department of Intensive Care, Adrz, Goes, The Netherlands.
³⁷ Department of Anesthesia and Intensive Care, Slingeland Ziekenhuis, Doetinchem, The Netherlands.
³⁸ Department of Anesthesiology, Intensive Care and Pain Medicine, Amphia Ziekenhuis, Breda, The Netherlands.
³⁹ Intensivist, LUMC, Leiden, The Netherlands.
⁴⁰ Department of Intensive Care Medicine, Laboratory for Critical Care Computational Intelligence, Amsterdam Medical Data Science, Amsterdam UMC, Vrije Universiteit, Amsterdam, The Netherlands.
⁴¹ Pacmed, Amsterdam, The Netherlands.
⁴² Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands; National Intensive Care Evaluation (NICE) foundation, Amsterdam, The Netherlands.

Abstract

Purpose: To assess, validate and compare the predictive performance of models for in-hospital mortality of COVID-19 patients admitted to the intensive care unit (ICU) over two different waves of infections. Our models were built with high-granular Electronic Health Records (EHR) data versus less-granular registry data.

Methods: Observational study of all COVID-19 patients admitted to 19 Dutch ICUs participating in both the national quality registry National Intensive Care Evaluation (NICE) and the EHR-based Dutch Data Warehouse (hereafter EHR). Multiple models were developed on data from the first 24 h of ICU admissions from February to June 2020 (first COVID-19 wave) and validated on prospective patients admitted to the same ICUs between July and December 2020 (second COVID-19 wave). We assessed model discrimination, calibration, and the degree of relatedness between development and validation population. Coefficients were used to identify relevant risk factors.

Results: A total of 1533 patients from the EHR and 1563 from the registry were included. With high granular EHR data, the average AUROC was 0.69 (standard deviation of 0.05) for the internal validation, and the AUROC was 0.75 for the temporal validation. The registry model achieved an average AUROC of 0.76 (standard deviation of 0.05) in the internal validation and 0.77 in the temporal validation. In the EHR data, age, and respiratory-system related variables were the most important risk factors identified. In the NICE registry data, age and chronic respiratory insufficiency were the most important risk factors.

Conclusion: In our study, prognostic models built on less-granular but readily-available registry data had similar performance to models built on high-granular EHR data and showed similar transportability to a prospective COVID-19 population. Future research is needed to verify whether this finding can be confirmed for upcoming waves.

Keywords: Covid-19 [C01.748.610.763.500]; Critical care [E02.760.190]; Electronic Health Record [E05.318.308.940.968.625.500]; In-hospital mortality [E05.318.308.985.550.400]; Machine learning [G17.035.250.500]; Prognosis [E01.789].

Publication types

Observational Study
Research Support, Non-U.S. Gov't

MeSH terms

COVID-19* / epidemiology
Electronic Health Records
Hospital Mortality
Humans
Intensive Care Units
Netherlands / epidemiology
Registries
Retrospective Studies