This study analysed the spatial distribution and clustering patterns of comorbidities in hospitalised pulmonary tuberculosis (PTB) patients in Guangdong Province and evaluated associations with economic burden. The work used large-scale inpatient data to characterise comorbidity spectra, detect geographic clusters for key PTB comorbidity pairs, and quantify how comorbidity burden and spatial clustering relate to direct and indirect costs.
Investigators used the Hospital Discharge Records Database of Guangdong Province covering the period 2016–2024. The analytic sample comprised 646,114 hospitalised PTB patients identified in that database. The authors report that 66.42% of included PTB patients had at least one comorbidity.
To examine spatial patterns, the study applied Global Moran’s I and local indicators of spatial association (LISA) to identify clustering of major PTB comorbidity pairs across cities in Guangdong. To evaluate economic consequences, generalised linear models were fitted to estimate associations between the number of comorbidities, spatial clustering patterns, and measures of economic burden. Both direct medical costs and indirect costs (productivity loss) were considered in the assessment of economic burden.
Among hospitalised PTB patients in the dataset, two-thirds had at least one comorbidity. The three most prevalent comorbid conditions reported were hypertension (HTN), diabetes mellitus (DM), and chronic obstructive pulmonary disease (COPD). The study describes temporal trends in the comorbidity spectrum over the 2016–2024 study period, although specific year-by-year prevalence figures beyond the overall proportions were not reported in the abstract.
Spatial autocorrelation analyses revealed distinct geographic patterns for different comorbidity pairs:
PTB–HTN: High-high clusters (areas where both PTB and HTN comorbidity rates were high) were mainly concentrated in the central Pearl River Delta.
PTB–DM: High-high clusters were concentrated in eastern Guangdong, while low-low clusters (areas with relatively low comorbidity rates) were found in western Guangdong.
PTB–COPD: High-high clusters were primarily located in eastern and northern Guangdong; low-low clusters remained relatively stable in several cities of the Pearl River Delta region.
These findings indicate heterogeneous spatial distribution of comorbidities among hospitalised PTB patients across Guangdong.
Using generalised linear models, the authors found that the number of comorbidities was significantly positively associated with both direct and indirect economic burden (P < 0.001). Economic burden increased progressively as the number of comorbidities rose. The study therefore links higher comorbidity counts with greater medical costs and productivity losses among hospitalised PTB patients.
Beyond comorbidity count, the effect of spatial clustering patterns on economic burden showed heterogeneity across different comorbidity types. In other words, the economic impact associated with comorbid conditions varied by geographic cluster and by the specific comorbidity pair examined. The abstract indicates that spatial context modifies economic outcomes but does not provide granular effect estimates for each comorbidity–cluster combination.
The authors conclude that the burden of comorbidities among hospitalised PTB patients in Guangdong is substantial and displays meaningful spatial clustering. They recommend that regional epidemiological characteristics of comorbidities be incorporated into PTB comorbidity management and into optimisation of health resource allocation. Prevention and control strategies should account for both direct medical expenditures and productivity losses to reduce the broader socioeconomic burden associated with PTB.
The abstract summarises key methods and principal findings, but several details were not reported in the abstract and therefore cannot be asserted here. Unreported items in the source abstract include: precise year-by-year comorbidity prevalence figures, definitions or ICD coding used to identify comorbidities, thresholds or geographic units used for spatial clustering maps, numeric effect estimates from the generalized linear models, model covariates, and magnitudes of direct versus indirect cost components. The full article would be required to obtain these specifics.
In a large inpatient dataset covering 2016–2024, two-thirds of hospitalised PTB patients in Guangdong had at least one comorbidity, with hypertension, diabetes mellitus, and COPD most common. Distinct spatial clustering patterns were identified for major comorbidity pairs across Guangdong, and both the number of comorbidities and their spatial clustering were associated with higher direct and indirect economic burden. The study supports regionally tailored comorbidity management and resource allocation, and highlights the importance of including productivity losses alongside medical costs when assessing the socioeconomic impact of PTB.