Prevalence and Etiological Characteristics of Norovirus Infection in China: A Systematic Review and Meta-Analysis

Norovirus is a common cause of sporadic cases and outbreaks of gastroenteritis worldwide, although its prevalence and the dominant genotypes responsible for gastroenteritis outbreaks remain obscure. A systematic review was conducted on norovirus infection in China between January 2009 and March 2021. A meta-analysis and beta-binomial regression model were used to explore the epidemiological and clinical characteristics of norovirus infection and the potential factors contributing to the attack rate of the norovirus outbreaks, respectively. A total of 1132 articles with 155,865 confirmed cases were included, with a pooled positive test rate of 11.54% among 991,786 patients with acute diarrhea and a pooled attack rate of 6.73% in 500 norovirus outbreaks. GII.4 was the predominant genotype in both the etiological surveillance and outbreaks, followed by GII.3 in the etiological surveillance, and GII.17 in the outbreaks, with the proportion of recombinant genotypes increasing in recent years. A higher attack rate in the norovirus outbreaks was associated with age group (older adults), settings (nurseries, primary schools, etc.) and region (North China). The nation-wide pooled positive rate in the etiological surveillance of norovirus is lower than elsewhere in the global population, while the dominant genotypes are similar in both the etiological surveillance and the outbreak investigations. This study contributes to the understanding of norovirus infection with different genotypes in China. The prevention and control of norovirus outbreaks during the cold season should be intensified, with special attention paid to and enhanced surveillance performed in nurseries, schools and nursing homes from November to March.


Introduction
Norovirus is a common cause of sporadic gastroenteritis in all age groups and is a major cause of acute gastroenteritis outbreaks in humans, accounting for approximately 50% of acute gastroenteritis outbreaks worldwide [1]. Human infection with norovirus brings a significant global disease burden, with about 685 million infections worldwide, and more than 200,000 deaths [2], USD 4.2 billion in direct health system costs, and USD 60.3 billion in social costs every year [3]. Norovirus infection is also an important cause of diarrhea death in children under 5 years of age, suggesting a significant threat to children [4]. Noroviruses have shown genetic diversity in 10 genomes (GI-GX) according to polymerase and capsid gene sequences, of which the GI, GII, GIV, GVII and GIX genomes can infect humans [5]. About 90% of global norovirus outbreaks and sporadic cases are associated with the GII genogroup [6]. Seasonality concerning outbreaks has been reported, with the majority occurring in cool months [7]. Common symptoms of norovirus infection include vomiting, diarrhea, abdominal pain, fever, chills, headache, and myalgia [8]. Norovirus is a self-limiting disease that usually causes mild symptoms [9], but it can lead to serious consequences in children, older adults, and other immunocompromised people infected with norovirus [10,11].
In China, norovirus has also caused a severe disease burden with about 20% of acute gastroenteritis cases attributed to its infection, and GII.4 and GII. 3 have been reported as the dominant genotypes [12]. During 2014-2017, norovirus infection became the most common cause of reported outbreaks of acute gastroenteritis, resulting in more than 30,000 cases reported by the National Public Health Emergency Event Surveillance System (PHEESS) [13]. Nurseries and schools are major settings for norovirus outbreaks and pose a high risk to children and adolescents [13,14]. Most norovirus outbreaks are caused by human-to-human transmission, and some recent outbreaks have been associated with newly emerging recombinant genotypes [14]. Recent studies have described the situation of norovirus outbreaks as epidemic based on epidemic surveillance systems, such as the Public Health Emergency Event Surveillance System and CaliciNet China [13,14]. Fan Yu et al. conducted a systematic review on the epidemiological characteristics, transmission mode and genotype distribution of norovirus outbreaks in China from 2000-2018 [15], suggesting that the epidemiological and clinical characteristics vary between regions, seasons, and genotypes. However, there is a lack of research on genotype distribution and the clinical characteristics of norovirus infection in etiological surveillance in combination with outbreak investigation, and no studies have focused on the factors that influence the outbreak attack rate on a national scale.
In order to strengthen the understanding of the epidemiology and clinical characteristics of human norovirus infection with different genotypes, and to guide the surveillance of sporadic cases and the control of outbreaks caused by norovirus, we conducted a systematic review of the reported outbreaks and etiological surveillance of norovirus infection in China. This study aims to reveal the epidemiological and clinical characteristics and dynamics of norovirus infection with different genotypes, as well as the patterns of norovirus outbreaks in different populations, seasons, regions and settings, and to analyze the factors that influence the attack rate of norovirus outbreaks.

