Hand, foot, and mouth disease in China, : an epidemiological study

Transcription

1 Hand, foot, and mouth disease in China, 28 12: an epidemiological study Weijia Xing*, Qiaohong Liao*, Cécile Viboud*, Jing Zhang*, Junling Sun, Joseph T Wu, Zhaorui Chang, Fengfeng Liu, Vicky J Fang, Yingdong Zheng, Benjamin J Cowling, Jay K Varma, Jeremy J Farrar, Gabriel M Leung, Hongjie Yu Lancet Infect Dis 214; 14: Published Online January 31, S (13) See Comment page 262 *Authors contributed equally Division of Infectious Disease, Key Laboratory of Surveillance and Early Warning on Infectious Disease, Chinese Center for Disease Control and Prevention, Beijing, China (W Xing PhD, Q Liao MD, J Zhang MD, J Sun PhD, Z Chang MD, F Liu MD, H Yu PhD); Fogarty International Center, National Institutes of Health, Bethesda, MD, USA (C Viboud PhD); Division of Epidemiology and Biostatistics, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China (B J Cowling PhD, J T Wu PhD, V J Fang MPhil, Prof G M Leung MD); School of Public Health, Peking University, Health Science Center, Beijing, China (Y Zheng PhD); New York City Department of Health and Mental Hygiene, New York, NY, USA (J K Varma MD); Oxford University Clinical Research Unit Wellcome Trust Major Overseas Programmes, Ho Chi Minh City, Vietnam (Prof J J Farrar PhD); International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC), Centre for Tropical Medicine, University of Oxford, Churchill Hospital, Oxford, UK (Prof J J Farrar); Department of Medicine, National University of Singapore, Singapore (Prof J J Farrar); and The Li Ka Shing Oxford Global Health Programme and Wellcome Trust, Oxford, UK (Prof J J Farrar) Summary Background Hand, foot, and mouth disease is a common childhood illness caused by enteroviruses. Increasingly, the disease has a substantial burden throughout east and southeast Asia. To better inform vaccine and other interventions, we characterised the epidemiology of hand, foot, and mouth disease in China on the basis of enhanced surveillance. Methods We extracted epidemiological, clinical, and laboratory data from cases of hand, foot, and mouth disease reported to the Chinese Center for Disease Control and Prevention between Jan 1, 28, and Dec 31, 212. We then compiled climatic, geographical, and demographic information. All analyses were stratified by age, disease severity, laboratory confirmation status, and enterovirus serotype. Findings The surveillance registry included probable cases of hand, foot, and mouth disease (annual incidence, 1 2 per 1 person-years from 21 12), of which (3 7%) were laboratory confirmed and 2457 (3%) were fatal. Incidence and mortality were highest in children aged months (38 2 cases per 1 person-years and 1 5 deaths per 1 person-years in 212). Median duration from onset to diagnosis was 1 5 days (IQR 5 2 5) and median duration from onset to death was 3 5 days ( ). The absolute number of patients with cardiopulmonary or neurological complications was (case-severity rate 1 1%), and 2457 of patients with severe disease died (fatality rate 3 %); 1617 of 1737 laboratory confirmed deaths (93%) were associated with enterovirus 71. Every year in June, hand, foot, and mouth disease peaked in north China, whereas southern China had semiannual outbreaks in May and September October. Geographical differences in seasonal patterns were weakly associated with climate and demographic factors (variance explained 8 23% and 3 19%, respectively). Interpretation This is the largest population-based study up to now of the epidemiology of hand, foot, and mouth disease. Future mitigation policies should take into account the heterogeneities of disease burden identified. Additional epidemiological and serological studies are warranted to elucidate the dynamics and immunity patterns of local hand, foot, and mouth disease and to optimise interventions. Funding China US Collaborative Program on Emerging and Re-emerging Infectious Diseases, WHO, The Li Ka Shing Oxford Global Health Programme and Wellcome Trust, Harvard Center for Communicable Disease Dynamics, and Health and Medical Research Fund, Government of Hong Kong Special Administrative Region. Introduction Hand, foot, and mouth disease is a common infectious disorder caused by various enteroviruses; enterovirus 71 (EV71) and Coxsackie virus A16 (CV-A16) are the most commonly reported. 1 Children younger than 5 years are especially prone to hand, foot, and mouth disease, and most patients show self-limiting illness typically including fever, skin eruptions on hands and feet, and vesicles in the mouth. However, some patients rapidly develop neurological and systemic complications that can be fatal, especially in cases associated with EV71. 2 EV71 was first identified in 1969 in California, USA, 3 and has become more predominant across the Asia Pacific region in the past 15 years China established a national enhanced surveillance system for hand, foot, and mouth disease in May, 28, partly in response to several large outbreaks in 27 and early Here we describe the enhanced hand, foot, and mouth disease surveillance system and characterise the epidemiology of the disease in China, focusing on age, seasonal, and geographical patterns from 28 to 212. Methods Enhanced national surveillance programme From Jan 1, 28, to May 1, 28, all probable and laboratory-confirmed cases of hand, foot, and mouth disease were reported on a voluntary basis to the Chinese Center for Disease Control and Prevention (China CDC) in Beijing. On May 2, 28, hand, foot, and mouth disease was made statutorily notifiable. Case definitions A probable case of hand, foot, and mouth disease was defined as a patient with papular or vesicular rash on hands, feet, mouth, or buttocks, with or without fever. A confirmed case was defined as a probable case with laboratory evidence of enterovirus infection (including 38 Vol 14 April 214

