Introduction
Tuberculosis (TB) remains a formidable global health challenge, ranking among the top ten causes of death worldwide. China bears a significant portion of this burden, accounting for approximately 9% of global cases in 2018. To address this, the World Health Organization (WHO) launched the "End TB Strategy," which established ambitious targets for reducing pulmonary TB (PTB) incidence: a 20% reduction by 2020 and a 50% reduction by 2025, relative to a 2015 baseline.
In Henan Province, one of China's most populous regions, public health officials must determine if current control measures are sufficient to meet these international benchmarks. This study analyzes 13 years of historical registration data to model TB trends and forecast whether Henan is on track to achieve the WHO milestones.
Methodology and Data Sources
The researchers utilized data from the Chinese Tuberculosis Information Management System (CTIMS), covering 854,672 active PTB cases registered in Henan between 2007 and 2019. Population data from the Henan Statistical Yearbook was used to calculate annual registration rates per 100,000 people.
To analyze these trends, the study employed the Exponential Smoothing (ES) Winter’s Multiplication Model. This time-series approach is particularly effective for TB data because it accounts for both a consistent downward trend and strong seasonal fluctuations. The model was trained on data from 2007–2016 and validated against observed data from 2017–2019 using the SPSS Expert Modeler to ensure the best fit based on stationary R-squared and the Bayesian Information Criterion (BIC).
Key Findings
Seasonal Patterns and Historical Decline
The analysis revealed a distinct seasonal pattern, with TB registration rates peaking consistently in the spring and early summer (March through June). The authors suggest this may be linked to increased indoor crowding during winter months followed by a delay in seeking care until after the Spring Festival, as well as the physical stress associated with school entrance exams during those months.
Despite these seasonal surges, Henan has seen a robust overall decline in TB. Between 2007 and 2019, the registration rate dropped at an average annual rate of 4.8%. This significantly outpaces the global average decline of 1.6% and the national Chinese average of 2.29%, highlighting the effectiveness of Henan’s local control programs, such as free diagnostic services and standardized treatment protocols.
Model Performance and Predictions
The exponential smoothing model demonstrated high predictive accuracy, with a Mean Absolute Percentage Error (MAPE) of 5.527%. As shown in Figure 1, the predicted values closely tracked the observed data during the validation period.
Based on this model, the researchers projected the following outcomes:
- 2020 Milestone: The predicted registration rate for 2020 was 49.1 per 100,000. This represents a 23.7% reduction from the 2015 baseline, successfully meeting the WHO’s 20% target.
- 2025 Milestone: The predicted rate for 2025 is 34.4 per 100,000. While this reflects a substantial 46.8% reduction from 2015, it falls just short of the WHO’s 50% goal.
Conclusions and Future Directions
The study suggests that while Henan Province likely achieved the first WHO End TB milestone in 2020, the 2025 target remains out of reach under current conditions. The "inertia" of the current public health trajectory is positive but insufficient for the more aggressive 2025 goal.
To bridge this gap, the authors advocate for intensified "early response measures," including expanded universal health coverage and more targeted interventions during high-risk seasons. Future research should investigate the specific biological and social drivers of seasonality to optimize the timing of public health campaigns. While time-series models are limited by their inability to predict sudden socio-economic shifts or medical breakthroughs, they provide a vital baseline for evaluating the adequacy of current TB control strategies.
Summary updated