Association of Temperature, Precipitation, and Humidity with Incidence of Infectious Diarrheal Disease in the Sichuan Province of China from 2005 to 2016 Open Access

McCain, Kelly (Spring 2020)

Permanent URL:


Infectious diarrhea is a major cause of morbidity and mortality globally, causing 1.6 million deaths and 49.8 million disability-adjusted life years lost in 2016. Diarrheal disease often has distinct seasonality, with viral cases more often occurring in the colder, drier winter months and bacterial cases more often in the warmer, wetter summer months. In Sichuan, China, and the capital city of Chengdu, while there has been an overall decreasing trend in diarrheal disease, seasonal patterns have shifted from annual to biannual peaks, beginning in January 2012. This study utilized China’s National Infectious Disease Reporting System to investigate the relationship between meteorological factors and diarrheal disease and how the shift in disease seasonality modified the impact of temperature, humidity, and precipitation on the incidence of ‘other infectious diarrhea’ in Chengdu. A negative binomial generalized linear model was fit to estimate the effects of extreme precipitation, relative humidity, and temperature on diarrheal disease, while controlling for season and autocorrelation. Relative humidity lagged 1 week was associated with diarrheal disease (IRR, 95% CI: 0.995 (0.990, 0.999) for each 1% increase in relative humidity). After the shift, for every 1% increase in non-lagged humidity, the rate of diarrheal disease decreased by 0.01 (IRR and 95% CI: 0.990 (0.983, 0.997)). No significant relationships were found between all other meteorological terms and diarrheal disease either before or after the shift to biannual peaks. The weekly rate of diarrheal disease after the shift in 2012 was 3.5 times higher than the rate beforehand (IRR, 95% CI: 3.547 (1.237, 10.170)). The significance of the diarrheal disease autocorrelation terms, the log of the count of cases lagged at 1-4 weeks and at 8 weeks, indicates that disease prevalence is more strongly associated with transmission of diarrheal disease in Chengdu than meteorological factors (IRR, 95% CI: 1-week lag: 1.198 (1.125,1.256); 2-week lag: 1.164 (1.091, 1.241); 3-week lag: 1.106 (1.037, 1.181); 4-week lag: 1.088 (1.020, 1.160); 8-week lag: (IRR, 95% CI: 0.927 (0.866, 0.991) for each additional case). Further research that includes both infectious disease dynamics and meteorological factors is needed to identify other drivers may have contributed to the shift in seasonality of diarrheal disease cases.

Table of Contents

1. Introduction 1

Infectious Diarrhea 1

Transmission Pathways of Diarrheal Disease 2

Seasonality of Diarrheal Disease 2

Climate, Weather, and Infectious Diarrhea 3

Temperature 4

Precipitation 5

Relative Humidity 6

Diarrheal Disease Dynamics 7

Weather and Diarrhea in Urban Settings 8

Sichuan Province and Chengdu City 8

2. Methods 12

Diarrheal disease case data 12

Meteorological data 12

Statistical analyses 13

3. Results 16

Description of diarrheal disease cases 16

Description of meteorological data 16

Spearman correlations between meteorological factors and diarrheal disease case counts 17

Negative binomial generalized linear model 17

4. Discussion 19

Limitations 22

Public Health Implications 23

Conclusion 24

References 25

Figures and Tables 30

Appendix. 37

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Last modified

Primary PDF

Supplemental Files