2 EV71, CV-A16, or other non-ev71 and non-cv-a16) detected by RT-PCR, real-time PCR, or virus isolation. Patients with hand, foot, and mouth disease, whether probable or confirmed, were classified as severe if they had any neurological complications (aseptic meningitis, encephalitis, encephalomyelitis, acute flaccid paralysis, or autonomic nervous system dysregulation), or cardiopulmonary complications (pulmonary oedema, pulmonary haemorrhage, or cardiorespiratory failure), or both; otherwise, patients were categorised as mild cases. Collection and testing of specimens In view of restricted laboratory capacity during the first year of enhanced surveillance (May, 28 June, 29), we collected clinical samples in all 31 Chinese provinces from the first five patients with mild, probable disease who visited hospital outpatient departments every week (appendix pp 14). We also attempted to collect specimens from all severe cases. With increased capacity after June, 29, we collected samples from the first five patients with mild, probable disease who visited hospital outpatient departments every month in each of the 374 counties or districts, and from all severe cases (appendix pp 14). 12 Depending on the symptoms and clinical status of each patient, we collected the appropriate clinical specimens, including throat swab, rectal swab, faecal sample, vesicular fluid, or cerebrospinal fluid. Sampling was systematic and did not follow community outbreaks. We placed specimens in sterile viral transport medium and sent them to provincial or prefecture CDCs (n=425) for PCR or virus isolation, according to standardised protocols disseminated by the China CDC. 13 Viral RNA was extracted with available commercial kits (most frequently, QIAamp Viral RNA Mini Kit, Qiagen, Valencia, CA, USA, and Geneaid Viral RNA Mini Kit, Geneaid Biotech, Taiwan, China) as per the manufacturer s protocols. We tested RNA from each sample for specific primers and probes that targeted panenterovirus, EV71, and CV-A16. Testing was done in biosafety level two facilities. We classified test results into four categories: enterovirus negative, EV71 positive, CV-A16 positive, or positive for another enterovirus without further serotype identification. Some samples underwent virus isolation at provincial CDCs. According to national guidelines, 12 at least ten enterovirus strains should be identified in each province every month. Specimens were inoculated into human rhabdomyosarcoma (susceptible to EV71 and CV-A16) and Hep-2 (susceptible to other enteroviruses) cells. If cultures showed cytopathic effect, we did sequencing analysis to identify serotype. 14 We did a blind passage with all cultures, which showed no cytopathic effect after 7 days. Clinical and epidemiological data Probable and confirmed cases were reported online to the China CDC within 24 h of diagnosis, by use of a standardised form including basic demographic information (sex, date of birth, and address), case classification (probable or confirmed), severity (mild or severe), death status, date of symptoms onset, date of diagnosis, and date of death (if applicable), and virus serotype (EV71, CV-A16, other enterovirus) for confirmed cases. Climate, geographical, and demographic data To analyse the potential seasonal drivers of hand, foot, and mouth disease, we collected demographic, economic, geographical, and climatic information for all provinces from 28 to 212, including: age-specific population denominators, population density, gross domestic product, and gross regional product; 15 latitude and longitude of province capitals; data for air, rail, road, and boat travel; 16 meteorological data including daily temperature, rainfall, relative humidity, air pressure, average vapour pressure, and hours of sunshine (appendix pp 2). 17 Climate variables were aggregated by season (spring: April June; autumn: September November) to show the seasonal peaks of hand, foot, and mouth disease in surveillance data. We further grouped the 31 provinces into six climatic regions (appendix pp 16). Data analysis We included all cases with illness onset from Jan 1, 28, to Dec 31, 212, in the analysis. We estimated age-specific rates of incidence, severe illness, and mortality by combining probable and confirmed cases. We estimated 95% CIs with Poisson methods. For rates with a denominator greater than 1 and numerator of 2 or more, we calculated 95% CIs assuming a normal distribution. We assessed the geographical distribution of cases across all 31 provinces, grouped into seven regions (appendix pp 16). We estimated the age-specific caseseverity risk (number of severe cases divided by number of probable and confirmed cases), case-fatality rate (number of deaths divided by number of probable and confirmed cases), and fatality rate of severe cases of hand, foot, and mouth disease (number of deaths divided by number of severe cases), overall, and stratified by serotype. In serotype-specific analyses, we estimated the number of serotype-specific cases of hand, foot, and mouth disease by applying the distribution of serotypes in specimens positive for enteroviruses and further serotyped to the universe of probable and confirmed cases. We calculated the distributions of times from illness onset to diagnosis, illness onset to death, and diagnosis to death by virus serotype. To identify predictors for severe or fatal outcomes, we applied logistic regression and backward selection of demographic and geographical variables, time from onset to diagnosis, and virus serotype (p value for exclusion 1), separately for laboratory-confirmed cases and probable cases. We did all analyses with SAS (version 9.1). Correspondence to: Prof Gabriel M Leung, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China gmleung@hku.hk or Dr Hongjie Yu, Division of Infectious Disease, Key Laboratory of Surveillance and Early Warning on Infectious Disease, Chinese Center for Disease Control and Prevention, Changping district, Beijing, China yuhj@chinacdc.cn See Online for appendix Vol 14 April