Materials and Methods
This study was registered with the International Prospective Register of Ongoing Systematic Reviews (PROSPERO), CRD42022297065, and was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Appendix S1) [16].

Search Strategy and Selection Criteria
A literature search was performed on four databases using a set of terms and Boolean operators, including the PubMed database (https://pubmed.ncbi.nlm.nih.gov/, accessed on 1 April 2021), China National Knowledge Infrastructure (CNKI, http://www.cnki. net/, accessed on 1 April 2021), Wanfang Database (http://www.wanfangdata.com.cn/, accessed on 1 April 2021), and Chongqing VIP Chinese Science and Technology Journal Database (CQVIP, http://www.cqvip.com.cn/, accessed on 1 April 2021). Different retrieval formulas were developed for these databases based on individual retrieval methods. The standardized medical subject heading (MeSH) term "Norovirus", "Norwalk Virus", and free word "China" were used for the PubMed database, while "Norovirus", "Norwalk virus" and "Norwalk-like virus" were set as the subject heading, title, and keywords to search in CNKI, Wanfang, and CQVIP (Appendix S2 p. 3). All articles published between January 2009 and March 2021 were searched without language restrictions.
Comprehensive inclusion and exclusion criteria were pre-defined to facilitate the screening process. To be included, published reports possessed the following characteristics: (1) the study subjects lived in the mainland of China; (2) noroviruses were reported as the etiologic agent; (3) the study period was between 2009 and 2021; (4) the types of articles included etiological surveillance and outbreak investigation. Reports of the laboratory methods, animal and plant studies, vaccines, model studies, cross-sectional studies in healthy people, health education, molecular mechanisms, and reviews related to norovirus were excluded from the final analysis (Appendix S2 pp. [4][5]. The deduplication and addition of the articles was carried out using Endnote X9 software and manual methods. Titles and abstracts of the retrieved studies were screened using Endnote X9 independently by two reviewers (TTL and MCL) to identify studies that might be eligible for inclusion, and then the full texts of the potentially eligible studies were retrieved and independently assessed for eligibility by the two reviewers. Discrepancies between reviews were resolved by consensus or a third reviewer (QX). Studies potentially describing overlapping data were noted and the duplications were removed (e.g., same hospital and population during an overlapping time period).

Data Extraction and Variable Definitions
The full texts of all the included literature were reviewed and the data were extracted using a standardized form. A total of 20 variables were included (details in Appendix S2 pp. 6-7). Based on the average monthly temperature of each city included in the study, the top six months were classified as the "warm season" and the rest as the "cold season" [17]. A norovirus outbreak was defined as >5 acute gastroenteritis cases within 3 days after exposure in a common setting where >2 samples (whole fecal, rectal swab, or vomitus) tested positive for norovirus. Acute gastroenteritis was defined as >3 events involving loose feces, vomiting, or both, within 24 h [14]. For articles containing information on more than one outbreak, the data of each outbreak were extracted separately. Moreover, outbreaks reported in multiple publications were recorded only once. The Microsoft Excel program (version 2019) was used for data entry.

Data Analysis
The descriptive statistics included a frequency analysis for the categorical variables, medians, and inter-quartile range (IQR) for the continuous variables. We further performed a meta-analysis to estimate the attack rate of the outbreaks, positive detection rate of norovirus for etiological surveillance, and the proportion of clinical manifestations for patients infected with norovirus. Briefly, the proportion of clinical manifestations was estimated by dividing the number of cases with a single clinical presentation by the total number of cases. Heterogeneity was estimated using the statistic of I 2 , of which >50% was considered as significant heterogeneity. If substantial heterogeneity existed, a random effect model was used. Otherwise, the fixed-effects model was preferred to summarize the pooled percentage, as well as a 95% CI [18] (Appendix S3). A subgroup analysis was performed in the meta-analysis to compare the categorical variables between groups. For the estimation of the attack rates in the subgroups, outbreaks with fewer than 30 at-risk individuals were excluded. The principles of the Transparent Reporting of Systematic Reviews and Meta-Analysis were implemented throughout the study. In addition, this investigation expanded upon previous research by exploring the association between norovirus infection and potential risk factors using the beta-binomial (BB) regression model; a p value <0.05 was considered statistically significant [19]. All data analyses were carried out using R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria), meta-analyses were performed using R package "meta" and graphical presentations were performed using R package "ggplot2" and the ArcGIS 10.7 (Esri Inc., Redlands, CA, USA).

