Influential factors and spatial–temporal distribution of tuberculosis in mainland China

Abstract

Tuberculosis (TB) is an infectious disease that threatens human safety. Mainland China is an area with a high incidence of tuberculosis, and the task of tuberculosis prevention and treatment is arduous. This paper aims to study the impact of seven influencing factors and spatial–temporal distribution of the relative risk (RR) of tuberculosis in mainland China using the spatial–temporal distribution model and INLA algorithm. The relative risks and confidence intervals (CI) corresponding to average relative humidity, monthly average precipitation, monthly average sunshine duration and monthly per capita GDP were 1.018 (95% CI 1.001–1.034), 1.014 (95% CI 1.006–1.023), 1.026 (95% CI 1.014–1.039) and 1.025 (95% CI 1.011–1.040). The relative risk for average temperature and pressure were 0.956 (95% CI 0.942–0.969) and 0.767 (95% CI 0.664–0.875). Spatially, the two provinces with the highest relative risks are Xinjiang and Guizhou, and the remaining provinces with higher relative risks were mostly concentrated in the Northwest and South China regions. Temporally, the relative risk decreased year by year from 2013 to 2015. It was higher from February to May each year and was most significant in March. It decreased from June to December. Average relative humidity, monthly average precipitation, monthly average sunshine duration and monthly per capita GDP had positive effects on the relative risk of tuberculosis. The average temperature and pressure had negative effects. The average wind speed had no significant effect. Mainland China should adapt measures to local conditions and develop tuberculosis prevention and control strategies based on the characteristics of different regions and time.

Identifying spatiotemporal patterns of COVID-19 transmissions and the drivers of the patterns in Toronto: a Bayesian hierarchical spatiotemporal modelling

“…While evaluating the trends of the spatiotemporal relative risks 65 we observed that only three neighbourhoods (Neighbourhoods # 40,137 and 49) had a decreasing trend of the relative risk during the study period (Fig. 4 b).…”

“…Figure 3 b presents the posterior mean and 95% uncertainty band of the temporal relative risk (( 65 during the study period, which shows that the highest risk (RR > 9) was observed between March 14, 2021, and April 17, 2021. A total of 46 (51%) weeks out of the total 89 weeks experienced a relative risk of less than one during our study period.…”

“…A map of the posterior means of the spatial relative risk was created using the ArcGIS Desktop software 10.8.1 46 . The temporal relative risk ( 65 at week t represents the average COVID-19 incidence rate for all neighbourhoods in week i compared to the average COVID-19 incidence rate of the entire study period. The posterior mean of the temporal relative risk with its corresponding 95% credible intervals was plotted using RStudio Software 60 version 2021.09.0.…”

“…The posterior mean of the temporal relative risk with its corresponding 95% credible intervals was plotted using RStudio Software 60 version 2021.09.0. The spatiotemporal effect term represents a change that cannot be reflected by spatial and temporal effects only 65 . The spatiotemporal relative risk 65 represent the risk of COVID incidence rate in neighbourhood i and in time t compared to the overall incidence rate in entire study area and entire time period.…”

“…The spatiotemporal effect term represents a change that cannot be reflected by spatial and temporal effects only 65 . The spatiotemporal relative risk 65 represent the risk of COVID incidence rate in neighbourhood i and in time t compared to the overall incidence rate in entire study area and entire time period.…”