Epidemiology and Treatment Outcomes of Pulmonary Tuberculosis in Dazu District, Chongqing, China, 2005-2024: Surveillance Study

Introduction

Pulmonary tuberculosis (PTB), caused by Mycobacterium tuberculosis (MTB), is a chronic infection disease primarily affecting the lungs and represents the most common form of TB. As an airborne pathogen, MTB poses a serious global public health threat [1]. Globally, TB remains the leading cause of death from a single infectious agent [2]. According to the World Health Organization (WHO) Global Tuberculosis Report 2024 [3], there were approximately 10.8 million new TB cases and 1.25 million deaths in 2023. In response, the WHO and the United Nations have set a goal to reduce TB deaths by 95% and new cases by 90% by 2035. China bears one of the highest TB burdens globally, reporting 741,000 new cases in 2023—an estimated incidence of 52 per 100,000 population. It ranks third among the 30 high-TB burden countries, accounting for 6.8% of the global cases. An estimated 25,000 TB-related deaths occurred, corresponding to a mortality rate of 2 per 100,000. The treatment success rate for drug-sensitive TB is approximately 88%. These figures indicate that China faces significant challenges in meeting the WHO’s goal to end the TB epidemic by 2035.

The prevalence of TB exhibits notable spatial heterogeneity and demographic disparities. In China, regions with higher TB burdens are often associated with factors such as economic development, health care accessibility, and population mobility. Previous studies have indicated higher incidence rates in western regions and rural areas, with farmers, older adults, and male individuals being disproportionately affected [4-7]. As a municipality in western China with both extensive urban and rural populations, Chongqing plays a critical role in regional TB control efforts. However, compared with national or provincial macro-level data, long-term dynamic monitoring studies at the district and county level remain relatively scarce. Although several studies at the urban and county scale in Chongqing have been published [8-11], they are often limited by short time spans, small sample sizes, and insufficient depth.

District- and county-level dynamic surveillance of TB is critical for elucidating local transmission dynamics and evaluating the effectiveness of disease control interventions. In this study, we leverage 2 decades (2005‐2024) of surveillance data from Dazu District, Chongqing, to track temporal trends with Joinpoint regression, map spatial hotspots through geospatial techniques, and identify key epidemiological and treatment-related factors. This study aims to provide a scientific basis for optimizing local TB control strategies and advancing the goal of TB elimination by elucidating the epidemiological characteristics and temporal dynamics of PTB at the district and county levels.

Methods

Data Sources

Patient data of PTB cases from Dazu District, Chongqing, with confirmed diagnosis dates between January 1, 2005, and December 31, 2024 were collected from the Tuberculosis Management Information System (TBIMS) subsystem of the China Information System for Disease Control and Prevention (retrieved 2025-03-28). Data fields included current address, gender, age, registration date, bacteriological results, population category, key populations, symptom onset date, initial consultation date, complications, diagnosis confirmation date, treatment initiation date, and reasons for treatment termination. The exclusion criteria comprised duplicate reports, errors in basic infectious disease report card information, missing essential data, diagnostic modifications, cases initially managed outside Dazu District, extrapulmonary TB, untreated cases, and nontuberculous mycobacterial infections.

Case management data were retrieved from the TBIMS, with initial registration management units being the Dazu District Tuberculosis Prevention and Control Center or Dazu District People’s Hospital of Chongqing (designated medical institution). The TBIMS was launched in 2005 using paper-based reporting and manual entry and was upgraded to a web-based real-time electronic system in 2010, greatly reducing latency. It covers all TB control institutions across 31 provinces, over 300 prefectures, and 3000 counties in mainland China, mandating reporting for all bacteriologically confirmed or clinically diagnosed TB cases. Now the world’s largest TB registration system, the TBIMS, is recommended by the WHO as a model for global TB surveillance.

