Norovirus Outbreak Surveillance, China, 2016–2018

CaliciNet China, a network of provincial, county, and city laboratories coordinated by the Chinese Centers for Disease Control and Prevention, was launched in October 2016 to monitor the epidemiology and genotype distribution of norovirus outbreaks in China. During October 2016–September 2018, a total of 556 norovirus outbreaks were reported, and positive fecal samples from 470 (84.5%) outbreaks were genotyped. Most of these outbreaks were associated with person-to-person transmission (95.1%), occurred in childcare centers or schools (78.2%), and were reported during November–March of each year (63.5%). During the 2-year study period, 81.2% of all norovirus outbreaks were typed as GII.2[P16]. In China, most norovirus outbreaks are reported by childcare centers or schools; GII.2[P16] is the predominant genotype. Ongoing surveillance by CaliciNet China will provide information about the evolving norovirus genotype distribution and outbreak characteristics important for the development of effective interventions, including vaccines.

from the National Outbreak Reporting System, yielding comprehensive surveillance in the United States that enables multifactorial characterization of norovirus outbreaks (9).
Since 2004, provincial and local Centers for Disease Control and Prevention (CDCs) in China have been required to report all acute gastroenteritis outbreaks of >20 cases, including those caused by noroviruses, to a passive national outbreak surveillance system called Public Health Emergency Event Surveillance System (PHEESS) (12). Although PHEESS provides useful information, detailed epidemiologic and molecular data from norovirus outbreaks are typically not included. In October 2016, the National Institute for Viral Disease Control and Prevention at the China CDC launched CaliciNet China as a surveillance network to monitor the epidemiology and molecular characteristics of norovirus outbreaks. We describe this new network and report data from the first 2 years of surveillance.

CaliciNet China
For the surveillance of norovirus outbreaks, CaliciNet China relies on contributions from laboratory and epidemiologic staff at county, city, and provincial CDCs. County CDCs are responsible for acute gastroenteritis outbreak investigations and collection of clinical specimens. Laboratories of city and provincial CDCs perform norovirus detection and genotyping on the specimens. The national laboratory at China CDC receives and aggregates findings and provides overall quality control. Laboratory staff at these CDCs received training on using the standardized methods of norovirus detection and typing (7). Each laboratory performs proficiency testing once a year for quality assurance.

Outbreak Reporting
CaliciNet China defines norovirus outbreaks as >5 acute gastroenteritis cases within 3 days after exposure in a common setting where >2 samples (whole fecal, rectal swab, or vomitus) had been laboratory confirmed as norovirus. Acute gastroenteritis was defined as >3 events involving loose feces, vomiting, or both within a 24-hour period. The data were collected in collaboration with the local epidemiologists investigating the outbreaks using national guidelines (13). Local CDCs aggregated epidemiologic data for each outbreak, including date of first illness onset, setting, transmission route, number of cases, and type of specimens collected. All data were entered into Microsoft Excel (Microsoft, https://www.microsoft.com). City, county, and provincial CDCs also submitted laboratory results from real-time reverse transcription PCR (rRT-PCR), conventional RT-PCR, and sequences into a BioNumerics version 6.6 database (Applied Maths, https://www.applied-maths.com) using CaliciNet scripts provided by CaliciNet USA (7). Each month, data were sent electronically to China CDC.

Data Management and Analysis
We merged epidemiologic and genotype outbreak data at China CDC and then imported them into SPSS Statistics 25 software (IBM, https://www.ibm. com) for data cleaning and analysis. We generated descriptive statistics on reported outbreaks by setting, transmission route, outbreak size, and province and examined differences between genotypes and mode of transmission, outbreak size, setting, and season using χ 2 tests. We used an α value of p<0.05 to assess statistical significance. Genotypes were assigned by phylogenetic analysis using the UPGMA method with reference sequences used by CaliciNet for capsid typing.

Ethics Review
The China CDC Ethical Review Committee approved CaliciNet China as routine surveillance for norovirus. Because CaliciNet China collects only aggregate data on norovirus outbreaks, we did not analyze any personal identifying information as part of this project. Linkage of specimens to patient name was maintained by provincial-level CDCs and not submitted into the CaliciNet China database.