3 Incidence rates (95% CI), per 1 million Severe illness rates (95% CI), per 1 million Mortality rates (95% CI), per 1 million Total ( ) <6 months ( ) 6 11 months ( ) months ( ) months ( ) 5 9 years 58 7 ( ) 1 14 years 68 4 ( ) 15 years 2 3 ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 4 1 (4 4 2) ( ) ( ) ( ) ( ) ( ) 1712 ( ) ( ) 6 5 ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 5 9 (5 8 6 ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ) 9 ( 9 1 ) 3 8 ( ) 27 3 ( ) 26 4 ( ) 8 8 (8 9 6) 7 ( 5 9) 1 1) ) 1 4 ( ) 35 9 (31 7 4) ( ) ( ) 93 2 ( ) 5 2 ( ) 3 ( 2 5) ) 21 3 ( ) 6 5 ( ) ( ) ( ) 24 8 ( ) 1 4 ( ) 1 1 ( 9 1 4) ) 14 1 ( ) 45 5 ( ) ( ) ( ) ( ) 8 8 ( ) 9 ( 7 1 2) ) 15 6 ( ) 65 ( ) 47 4 ( ) ( ) ( ) 1 2 ( ) 8 ( 6 1 ) ) 1 ( 1 1) 6 ( 2 1 5) 3 8 ( ) 3 2 ( ) 6 ( 4 9) 1) ) ) 3 ( 2 3) 2 ( ) 1 5 ( ) 8 6 (7 2 1) 1 8 ( ) 1 1) ) ) 7 ( 6 7) 3 3 ( ) 27 1 ( ) 2 2 ( ) 5 2 ( ) 1 ( 1 3) 1) ) 4 ( 4 4) 2 4 ( ) 11 5 ( ) 12 9 ( ) 3 9 ( ) 1 2) 1) ) 4 ( 4 5) 1 5 ( 7 2 7) 13 3 ( ) 15 3 ( ) 4 1 ( ) 1 ( 1 2) 1) ) Table 1: Estimated rates of incidence, severe illness, and mortality in all cases of hand, foot, and mouth disease by age group A Mild B Severe C Fatal Number of cases Number of cases Number of cases Proportions Year Year Year EV71 CV-A16 Other enterovirus Figure 1: Proportions of enterovirus serotypes in laboratory-confirmed cases of hand, foot, and mouth disease by clinical severity in China, (A) Based on mild cases. (B) Based on severe cases who survived. (C) Based on fatal cases. EV71=enterovirus 71. CV-A16=Coxsackie virus A16. To quantify seasonal patterns of hand, foot, and mouth disease by province, we created heat maps of the proportion of cases identified in each week of the year. 18 We also fitted seasonal multiple linear regression models to weekly hand, foot, and mouth disease time series, including time trends and harmonic terms 31 Vol 14 April 214

4 representing annual and semiannual periodic cycles (appendix pp 3 5). 18,19 We extracted the amplitude and peak timing of the annual and semiannual cycles based on model coefficients. 18 The amplitude measures the difference between the maximum and minimum of a seasonal curve. 19 We used multivariate stepwise linear regression to study the association between seasonal characteristics of hand, foot, and mouth disease and putative seasonal drivers, including geography, human mobility patterns, demographic, economic, and climate variables. 18 We identified epidemiologically relevant disease regions by hierarchical clustering with Ward s minimum variance method, and identified predictors of these regions by discriminant analysis. 18 Time series were standardised for differences in sampling intensity over time and geography by dividing weekly case counts by the annual number of cases. Role of the funding source The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding authors had full access to all the data in the study and had final responsibility for the decision to submit for publication. Results probable cases of hand, foot, and mouth disease were reported to the China CDC surveillance system during 28 12, of which (3 7%) were laboratory Age distribution (%) A EV71 CV-A16 Other enterovirus Probable Case fatality risk (%) B C 3 5 D Case severity risk (%) Severity fatality risk (%) <6 months 6 11 months 1 year 2 4 years Age 5 9 years 1 14 years >14 years <6 months 6 11 months 1 year 2 4 years Age 5 9 years 1 14 years >14 years Figure 2: Age distribution and clinical severity of probable and laboratory-confirmed cases of hand, foot, and mouth disease in China, (A) Age distribution of probable and laboratory-confirmed cases. (B) Risk of fatality in cases by age group and serotype of virus. (C) Risk of severe illness in cases by age group and serotype of virus. (D) Risk of fatality in severe cases by age group and serotype of virus. We calculated severity estimates in (B) (D) by extrapolating the serotype distribution in test-positive cases to untested and test-negative cases. The number of mild cases with serotype X (EV71, CV-A16, or other enteroviruses) was estimated to be number of mild cases test-positive for serotype X divided by number of test-positive mild cases times number of mild cases; severe cases were similarly analysed. Results were only similar if the or 212 data were used (appendix pp 2, 21). EV71=enterovirus 71. CV-A16=Coxsackie virus A16. Vol 14 April

5 Probability density function A B C 4 EV71 CV-A16 Other enterovirus Probable cases Time from onset to diagnosis (days) Time from onset to death (days) Time from diagnosis to death (days) 1 Figure 3: Estimates of onset-to-diagnosis, onset-to-death, and diagnosis-to-death distributions of probable and laboratory-confirmed cases of hand, foot, and mouth disease in China, (A) Onset-to-diagnosis distribution by serotype of virus (n=7 2 92). (B) Onset-to-death distribution by serotype of virus (n=2457). (C) Diagnosis-to-death distribution by serotype of virus (n=2457). Some intervals are negative because diagnosis was done after death. EV71=enterovirus 71. CV-A16=Coxsackie virus A16. confirmed and 2457 (3%) were fatal. Since the initiation of surveillance in 28, incidence of reported cases of hand, foot, and mouth disease has sharply increased (table 1), because of improvements in reporting and surveillance. Reported disease rates reached a steady state at more than 1 2 cases per 1 person-years