Proposed Clinical Prediction Rules in the Medical Literature
| Study | Predictor Variables | Outcome Predicted | Validation (Country) | Type of CPR With Outcome |
|---|---|---|---|---|
| Lu, 2020* | Age, CRP | Death | Internal validation (China) | Classification tree: |
| Low: 0% mortality | ||||
| Mod: 6% mortality | ||||
| High: 33% mortality | ||||
| Xie, 2020 | Age, LDH, lymphocytes, SpO2 | Death | External validation in 1 hospital (China) | Probability assessment using full logistic model as nomogram |
| Yan, 2020 | LDH, CRP, lymphocytes | Death | Internal validation (China) | Classification tree |
| Yu, 2020 | Age, sex, diabetes mellitus, lymphocytes, procalcitonin | Death | Internal validation (China) | Risk score |
| High risk (>3 points): 22.8% | ||||
| Low risk (≤3 points): 5.4%) | ||||
| Shi, 2020 | Age, sex, hypertension | Death or severe disease | Internal validation (China) | 0 factors: 0% |
| 1 factor: 6% | ||||
| 2 factors: 19% | ||||
| 3 factors: 40% | ||||
| Galloway, 2020 | Age, sex, race, oxygen saturation, chest radiograph, neutrophils, CRP, albumin, creatinine, diabetes mellitus, hypertension, chronic lung disease | Death or critical care | Internal validation (UK) | Risk score |
| High risk (≥4 points): 40.7% | ||||
| Low risk (<4 points): 12.4% | ||||
| Petrilli, 2020 | Age, SpO2, procalcitonin, troponin, CRP, hypertension | Severe disease | Internal validation (US) | Classification tree |
| Liang, 2020 | Age, neutrophil/lymphocyte ratio, LDH, direct bilirubin, chest radiograph, hemoptysis, dyspnea, unconsciousness, comorbidities, cancer | Severe disease | External validation in 3 hospitals (China) | Logistic regression model requiring online calculator |
| Zhou, 2020 | Temperature, cough, dyspnea, hypertension, chronic liver disease, chronic kidney disease, cardiovascular disease | Severe disease | Internal validation (China) | Nomogram |
| Kaeuffer, 202020 | Age, BMI, sex, dyspnea, neutrophil count, lymphocyte count, CRP | Severe disease | External validation | Risk score: |
| Low risk (≤6): 13% | ||||
| Moderate risk (6–14): NR | ||||
| High risk (>14): 66% | ||||
| Knight, 202021 | Age, sex, number of comorbidities, respiratory rate, SpO2, level of consciousness, urea level, CRP | Mortality | External validation | Risk score from 0 to 21 points |
| Low risk (0–3): 1.2% | ||||
| Intermediate risk (4–8): 9.9% | ||||
| High risk (9–14): 31.4% | ||||
| Very high risk (≥15): 61.5% |
↵* Study included some patients with suspected but not confirmed COVID-19. Preprint at MedRxIV: Lu J, et al. ACP risk grade: a simple mortality index for patients with confirmed or suspected severe acute respiratory syndrome coronavirus 2 disease (COVID-19) during the early stage of outbreak in Wuhan (https://doi.org/10.1101/2020.02.20.20025510).
CPR, clinical prediction rule; CRP, c-reactive protein; LDH, lactate dehydrogenase; BMI, body mass index.