This is an openaccess article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on https://publichealth.jmir.org, as well as this copyright and license information must be included.
In the past few decades, liver disease has gradually become one of the major causes of death and illness worldwide. Hepatitis is one of the most common liver diseases in China. There have been intermittent and epidemic outbreaks of hepatitis worldwide, with a tendency toward cyclical recurrences. This periodicity poses challenges to epidemic prevention and control.
In this study, we aimed to investigate the relationship between the periodic characteristics of the hepatitis epidemic and local meteorological elements in Guangdong, China, which is a representative province with the largest population and gross domestic product in China.
Time series data sets from January 2013 to December 2020 for 4 notifiable infectious diseases caused by hepatitis viruses (ie, hepatitis A, B, C, and E viruses) and monthly data of meteorological elements (ie, temperature, precipitation, and humidity) were used in this study. Power spectrum analysis was conducted on time series data, and correlation and regression analyses were performed to assess the relationship between the epidemics and meteorological elements.
The 4 hepatitis epidemics showed clear periodic phenomena in the 8year data set in connection with meteorological elements. Based on the correlation analysis, temperature demonstrated the strongest correlation with hepatitis A, B, and C epidemics, while humidity was most significantly associated with the hepatitis E epidemic. Regression analysis revealed a positive and significant coefficient between temperature and hepatitis A, B, and C epidemics in Guangdong, while humidity had a strong and significant association with the hepatitis E epidemic, and its relationship with temperature was relatively weak.
These findings provide a better understanding of the mechanisms underlying different hepatitis epidemics and their connection to meteorological factors. This understanding can help guide local governments in predicting and preparing for future epidemics based on weather patterns and potentially aid in the development of effective prevention measures and policies.
Hepatitis, which describes an inflammation of the liver, is one of the most common diseases in the world. Hepatitis can be caused by a variety of factors. However, viruses are the most prevalent cause. Hepatitis A [
Viruses can be transmitted in 2 ways: horizontal and vertical transmission. Horizontal transmission involves the transmission of a virus from one person to another within the same generation, while vertical transmission refers to the transmission of viruses from mother to child. Hepatitis infections often show an agedependent property, with more severe symptoms observed in adults [
Recrudescence at a fixed frequency is a common feature of infectious diseases throughout the world [
In light of the above, in this study, we aimed to explore the oscillatory properties of the hepatitis epidemic in Guangdong (20°13'25°31' N and 109°39'117°19' E) as well as the potential natural contributors to the oscillatory outbreaks. We obtained data for the period from January 2013 to December 2020 on 4 notifiable infectious diseases caused by hepatitis viruses (including HAV, HBV, HCV, and HEV) in Guangdong and the meteorological elements (eg, temperature, precipitation, and humidity) in the same time frame. Power spectrum analysis was conducted on these data to capture the oscillatory strength of the outbreaks. We then explored the relationship between meteorological elements and oscillatory properties based on the regression and correlation analyses.
Time series data on available monthly reported and confirmed cases of 4 hepatitis diseases (A, B, C, and E) were obtained for Guangdong province in China’s mainland, from January 2013 to December 2020, from the Health Commission of Guangdong. The data set is available to the public around the world and is reported monthly. Monthly reported data of meteorological elements (eg, temperature, precipitation, and humidity) of Guangdong province from January 2013 to December 2020 were obtained from the China Statistical Yearbook 20142021. The meteorological factors included in this study are continuous values and vary with time, which would directly reflect the actual natural conditions each month. This data set is also available to the public around the world and is reported annually.
For this study, the data we used are open to the public. Our study did not involve any interventions in human participants. This study was approved by the ethics committee of Beijing Sport University, China (2022142H).
We used spectrum analysis to quantify fluctuations and the recurrence of epidemics. Similar methods have been used in classic and modern studies in the field of public health [
The tuning curve of monthly infected cases depicts the basic character of disease outbreaks, providing a direct view of the situation each month based on the historical data. We took the monthly average number of infected cases of each hepatitis epidemic and computed them into a tuning curve (equation 1). Each type of hepatitis epidemic in this study has a tuning curve, and the periodic pattern within a year would be obvious based on it.
