Machine learning prediction for mortality of patients diagnosed with COVID-19: a nationwide Korean cohort study

Abstract

The rapid spread of COVID-19 has resulted in the shortage of medical resources, which necessitates accurate prognosis prediction to triage patients effectively. This study used the nationwide cohort of South Korea to develop a machine learning model to predict prognosis based on sociodemographic and medical information. Of 10,237 COVID-19 patients, 228 (2.2%) died, 7772 (75.9%) recovered, and 2237 (21.9%) were still in isolation or being treated at the last follow-up (April 16, 2020). The Cox proportional hazards regression analysis revealed that age > 70, male sex, moderate or severe disability, the presence of symptoms, nursing home residence, and comorbidities of diabetes mellitus (DM), chronic lung disease, or asthma were significantly associated with increased risk of mortality (p ≤ 0.047). For machine learning, the least absolute shrinkage and selection operator (LASSO), linear support vector machine (SVM), SVM with radial basis function kernel, random forest (RF), and k-nearest neighbors were tested. In prediction of mortality, LASSO and linear SVM demonstrated high sensitivities (90.7% [95% confidence interval: 83.3, 97.3] and 92.0% [85.9, 98.1], respectively) and specificities (91.4% [90.3, 92.5] and 91.8%, [90.7, 92.9], respectively) while maintaining high specificities > 90%, as well as high area under the receiver operating characteristics curves (0.963 [0.946, 0.979] and 0.962 [0.945, 0.979], respectively). The most significant predictors for LASSO included old age and preexisting DM or cancer; for RF they were old age, infection route (cluster infection or infection from personal contact), and underlying hypertension. The proposed prediction model may be helpful for the quick triage of patients without having to wait for the results of additional tests such as laboratory or radiologic studies, during a pandemic when limited medical resources must be wisely allocated without hesitation.

A meta-analysis: The mortality and severity of COVID-19 among patients with mental disorders

“… Death 9 32/734 (4.4) 98/2,817 (3.5) 10/159 (6.3) 12/273 (4.4) 1.15(0.73–1.82) - Severity (10) 9 ( Hirashima et al, 2021 ) Japan Retrospective cohort 47.51 61 2/61 (3.3) - 0/2 16/59 (27.1) - - - 0.53 (0.02-11.57) - Severity (11) 7 ( Yang et al, 2020 ) UK Prospective cohort 67.8±8.12 1,951 442/1,951 (22.7) Depression= 231/442 (52.3) Anxiety= 165/442 (37.3) Stress Related Disorder=10/442 (2.3) Substance Misuse=212/442 (50.0) Psychotic Disorders=19/442 (4.3) 120/442 (27.1) 256/1,509 (17.0) - 68/231 (29.4) 2.03 (1.59−2.59) - Year, sex, race or ethnicity, Townsend deprivation index, educational attainment, annual household income, body-mass index, smoking status, and co-morbidities(12). Death 9 ( Lee et al, 2020a ) South Korea Retrospective cohort - 781 236/781 (30.21%) Depressive Disorder=228/497 (45.9) Anxiety=205/497 (41.2) Bipolar Disorder= 72/497 (14.5) Psychotic Disorder=62/497 (12.5) Schizophrenia=41/497 (8.2) Substance Use Dis Related Diseases=18/497 (3.6) 30/236 (12.7) 37/545 (6.8) - - - 2.00 (1.20-3.32) Age, sex, and index month Death 7 ( An ...…”

“…This analysis provides evidence suggesting that the need of mechanical ventilation, ICU admission and mortality is higher among COVID-19 patients who were diagnosed with mental disorder by the provided odds ratio by 63.8% (OR= 1.76, 95% CI=1.29-2.41). ( An et al, 2020 ; Batty et al, 2021 ; Cavallaro et al, 2020 ; Collaborative, 2021 ; Cummins et al, 2021 ; Fond et al, 2020 ; Giannoglou et al, 2020 ; Hirashima et al, 2021 ; Jeon et al, 2020 ; Lee et al, 2020a ; Lee et al, 2020b ; McKeigue et al, 2020 ; Nemani et al, January 2021 in JAMA psychiatry; Wang et al, 2020 ; Yang et al, 2020 ; Yanover et al, 2020 ) This was also suggested in the fully adjusted as patients with mental disorders had 60.3% higher risk (OR=1.52, 95% CI= 1.20-1.93). ( Cavallaro et al, 2020 ; Fond et al, 2020 ; Jeon et al, 2020 ; Lee et al, 2020b ; Yang et al, 2020 )…”

“…Several observational studies ( An et al, 2020 ; Batty et al, 2021 ; Cavallaro et al, 2020 ; Collaborative, 2021 ; Cummins et al, 2021 ; Fond et al, 2020 ; Giannoglou et al, 2020 ; Hirashima et al, 2021 ; Jeon et al, 2020 ; Lee et al, 2020a ; Lee et al, 2020b ; McKeigue et al, 2020 ; Nemani et al, January 2021 in JAMA psychiatry; Wang et al, 2020 ; Yang et al, 2020 ; Yanover et al, 2020 ) investigated whether people with mental disorders have poorer COVID-19 outcomes or not, thus we have decided to conduct this meta-analysis that aims to investigate the relationship between pre diagnosed psychiatric illness and COVID-19 mortality and severity suggesting that people with mental disorders might have worse outcome of COVID-19.…”

Developing and validating COVID-19 adverse outcome risk prediction models from a bi-national European cohort of 5594 patients

“…In this study, we analyse prognostic and factors associated with disease progression in 3944 SARS-CoV-2 positive patients by constructing an interpretable ML framework. In contrast to previous studies 1 , 2 , these included diagnosed patients outside hospitals, and thus included the entire spectrum of SARS-CoV-2 positive patients in the 2.6 million regional population.…”

Development and validation of a nomogram to predict pulmonary function and the presence of chronic obstructive pulmonary disease in a Korean population

“…The diagnosis claimed by the healthcare providers and the actual diagnosis may differ because the dataset was established for recording claims and reimbursements. Thus, operational definition was used for determining the presence of medical conditions, as previously reported [ 21 ]. For example, hypertension was diagnosed when a patient on antihypertensive medication was admitted for the first time or visited outpatient clinic for a second time with the International Classification of Diseases 10 th revision (ICD-10) codes for hypertensive disease.…”