Comparison of static and rolling logistic regression models on predicting invasive mechanical ventilation or death from COVID-19-A retrospective, multicentre study
© 2022 The Authors. The Clinical Respiratory Journal published by John Wiley & Sons Ltd..
INTRODUCTION: COVID-19 virus has undergone mutations, and the introduction of vaccines and effective treatments have changed its clinical severity. We hypothesized that models that evolve may better predict invasive mechanical ventilation or death than do static models.
METHODS: This retrospective study of adult patients with COVID-19 from six Michigan hospitals analysed 20 demographic, comorbid, vital sign and laboratory factors, one derived factor and nine factors representing changes in vital signs or laboratory values with time for their ability to predict death or invasive mechanical ventilation within the next 4, 8 or 24 h. Static logistic regression was constructed on the initial 300 patients and tested on the remaining 6741 patients. Rolling logistic regression was similarly constructed on the initial 300 patients, but then new patients were added, and older patients removed. Each new construction model was subsequently tested on the next patient. Static and rolling models were compared with receiver operator characteristic and precision-recall curves.
RESULTS: Of the 7041 patients, 534 (7.6%) required invasive mechanical ventilation or died within 14 days of arrival. Rolling models improved discrimination (0.865 ± 0.010, 0.856 ± 0.007 and 0.843 ± 0.005 for the 4, 8 and 24-h models, respectively; all p < 0.001 compared with the static logistic regressions with 0.827 ± 0.011, 0.794 ± 0.012 and 0.735 ± 0.012, respectively). Similarly, the areas under the precision-recall curves improved from 0.006, 0.010 and 0.021 with the static models to 0.030, 0.045 and 0.076 for the 4-, 8- and 24-h rolling models, respectively, all p < 0.001.
CONCLUSION: Rolling models with contemporaneous data maintained better metrics of performance than static models, which used older data.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:17 |
|---|---|
| Enthalten in: |
The clinical respiratory journal - 17(2023), 1 vom: 22. Jan., Seite 40-49 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Engoren, Milo [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
COVID |
|---|
| Anmerkungen: |
Date Completed 11.01.2023 Date Revised 12.01.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1111/crj.13560 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM349248788 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM349248788 | ||
| 003 | DE-627 | ||
| 005 | 20231226042152.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1111/crj.13560 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1164.xml |
| 035 | |a (DE-627)NLM349248788 | ||
| 035 | |a (NLM)36411722 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Engoren, Milo |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Comparison of static and rolling logistic regression models on predicting invasive mechanical ventilation or death from COVID-19-A retrospective, multicentre study |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 11.01.2023 | ||
| 500 | |a Date Revised 12.01.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2022 The Authors. The Clinical Respiratory Journal published by John Wiley & Sons Ltd. | ||
| 520 | |a INTRODUCTION: COVID-19 virus has undergone mutations, and the introduction of vaccines and effective treatments have changed its clinical severity. We hypothesized that models that evolve may better predict invasive mechanical ventilation or death than do static models | ||
| 520 | |a METHODS: This retrospective study of adult patients with COVID-19 from six Michigan hospitals analysed 20 demographic, comorbid, vital sign and laboratory factors, one derived factor and nine factors representing changes in vital signs or laboratory values with time for their ability to predict death or invasive mechanical ventilation within the next 4, 8 or 24 h. Static logistic regression was constructed on the initial 300 patients and tested on the remaining 6741 patients. Rolling logistic regression was similarly constructed on the initial 300 patients, but then new patients were added, and older patients removed. Each new construction model was subsequently tested on the next patient. Static and rolling models were compared with receiver operator characteristic and precision-recall curves | ||
| 520 | |a RESULTS: Of the 7041 patients, 534 (7.6%) required invasive mechanical ventilation or died within 14 days of arrival. Rolling models improved discrimination (0.865 ± 0.010, 0.856 ± 0.007 and 0.843 ± 0.005 for the 4, 8 and 24-h models, respectively; all p < 0.001 compared with the static logistic regressions with 0.827 ± 0.011, 0.794 ± 0.012 and 0.735 ± 0.012, respectively). Similarly, the areas under the precision-recall curves improved from 0.006, 0.010 and 0.021 with the static models to 0.030, 0.045 and 0.076 for the 4-, 8- and 24-h rolling models, respectively, all p < 0.001 | ||
| 520 | |a CONCLUSION: Rolling models with contemporaneous data maintained better metrics of performance than static models, which used older data | ||
| 650 | 4 | |a Multicenter Study | |
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a COVID | |
| 650 | 4 | |a death, clinical decision support, clinical prediction models | |
| 650 | 4 | |a logistic regression | |
| 650 | 4 | |a mechanical ventilation | |
| 700 | 1 | |a Pancaro, Carlo |e verfasserin |4 aut | |
| 700 | 1 | |a Yeldo, Nicholas S |e verfasserin |4 aut | |
| 700 | 1 | |a Kerzabi, Lotfi S |e verfasserin |4 aut | |
| 700 | 1 | |a Douville, Nicholas |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t The clinical respiratory journal |d 2007 |g 17(2023), 1 vom: 22. Jan., Seite 40-49 |w (DE-627)NLM19646837X |x 1752-699X |7 nnns |
| 773 | 1 | 8 | |g volume:17 |g year:2023 |g number:1 |g day:22 |g month:01 |g pages:40-49 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1111/crj.13560 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 17 |j 2023 |e 1 |b 22 |c 01 |h 40-49 | ||