Genotypes
We obtained genotype information for 470 (84.5%) of the 556 outbreaks. Of the 86 (15.5%) outbreaks with no genotyping results, 72 were not further genotyped by network laboratories and 14 were positive by rRT-PCR but negative by conventional RT-PCR and thus could not be genotyped. Of the typed outbreaks, 430 (91.5%) were GII, 26 (5.5%) were GI, and 14 (2.5%) comprised both GI-and GII-positive samples. Overall, 5 GI genotypes and 12 GII genotypes were detected (Appendix Table). Of 470 genotyped outbreaks, GI genotypes included GI.     Table). Of the 14 outbreaks with samples containing multiple genotypes, 4 occurred in universities, 6 in middle schools, 1 in a primary school, and 3 in childcare centers (Appendix Table).   As found in studies in other countries in the Northern Hemisphere, we found the number of norovirus outbreaks was highest during October-March (17,18). Person-to-person transmission was the dominant route (95.1%), even higher than has been reported in other countries (11,19). Almost all (93.7%) outbreaks in our study occurred in childcare centers and schools (primary schools, middle schools, and universities). These outbreak settings have likewise been identified in other Asia countries, such as Japan and South Korea (18,20), as the most common settings for norovirus outbreaks. In contrast, in the United States and Europe, healthcare facilities (primarily nursing homes and hospitals) are the most commonly reported setting for norovirus outbreaks (11,21,22). The high proportion of norovirus outbreaks in childcare centers and schools seems unique to China and might be associated with the high population density in these settings (23) and the enhanced monitoring and reporting of any outbreaks in schools in China. In 2006, the national government started to require school officials to check and screen children attending kindergarten, primary school, and middle schools each morning for fever, vomiting, or diarrhea (24). This program has helped facilitate the detection and reporting of infectious diseases, including norovirus, at kindergartens and schools. However, we received few norovirus outbreak reports from other settings involving more adults and elderly persons (such as companies, restaurants, and nursing homes). This difference in the collection of norovirus outbreaks between school and other settings may cause potential bias of the strain patterns of norovirus.

Discussion
Previous data from CaliciNet and NoroNet have suggested a correlation of certain genotypes with    (27). The outbreaks caused by multiple genotypes were more often associated with waterborne transmission than foodborne or person-to-person transmission.
The dominant genotype in our study, GII.2[P16], was mainly associated with outbreaks in childcare centers and schools, consistent with a study in Japan, which found GII.2 was the most prevalent genotype in childcare facilities and schools for multiple years (28). In other parts of the world, GII.4 viruses have been reported as the dominant genotype among adults and elderly persons, especially in outbreaks in long-term care facilities (29,30). Our study had only 11 GII.4 outbreaks, and they mainly occurred in childcare centers and primary schools. GII.3[P12] mainly occurred among children in this study, similar to other studies where GII.3 was one of the most common genotypes associated with sporadic norovirus infection, particularly among children (29,(31)(32)(33)(34). We also found that GII.17 outbreaks tended to infect older children and adults rather than younger children. Previous reports indicated that the median age of persons infected by GII.17 (49, range 9-75 years) was significantly higher than that of GII.4 cases (1, range 1-8 years) (35).
Since   (36,45). These observations highlight the need for enhanced global surveillance for potential epidemics of emerging norovirus genotypes, which may have different regional impacts. CaliciNet China uses database scripts provided by CaliciNet USA (16) and updated detection and dual genotyping protocols (14). The dual-typing method is helpful in clarifying the molecular epidemiology of noroviruses, including the identification of newly emerging recombinant viruses. For example, GII.4 Sydney[P31] predominated since it was first recognized in 2012 and subsequently became one of the most successful genotypes causing epidemics globally (8). In November 2015, this virus was gradually, although not completely, replaced by a new recombinant GII.4 virus, GII.P16-GII.4 Sydney (7), and has since caused ≈50% of all norovirus outbreaks in the United States (https://www.cdc.gov/norovirus/ reporting/calicinet/data.html).
Our study has several limitations. First, participation in the network was voluntary, so the distribution of the network does not represent the entire country. Currently, network laboratories are located primarily in eastern and southern China, which generally have better infrastructure. Second, the epidemiologic information for each outbreak collected by CaliciNet China is still limited; however, China CDC is making efforts to include more epidemiologic staff in the network and collect more complete and accurate epidemiologic data. Third, the timeliness of the data reporting in CaliciNet needs to be improved. For a variety of reasons, not all network laboratories were able to submit data to China CDC on a monthly basis. In the future, submission of epidemiologic and laboratory data through a Web-based information system will enable uploading of data in near real time.
CaliciNet China's use of the same protocols as other norovirus surveillance networks, such as Cali-ciNet USA, enables comparison of data internationally and potentially provides an early warning when new strains emerge that have the potential to cause global epidemics. In addition, monitoring changes in the distribution of genotypes can help inform the development and assessment of norovirus vaccines, several of which are in clinical trials (46,47).
In conclusion, we collected information about 556 norovirus outbreaks using standardized epidemiologic definitions and laboratory testing procedures in 6 provinces in China. Person-to-person was the predominant transmission route, and childcare centers and schools were the most common settings for reported norovirus outbreaks. The large number of outbreaks during winter 2016-17 was attributable at least in part to the emergence of the new recombinant genotype GII.2[P16]. Ca-liciNet China provides essential information about the evolving strain distribution and epidemiologic characteristics of norovirus outbreaks, which can contribute to the development of effective vaccines.