In this equation, N represents the number of the year.
The tuning curve of meteorological factors (eg, temperature, precipitation, and humidity) depicts the basic character of natural conditions, providing a direct view of the situation each month based on historical data. We computed the monthly average of meteorological factors into a tuning curve (equation 2). Each meteorological factor in this study has a tuning curve, and the periodic pattern within a year is obvious based on it.
In this equation, N represents the number of the year.
The regression model was shown as equation 3.
In this equation,
We used the Pearson correlation to measure the relationship between infected cases and meteorological elements. The correlation analysis was conducted using the corr function in MATLAB (2020a).
This study analyzed the monthly data of confirmed cases of 4 hepatitis diseases in Guangdong from January 2013 to December 2020 (
Periodic phenomena of hepatitis epidemics with power spectrum. Left map shows the geological location of Guangdong. (A) shows the monthly incidences of hepatitis A, B, C, and E viruses (HAV, HBV, HCV, and HEV) and its spectrogram from January 2013 to December 2020 in Guangdong. (B) The power spectrum of time series data shown in panel A. The yaxis is the relative power, which is defined as power(f)/max(power(f)).
Meteorological elements might be a potential contributor to the periodic phenomenon of the epidemic; therefore, we also conducted the power spectrum analysis of the time series data of temperature, precipitation, and humidity in Guangdong province (
Meteorological elements of hepatitis epidemics with power spectrum. Left map shows the geological location of Guangdong. (A) The monthly time series meteorological elements (temperature, precipitation, and humidity) and its spectrogram from January 2013 to December 2020 in Guangdong. (B) The power spectrum of time series data shown in panel A. The yaxis is the relative power, which is defined as power(f)/max(power(f)).
From the observation of the average infected cases of the hepatitis epidemic (first row in
Relationship between infected cases of hepatitis epidemic and meteorological elements. Top row: the tunning curves for the number of infected cases of 4 hepatitis epidemics over a year, with the grey area denoting the standard error of mean. Left column: the tunning curves for an index of natural factors for each month of each year. Rows 24 show scatter plots between the number of cases of each hepatitis type and temperature, precipitation, and humidity, respectively. Each column represents a different type of hepatitis. The red square in each plot indicates conditions that showed significant results in both correlation and regression analyses. HAV: hepatitis A virus; HBV: hepatitis B virus; HCV: hepatitis C virus; HEV: hepatitis E virus.
From the correlation analysis, we discovered a substantial positive association between temperature and the incidence of HAV (
From the regression analysis between the number of hepatitis infections and meteorological factors (
All the variance inflation factors are in the range from 1.5 to 2 (ie, 1.64, 1.93, and 1.78). The correlation coefficients among the 3 meteorological factors (ie, temperature, precipitation, and humidity) were lower than 0.8 (temperature and precipitation: 0.59; temperature and humidity: 0.54; and precipitation and humidity: 0.63, respectively), which would not be affected by collinearity. In the first regression model, using 3 meteorological factors to predict the number of hepatitis infections, the coefficients of temperature in the prediction of hepatitis A, B, and C were significantly positive (hepatitis A:
In sum, we found that warmer temperatures can predict a higher prevalence of hepatitis A, B, and C in Guangdong. Meanwhile, higher humidity and lower temperature can both predict a higher prevalence of hepatitis E.
Statistics for the regression analysis (temperature, precipitation, and humidity).
Regression anlysis  Hepatitis A virus  Hepatitis B virus  Hepatitis C virus  Hepatitis E virus  

Coefficient  Coefficient  Coefficient  Coefficient  
β_{0}  106.94  —^{a}  10320.51  —  1425.52  —  49.96  —  
β_{1(temperature)}  1.809  3.71^{b}  127.78  3.29^{c}  27.59  4.41^{b}  –2.05  2.2^{d}  
β_{2}_{(precipitation)}  0.001  0.04  –1.04  0.77  –0.22  1.02  0.025  0.76  
β_{3(humidity)}  –0.155  0.31  20.01  0.51  1.57  0.24  2.80  2.93^{c} 
^{a}Not applicable.
