Early prediction of level-of-care requirements in patients with COVID-19
This study examined records of 2566 consecutive COVID-19 patients at five Massachusetts hospitals and sought to predict level-of-care requirements based on clinical and laboratory data. Several classification methods were applied and compared against standard pneumonia severity scores. The need for...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
eLife Sciences Publications Ltd
2020-10-01
|
| Series: | eLife |
| Subjects: | |
| Online Access: | https://elifesciences.org/articles/60519 |
| _version_ | 1828384527147335680 |
|---|---|
| author | Boran Hao Shahabeddin Sotudian Taiyao Wang Tingting Xu Yang Hu Apostolos Gaitanidis Kerry Breen George C Velmahos Ioannis Ch Paschalidis |
| author_facet | Boran Hao Shahabeddin Sotudian Taiyao Wang Tingting Xu Yang Hu Apostolos Gaitanidis Kerry Breen George C Velmahos Ioannis Ch Paschalidis |
| author_sort | Boran Hao |
| collection | DOAJ |
| description | This study examined records of 2566 consecutive COVID-19 patients at five Massachusetts hospitals and sought to predict level-of-care requirements based on clinical and laboratory data. Several classification methods were applied and compared against standard pneumonia severity scores. The need for hospitalization, ICU care, and mechanical ventilation were predicted with a validation accuracy of 88%, 87%, and 86%, respectively. Pneumonia severity scores achieve respective accuracies of 73% and 74% for ICU care and ventilation. When predictions are limited to patients with more complex disease, the accuracy of the ICU and ventilation prediction models achieved accuracy of 83% and 82%, respectively. Vital signs, age, BMI, dyspnea, and comorbidities were the most important predictors of hospitalization. Opacities on chest imaging, age, admission vital signs and symptoms, male gender, admission laboratory results, and diabetes were the most important risk factors for ICU admission and mechanical ventilation. The factors identified collectively form a signature of the novel COVID-19 disease. |
| first_indexed | 2024-12-10T05:04:47Z |
| format | Article |
| id | doaj.art-29e3eae1bf974ebbb0ee81ea472657a3 |
| institution | Directory Open Access Journal |
| issn | 2050-084X |
| language | English |
| last_indexed | 2024-12-10T05:04:47Z |
| publishDate | 2020-10-01 |
| publisher | eLife Sciences Publications Ltd |
| record_format | Article |
| series | eLife |
| spelling | doaj.art-29e3eae1bf974ebbb0ee81ea472657a32022-12-22T02:01:16ZengeLife Sciences Publications LtdeLife2050-084X2020-10-01910.7554/eLife.60519Early prediction of level-of-care requirements in patients with COVID-19Boran Hao0Shahabeddin Sotudian1https://orcid.org/0000-0002-5864-6192Taiyao Wang2https://orcid.org/0000-0002-0331-3892Tingting Xu3Yang Hu4Apostolos Gaitanidis5Kerry Breen6George C Velmahos7Ioannis Ch Paschalidis8https://orcid.org/0000-0002-3343-2913Center for Information and Systems Engineering, Boston University, Boston, United StatesCenter for Information and Systems Engineering, Boston University, Boston, United StatesCenter for Information and Systems Engineering, Boston University, Boston, United StatesCenter for Information and Systems Engineering, Boston University, Boston, United StatesCenter for Information and Systems Engineering, Boston University, Boston, United StatesDivision of Trauma, Emergency Services, and Surgical Critical Care Massachusetts General Hospital, Harvard Medical School, Boston, United StatesDivision of Trauma, Emergency Services, and Surgical Critical Care Massachusetts General Hospital, Harvard Medical School, Boston, United StatesDivision of Trauma, Emergency Services, and Surgical Critical Care Massachusetts General Hospital, Harvard Medical School, Boston, United StatesCenter for Information and Systems Engineering, Boston University, Boston, United StatesThis study examined records of 2566 consecutive COVID-19 patients at five Massachusetts hospitals and sought to predict level-of-care requirements based on clinical and laboratory data. Several classification methods were applied and compared against standard pneumonia severity scores. The need for hospitalization, ICU care, and mechanical ventilation were predicted with a validation accuracy of 88%, 87%, and 86%, respectively. Pneumonia severity scores achieve respective accuracies of 73% and 74% for ICU care and ventilation. When predictions are limited to patients with more complex disease, the accuracy of the ICU and ventilation prediction models achieved accuracy of 83% and 82%, respectively. Vital signs, age, BMI, dyspnea, and comorbidities were the most important predictors of hospitalization. Opacities on chest imaging, age, admission vital signs and symptoms, male gender, admission laboratory results, and diabetes were the most important risk factors for ICU admission and mechanical ventilation. The factors identified collectively form a signature of the novel COVID-19 disease.https://elifesciences.org/articles/60519COVID-19SARS-CoV-2risk predictionartificial intelligencemachine learningcritical care |
| spellingShingle | Boran Hao Shahabeddin Sotudian Taiyao Wang Tingting Xu Yang Hu Apostolos Gaitanidis Kerry Breen George C Velmahos Ioannis Ch Paschalidis Early prediction of level-of-care requirements in patients with COVID-19 eLife COVID-19 SARS-CoV-2 risk prediction artificial intelligence machine learning critical care |
| title | Early prediction of level-of-care requirements in patients with COVID-19 |
| title_full | Early prediction of level-of-care requirements in patients with COVID-19 |
| title_fullStr | Early prediction of level-of-care requirements in patients with COVID-19 |
| title_full_unstemmed | Early prediction of level-of-care requirements in patients with COVID-19 |
| title_short | Early prediction of level-of-care requirements in patients with COVID-19 |
| title_sort | early prediction of level of care requirements in patients with covid 19 |
| topic | COVID-19 SARS-CoV-2 risk prediction artificial intelligence machine learning critical care |
| url | https://elifesciences.org/articles/60519 |
| work_keys_str_mv | AT boranhao earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT shahabeddinsotudian earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT taiyaowang earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT tingtingxu earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT yanghu earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT apostolosgaitanidis earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT kerrybreen earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT georgecvelmahos earlypredictionoflevelofcarerequirementsinpatientswithcovid19 AT ioannischpaschalidis earlypredictionoflevelofcarerequirementsinpatientswithcovid19 |