This study analyzed incidence characteristics (temporal, geographical, and demographic distributions) and investigated influencing factors of treatment outcomes. Annual notification rates used end-of-year permanent-resident denominators from the Dazu District Statistical Yearbooks published by Dazu District, Chongqing, during 2005-2024 (retrieved 2025-03-31). Age-standardized incidence rates were calculated using the 2020 Seventh National Population Census data (National Bureau of Statistics) as the standard population. The names of the 27 townships or streets involved in this study are all standardized administrative division units, which do not reflect the differences in special geographical or social attributes. Its division is based on the “Chongqing Urban and County Administrative Division Code” (GB/T 2260), which only reflects the administrative management level.

Research Methods

Diagnosis and Treatment of PTB

Patients were diagnosed according to the 《WS 288‐2008 Pulmonary Tuberculosis Diagnostic Criteria》, 《WS 288‐2017 Pulmonary Tuberculosis Diagnosis 》 [12,13], and 《WS 196‐2017 Tuberculosis Classification》 [14] standards. All TB treatments were implemented in accordance with the 《Technical Guidelines for Tuberculosis Prevention and Control in China (2021 Edition)》 [15]. To maintain data comparability across the study period, TB cases in this research included tuberculous pleuritis cases throughout all calendar years.

Definitions

Definitions are as follows: (1) case detection methods: referral, active screening, direct consultation or referral, contact tracing, health examination, and others (cases with undetermined or unclassifiable sources). (2) Discovery delay: the total time interval between a patient’s initial clinical symptom and confirmed TB diagnosis, encompassing both care-seeking and diagnostic confirmation delays. This study employed a 28-day threshold as the criterion for determining discovery delay. (3) Bacteriological results: from 2005 to 2018, smear-positive or culture-positive cases were classified as bacteriologically positive, while cases with negative results in both smear microscopy and culture were classified as bacteriologically negative. From 2019 to 2024, cases with any positive result in smear microscopy, culture, or molecular biological testing were classified as bacteriologically positive, whereas cases with negative results in all 3 methods were classified as bacteriologically negative. Patients with tuberculous pleurisy were uniformly considered bacteriologically negative. (4) Number of patients with PTB: it refers to cases registered and reported between 2005 and 2024 with diagnostic confirmation of TB (including clinical diagnoses) and including patients with tuberculous pleurisy. (5) Treatment outcome determination: according to standards established in the 《Technical Specifications for Tuberculosis Prevention and Control in China (2020 Edition) 》 [16], treatment outcomes are categorized into either treatment success (including cure and treatment completion) or unfavorable outcome (including mortality, loss to follow-up, treatment discontinuation, and revision of therapy to a multidrug-resistant TB regimen).

Observation Indicators

Observation indicators are as follows: (1) notification incidence rate: it refers to the proportion of patients with registered and reported PTB to the total population within a specified period. (2) Age-standardized incidence rate: it is an incidence rate calculated by weighting age-specific rates of a study population to a standard population structure, enabling comparisons unaffected by differences in age distribution. (3) Calculation formula for average annual decline rate of PTB notification incidence rate: average annual decline rate=(1 - $\sqrt[\mathrm{n}]{{\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_0}$)×100%, where P_n represents the registered incidence rate in the nth year, P₀ denotes the registered incidence rate in the base year, and n indicates the number of interval years. (4) Successful treatment rate = (Cured+Treatment completion)/(Number of pulmonary tuberculosis patients–Transferred to rifampicin-resistant treatment–Diagnosis changed)×100%.

Ethical Considerations

This study was submitted to the Ethics Commission of the People’s Hospital of Dazu District. This study does not involve human participants or animals. Ethics approval was not required for this study because we did not include any identifiable data of patients’ personal information, including name, identity information, address, and telephone number. This study reviewed only secondary aggregated data at the population level; therefore, the need for written informed consent was waived. The Ethics Committee of the People’s Hospital of Dazu District approved this secondary analysis of deidentified TBIMS data (Approval 2025LLSC158; 2025-03-12) and granted a waiver of informed consent.