during Incidence of hand, foot, and mouth disease varied greatly with age, with highest rates in children aged 6 months to 2 years (table 1). The median age of reported cases was 27 months (IQR 18 43). Most cases were in children younger than 5 years, and incidence was very low in infants younger than 6 months, older children (age 5 14 years), and adults (age 15 years). The incidence of hand, foot, and mouth disease was 1 6 times higher in boys younger than 5 years than in girls of the same age (2 1 vs 1 3 cases per 1 person-years, p<1). EV71 predominated in laboratory-confirmed cases, accounting for 45% of mild, 8% of severe, and 93% of fatal cases (figure 1). EV71 predominance did not vary substantially by age (figure 2A). EV71, CV-A16, and other enteroviruses co-circulated throughout the study period in all regions (appendix pp 18). The median time from illness onset to diagnosis was 1 5 days (IQR 5 2 5), 3 5 days ( ) for illness onset to death, and 5 day ( 5 1 5) for diagnosis to death. The probability density distributions of the onset-to-diagnosis, onset-to-death, and diagnosisto-death intervals were similar between probable and laboratory-confirmed cases of EV71 or other enteroviruses (figure 3). However, density distributions of CV-A16 showed attenuated signal-to-noise ratio in view of the small number of deaths (n=25). Overall, the case-fatality rate was 3% (range across years 3 5%), the case-severity rate was 1 1% ( 2 1 6%), and the severe case-fatality rate was 3 % ( %). Illness severity and death were inversely related to age (table 2, figure 2B D). Consistently across all age groups, EV71 infections were much more severe than were CV-A16 infections (figure 2B D). We did sensitivity analyses on severity estimates by age and viral serotypes for (appendix pp 2) and 212 (appendix pp 21), and recorded similar results. With multivariate backward logistic regression of laboratoryconfirmed cases, we identified several mortality risk factors beyond young age and infection with EV71, including rural residence and long onset-to-diagnosis interval. Further, males and rural residents had increased risk of severe disease. Inclusion of probable cases in the analysis yielded similar risk predictors (table 2). Nationally, hand, foot, and mouth disease showed semiannual peaks of activity, including a major peak in spring and early summer followed by a smaller peak in autumn, consistent between probable and confirmed time series (figure 4). Although southern China had two outbreaks of hand, foot, and mouth disease peaking in May and October of each year, northern China had one annual peak in June (figures 4 and 5). The annual amplitude of hand, foot, and mouth disease epidemics increased with increasing latitude and semi-annual periodicity was strongest in the south (figures 5 and 6). All enteroviruses showed latitudinal gradients in the amplitude of annual and semiannual epidemics (appendix pp 22 25). Annual peak timing of CV-A16 activity happened earlier than for EV71 and other enteroviruses (p<1). We identified putative predictors of seasonal characteristics of hand, foot, and mouth disease with multivariate stepwise regression (appendix pp 6 13). We noted an association between the annual amplitude of epi demics and spring rainfall and spring sunshine in overall analysis (p=3 and p=4, respectively, partial r² 5 23%) and serotype-stratified analysis Vol 14 April 214

6 (association between the annual amplitude of epidemics and spring rainfall in EV71 p=31, in CV-A16 p=2, in other enterovirus p=5; association between the annual amplitude of epidemics and spring sunshine in EV71 p=18, in CV-A16 p=2, in other enterovirus p=6, partial r² 8 23%). Annual peak timing of hand, foot, and mouth disease was associated with maximum spring air pressure in overall and serotype-stratified analyses (association between the annual peak timing and maximum spring air pressure in overall p=19, in EV71 p=16, in CV-A16 p=34, in other enterovirus: p=1, partial r² 4 12%). Predictors of semiannual periodicity of hand, foot, and mouth disease epidemics were more varied. Population and transportation factors were moderately associated with seasonal characteristics of hand, foot, and mouth disease, explaining 1 19% of the variance in epidemic timing and periodicity. Overall, multivariate models explained a moderate-to-high fraction of the variance in seasonal characteristics of disease through many factors (r² 4 92%). We identified two main geographical regions that share similar epidemiological characteristics of hand, foot, and mouth disease: a northern region (latitude range N, plus Tibet, 14 provinces), and a southern region ( N, 17 provinces; figure 7). Serotype-specific analysis showed a broadly similar split between northern and southern China, with a third epidemiological region including western provinces identified for EV71 and CV-A16 (appendix pp 26 29). Discriminant analysis confirmed that climate factors were the main predictors of epidemiological regions of hand, foot, and mouth disease (figure 7C, appendix pp 26 29). Discussion Our study of more than 7 2 million cases of hand, foot, and mouth disease reported to the national enhanced surveillance system during in China gives the first comprehensive account of the national burden and epidemiology of the disease (panel). The estimated incidence of hand, foot, and mouth disease is 1 2 per 1 person-years from in China, and the disease is responsible for 5 9 reported deaths every year, mainly in young children. Like in other countries, we recorded a predominance of EV71 serotype in severe cases, with a particularly young median age of infection at 2 3 years. Our large study covers 31 climatologically diverse provinces and suggests that although cases of hand, foot, and mouth disease tend to arise in warmer months of the year throughout the country, peak timing and periodicity of epidemics varies with latitude. All serotypes causing hand, foot, and mouth disease seem to broadly follow similar geographical gradients, which are moderately associated with climate differences. Laboratory-confirmed cases n= Number of cases Adjusted OR (95% CI) Our age profile of infection is in agreement with reports from other countries, 4,2 22 and the particularly young mean age at infection suggests that enteroviruses causing hand, foot, and mouth disease are highly transmissible. Relative sparing of infants younger than 6 months was most likely due to protection by maternal antibodies. 