Epidemiological characteristics and spatiotemporal analysis of mumps at township level in Wuhan, China, 2005-2019.

The resurgence and outbreaks of mumps occur frequently in many countries worldwide in recent years, even in countries with high vaccination coverage. In this study, a descriptive and spatiotemporal clustering analysis at the township level was conducted to explore the dynamic spatiotemporal aggregation and epidemiological characteristics of mumps in Wuhan. During 2005 and 2019, there were 40 685 cases reported in Wuhan, with an average annual morbidity of 28.11 per 100 000 populations. The morbidity showed a fluctuating tendency, and peaked in 2010 and 2018. Bimodal seasonality was found, with a large peak between May and July, and a mild peak from November to January in the following year. Male students aged 5-9-year-old were the main risk group of mumps infection. Significant global spatial auto-correlation was detected except in 2007, 2009 and 2015. The spatial and temporal scan statistics indicated that the hot-spots mainly located at the western and southern areas of Wuhan with variations almost every year. Our findings could assist the public health authorities to develop and improve targeted health strategies, and allocate health resources rationally.

Identifying low-PM2.5 exposure commuting routes for cyclists through modeling with the random forest algorithm based on low-cost sensor measurements in three Asian cities.

Cyclists can be easily exposed to traffic-related pollutants due to riding on or close to the road during commuting in cities. PM2.5 has been identified as one of the major pollutants emitted by vehicles and associated with cardiopulmonary and respiratory diseases. As routing has been suggested to reduce the exposures for cyclists, in this study, PM2.5 was monitored with low-cost sensors during commuting periods to develop models for identifying low exposure routes in three Asian cities: Taipei, Osaka, and Seoul. The models for mapping the PM2.5 in the cities were developed by employing the random forest algorithm in a two-stage modeling approach. The land use features to explain spatial variation of PM2.5 were obtained from the open-source land use database, OpenStreetMap. The total length of the monitoring routes ranged from 101.36 to 148.22 km and the average PM2.5 ranged from 13.51 to 15.40 µg/m³ among the cities. The two-stage models had the standard k-fold cross-validation (CV) R2 of 0.93, 0.74, and 0.84 in Taipei, Osaka, and Seoul, respectively. To address spatial autocorrelation, a spatial cross-validation approach applying a distance restriction of 100 m between the model training and testing data was employed. The over-optimistic estimates on the predictions were thus prevented, showing model CV-R2 of 0.91, 0.67, and 0.78 respectively in Taipei, Osaka, and Seoul. The comparisons between the shortest-distance and lowest-exposure routes showed that the largest percentage of reduced averaged PM2.5 exposure could reach 32.1% with the distance increases by 37.8%. Given the findings in this study, routing behavior should be encouraged. With the daily commuting trips expanded, the cumulative effect may become significant on the chronic exposures over time. Therefore, a route planning tool for reducing the exposures shall be developed and promoted to the public.

Application of seasonal auto-regressive integrated moving average model in forecasting the incidence of hand-foot-mouth disease in Wuhan, China.

Outbreaks of hand-foot-mouth disease (HFMD) have occurred many times and caused serious health burden in China since 2008. Application of modern information technology to prediction and early response can be helpful for efficient HFMD prevention and control. A seasonal auto-regressive integrated moving average (ARIMA) model for time series analysis was designed in this study. Eighty-four-month (from January 2009 to December 2015) retrospective data obtained from the Chinese Information System for Disease Prevention and Control were subjected to ARIMA modeling. The coefficient of determination (R 2), normalized Bayesian Information Criterion (BIC) and Q-test P value were used to evaluate the goodness-of-fit of constructed models. Subsequently, the best-fitted ARIMA model was applied to predict the expected incidence of HFMD from January 2016 to December 2016. The best-fitted seasonal ARIMA model was identified as (1,0,1)(0,1,1)12, with the largest coefficient of determination (R 2=0.743) and lowest normalized BIC (BIC=3.645) value. The residuals of the model also showed non-significant autocorrelations (P Box-Ljung (Q)=0.299). The predictions by the optimum ARIMA model adequately captured the pattern in the data and exhibited two peaks of activity over the forecast interval, including a major peak during April to June, and again a light peak for September to November. The ARIMA model proposed in this study can forecast HFMD incidence trend effectively, which could provide useful support for future HFMD prevention and control in the study area. Besides, further observations should be added continually into the modeling data set, and parameters of the models should be adjusted accordingly.