Statistical Analysis

Descriptive Analysis

Data were organized, and a database was established using Microsoft Excel 2021. Statistical analysis was performed using SPSS (version 27.0; IBM SPSS Statistics,) and Joinpoint (version 5.2.0; National Cancer Institute of America) software. Descriptive epidemiologic methods were employed. Count data were described using “number of cases, incidence rate (/100,000), and composition ratio (%).“ Intergroup comparisons were analyzed using the χ² test.

Joinpoint Regression

The temporal trend change points in age-standardized incidence rates were identified using the Joinpoint Regression Program (version 5.40; National Cancer Institute of America). This method fits the natural logarithm of the incidence rate as a piecewise linear function. The optimal number of joinpoints was determined and adopted using a Monte Carlo permutation test (10,000 iterations, α=.05). The independent variable was the calendar year, the interval type was set to annual, and the data point offset was 0. The dependent variables were the annual notification incidence rate and the age-standardized notification incidence rate, with data points where the incidence rate was 0 being excluded. The annual percent change (APC) along with its 95% CI was calculated for each trend segment, and the average annual percent change (AAPC) for the entire study period was also computed. The overall significance level for the permutation test was set at 0.0500. The maximum number of joinpoints was constrained to 3 based on the 19-year observation period, and model selection was optimized using the Bayesian information criterion.

Spatial Analysis

Global and local spatial autocorrelation analyses were performed using Moran I and Getis-Ord Gi* statistics, respectively. The Moran I index ranges from −1 to 1, where values approaching +1 indicate positive spatial clustering (similar values cluster together), values near 0 suggest a random distribution, and values approaching −1 indicate dispersion. Statistically significant Giclusters were classified as hotspots (high-value clusters; P<.05, |Z|>1.96) or coldspots (low-value clusters; P<.05, |Z|<1.96).

Logistic Regression

A binary multivariate logistic regression analysis model was used to analyze the influencing factors of treatment outcomes; a P<.05 was considered statistically significant.

Results

Reported Incident Case Count, Notification Incidence Rates, and Bacteriological Positivity Rates

All regions share the same case reporting system and diagnostic criteria (Figure 1 and Checklist 1). A total of 10,236 cases were reported. The highest reported case count was in 2005 with 1000 cases, followed by a gradual downward trend, reaching a nadir of 234 cases in 2024. The notification incidence similarly decreased from 131.2/100,000 in 2005 to 28.4/100,000 in 2024, with an average annual notification incidence of 65.2/100,000. The average annual decline rate of reported TB incidence was 7.7%. The bacteriological positivity rate gradually declined from 46.9% in 2005 to 31.6% in 2017 and then subsequently increased to 81.6% by 2024. Monthly analysis revealed that March had the highest cumulative reported case count (1098 cases), while October, November, and December showed relatively lower counts, with 720, 750, and 748 cases, respectively (Tables 1 and 2).

‎

The Joinpoint regression analysis revealed that the annual notification incidence of PTB in Dazu District exhibited a significant overall downward trend (AAPC=−6.81; 95% CI −7.25 to −6.30; P<.001). This overall trend was characterized by a fluctuating decline across 3 distinct phases: a slight decrease from 2005 to 2015 (APC=−4.76; 95% CI −5.58 to −3.63; P=.002), an accelerated decline from 2015 to 2018 (APC=−14.82; 95% CI −16.96 to −10.04; P=.002), followed by a moderated rate of decline from 2018 to 2024 (APC=−5.99; 95% CI −7.83 to −0.38; P=.04; Figure 2A).

