ObjectiveThe COVID-19 outbreak was first reported in Wuhan, China, and has been acknowledged as a pandemic due to its rapid spread worldwide. Predicting the trend of COVID-19 is of great significance for its prevention. A comparison between the autoregressive integrated moving average (ARIMA) model and the eXtreme Gradient Boosting (XGBoost) model was conducted to determine which was more accurate for anticipating the occurrence of COVID-19 in the USA.DesignTime-series study.SettingThe USA was the setting for this study.Main outcome measuresThree accuracy metrics, mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE), were applied to evaluate the performance of the two models.ResultsIn our study, for the training set and the validation set, the MAE, RMSE and MAPE of the XGBoost model were less than those of the ARIMA model.ConclusionsThe XGBoost model can help improve prediction of COVID-19 cases in the USA over the ARIMA model.

This study sought to identify the spatial, temporal, and spatiotemporal clusters of COVID-19 cases in 366 cities in mainland China with the highest risks and to explore the possible influencing factors of imported risks and environmental factors on the spatiotemporal aggregation, which would be useful to the design and implementation of critical preventative measures. The retrospective analysis of temporal, spatial, and spatiotemporal clustering of COVID-19 during the period (January 15 to February 25, 2020) was based on Kulldorff’s time-space scanning statistics using the discrete Poisson probability model, and then the logistic regression model was used to evaluate the impact of imported risk and environmental factors on spatiotemporal aggregation. We found that the spatial distribution of COVID-19 cases was nonrandom; the Moran’s I value ranged from 0.017 to 0.453 ( P < 0.001). One most likely cluster and three secondary likely clusters were discovered in spatial cluster analysis. The period from February 2 to February 9, 2020, was identified as the most likely cluster in the temporal cluster analysis. One most likely cluster and seven secondary likely clusters were discovered in spatiotemporal cluster analysis. Imported risk, humidity, and inhalable particulate matter PM 2.5 had a significant impact on temporal and spatial accumulation, and temperature and PM 10 had a low correlation with the spatiotemporal aggregation of COVID-19. The information is useful for health departments to develop a better prevention strategy and potentially increase the effectiveness of public health interventions.

The COVID-19 outbreak emerged in Wuhan, China, and was declared a global pandemic in March 2020. This study aimed to explore the association of daily mean temperature with the daily counts of COVID-19 cases in Beijing, Shanghai, Guangzhou, and Shenzhen, China. Data on daily confirmed cases of COVID-19 and daily mean temperatures were retrieved from the 4 first-tier cities in China. Distributed lag nonlinear models (DLNMs) were used to assess the association between daily mean temperature and the daily cases of COVID-19 during the study period. After controlling for the imported risk index and long-term trends, the distributed lag nonlinear model showed that there were nonlinear and lag relationships. The daily cumulative relative risk decreased for every 1.0 °C change in temperature in Shanghai, Guangzhou, and Shenzhen. However, the cumulative relative risk increased with a daily mean temperature below − 3 °C in Beijing and then decreased. Moreover, the delayed effects of lower temperatures mostly occurred within 6–7 days of exposure. There was a negative correlation between the cumulative relative risk of COVID-19 incidence and temperature, especially when the temperature was higher than − 3 °C. The conclusions from this paper will help government and health regulators in these cities take prevention and protection measures to address the COVID-19 crisis and the possible collapse of the health system in the future.