Analyzing seasonality of tuberculosis across Indian states and union territories

Pankaj Narula, Praveer Sihota, Sarita Azad, Píetro LióJournal of Epidemiology and Global Health 2015 · DOI

Abstract

A significant seasonal variation in tuberculosis (TB) is observed in north India during 2006-2011, particularly in states like Himachal Pradesh, Haryana and Rajasthan. To quantify the seasonal variation, we measure average amplitude (peak to trough distance) across seasons in smear positive cases of TB and observe that it is maximum for Himachal Pradesh (40.01%) and minimum for Maharashtra (3.87%). In north India, smear positive cases peak in second quarter (April-June) and reach a trough in fourth quarter (October-December), however low seasonal variation is observed in southern region of the country. The significant correlations as 0.64 (p-value<0.001), 0.54 (p-value<0.01) and 0.42 (p-value<0.05) are observed between minimum temperature and seasonality of TB at lag-1 in north, central and northeast India respectively. In south India, this correlation is not significant.

Introduction

India carries roughly 26% of the global tuberculosis burden, yet the temporal dynamics of the disease across its diverse climatic zones remain poorly understood. While diseases like malaria or cholera have well-known seasonal peaks, TB is often viewed as a constant threat. This study investigates whether TB incidence in India follows a distinct seasonal pattern and how this seasonality varies by geography. Understanding these patterns is critical for public health: if health agencies can predict "peak seasons," they can better allocate diagnostic kits, staff, and mobile screening units to the regions that need them most at specific times of the year.

Methodology

The researchers analyzed quarterly notification data for "smear-positive" TB cases—the most infectious form—across 34 Indian states and union territories from 2006 to 2011. Data was sourced from the Revised National Tuberculosis Control Program (RNTCP) reports. To quantify seasonality, they employed a Generalized Linear Model (GLM) using a Poisson distribution, which is the standard statistical approach for modeling count data. The core of their model was a sinusoidal function to fit the cyclical rises and falls in the data. Key metrics included Average Amplitude (the distance between the seasonal peak and trough) and Spearman Rank Correlation to check the relationship between temperature and TB cases at various lags.

Results and Geographic Variation

The analysis revealed a distinct North-South divide in TB seasonality. Northern states, such as Himachal Pradesh and Rajasthan, showed dramatic seasonal swings, whereas Southern states like Karnataka and Kerala exhibited minimal variation. This suggests that extreme climatic shifts, rather than just national healthcare policy, drive TB timing.

Comparison of quarterly TB cases in Delhi (North) and Karnataka (South) from 2006 to 2011.

Himachal Pradesh recorded the highest seasonal amplitude at 40.01%, while Maharashtra showed the lowest at 3.87%. In North India, cases consistently peaked in the second quarter (April–June) and hit their lowest point in the fourth quarter (October–December). Furthermore, the annual total case detection rate in Delhi (North) was 288 per 100,000, significantly higher than Kerala (South) at 75 per 100,000.

Climatic Correlations and the "Lag" Effect

The strongest correlation between temperature and TB occurs at a one-quarter lag (3 months). Low temperatures in the winter (Q4) are strongly associated with a spike in reported cases in the following spring/summer (Q2). Significant correlations (up to 0.64) were found between minimum temperature and TB seasonality at a lag-1 in Northern, Central, and Northeastern India. In the South, this correlation was statistically insignificant.

Discussion and Hypotheses

The authors suggest that lower ultraviolet radiation in Northern winters leads to vitamin D deficiency, which weakens the immune system and makes individuals more susceptible to activating latent TB. Additionally, harsh winters in the Himalayas and North India may discourage patients from seeking care or make it harder for health workers to reach remote areas. This creates a "backlog" of undiagnosed cases that flood the system once the weather warms up in April.

However, the study has limitations. The data comes from the public RNTCP system, potentially missing nearly 50% of TB cases treated in the private sector. Furthermore, while the correlation with temperature is strong, it doesn't prove causation; factors like indoor air pollution from winter heating or seasonal migration patterns could also play a role.

Conclusion

TB in India follows a clear seasonal pattern, particularly in the North, influenced by climatic factors and temperature lags. These findings provide a framework for health authorities to optimize the timing of public health interventions and resource distribution across different regions of the country.

Summary updated

Abstract

A significant seasonal variation in tuberculosis (TB) is observed in north India during 2006-2011, particularly in states like Himachal Pradesh, Haryana and Rajasthan. To quantify the seasonal variation, we measure average amplitude (peak to trough distance) across seasons in smear positive cases of TB and observe that it is maximum for Himachal Pradesh (40.01%) and minimum for Maharashtra (3.87%). In north India, smear positive cases peak in second quarter (April-June) and reach a trough in fourth quarter (October-December), however low seasonal variation is observed in southern region of the country. The significant correlations as 0.64 (p-value<0.001), 0.54 (p-value<0.01) and 0.42 (p-value<0.05) are observed between minimum temperature and seasonality of TB at lag-1 in north, central and northeast India respectively. However, in south India, this correlation is not significant.
BibTeX
@article{narula2015analyzing,
  author = {Pankaj Narula and Praveer Sihota and Sarita Azad and Píetro Lió},
  title = {{Analyzing seasonality of tuberculosis across Indian states and union territories}},
  journal = {Journal of Epidemiology and Global Health},
  year = {2015},
  doi = {10.1016/j.jegh.2015.02.004},
}
Figure