‎

The trends in age-standardized notification incidence of PTB in Dazu District varied across different age groups. The <15 years age group experienced a significant initial decline (2005‐2014: APC=−23.95; 95% CI −37.04 to −18.47; P<.001), which then plateaued (2014‐2023: APC=−3.18; 95% CI −9.89 to 13.39; P=.48). The 15‐64 years age group showed the largest overall decrease, with an accelerating downward trend (2005‐2014: APC=−5.00; 95% CI −6.51 to −2.73; P=.004; 2014‐2024: APC=−11.64; 95% CI −13.47 to −10.36; P<.001). In contrast, the ≥65 years age group, after a period of fluctuating decline (2005‐2015: APC=−3.07; 95% CI −5.32 to −10.80; P=.14; 2015‐2018: APC=−11.68; 95% CI −17.20 to 0.15; P=.05), transitioned to a slow increase (2018‐2024: APC=4.65; 95% CI −1.78 to 18.40; P=.08; Figure 2B).

Demographic Distribution

Reported patients comprised 7998 cases among male patients and 2238 cases among female patients, yielding a male-to-female ratio of 3.57:1. When categorized in 5-year intervals, male and female cases predominantly occurred in the 15‐64 age group, with relatively uniform case distribution across subintervals within this range (Figure 3). Throughout the observation period and across all populations, children <15 years accounted for 102 cases (0.9%), young adults aged 15‐44 years comprised 4028 cases (39.4%), middle-aged adults 45‐64 years constituted 4000 cases (39.1%), and older patients ≥65 years represented 2106 cases (20.6%). The proportion of patients ≥65 years showed a gradually increasing trend, while the incidence proportion among young adults 15‐44 years progressively declined. The proportion among middle-aged adults 45‐64 years remained stable (Table 3, Figure 3).

‎

The top 3 occupational groups were farmers (77.6%, 7948/10,236), homemakers or unemployed individuals (5.6%, 570/10,236), and students (5.2%, 532/10,236), accounting for 88.4% (9050/10,236) of the total reported cases. These were followed by workers and migrant workers (3.0%, 303/10,236), commercial service personnel (1.1%, 116/10,236), and government employees (0.9%, 93/10,236). The remaining occupations accounted for less than 7% collectively (Multimedia Appendix 1).

Geographical Distribution

All towns or subdistricts in Dazu District except for Longtanzi Subdistrict reported cases annually. The highest reported case count was in Longshui Town (13.1%, 1339/10,236), followed by Longgang Subdistrict (9.1%, 933/10,236) and Tangxiang Subdistrict (7.9%, 811/10,236). The lowest case count occurred in Longtanzi Subdistrict (0.7%, 72/10,236).

Due to the difficulty in obtaining accurate annual resident population data for each town and street, the proportion of reported cases in each town relative to the district’s total annual cases was used as a substitute for the annual incidence rate (Table 4 and Figure 4). Longshui Town had the highest proportion of the reported cases in most years, exceeding 10% (47/234) in all years except 2011, although its case numbers showed a clear downward trend; it was followed by Longgang Street. In Tangxiang Street, the annual case count declined from 2005 to 2017 and then gradually fluctuated upward. By 2024, it had the highest number of reported cases in the district, accounting for 20.1% (Table 5, Figures 5 and 6).

‎

‎

‎

We selected annual reported cases from 2005, 2015, and the most recent year (2024) to create geographical heatmaps (Figure 6). As shown, in 2005, Dazu District exhibited widespread high-intensity clustering of cases (>50 cases), with 7 towns and streets along the central axis (including Longshui, Longgang, and Tangxiang) each reporting over 50 cases. By 2015, the TB burden had decreased significantly across the region. Only Longshui Town remained a high-intensity cluster, while the other 6 areas had declined to medium (31‐50 cases) or medium-low intensity (11‐30 cases). By 2024, except for Tangxiang Street and Longshui Town, which were classified as medium-intensity areas, all other towns and streets fell into medium-low or low-intensity categories, of which 20 are low-intensity areas (<10 cases), with a coverage rate of more than 74% (Figure 6).

