Prediction models for COVID-19 disease outcomes

Abstract

SARS-CoV-2 has caused over 6.9 million deaths and continues to produce lasting health consequences. COVID-19 manifests broadly from no symptoms to death. In a retrospective cross-sectional study, we developed personalized risk assessment models that predict clinical outcomes for individuals with COVID-19 and inform targeted interventions. We sequenced viruses from SARS-CoV-2-positive nasopharyngeal swab samples between July 2020 and July 2022 from 4450 individuals in Missouri and retrieved associated disease courses, clinical history, and urban-rural classification. We integrated this data to develop machine learning-based predictive models to predict hospitalization, ICU admission, and long COVID.The mean age was 38.3 years (standard deviation = 21.4) with 55.2% (N = 2453) females and 44.8% (N = 1994) males (not reported, N = 4). Our analyses revealed a comprehensive set of predictors for each outcome, encompassing human, environment, and virus genome-wide genetic markers. Immunosuppression, cardiovascular disease, older age, cardiac, gastrointestinal, and constitutional symptoms, rural residence, and specific amino acid substitutions were associated with hospitalization. ICU admission was associated with acute respiratory distress syndrome, ventilation, bacterial co-infection, rural residence, and non-wild type SARS-CoV-2 variants. Finally, long COVID was associated with hospital admission, ventilation, and female sex.Overall, we developed risk assessment models that offer the capability to identify patients with COVID-19 necessitating enhanced monitoring or early interventions. Of importance, we demonstrate the value of including key elements of virus, host, and environmental factors to predict patient outcomes, serving as a valuable platform in the field of personalized medicine with the potential for adaptation to other infectious diseases.

Keywords: COVID-19 prediction; Long COVID; disease outcome prediction; machine learning; personalized medicine; predictive model for COVID-19.

Graphical abstract

Figure 1.

Features used for model training and selection. The model inputs included host, virus, and environment data. Host features were collected from electronic health records, virus features were collected using whole genome sequencing, and environmental features included the individual’s urban-rural classification based on their home ZIP Code. The model outcomes included hospital admission, ICU admission, long COVID, evaluated separately.

Figure 2.

Statistical Analyses of Predictors. Evaluation of statistical associations between the selected predictive features and (A) hospitalization, (B) ICU admission, and (C) long COVID. Each dot represents the odds ratio, and each line represents the 95% confidence interval (CI) for each feature. Statistically significant associations are highlighted in red and defined as a 95% CI that does not overlap 1 (illustrated by the vertical dotted line). The left panel illustrates unadjusted odds ratios, and the right panel illustrates odds ratios adjusted for age, sex, and COVID-19 vaccination. ARDS, acute respiratory distress syndrome.

Figure 3.

Significant predictors associated with disease outcomes. Features selected from the best-performing models for each outcome (hospitalization, ICU admission, and long COVID) were further analyzed using conventional logistic regression, and significant features were shown. Significant predictive features of multiple outcomes are shown between the respective outcome panels. ARDS, acute respiratory distress syndrome; Symptoms listed under hospitalization included gastrointestinal, constitutional, and cardiac symptoms.