Objective To describe the the temporal changing trend and spatiotemporal distribution of the active screening detection rate of pulmonary tuberculosis (TB) in diabetes mellitus patients in China from 2020 to 2023, to identify the clustering time and area of pulmonary TB cases in the active screening in diabetes mellitus patients, and to provide reference for the active screening, prevention and control of pulmonary TB in the diabetes mellitus patients in the future.

Methods Active screening data of pulmonary TB in diabetes mellitus patients in China from 2020 to 2023 were collected. Software Joinpoint 5.4.0 was used to conduct temporal trend analysis of the detection rate of pulmonary TB in diabetes mellitus patients, the package spdep of software R 4.3.3 was used to conduct spatial auto-correlation analysis and software SaTScan 10.3.2 were used to conduct spatiotemporal scanning.

Results A total of 293 856 968 visits of pulmonary TB symptoms detections in diabetes mellitus patients were conducted from 2020 to 2023, in which 58 743 diabetes mellitus patients were diagnosed with pulmonary TB, with a rate of 19.99/100 000. The detection rate of pulmonary TB in diabetes mellitus patients decreased from 36.25/100 000 to 12.36/100 000 from 2020 to 2023, with an average annual change percentage of −31.56% (95% CI: −22.62% to −40.01%), showed a significant downward trend (P<0.001). The thematic map showed that the detection rate of pulmonary TB in diabetes mellitus patients was higher in northwestern and northeastern while lower in southeastern China. Global spatial auto-correlation analysis revealed the detection rate of pulmonary TB in diabetes mellitus patients was not randomly distributed in China in 2023, showing a significant positive spatial auto-correlation (Moran's I=0.24, Z=3.028, P=0.001). Local spatial auto-correlation analysis showed that pulmonary TB cases in diabetes mellitus patients were distributed in the high-high clustering areas in northwestern and northeastern China (Qinghai Province, Xizang Autonomous Region, Xinjiang Uygur Autonomous Region, and Liaoning Province). Spatiotemporal scanning analysis detected four spatiotemporal clustering areas in China. The primary clustering area was mainly in Hunan Province, involving a total of 10 provinces, from 2020 to 2021 relative risk (RR)=3.40, log-likelihood ratio (LLR)= 8 338.08, P<0.001. There are two secondary clustering areas, one was in Liaoning Province and Jilin Province from 2020 to 2021 (RR=8.42, LLR=2 870.75, P<0.001), another one was in Xinjiang Uygur Autonomous Region and Xizang Autonomous Region from 2020 to 2021 (RR=7.82, LLR=2 112.43, P<0.001). The third clustering area was in Shandong Province in 2020 (RR=1.32, LLR=29.44, P<0.001).

Conclusion The detection rate of pulmonary TB in active screening in diabetes mellitus patients in China decreased year by year. The detection rate of pulmonary TB in diabetes mellitus patients was relatively high in northwestern and northeastern China, and there was a high level clustering. The detection rate of pulmonary TB in diabetes mellitus patients in southeastern China was not high, but similar to the surrounding areas. In the future, efforts should be focused on strengthening health education and screening programs for pulmonary TB in diabetes mellitus patients in areas with high incidence of pulmonary TB.