We calculated the notification incidence rate for each township and subdistrict in Dazu District from 2021 to 2024, using the permanent resident population data from the 2020 Seventh National Population Census of China as the denominator, and subsequently performed a spatial autocorrelation analysis. The global spatial autocorrelation analysis (Table 6) indicated that the spatial clustering of township-level PTB incidence was not statistically significant overall during the 2021‐2024 period, with the exception of the year 2023, which exhibited significant spatial clustering rather than a random distribution (P<.001). Local spatial autocorrelation analysis (Figure 7) revealed indistinct spatial segregation of cold-spot or hot-spot areas for PTB incidence among the townships and subdistricts of Dazu District, with the identified cold-spot and hot-spot areas varying across different years.

‎

Univariate Analysis of Treatment Outcomes

Among 10,236 reported cases, excluding 159 cases still under treatment, a total of 10,077 cases were included in the treatment outcome analysis. There were 9203 successfully treated cases, yielding a treatment success rate of 91.3% (9203/10,077). Univariate analysis revealed that age (χ²₃=78.265, P<.001), occupation (χ²₃=10.228, P=.02), patient origin (χ²₅=11.59, P=.04), treatment classification (initial treatment vs retreatment, χ²₁=10.205, P=.001), diagnostic results (bacteriologically negative vs positive, χ²₁=52.723, P<.001), HIV coinfection status (χ²₁=14.724, P<.001), and treatment management modality (intensive phase supervision vs full-course supervision, χ²₂=430.013, P<.001) were risk factors influencing treatment outcomes (Table 7).

Multivariate Analysis of Treatment Outcomes

Collinearity analysis was performed for all variables. All variables showed tolerance >0.1, variance inflation factor <10, eigenvalues substantially greater than 0, and condition indices <30, indicating no significant multicollinearity issues. Using treatment outcome as the dependent variable (1=Treatment success, 0=Unfavorable treatment outcome), variables showing significance in univariate analysis were included as independent variables in multivariate binary logistic regression with the enter method. Multivariate logistic regression revealed that age, treatment classification, HIV coinfection status, bacteriologic test results, and treatment management modality were significant determinants of treatment success (Table 8).

Discussion

Our analysis of 2 decades of surveillance data yielded 3 main findings that align with the study objectives. First, Joinpoint regression confirmed a significant decline in the incidence of PTB (AAPC=–6.81; 95% CI –7.25 to –6.30%; P<.0001), with bacteriological positivity rates following a U-shaped trend over the study period. Second, spatial analysis revealed persistent high-burden areas, with Longshui Town and Tangxiang Subdistrict exhibiting the highest disease burden. Third, treatment success was significantly associated with younger age, initial treatment, HIV-negative status, bacteriologically negative results, and full-course supervised management.

Accurate data on TB incidence are difficult to obtain; consequently, reported incident case numbers serve as crucial indicators of disease burden and epidemiologic trends. From 2005 to 2024, a total of 10,236 patients with PTB were reported in our district. The annual notification incidence rate decreased from an initial rate of 131.2/100,000 population to 28.4/100,000 population in 2024, with an average annual rate of 65.2/100,000 population. This rate is slightly higher than the national average (60.77/100,000 population) [4] and the average annual notification incidence rate of Chongqing (54.6/100,000 population) [17] but lower than that of the Western China region (87.35/100,000 population) [4]. The notification incidence demonstrated an overall downward trend with an average annual decline rate of 7.7%, comparable to the annual decline rate observed in Chengdu City, China (7.75%) [18]. From an annual perspective, the TB case count in our district was particularly high in 2005, reaching 1000 cases, or 9.8% of the total cases reported during the entire study period. This historic peak is attributed to the full implementation of the Directly Observed Therapy-Short-course strategy [19] and improvements of the infectious disease reporting system that year. However, with the refinement of PTB prevention and control measures, annual case counts have gradually decreased, dropping to only 234 cases in 2024. The bacteriological positivity rate in our district initially decreased from 46.9% to 24.4% in 2016 and then gradually increased to 81.6% by 2024. This trend aligns with national reports on positivity rate changes [4], reflecting that our district’s epidemic prevention and control strategies have kept pace with national efforts. The increase is also associated with the widespread adoption of molecular biology detection technologies in recent years, particularly second-generation gene sequencing of bronchoscopic alveolar lavage fluid, which has significantly enhanced the bacteriological positivity rate of suspected PTB cases [20].

