BACKGROUND: COVID-19 and influenza A can cause severe respiratory illness. Differentiating between the two diseases and identifying critically ill patients in times of epidemics become a challenge for frontline medical staff. We sought to investigate whether both diseases and their severity could be recognized by routine blood parameters. METHODS: Our retrospective study analysed the clinical data and first-time routine blood parameters of 80 influenza A patients and 123 COVID-19 patients. COVID-19 patients were divided into three groups according to treatment modalities and outcomes: outpatient group, inpatient without invasive mechanical ventilation (IMV) group, and inpatient with IMV group. We used the Mann-Whitney and Kruskal-Wallis tests to analyze the differences in routine blood parameters between the two or three groups. Receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) were used to assess the diagnostic accuracy. RESULTS: Compared with outpatient influenza A patients, outpatient COVID-19 patients had a higher neutrophil to lymphocyte ratio (NLR) (6.63 vs 3.55). ROC analysis showed that the NLR had a high diagnostic value for differentiating COVID-19 from influenza A (AUC = 0.739). The best cut-off point of the NLR was 6.48, the diagnostic sensitivity was 0.523, and the specificity was 0.925. The median platelet (PLT) count in the different COVID-19 groups was as follows: outpatient group (189×109/L), inpatient without IMV group (161×109/L), and inpatient with IMV group (94×109/L). Multivariate logistic regression analysis found a significant association between PLT and treatment modality and outcome in COVID-19 patients (p<0.001). CONCLUSIONS: NLR can be used as a potential biological indicator to distinguish COVID-19 and influenza A. Decreased PLT predicts the critical condition of COVID-19 patients and helps stratify the treatment of COVID-19 patients.