A Spatio-temporal Bayesian model to estimate risk and influencing factors related to tuberculosis in Chongqing, China, 2014–2020

In this study, a Bayesian spatio-temporal model was developed to analyze the risk of TB incidence in Chongqing from 2014 to 2020, to explore the spatio-temporal incidence patterns of TB in different populations, and to identify potential correlates of TB incidence.

Chongqing reported a total of 170,934 TB cases from 2014 to 2020, with the morbidity per 100,000 people decreasing from 82.81 in 2014 to 70.19 in 2020, but higher than the national average in all years [4]. The elderly population aged over 60 years old was a key population with a high incidence of TB, and the risk of TB incidence was on the rise in the elderly population in southeast Chongqing. This might be related to the lower literacy level and poor health literacy of the elderly, especially in economically backward areas such as southeastern Chongqing, where with the exodus of young and middle-aged groups, aging and empty-nesting were becoming more serious, and the local elderly left behind had even less health awareness and lacked of knowledge about TB prevention and treatment [29]. In this regard, it is recommended that primary healthcare institutions should popularize knowledge about TB disease among the elderly, provide free consultations to empty nesters regularly, and implement early detection and treatment of TB in this group. In addition, there was an increasing trend in the risk of TB incidence in main urban areas among those aged 30 to 44 years. This might be due to the dramatic increase in youth mobility in main urban areas in recent years [29], but the corresponding housing conditions and medical resources were relatively poor. Therefore, it is recommended that the government improves the living environment of the urban population, especially the mobile population, and improves the primary health service and medical security system to reduce the burden of TB disease in this population.

The study found a positive effect of SO2 on the development of TB, which was consistent with the study in Hubei, China, and this might be related to the fact that SO2 caused death of alveolar macrophages and reduced the ability of the immune system to clear Mycobacterium TB [11]. In China, SO2 emissions from coal combustion were a major problem, especially in winter when heating increases SO2 emissions [30]. In Chongqing, due to the mountainous terrain, the low wind speed in winter made it difficult for air pollutants to escape, which aggravates the impact of SO2 emissions on the development of TB. This study also found that there was a clear seasonal trend in the number of TB cases, mainly concentrated in January–March each year, which might be related to the increased SO2 concentration in winter. Therefore, in order to control air pollution at the source and reduce the possibility of TB incidence, it is recommended to vigorously promote “coal to gas” or “coal to electricity” in all districts and counties, especially in rural areas, or to use sulfur-free coal to reduce SO2 emissions [30].

High UR was also found in the study to have a significant negative effect on the increased risk of TB incidence in the whole population. This might be attributed to areas with high UR had more knowledge of TB control, good health awareness, and better health care for their residents. However, further subpopulation analysis found a significant positive effect of high UR on TB incidence in the female population. Analysis of its possible causes: First, women in areas with high UR have higher employment rates [31], and employment might increase their chances of contracting Mycobacterium TB, especially in main urban areas with high mobile populations. Our findings also showed that women had a higher risk of TB incidence in main urban areas. Secondly, environmental pollution, traffic congestion and deteriorating living conditions brought about by urbanization may also accelerate the infection and spread of Mycobacterium TB. In addition, high UR increased the risk of morbidity in people aged 0–29 years. Students accounted for a large portion of the population, and the concentration of classes and housing on campus makes it extremely easy for the spread of Mycobacterium TB and the spread of infection. At the same time, the increased work pressure and work style changes brought by urbanization might bring more physical and psychological stress to the youth population [32], and a chronic state of physical suboptimal health, which may increase the risk of TB. In this regard, the government should emphasize the coordination between the speed of development and the quality of people’s living standards, optimize the spatial layout of cities and towns, improve the living and working environment of residents, and realize healthy urbanization.

In the Global TB Report 2021, it was suggested that poverty was a determinant of the number of people infected with TB disease [2, 33]. Similar results were obtained in the present study, where we found a significant positive effect of the increase in LINA on the increase in TB risk. Some studies have shown that people in poor areas had lower access to health services and were more likely to have poor nutrition and poor behavioral habits such as smoking and alcohol consumption [34], which increased the risk of TB infection. Therefore, it is suggested that the government should increase the medical protection and assistance for people in poor areas, and strengthen the improvement of their diet and nutrition and the guidance of exercise, so as to promote the development of healthy habits and enhance the physical quality of such people. Subgroup analysis further revealed that the female population and the older age group over 60 were more affected by poverty. In resource-limited conditions, elderly and female populations may be more vulnerable, especially due to folk customs and patriarchal beliefs, and they may have poorer health status and lower access to health services [35, 36], thus increasing the risk of TB incidence.

We had several strengths in our research. First, the Bayesian spatio-temporal model used in this paper took the spatio-temporal structure as a priori, considered the influence of adjacent temporal and spatial structures, realized the simultaneous analysis of temporal, spatial, and spatio-temporal interactions and potential influencing factors of TB, and considered more uncertainties, so that the risk of TB incidence could be estimated more accurately, and the pattern of spatio-temporal variation of its incidence and its influencing factors could be analyzed. Second, differences in the spatial and temporal distribution of TB risk in different populations were found in the results of this study. The female population and the young people aged 0–29 years were at higher risk of TB in the main urban area, and there was an increasing risk for people aged 30–44 years in the main urban area and for people over 60 years old in the southeast of Chongqing. These results provided additional evidence and insight to specifically analyze the distribution patterns of TB incidence in different populations. Third, the results of this paper confirmed the effects of air pollution factors and socio-economic factors on the incidence of TB, and further explored the differences in the effects of these factors on different populations. The increase in UR had a significant positive effect on the increased risk of TB in the female population and the youth population, while it had a negative effect on the rest of the population. The LINA per 1000 persons had a greater impact on the risk of TB in the female population and in the population aged 60 years or older than in other populations.

There were also some limitations in our analysis. First, although several environmental and socio-economic variables were included in our study, there were a number of potential influences on TB incidence that were not included, such as smoking rates, alcohol consumption rates, etc. Second, the TB cases reported in this study were from the TB surveillance system, which might have been influenced by the lack of reporting and underdiagnosis in some poor counties, leading to an underestimation of TB incidence. Finally, the analysis level of this study was divided according to districts (counties). We can consider using smaller geographical units, such as streets and towns to explore the districts with a high risk of TB incidence like Southeast Chongqing and Northeast Chongqing in a future study.

留言 (0)

沒有登入
gif