Spatiotemporal Epidemiology of Tuberculosis in Thailand from 2011 to 2020

Abstract

Tuberculosis is a leading cause of infectious disease globally, especially in developing countries. Better knowledge of spatial and temporal patterns of tuberculosis burden is important for effective control programs as well as informing resource and budget allocation. Studies have demonstrated that TB exhibits highly complex dynamics in both spatial and temporal dimensions at different levels. In Thailand, TB research has been primarily focused on surveys and clinical aspects of the disease burden with little attention on spatiotemporal heterogeneity. This study aimed to describe temporal trends and spatial patterns of TB incidence and mortality in Thailand from 2011 to 2020. Monthly TB case and death notification data were aggregated at the provincial level. Age-standardized incidence and mortality were calculated; time series and global and local clustering analyses were performed for the whole country. There was an overall decreasing trend with seasonal peaks in the winter. There was spatial heterogeneity with disease clusters in many regions, especially along international borders, suggesting that population movement and socioeconomic variables might affect the spatiotemporal distribution in Thailand. Understanding the space-time distribution of TB is useful for planning targeted disease control program activities. This is particularly important in low- and middle-income countries including Thailand to help prioritize allocation of limited resources.

Spatial patterns of tuberculosis in Aceh Province during the COVID-19 pandemic: a geospatial autocorrelation assessment

“…indicate that the TB cases in Aceh Province were geographically dispersed at the districts level. Our findings contradicted with previous studies in Thailand [23] and China [24] where the TB incidence were geographically clustered due to the TB incidence in both a particular spatial unit and its neighbouring unit either surpassing or falling below the mean. While in our study, the TB incidence in a particular spatial unit and its neighbouring unit diverged on the opposite side of the mean.…”

“…Although Langsa is not the capital city of Aceh Province, it is considered to be one of the major cities in Aceh Province with better healthcare facilities. A study in Thailand found a spatial pattern of TB incidence in Thai borderland and seemingly related to the migrants' behaviour from Laos to find a better healthcare facility in Thai borderland [23]. However, further investigation on the association between spatial pattern of TB case distribution and mobility behaviour of the population is highly suggested to understand the spatial pattern of TB incidence in borderland of Aceh Province.…”

“…The global spatial autocorrelation used to estimate the overall spatial dependency of the TB case. In this study, we used Global Moran's I which already been used widely to estimate the spatial autocorrelation of TB cases [23][24][25]. Global Moran's I have a value range between -1 to 1 where 0 indicates the TB cases are randomly scattered while positive and negative value indicate the observation are spatially clustered and dispersed, respectively [26,27].…”