23,24 Infection with the causative viruses seems to confer lifelong immunity at least to disease and possibly infection given the almost complete absence of adult cases. Probable cases n= Number of cases Adjusted OR (95% CI) Death Enterovirus serotype CV-A Ref NA NA Other enterovirus ( ) NA NA EV ( ) NA NA Age group 5 years Ref Ref months ( ) ( ) 6 23 months ( ) ( ) <6 months ( ) ( ) Residence Urban Ref Ref Rural ( ) ( ) Per day increase of onsetto-diagnosis NA 1 1 ( ) NA 1 2 ( ) interval Sex Female Ref Ref Male ( ) ( ) Severe illness Enterovirus serotype CV-A Ref NA NA Other enterovirus ( ) NA NA EV ( ) NA NA Age group 5 years Ref Ref months ( ) ( ) 6 23 months ( ) ( ) <6 months ( ) ( ) Residence Urban Ref Ref Rural ( ) ( ) Per day increase of onsetto-diagnosis NA 1 (1 1 ) NA 1 2 ( ) interval Sex Female Ref Ref Male ( ) ( ) Results from multivariate logistic regression with backward selection of demographic and geographical variables, time from onset to diagnosis, and virus serotype (p value for exclusion 1). OR=odds ratio. EV71=enterovirus 71. CV-A16= Coxsackie virus A16. Ref=reference. NA=not applicable. Table 2: Risk factors associated with severe illness and death in probable cases and laboratory-confirmed cases of hand, foot, and mouth disease Vol 14 April

7 A Heilongjiang Jilin Xinjiang Inner Mongolia Liaoning Beijing Tianjin Hebei Shanxi Ningxia Qinghai Shandong Gansu Henan Shaanxi Jiangsu Anhui Shanghai Hubei Sichuan Zhejiang Tibet Chongqing Jiangxi Hunan Guizhou Fujian Yunnan Guangdong Guangxi Hainan F A J A O F A J A O F A J A O F A J A O F A J A O Number of cases C May 2, 28 statutory requirement Hand, foot, and mouth disease weekly cases B Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec 6 EV71 5 CV-A16 4 Other enterovirus 3 Probable Seasonal hand, foot, and mouth disease distribution 2 F A J A O D F A J A O D F A J A O D F A J A O D F A J A O D Jan 1 to May 1 voluntary enhanced surveillance May 28 to June 29 limited laboratory testing July 29 onwards steady-state laboratory capacity Figure 4: Heat map of surveillance data for hand, foot, and mouth disease by Chinese province, (A) Time series of weekly probable and laboratory-confirmed cases of HFMD, standardised by the number of annual cases. (B) Seasonal distribution of cases of HFMD, plotted as the median value of proportion of cases in each week of the year from 28 to 212. For (A) and (B), the provinces were ordered by latitude from northermost (top) to southernmost (bottom). (C) Number of cases of HFMD by week of illness onset. The insert is a superposition of the number of cases without probable cases of HFMD by week of illness onset. HFMD=hand, foot, and mouth disease. EV71=enterovirus 71. CV-A16=Coxsackie virus A16. Our large-scale analysis of seasonal patterns of hand, foot, and mouth disease identified two main epidemiological regions corresponding to northern China, where hand, foot, and mouth disease peaks in summer, and southern China, which has two hand, foot, and mouth disease peaks in spring and autumn. A highly transmissible, strongly immunising disease needs constant replenishment of susceptibles to sustain semiannual, or even annual, epidemics. Findings from Asia Pacific countries have shown biennial 25 or triennial 2,21 outbreaks against a background total fertility rate of births per woman, 26 compared with China s 1 6 births per woman under the longstanding one-child policy. 27 By contrast, annual epidemics were reported in Japan and Malaysia, 6, Vol 14 April 214

8 consistent with patterns seen in northern China, whereas semiannual epidemics were reported in Hong Kong, south Taiwan of China, and Vietnam, 8,1,21 in line with data for southern China. Further, our data suggest that EV71 epidemics arise annually in China, by contrast with Japan and Malaysia where this serotype circulates every 3 or 4 years. 6,25 Hand, foot, and mouth disease periodicity might be complicated by interference between the causative enterovirus serotypes, perhaps associated with cross-serotype immunity. Overall, the drivers of hand, foot, and mouth disease periodicity are not fully understood and need further study. Although seasonal patterns of hand, foot, and mouth disease were associated with precipitations, sunshine, temperature, and air pressure, no one climatic variable explained the complexity of hand, foot, and mouth disease seasonality across China. Our results confirm those of previous time-series analyses suggesting an association between hand, foot, and mouth disease and climatic factors in Singapore, Hong Kong, southern China, and Japan Moreover, the occurrence of poliomyelitis, a disease caused by another enterovirus, might be associated with humidity. 