The incidence of PTB exhibits seasonal patterns [21], with the highest numbers of reported cases in our district occurring in March, predominantly concentrated during spring and summer, while winter case counts remained relatively lower. This phenomenon aligns with the findings of an international study [22]. Potential contributing factors include (1) cold winter weather increases indoor congregation with poor ventilation, facilitating MTB transmission; (2) reduced sunlight exposure during the winter decreases vitamin D synthesis; low vitamin D levels may increase susceptibility to TB [23]; (3) January and February coincide with China’s Spring Festival period. As TB symptoms often present atypically during early disease, many patients delay medical consultations until post-festival periods, leading to a surge in reported cases by March.

Our finding that cases were predominantly among male patients (male-to-female ratio of 3.57:1) was consistent with the findings of domestic and international studies [24,25]. This phenomenon may be attributed to gender-specific differences in hormone levels and genetic polymorphisms that influence susceptibility to MTB [26]. Many studies in humans and experimental animals have established clear links between sex-specific factors and the differential susceptibility of male and female individuals to a number of infectious diseases. Sex-specific determinants of Semin Immunopathol immunity include effects of sex steroid hormones as well as sex chromosome–encoded genes and microRNA [27]. Additionally, male patients often face systemic disadvantages in seeking or accessing TB care across various social contexts [28]. These findings highlight the necessity of incorporating gender-sensitive approaches into PTB prevention and control measures, particularly by enhancing diagnostic accessibility and screening services for male populations, to ensure equitable health care delivery across genders. This study revealed that reported PTB cases, in both male and female patients, were predominantly concentrated in the 15‐ to 64-year-old age group, consistent with previous research [29]. Notably, the proportion of cases among children less than 15 years of age is gradually decreasing in our district, while reported cases among those over the age of 65 years are progressively increasing. This trend suggests that our district must place greater emphasis on TB prevention and control in the older population. The observed pattern may be attributed to accelerated population aging, declining physiological function, and higher comorbidity rates among older adults, increasing susceptibility to TB. Conversely, the gradual reduction in cases among children under 15 years of age likely results from protection conferred by Bacillus Calmette Guerin vaccination and enhanced TB control measures targeting student populations. These findings highlight the necessity to increase the frequency of PTB screening among the older adults, particularly among male seniors, to reduce incidence rates in this vulnerable demographic.

Our finding that farmers constituted the majority (77.6%, 7948/10,236) of reported cases in our district is consistent with the findings of a previous study [30]. Potential explanations for this pattern include the following: (1) China’s status as an agricultural nation with a large rural population and high demographic proportion; (2) rural areas may experience relatively underdeveloped economic conditions, limited access to medical services, and inadequate prevention and control measures; (3) a generally lower educational status and limited awareness of PTB-related knowledge among farmers. TB-related knowledge, educational levels, and awareness of transmission risk factors correlate with disease incidence [31,32]. Therefore, enhanced education regarding PTB prevention and control measures is essential in rural communities.

Our observation of the second-highest case count (5.6%, 570/10,236) among homemakers and unemployed may reflect an association between relatively low income levels and unwillingness to seek medical care when symptoms emerge, which may facilitate TB transmission [33]. The third-highest case count (5.2%, 532/10,236) among students may be related to school population densities, intense academic pressure, and sleep deprivation [34] that contribute to both MTB transmission and increased susceptibility to post-infection disease onset and progression. These findings collectively highlight the necessity for targeted prevention and control measures focused on these high-risk populations in our district.

