Can predicting COVID-19 mortality in a European cohort using only demographic and comorbidity data surpass age-based prediction: An externally validated study

Chatterjee, A.; Wu, Guangyao; Primakov, Sergey; Oberije, Cary; Woodruff, Henry C.; Kubben, Pieter; Henry, Ronald M.A.; Aries, Marcel; Beudel, Martijn; Noordzij, Peter G.; Dormans, Tom; Oever, Niels C. Gritters van den; Bergh, Joop van den; Wyers, Caroline E.; Şimşek, Suat; Douma, Renée A.; Reidinga, Auke C.; Kruif, Martijn D. de; Guiot, Julien; Frix, Anne‐Noëlle; Louis, Renaud; Moutschen, Michel; Lovinfosse, Pierre; Lambin, Philippe

doi:10.1371/journal.pone.0249920

Cited by 17 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consistent with earlier studies conducted in a range of different populations infected with SARS-CoV-2 [ 8 , 9 , 14 – 29 , 71 ], the age gradient for crude mortality was extremely steep and mortality rates increased exponentially while slopes for non-death outcomes plateaued at older ages. Age-related increases in rates of ICU admission and mechanical ventilation among members of our cohort were much less steep than those for mortality and hospitalization.…”

Section: Discussionsupporting

confidence: 87%

Age differences in the association of comorbid burden with adverse outcomes in SARS-CoV-2

et al. 2021

View full text Add to dashboard Cite

Background Older age and comorbid burden are both associated with adverse outcomes in SARS-CoV-2, but it is not known whether the association between comorbid burden and adverse outcomes differs in older and younger adults. Objective To compare the relationship between comorbid burden and adverse outcomes in adults with SARS-CoV-2 of different ages (18–64, 65–79 and ≥ 80 years). Design, setting, and participants Observational longitudinal cohort study of 170,528 patients who tested positive for SARS-CoV-2 in the US Department of Veterans Affairs (VA) Health Care System between 2/28/20 and 12/31/2020 who were followed through 01/31/2021. Measurements Charlson Comorbidity Index (CCI); Incidence of hospitalization, intensive care unit (ICU) admission, mechanical ventilation, and death within 30 days of a positive SARS-CoV-2 test. Results The cumulative 30-day incidence of death was 0.8% in cohort members < 65 years, 7.1% in those aged 65–79 years and 20.6% in those aged ≥80 years. The respective 30-day incidences of hospitalization were 8.2, 21.7 and 29.5%, of ICU admission were 2.7, 8.6, and 11% and of mechanical ventilation were 1, 3.9 and 3.2%. Median CCI (interquartile range) ranged from 0.0 (0.0, 2.0) in the youngest, to 4 (2.0, 7.0) in the oldest age group. The adjusted association of CCI with all outcomes was attenuated at older ages such that the threshold level of CCI above which the risk for each outcome exceeded the reference group (1st quartile) was lower in younger than in older cohort members (p < 0.001 for all age group interactions). Limitations The CCI is calculated based on diagnostic codes, which may not provide an accurate assessment of comorbid burden. Conclusions Age differences in the distribution and prognostic significance of overall comorbid burden could inform clinical management, vaccination prioritization and population health during the pandemic and argue for more work to understand the role of age and comorbidity in shaping the care of hospitalized patients with SARS-CoV-2.

show abstract

Section: Discussionsupporting

confidence: 87%

Age differences in the association of comorbid burden with adverse outcomes in SARS-CoV-2

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In this study we have analyzed the different COVID-19 patient types in Southeastern Spain (n=86867). In contrast to most COVID-19 studies that developed predictive models in the literature that handle less than 5000 patients [17][18][19][20][21] . In addition, we have presented a technique specially designed to treat imbalance problems (IPIP), with which we have developed machine learning models to predict the final state of the patient and the need for hospitalization of those.…”

Section: Discussionmentioning

confidence: 99%

“…Using LASSO and a predictive equation with binary logistic regression based on pre-existing comorbidities and demographic data it was concluded that these variables demonstrated a good ability to discriminate severe from non-serious outcomes using only this historical information with an AUC of 0.76 19 . A further study developed models based on machine learning with different techniques, LASSO, novel univariate and pairwise, but concluded that no model was able to outperform a model based solely on age, where age had an AUC of 0.85 and balanced accuracy of 0.77 20 . Another model was able to predict the risk of hospital/ICU admission and death already at diagnosis with a ROC-AUC of 0.902 by focusing only on a limited number of comorbidities and demographic variables, such as age, sex, and BMI 21 .…”