32 Humidity was a weak predictor of seasonal patterns of hand, foot, and mouth disease in our large Chinese study, which encompasses a greater diversity of climate zones and populations than previously studied. The epidemiological patterns of hand, foot, and mouth disease in Hainan and Tibet were different from those in the other provinces. This divergence could be due to the extreme, unique climate of these two outliers; Hainan is the only tropical region of China and the climate in Tibet is very particular and complex because of the great altitude. We cannot rule out artefactual surveillance biases, in part due to scarce financial and human resources, especially in Tibet. More work is needed to clarify the local drivers of hand, foot, and mouth disease seasonality in southeast Asia. Systematic, population-representative sero epi demiology studies across the age spectrum from mother infant pairs through the first 1 years of life would greatly refine understanding of hand, foot, and mouth disease dynamics in different regions. Previous crosssectional serosurveys have provided age-specific estimates of the cumulative incidence of infection at certain ages Prospective longitudinal serological studies could provide more detailed information about age-specific incidence of infection and disease, risk factors for infection, and the protection against new infections conferred by previous infections with the same or different viruses. Longitudinal serosurveys were instrumental to elucidate the acquisition of infection and immunity in other complex disease systems including competition between viruses, which in turn informed vaccination strategies. 37 EV71 infection in early infancy conferred the most serious risks of clinical severity and fatality. 21,38 Several candidate vaccines against EV71 are undergoing Latitude Latitude A Annual amplitude vs latitude p< Amplitude (%) B Semiannual amplitude vs latitude p= Amplitude (%) C Ratio semiannual component vs latitude 5 Figure 5: Latitudinal gradients in periodicity of hand, foot, and mouth disease (A) Amplitude of the annual periodicity. (B) Amplitude of the semiannual periodicity. (C) Contribution of the semiannual periodicity, measured by the ratio of the amplitude of the semiannual periodicity to the sum of the amplitudes of annual and semiannual periodicities (higher ratio suggests a stronger semiannual periodicity). Symbol size is proportional to the number of cases in each province. Black solid lines represent linear regression fit (regression weighted by mean annual number of cases of hand, foot, and mouth disease). p values are given on the graphs. Colours represent different climatic zones (black=cold-temperate, blue=mid-temperate, green=warm-temperate, red=subtropical, turquoise=tropical). regulatory vetting in China. 39 Whether and how these vaccines should be deployed needs further assessment of cost-effectiveness, feasibility, and contextual considerations in different regions. Transmission models are needed to fully assess the direct and indirect benefits of vaccination, and the risk of serotype replacement in view of the use of a monovalent vaccine. Our severity data suggest two independent risk factors: each extra day since symptom onset was associated with a 1% higher risk of mortality; and those living in a rural area had a quarter excess risk of death from infection. In addition to disparity in health-care access and access to advanced life-sustaining technology between rural and urban settings, 4 widespread and inappropriate use of glucocorticoids and pyrazolones in mild hand, foot, and mouth disease stages 41,42 could have precipitated critical illness and death. This study has several limitations. First, as for any common, self-limiting illness, surveillance is not complete and most cases go undetected because the disorder is asymptomatic, or the patient does not seek formal care or is not diagnosed and reported. Second, access to and provision of health care and technical Latitude p< Ratio semiannual component Vol 14 April

9 Articles Annual amplitude (%) Semiannual ratio Annual phase (weeks) < Figure 6: Periodicity and peak timing of hand, foot, and mouth disease epidemics in China (A) Amplitude of the annual cycle from pale red (low) to red (high). (B) Importance of the semiannual periodicity, measured by the ratio of the amplitude of the semiannual cycle to the sum of the amplitudes of annual and semiannual cycles. Pale green shows strongly annual epidemics; dark green shows dominant semiannual activity. (C) Timing of primary annual hand, foot, and mouth disease peak, in weeks from Jan 1. Timing is colour coded from pale blue to dark blue. A Beijing Shandong Qinghai Hebei Xinjiang Gansu Tianjin Shanxi Ningxia Liaoning Heilongjiang Hunan Guangxi Yunnan Shaanxi Zhejiang Guizhou Shanghai Guangdong Fujian Sichuan Jiangsu Chongqing Anhui Hubei Jiangxi Hainan Henan 1 Inner Mongolia Tibet 2 Jilin Epidemiological distances 5 2 B C Epidemiological region Hours of autumn sunshine Minimum spring relative humidity (%) Figure 7: Hand, foot, and mouth disease epidemiological regions and predictors (A) Epidemiological regions based on hierarchical clustering, identified with the Euclidian distance between weekly standardised hand, foot, and mouth disease time series. Provinces are colour-coded by climatic region (black=cold-temperate, blue=mid-temperate, green= warm temperate, red=subtropical, turquoise=tropical). (B) Map of the two epidemiological regions identified in (A). (C) Climate predictors of the two main clusters identified in (A), based on stepwise discriminant analysis. capacity varied between and sometimes within provinces, with no formal quality assurance or systematic audit for disease surveillance. In particular, 316 we cannot rule out that surveillance bias explained the marked differences in hand, foot, and mouth disease patterns recorded in Hainan and Tibet, relative to other Vol 14 April 214