Application of Seasonal Auto-regressive Integrated Moving Average Model in Forecasting the Incidence of Hand-foot-mouth Disease in Wuhan, China / 华中科技大学学报（医学）（英德文版）

Outbreaks of hand-foot-mouth disease (HFMD) have occurred many times and caused serious health burden in China since 2008.Application of modem information technology to prediction and early response can be helpful for efficient HFMD prevention and control.A seasonal auto-regressive integrated moving average (ARIMA) model for time series analysis was designed in this study.Eighty-four-month (from January 2009 to December 2015) retrospective data obtained from the Chinese Information System for Disease Prevention and Control were subjected to ARIMA modeling.The coefficient of determination (R2),normalized Bayesian Information Criterion (BIC) and Q-test P value were used to evaluate the goodness-of-fit of constructed models.Subsequently,the best-fitted ARIMA model was applied to predict the expected incidence of HFMD from January 2016 to December 2016.The best-fitted seasonal ARIMA model was identified as (1,0,1)(0,1,1)12,with the largest coefficient of determination (R2=0.743) and lowest normalized BIC (BIC=3.645) value.The residuals of the model also showed non-significant autocorrelations (PBox-Ljung (Q)=0.299).The predictions by the optimum ARIMA model adequately captured the pattern in the data and exhibited two peaks of activity over the forecast interval,including a major peak during April to June,and again a light peak for September to November.The ARIMA model proposed in this study can forecast HFMD incidence trend effectively,which could provide useful support for future HFMD prevention and control in the study area.Besides,further observations should be added continually into the modeling data set,and parameters of the models should be adjusted accordingly.

[Investigation on the risk factors for reported rising of measles incidence during November, 2011 to February, 2012, in some areas of Hubei province].

OBJECTIVE: To identify the risk factors for reported sudden rise of measles incidence in November, 2011, in some areas of Hubei province. METHODS: We analyzed all measles patients reported in the measles surveillance system from November 2011 to February 2012 in two prefectures with highest attack rates and their vaccination history by reviewing vaccination records. We interviewed patients' parents by telephone to obtain the history of visiting health care within 7 to 21 days before onset. We also used case-crossover study to estimate the relative risk (RR) of hospital acquired infection and to compare the exposure to health care between 7-21 days before onset to 37-51 days before onset among measles patients. RESULTS: Totally 140 patients were reported in the two prefectures. Reported measles incidence rates among the population aged <8 m(69/100 000) and 8 m to 17 m (72/100 000) were higher than other age groups (rang from 0 to 5.8 per 100 000). Among the population aged 8 m to 17 m, estimated vaccination coverage was lower than 75%, and it was lower than 90% among those aged 18 m to 3 yrs. During 7-21 days before onset, 58% (29/50) of the patients had an exposure to health care settings, compared to 14% (7/50) of patients during 37 to 51 days before onset (MH RR = 5.4, 95% confidence interval = 2.1-14.0). CONCLUSION: Under the condition of measles vaccination coverage lower than 95%, iatrogenic infection was a risk factor for measles in Hubei.

[Analysis of sequence and genotype of E gene of the newly isolated Japanese encephalitis virus strains in Wuhan, Hubei Province].