^{b}
^{c}
^{d}
Statistics for the regression analysis (temperature and humidity).
Regression anlysis  Hepatitis A virus  Hepatitis B virus  Hepatitis C virus  Hepatitis E virus  

Coefficient  Coefficient  Coefficient  Coefficient  
β_{0}  106.23  —^{a}  11489.05  —  1675.11  —  21.88  —  
β_{1(temperature)}  1.816  4.06^{b}  116.365  3.25^{c}  25.15  4.34^{b}  –1.78  –2.03^{d}  
β_{2(humidity)}  –0.146  –0.34  6.113  0.18  –1.39  –0.25  3.13  3.70^{b} 
^{a}Not applicable.
^{b}
^{c}
^{d}
In this study, we found that hepatitis epidemics (A, B, C, and E) have different oscillatory properties, and hepatitis A, B, and C in Guangdong have a stronger association with temperature, while hepatitis E showed a stronger association with humidity. Different types of hepatitis will have different periodic characteristics and different associations with different natural environments, which is crucial for epidemic prevention and management. Understanding the temporal characteristics of infectious diseases is essential for effective epidemic prevention, as it may inform the development and implementation of appropriate policies and strategies.
To our knowledge, this is the first study to investigate the periodic characteristics of the hepatitis epidemics and their relationship with natural factors, which fills a gap in this field. Previous works have mainly focused on descriptive statistics, detailing the number of infections without describing the precise time characteristics [
Another issue is the influence of meteorological factors on hepatitis epidemics. HAV and HEV are 2 distinct viruses. Although HAV and HEV share similarities in their fecaloral transmission route [
First, the host range of HAV is limited to humans and several nonhuman primates [
Second, there are 4 main genotypes of HEV. Genotypes 1 and 2 cause outbreaks or epidemics in humans [
Further research is needed to confirm these hypotheses using detailed biological experiments. Previous studies on how temperature affects viral transmission have been equivocal. For example, norovirus prevalence was associated with low water temperatures [
One limitation of our study is the limited data. Despite being a representative province, our study only includes a single province in China. In the future, we will further obtain more data and apply similar methods to assess infectious diseases in more regions with varying climates. Even though we have made explorations, several factors remain unclear and cannot be addressed at present. For example, hepatitis A and E have similar propagation principles, but their oscillatory strength is very different in Guangdong Province. In the future, higherdimensional data will be required to provide more clarity. Another potential limitation is that we did not explore the interaction of the variables and the potential nonlinear relationships in the regression model, which should be considered in other studies in the future, where necessary.
Our study makes a link between climate change and the recurrence of infectious disease epidemics, enabling us to predict the magnitude of the epidemics based on various weather conditions. It is essential for local meteorological and medical departments to collaborate closely to prepare for the prevention and control of various epidemics. Additionally, the general public should be informed about the impact of different weather conditions on the spread of disease so that they can take appropriate measures to reduce the spread of viruses.
hepatitis A virus
hepatitis B virus
hepatitis C virus
hepatitis E virus
This work was funded by the China Postdoctoral Science Foundation (2022M723299), the Beijing Municipal Hospital Clinical Technology Innovation and Research Plan (XMLX201805), the Beijing Municipal Hospital Research and Development Project (PX2021068), the Advanced Innovation Center for Human Brain Protection Project (350012020137), and the Guangdong Province Basic Research Grant (2019A1515110870).
The data set is available to the public, which can be found on the official website of the Guangdong Provincial Health Commission [
CH, ML, and XZ conceived and designed the study. CH, ML, NH, BW, MS, and XW contributed to the literature search. CH and ML contributed to data collection. CH and ML contributed to the data analysis and the interpretation of results. All authors contributed to writing the paper.
None declared.