10 Panel: research in context Systematic review We searched PubMed with the keywords hand, foot, and mouth disease, EV71, enterovirus 71, and coxsackie virus A for all journal articles written in English between Jan 1, 1995, and June 3, 213. In the past 15 years, EV71- associated hand, foot, and mouth disease epidemics have been increasingly reported across the Asia-Pacific region, chronologically including Malaysia, 4 Taiwan, 5 Japan, 6 Singapore, 7 Vietnam, 8 mainland China, 9 Hong Kong, 1 and Cambodia. 11 These reports were mostly descriptive, with very few exceptions that estimated epidemiological parameters, assessed severity, or analysed seasonality. Even fewer had sufficient power to yield definitive conclusions. No previous report had addressed all three sets of interrelated questions. Interpretation We described the enhanced hand, foot, and mouth disease surveillance system, characterised host characteristics, periodicity of epidemics, and geographical patterns from 28 to 212. Incidence of hand, foot, and mouth disease was 1 2 per 1 person-years with 5 9 reported deaths annually during , predominantly among young children. There was strong age dependency, with children aged 6 months to 2 years showing the highest rates. Boys younger than 5 years were 1 6 times more likely than girls of the same age to register disease. Overall, the case-fatality, case-severity, and severe case-fatality rates were 3%, 1 1%, and 3 %, respectively. Illness severity and death by enterovirus serotype were inversely related to age. The median time from illness onset to diagnosis, from illness onset to death, and from diagnosis to death were 1 5 days (IQR days), 3 5 days (IQR days), and 5 day (IQR days) respectively. Hand, foot, and mouth disease tended to arise during the warmer months of the year throughout the country, although peak timing and the periodicity of epidemics varied substantially with latitude, in particular demonstrating stronger semiannual cycles in southern China compared with annual outbreaks in the north. Seasonal variation was associated with precipitation, sunshine, temperature, and barometric pressure. However, no single climatic variable was sufficient to explain the complexity of hand, foot, and mouth disease seasonality across China. provinces. Third, data collected during the first year of rollout are probably less reliable than in more recent years. Fourth, we did not have data for mixed enterovirus infections 43 and were unable to monitor serotype enteroviruses beyond CV-A16 and EV71. In particular, we were unable to monitor serotype Coxsackie A virus 6, which has become more predominant in southern China. 44 Fifth, some severe neurological illnesses caused by EV71 might have been missed by using a strict hand, foot, and mouth disease case definition. Sixth, our study was based on a descriptive analysis of surveillance data. Unfortunately we cannot establish or estimate a threshold for yearly epidemic; this is beyond the scope of our study. In conclusion, we found substantial hand, foot, and mouth disease illness and mortality associated with cocirculating EV71, CV-A16, and other enteroviruses in China, disproportionately affecting children younger than 5 years. Younger age, infection with EV71, and living in a rural area are risk factors for severe disease. We also recorded differences in hand, foot, and mouth disease seasonality across China, which might be associated with climate. Overall, our study provides robust, populationbased, national data to support prioritisation of EV71 vaccination and optimisation of the timing of regional routine vaccination. Further, our results will serve as a prevaccination baseline against which future interventions can be compared. More broadly, further studies are warranted to elucidate the dynamics and immunity patterns of hand, foot, and mouth disease, a rising public health problem in Asia. Contributors HY and GML convieved, designed and supervised the study. WX, QL, CV and JZ designed the study, finalised the analysis, and interpreted the findings. JTW, BJC, VJF, and YZ assisted in data analysis. HY, GML, WX, and QL wrote the drafts of the manuscript. CV, JJF, and JKV interpreted the findings, commented on and helped revise drafts of the manuscript. JS, ZC, and FL participated in collection and management of data. Conflicts of interest We declare that we have no conflicts of interest. Acknowledgments We thank the hospitals, local health departments, and China CDC for assistance in coordinating data collection. We also thank Lin Wang for the data cleaning, and the National Health and Family Planning Commission for supporting this study. The views expressed are those of the authors and do not necessarily represent the policy of the China CDC. References 1 Melnick JK. Enteroviruses: polioviruses, coxsackieviruses, echoviruses, and newer enteroviruses. In: Fields BN, Knipe DM, Howley PM, et al, eds. Field s virology. 3rd edn. Philadelphia: Lippincott-Raven, Ooi MH, Wong SC, Lewthwaite P, Cardosa MJ, Solomon T. Clinical features, diagnosis, and management of enterovirus 71. Lancet Neurol 21; 9: Schmidt NJ, Lennette EH, Ho HH. An apparently new enterovirus isolated from patients with disease of the central nervous system. J Infect Dis 1974; 129: Chan LG, Parashar UD, Lye MS, et al, for the Outbreak Study Group. Deaths of children during an outbreak of hand, foot, and mouth disease in Sarawak, Malaysia: clinical and pathological characteristics of the disease. Clin Infect Dis 2; 31: Ho M, Chen ER, Hsu KH, et al. An epidemic of enterovirus 71 infection in Taiwan. Taiwan Enterovirus Epidemic Working Group. N Engl J Med 1999; 341: Infectious Agents Surveillance Report. Hand, foot and mouth disease in Japan, IASR 212; 33: Chan KP, Goh KT, Chong CY, Teo ES, Lau G, Ling AE. Epidemic hand, foot and mouth disease caused by human enterovirus 71, Singapore. Emerg Infect Dis 23; 9: Tu PV, Thao NT, Perera D, et al. Epidemiologic and virologic investigation of hand, foot, and mouth disease, southern Vietnam, 25. Emerg Infect Dis 27; 13: Zhang Y, Zhu Z, Yang W, et al. An emerging recombinant human enterovirus 71 responsible for the 28 outbreak of hand foot and mouth disease in Fuyang city of China. Virol J 21; 7: Ma E, Chan KC, Cheng P, Wong C, Chuang SK. The enterovirus 71 epidemic in 28: public health implications for Hong Kong. Int J Infect Dis 21; 14: e Vol 14 April