Section: Introductionmentioning

confidence: 99%

A predictive model for hospitalization and survival to COVID-19 in a retrospective population-based study

Cisterna-García

Guillen-Teruel

Caracena

et al. 2022

Preprint

View full text Add to dashboard Cite

The severe acute respiratory syndrome coronavirus (SARS-CoV-2) causing coronavirus disease 2019 (COVID-19) is highly transmissible and has been responsible for a pandemic associated with a high number of deaths. The clinical management of patients and the optimal use of resources are two important factors in reducing this mortality, especially in scenarios of high incidence. To this end, it is necessary to develop tools that allow early triage of patients with the minimal use of diagnostic tests and based on readily accessible data, such as electronic medical records. This work proposes the use of a machine learning model that allows the prediction of mortality and risk of hospitalization using simple demographic characteristics and comorbidities, using a COVID-19 dataset of 86867 patients. In addition, we developed a new method designed to deal with data imbalance problems. The model was able to predict with high accuracy (89-93%, ROC-AUC = 0.94) the patient's final status (expired/discharged) and with medium accuracy the risk of hospitalization (71-73%, ROC-AUC = 0.75). These models were obtained by assembling and using easily obtainable clinical characteristics (2 demographic characteristics and 19 predictors of comorbidities). The most relevant features of these models were the following patient characteristics: age, sex, number of comorbidities, osteoarthritis, obesity, depression, and renal failure.

show abstract

“…Older age is associated with adverse outcomes in COVID-19 patients [1][2][3]. Frailty, on the other hand, appears to be a predictor for adverse outcomes in hospitalised COVID-19 patients [4][5][6][7].…”

Section: Introductionmentioning

confidence: 97%

Association of Frailty with Adverse Outcomes in Patients with Suspected COVID-19 Infection

Simon

Jauslin

Rueegg

et al. 2021

JCM

View full text Add to dashboard Cite

Older age and frailty are predictors of adverse outcomes in patients with COVID-19. In emergency medicine, patients do not present with the diagnosis, but with suspicion of COVID-19. The aim of this study was to assess the association of frailty and age with death or admission to intensive care in patients with suspected COVID-19. This single-centre prospective cohort study was performed in the Emergency Department of a tertiary care hospital. Patients, 65 years and older, with suspected COVID-19 presenting to the Emergency Department during the first wave of the pandemic were consecutively enrolled. All patients underwent nasopharyngeal SARS-CoV-2 PCR swab tests. Patients with a Clinical Frailty Scale (CFS) > 4, were considered to be frail. Associations between age, gender, frailty, and COVID-19 status with the composite adverse outcome of 30-day-intensive-care-admission and/or 30-day-mortality were tested. In the 372 patients analysed, the median age was 77 years, 154 (41.4%) were women, 44 (11.8%) were COVID-19-positive, and 125 (33.6%) were frail. The worst outcome was seen in frail COVID-19-patients with six (66.7%) adverse outcomes. Frailty (CFS > 4) and COVID-19-positivity were associated with an adverse outcome after adjustment for age and gender (frailty: OR 5.01, CI 2.56–10.17, p < 0.001; COVID-19: OR 3.47, CI 1.48–7.89, p = 0.003). Frailty was strongly associated with adverse outcomes and outperformed age as a predictor in emergency patients with suspected COVID-19.

show abstract

Can predicting COVID-19 mortality in a European cohort using only demographic and comorbidity data surpass age-based prediction: An externally validated study

Cited by 17 publications

References 30 publications

Age differences in the association of comorbid burden with adverse outcomes in SARS-CoV-2

Age differences in the association of comorbid burden with adverse outcomes in SARS-CoV-2

A predictive model for hospitalization and survival to COVID-19 in a retrospective population-based study

Association of Frailty with Adverse Outcomes in Patients with Suspected COVID-19 Infection

Contact Info

Product

Resources

About