Individualizing Risk Prediction for Positive Coronavirus Disease 2019 Testing

Abstract

BACKGROUND: Coronavirus disease 2019 (COVID-19) is sweeping the globe. Despite multiple case-series, actionable knowledge to tailor decision-making proactively is missing. RESEARCH QUESTION: Can a statistical model accurately predict infection with COVID-19? STUDY DESIGN AND METHODS: We developed a prospective registry of all patients tested for COVID-19 in Cleveland Clinic to create individualized risk prediction models. We focus here on the likelihood of a positive nasal or oropharyngeal COVID-19 test. A least absolute shrinkage and selection operator logistic regression algorithm was constructed that removed variables that were not contributing to the model's cross-validated concordance index. After external validation in a temporally and geographically distinct cohort, the statistical prediction model was illustrated as a nomogram and deployed in an online risk calculator. RESULTS: In the development cohort, 11,672 patients fulfilled study criteria, including 818 patients (7.0%) who tested positive for COVID-19; in the validation cohort, 2295 patients fulfilled criteria, including 290 patients who tested positive for COVID-19. Male, African American, older patients, and those with known COVID-19 exposure were at higher risk of being positive for COVID-19. Risk was reduced in those who had pneumococcal polysaccharide or influenza vaccine or who were on melatonin, paroxetine, or carvedilol. Our model had favorable discrimination (c-statistic ¼ 0.863 in the development cohort and 0.840 in the validation cohort) and calibration. We present sensitivity, specificity, negative predictive value, and positive predictive value at different prediction cutoff points. The calculator is freely available at https://riskcalc.org/COVID19. INTERPRETATION: Prediction of a COVID-19 positive test is possible and could help direct health-care resources. We demonstrate relevance of age, race, sex, and socioeconomic characteristics in COVID-19 susceptibility and suggest a potential modifying role of certain common vaccinations and drugs that have been identified in drug-repurposing studies.

Factors Associated With Death in Critically Ill Patients With Coronavirus Disease 2019 in the US

“…For example, other than data on homelessness (0.6% prevalence in our cohort), we did not collect data on the socioeconomic status of the patients. Socioeconomic status is increasingly recognized as an important factor associated with health outcomes in patients with COVID-19 25 and could have influenced our findings with respect to variation in mortality across hospitals. We also did not collect detailed data on ventilator management strategies, hospital or ICU patient volume, or physician and nurse availability.…”

The association between influenza vaccination and the risk of SARS-CoV-2 infection, severe illness, and death: a systematic review of the literature

“…Jehi et al [9] Subjects tested in several clinics in Ohio, USA (n = 11,672) (a) RT-PCR COVID-19 cases were significantly less likely to be vaccinated than controls (p < 0.001) None Subjects tested in several clinics in Florida, USA (n = 2,295) (a) COVID-19 cases were significantly less likely to be vaccinated than controls (p = 0.011)…”

“…Finally, twelve independent articles met all inclusion criteria and were retained ( Figure 1). Seven articles [6][7][8][9][10][11][12] focused on the association between influenza vaccination and the risk of SARS-CoV-2 infection (Table 1): these encompassed a total of 242,323 subjects, of which 56.6% were contributed by Pawlowski et al [11] and 32.6% by Vila-Córcoles A et al [12]. Most studies were based on subjects from the general population, except the two smallest studies, which included 203 firefighters and paramedics from the USA [7], and 640 liver transplant patients from Italy [8].…”

Impact of the COVID-19 pandemic on healthcare workers risk of infection and outcomes in a large, integrated health system.

“…For the outcome of being test positive for COVID-19, the propensity score logistic regression model included covariates that were found to be associated with a positive COVID-19 test outcome in our previous work 23. For the outcomes of hospital and intensive care unit (ICU) admission of COVID-19 test positive patients, the propensity score covariates are those that…”