Significant geographical disparities of incidence were observed across towns and subdistricts. The highest reported case count (13.1%, 1339/10,236) observed in Longshui Town, followed by Longgang Subdistrict and Tangxiang Subdistrict, may be related to the large population bases of these 3 towns and subdistricts. Guoliang Town exhibited the highest average annual incidence rate (314.4/100,000), while Tongqiao Subdistrict showed the lowest (16.8/100,000). A total of 21 towns or subdistricts had incidence rates exceeding 100/100,000, highlighting the persistently severe epidemic in our district. This necessitates intensified prevention and control measures in these areas, with particular attention required to mitigate future epidemic trends in Guoliang Town. Global spatial autocorrelation analysis of township-level TB incidence in Dazu District (2021‐2024) showed significant clustering only in 2023 (P<.001). Local spatial autocorrelation revealed indistinct hot-spot or cold-spot segregation. We speculate that spatial autocorrelation may have limited sensitivity at this fine scale due to low case counts leading to unstable rates and high geographical proximity with shared risk factors among adjacent townships. Creating geographic heatmaps using annual case counts may better illustrate spatial variations in the epidemic.

Our treatment success rate of 91.3% was slightly lower than reported in previous studies [35,36]. Multivariate analysis revealed that age, diagnostic criteria, management approaches, HIV coinfection status, and bacteriological test results were risk factors that influenced treatment success. Variations in economic status, external environmental exposures, and stress levels across different age groups, combined with age-related differences in immune function, contribute to the heterogeneous manifestations of PTB pathogenesis. Compromised immune function and chronic comorbidities among people over 65 years of age increase susceptibility to TB [37].

Positive bacteriologic test results indicate a higher bacterial load that may be more pathogenic and pose greater therapeutic challenges. Retreatment cases exhibited lower treatment success rates due to higher drug resistance rates, poor adherence, prolonged treatment duration, and more extensive lung pathology. Patients enrolled solely in intensive phase supervision programs may initially benefit from health care worker engagement that may facilitate more regular medication intake and heightened awareness of their condition. However, once the intensive phase concludes without continued supervision, patients may mistakenly believe they are cured of TB, resulting in reduced adherence, irregular medication intake, and ultimately unfavorable treatment outcomes. Individuals with HIV coinfection experience varying degrees of immune suppression, resulting in higher susceptibility to MTB infection, lower treatment success rates, and higher mortality rates [38,39].

This study has certain limitations. First, when analyzing annual incidence rates across towns and subdistricts, we were unable to obtain specific permanent resident population figures for corresponding years, and the proportion of reported cases in each town relative to the district’s total annual cases was used as a substitute for the annual incidence rate. Second, potentially relevant variables such as nutritional status, silicosis comorbidity, and specific extrapulmonary TB manifestations were not included in the analysis of treatment outcomes. Third, underreporting inherent to passive surveillance systems, particularly among asymptomatic cases, may lead to an underestimation of true incidence. Fourth, the evolution of diagnostic techniques (including the introduction of molecular assays like Xpert MTB or rifampicin) over the 20-year period may influence the temporal comparability of bacteriological results. Fifth, univariate analyses involved multiple comparisons without statistical adjustment (eg, Bonferroni), increasing the risk of type I errors; thus, multivariate results should be prioritized. Finally, the findings, derived from Dazu District—an urban-rural transition zone in western Chongqing with a high farmer proportion (77.6%) and aging trend—may be most applicable to agricultural regions in central and western China and should be generalized to metropolitan core areas with caution.

In conclusion, the incidence of PTB in our district gradually decreased. Special focus should be directed toward monitoring epidemic trends among farmers and male individuals. Concurrently, treatment protocols should be enhanced for patients with PTB across different age groups, those with HIV coinfection, individuals receiving only intensive phase supervision, patients undergoing retreatment, and bacteriologically confirmed cases. These measures will facilitate the reduction of PTB incidence and the improvement of treatment success rates for patients in our district. Future research could focus on high-burden towns, explore and construct an active screening model for TB suitable for the grassroots level, and deeply study the impact mechanism and intervention strategies of labor mobility on the cross-regional transmission of TB, so as to provide assistance for the formation of collaborative prevention and control strategies at the district and county scale.