OBJECTIVE: To analyze the molecular characteristics of the newly isolated two Japanese encephalitis virus strains (JEV) in Wuhan. METHODS: The mosquitoes were collected in Wuhan from April to October in 2009. The envelope (E) protein gene of JEV was detected using RT-PCR and sequenced. Sequence comparisons and phylogenetic analysis were conducted using DNAstar and MegAlign. RESULTS: Two Japanese encephalitis virus (JEV) strains (WHJX09-9, WHJX09-10) were isolated from Culex tritaeniorhynchus among 16 mosquito pools and identified as genotype I. The result showed that the homology of the two strains was 98. 9% in nucleotides and 100% in deduced amines. The comparison between the new genotype 1 JEV strains and live attenuated vaccine strain SA14-14-2 in E gene showed that the homology of nucleotide sequence was 87.4% and 87.9%, the homology of amino acid was 96.9% (total 15 amino acid were different) in E gene. The mutation sites of amino acid distributed among three different coding domain, but no antigen binding site and neurotoxin-involved site of amino acid were changed. CONCLUSION: Wuhan had appeared a new genotype of JEV which was different from the former strain isolated in Wuhan, the new JEV strains still had neurotoxicity but had high homology with the vaccine strains adopted in Wuhan. The vaccine could still be adopted to prevent Japanese encephalitis if steps were take to eradicate mosquitos at the same time. laboratory surveillance were also an important task to build an early-warning mechanism against JEV.

[Early warning on measles through the neural networks].

To discuss the effects on early warning of measles, using the neural networks. Based on the available data through monthly and weekly reports on measles from January 1986 to August 2006 in Wuhan city. The modal was developed using the neural networks to predict and analyze the prevalence and incidence of measles. When the dynamic time series modal was established with back propagation (BP) networks consisting of two layers, if p was assigned as 9, the convergence speed was acceptable and the correlation coefficient was equal to 0.85. It was more acceptable for monthly forecasting the specific value, but better for weekly forecasting the classification under probabilistic neural networks (PNN). When data was big enough to serve the purpose, it seemed more feasible for early warning using the two-layer BP networks. However, when data was not enough, then PNN could be used for the purpose of prediction. This method seemed feasible to be used in the system for early warning.

[Investigation on the distribution of avian influenza virus in external environment and the level of H5N1 antibody in poultry-exposed population in Wuhan].

OBJECTIVE: To survey the distribution of influenza A subtypes in external environment and investigate the infectious status of highly pathogenic avian influenza (H5N1) in poultry-exposed population in Wuhan. METHODS: Seventy-eight external environmental samples (water, cage surface and fecal samples) were collected from 3 habitats of wild migratory birds and 5 urban live-poultry markets in 2010. In 13 avian influenza monitoring points, 249 serum samples were collected from people living around habitats of wild migratory birds or working in live poultry markets. Real-time RT-PCR method was adopted to detect influenza A virus from external environmental samples; and multiple RT-PCR method and specific H3, H5, H7 and H9 primers were then applied to analyze the subtypes of the positive samples. The levels of H5N1 antibody in poultry-exposed population were tested by horse hemagglutination inhibition test and two avian influenza inactivated antigens: A/Hubei/1/10 and A/Anhui/1/05. RESULTS: Of the 50 external environmental samples collected from live poultry markets, 17 samples were determined to be influenza A virus positive (positive rate 34.0%), including specific subtypes as follows: 4 samples of H5 single-positive subtype, 3 samples of H9 single-positive subtype, 4 samples of H3 and H5 mixed-positive subtype, 2 samples of H3 and H9 mixed-positive subtype, 2 samples of H5 and H9 mixed-positive subtype, 2 samples of H3, H5 and H9 mixed-positive subtype, but no H7 positive subtype was found. The 28 external environmental samples collected from habitats of wild migratory birds were all influenza A virus negative. Considering different types of external environmental samples, the influenza A virus positive rates in water, cage surface and fecal samples were 37.5% (6/16), 16.7% (5/30) and 18.8% (6/32), respectively. There were total 100 samples of serum whose A/Hubei/1/10 antigen inhibiting titers ≥ 40, accounting for 40.2%; while 36 samples of serum (14.5%) whose A/Anhui/1/05 antigen inhibiting titers ≥ 40 were found. The difference had statistical significance (χ(2) = 41.433, P < 0.05). Among the 249 serum samples collected from poultry-exposed population, 5 samples were H5N1 antibody positive against A/Hubei/1/10 antigen (inhibition titer ≥ 160), which came from 4 different live poultry markets, however, no positive serum sample against A/Anhui/1/05 antigen was found. CONCLUSION: Multiple subtypes of avian influenza virus simultaneously prevailed in Wuhan urban poultry markets. Moreover, results from the distribution of avian influenza virus in external environment were consistent with the level of H5N1 antibody in poultry-exposed population.