11 11 Seiff A. Cambodia unravels cause of mystery illness. Lancet 212; 38: China Ministry of Health. Guideline for HFMD public health response jszl_2275/296/t29612_2477.html (accessed Jan 9, 213). 13 China Ministry of Health. Protocol of sample collection and laboratory tests for HFMD cases jkzt/crb/szkb/jszl_2275/296/t29612_2477.htm (accessed Jan 9, 213). 14 Zhang Y, Wang J, Guo W, et al. Emergence and transmission pathways of rapidly evolving evolutionary branch C4a strains of human enterovirus 71 in the central plain of China. PLoS One 211; 6: e National Bureau of Statistics of China. National census in China in (accessed Jan 9, 214). 16 China Communication and Transportation Association. Yearbook of China integrated transport. (accessed Jan 9, 214). 17 Met. Data Services. China meteorological data sharing service system. Climatic data. (accessed Nov 11, 213). 18 Yu H, Alonso WJ, Feng L, et al. Characterization of regional influenza seasonality patterns in china and implications for vaccination strategies: spatio-temporal modeling of surveillance data. PLoS Med 213; 1: e Naumova EN, Jagai JS, Matyas B, DeMaria A Jr, MacNeill IB, Griffiths JK. Seasonality in six enterically transmitted diseases and ambient temperature. Epidemiol Infect 27; 135: Ang LW, Koh BK, Chan KP, Chua LT, James L, Goh KT. Epidemiology and control of hand, foot and mouth disease in Singapore, Ann Acad Med Singapore 29; 38: Chen SC, Chang HL, Yan TR, Cheng YT, Chen KT. An eight-year study of epidemiologic features of enterovirus 71 infection in Taiwan. Am J Trop Med Hyg 27; 77: Momoki ST. Surveillance of enterovirus infections in Yokohama city from 24 to 28. Jpn J Infect Dis 29; 62: Wang SM, Ho TS, Lin HC, Lei HY, Wang JR, Liu CC. Reemerging of enterovirus 71 in Taiwan: the age impact on disease severity. Eur J Clin Microbiol Infect Dis 212; 31: Zhu FC, Liang ZL, Meng FY, et al. Retrospective study of the incidence of HFMD and seroepidemiology of antibodies against EV71 and CoxA16 in prenatal women and their infants. PLoS One 212; 7: e Podin Y, Gias EL, Ong F, et al. Sentinel surveillance for human enterovirus 71 in Sarawak, Malaysia: lessons from the first 7 years. BMC Public Health 26; 6: The World Bank. Fertility rate, total (births per woman). worldbank.org/indicator/sp.dyn.tfrt.in (accessed Nov 11, 213). 27 Hesketh T, Lu L, Xing ZW. The effect of China s one-child family policy after 25 years. N Engl J Med 25; 353: Hii YL, Rocklov J, Ng N. Short term effects of weather on hand, foot and mouth disease. PLoS One 211; 6: e Ma E, Lam T, Wong C, Chuang SK. Is hand, foot and mouth disease associated with meteorological parameters? Epidemiol Infect 21; 138: Onozuka D, Hashizume M. The influence of temperature and humidity on the incidence of hand, foot, and mouth disease in Japan. Sci Total Environ 211; : Huang Y, Deng T, Yu S, et al. Effect of meteorological variables on the incidence of hand, foot, and mouth disease in children: a timeseries analysis in Guangzhou, China. BMC Infect Dis 213; 13: Nathanson N, Kew OM. From emergence to eradication: the epidemiology of poliomyelitis deconstructed. Am J Epidemiol 21; 172: Zhu Z, Zhu S, Guo X, et al. Retrospective seroepidemiology indicated that human enterovirus 71 and coxsackievirus A16 circulated wildly in central and southern China before large-scale outbreaks from 28. Virol J 21; 7: Zhu FC, Liang ZL, Meng FY, et al. Retrospective study of the incidence of HFMD and seroepidemiology of antibodies against EV71 and CoxA16 in prenatal women and their infants. PLoS One 212; 7: e Zeng M, El Khatib NF, Tu S, et al. Seroepidemiology of enterovirus 71 infection prior to the 211 season in children in Shanghai. J Clin Virol 212; 53: Yu H, Wang M, Chang H, et al. Prevalence of antibodies against enterovirus 71 in children from Lu an City in Central China. Jpn J Infect Dis 211; 64: Velazquez FR, Matson DO, Calva JJ, et al. Rotavirus infections in infants as protection against subsequent infections. N Engl J Med 1996; 335: Chang LY, King CC, Hsu KH, et al. Risk factors of enterovirus 71 infection and associated hand, foot, and mouth disease/herpangina in children during an epidemic in Taiwan. Pediatrics 22; 19: e Liang Z, Mao Q, Gao F, Wang J. Progress on the research and development of human enterovirus 71 (EV71) vaccines. Front Med 213; 7: Shi L. Health care in China: a rural-urban comparison after the socioeconomic reforms. Bull World Health Organ 1993; 71: Jiang Q, Yu BN, Ying G, et al. Outpatient prescription practices in rural township health centers in Sichuan Province, China. BMC Health Serv Res 212; 12: Ma H, He F, Wan J, et al. Glucocorticoid and pyrazolone treatment of acute fever is a risk factor for critical and life-threatening human enterovirus 71 infection during an outbreak in China, 28. Pediatr Infect Dis J 21; 29: Yan XF, Gao S, Xia JF, Ye R, Yu H, Long JE. Epidemic characteristics of hand, foot, and mouth disease in Shanghai from 29 to 21: Enterovirus 71 subgenotype C4 as the primary causative agent and a high incidence of mixed infections with coxsackievirus A16. Scand J Infect Dis 212; 44: He YQ, Chen L, Xu WB, et al. Emergence, circulation, and spatiotemporal phylogenetic analysis of coxsackievirus a6- and coxsackievirus a1-associated hand, foot, and mouth disease infections from 28 to 212 in Shenzhen, China. J Clin Microbiol 213; 51: Vol 14 April 214