Application of a clinical decision support system to assess the severity of the new coronavirus infection COVID-19. (14th October 2021)

• Record Type: Journal Article
• Authors: Gavrilov, D; Kuznetsova, T; Gusev, A; Korsakov, N; Novitskiy, R
• Is Part Of: European heart journal. Volume 42(2021)Supplement 1
• Publication Date: 2021-10-14
• Subjects: Artificial Intelligence (Machine Learning, Deep Learning); Cardiology -- Periodicals; Heart -- Diseases -- Periodicals; 616.12005
• DOI: 10.1093/eurheartj/ehab724.3054
• Languages: English
• ISSNs: 0195-668X

Abstract

Abstract: Purpose: To apply the clinical decision support system (CDSS) and evaluate its effectiveness in determining the prognosis of the new coronavirus infection COVID-19. Methods: 7118 outpatient and hospitalized cases with COVID-19 were analyzed, mean age 39.4±18.3 years, 52% men. The data was accumulated in the the Webiomed.DataSet service, which allows to accumulate a base of de-identified biomedical data from electronic health records. To test the severity of the COVID-19, the CDSS was connected to the 8 medical information systems in one region of the Russian Federation. For each risk factor (RF) of the unfavorable COVID-19 outcome the contribution to the risk was determined in points indicated in brackets: age over 60 (1), age over 80 (3), BMI of 30–34.9 kg/m 2 (1), BMI of ≥35 kg/m 2 (2), arterial hypertension (HTN) (1), diabetes mellitus (DM) (1), coronary artery disease (CAD) (2), cerebrovascular accident (CVA) (1), atrial fibrillation (AF) (1), pulmonary disease (1), cancer (1). Hospitalization and death were considered as unfavorable outcome. Each patient had risk level (high – two or more points, moderate – one, low – zero). Results: 64.2% was outpatient, age 34.8±17.3 years. 35.8% was hospitalized (mean age 47.8±15.1 y), 50 patients died (mean age 61.3±14.4 y, mortality 0.7%). Low risk had 74.9% outpatient treated patients, 26.4% – hospitalized, 26% – dead; average risk – 12.6%, 17.3%, 24%, high risk – 12.5%, 56.3%, 50% (respectively for subgroups). The RF incidence of poor prognosis in the groups: age over 60 years old – 9%, over 80 – 0.5%, HTN – 18.9%, DM – 5.2%, CAD – 3.9%, CVA – 1%, AF – 1.4%, COPD/asthma – 1.9%, cancer – 1.7%, obesity – 15.5%. In the hospitalized group: age over 60 years – 11.1%, over 80 – 1.6%, HTN – 13.3%, DM – 4.2%, CAD – 7.6%, CVA – 1.1%, AF – 1%, COPD/asthma – 1.3%, cancer – 1.3%, obesity – 13.2%. Among patient who died: age over 60 years – 54%, over 80 – 6%, AH – 50%, DM – 18%, CAD – 36%, CVA – 4%, AF – 12%, COPD/asthma – 6%, cancer – 4%, obesity – 30%. When comparing the incidence of RF in the high-risk group, a significant difference in hospitalized, dead, and patients treated outpatient was obtained for the following RF: age over 60 years (p<0.001), HTN (p<0.001), DM (0.004), CAD (p<0.001), AF (p<0.001), COPD and AD (p=0.043), obesity (p=0.031). In the moderate-risk group, the main RFs influencing the prognosis were age over 60 years (p<0.001), HTN (p=0.03) and obesity (p=0.004). Conclusions: The created CDSS allowed to stratify the risk of COVID-19 by the presence of cardiovascular risk factors and diseases, as well as by the presence of bronchopulmonary pathology and oncological diseases. The use of this CDSS allowed to route COVID-19 patients more effective. In addition to clinical criteria of the disease severity, the system allows to assess the prognosis quickly and hospitalize high-risk patients, or organize their careful monitoring in case of outpatient treatment. Funding Acknowledgement: Type of funding sources: None.