Overview of the Publications and Spatiotemporal Characteristics of Patients
The literature search identified a total of 6198 articles. After discarding the duplicates, 3237 articles were screened by the title and abstract, of which 1713 articles were screened by the full text. A total of 1132 articles were included in this study according to the inclusion The literature search identified a total of 6198 articles. After discarding the duplicates, 3237 articles were screened by the title and abstract, of which 1713 articles were screened by the full text. A total of 1132 articles were included in this study according to the inclusion and exclusion criteria, with 632 articles on etiological surveillance and 500 articles on outbreak investigation ( Figure 1, Appendix S4). A total of 155,865 confirmed cases were reported, with the number of cases increasing from 2009 and peaking in 2018 followed by a pattern of decline. The numbers of published articles and outbreak events also showed a similar pattern, both peaking in 2018 ( Figure 2A). A pooled test positive rate of 11.54% (104,896 patients infected with norovirus) was shown among 991,786 patients with acute diarrhea based on the etiological surveillance, with no difference between North and South China. The highest positive rate was shown in adolescents (14.74%) and adults (14.74%), followed by children (14.17%) and older adults (14.05%). The pooled attack rate of norovirus was 6.73% based on the data of 899 outbreak events, with a higher attack rate in North than in South China (10.10% vs. 4.81%, p < 0.05). The largest number of outbreak events was found among adolescents, followed by children, adults and older adults, while the highest attack rate was found among older adults Viruses 2023, 15, 1336 5 of 13 (11.85%), followed by children (9.48%), adolescents (5.53%) and adults (4.55%). Most of the norovirus outbreaks occurred in primary schools, followed by nurseries, secondary schools and universities. Human-to-human transmission was the main transmission model in the o utbreaks (44.65%), which occurred more frequently in the cold season than the warm season ( Table 1, Appendix S2 p. 33). Most outbreak events were published during the period from 2017-2020, with 83.11% (246/296) in North China and 61.27% (367/599) in South China. The highest attack rate was shown in 2016 in North China (15.77%) and in 2012 in South China (7.69%) ( Figure 3A, Appendix S2 p. 12). Based on the 866 outbreak events with monthly information, significant seasonality of norovirus outbreaks was shown, with a peak from November to March of the next year in South China (70.14%, 404/576) and dual peaks, respectively, from March to June and November to December in North China (80%, 232/290) ( Figure 3B, Appendix S2 p. 13). Diarrhea and vomiting were the most common symptoms of norovirus infection in adults and other age groups, respectively, and these also varied between genotypes (details in Appendix S2 pp. 22-28, Appendix S2 p. 34).

Figure 4.
Genotype characteristics of norovirus outbreaks in China by region, age and setting. The monthly number of outbreaks and the attack rate were estimated using a meta-analysis based on the outbreak investigation by virus genotypes (A), and the numbers of outbreaks and reported cases for different norovirus genotypes norovirus by region (B), by age (C), and setting (D). N (n) indicates the number of articles (outbreaks) included in this analysis. For the attack rate, the mean and a 95% CI are presented. * Indicates either a quite high mean attack rate or a wide 95% CI, which was presented for GII.

Influencing Factors of Attack Rate in Norovirus Outbreaks
A total of 216 articles involving 302 outbreak events were included in this analysis. The attack rates varied from 0.06% to 88.89% between the different outbreaks. The betabinomial model showed that norovirus outbreaks had lower attack rates in children (RR: 0. 13

Influencing Factors of Attack Rate in Norovirus Outbreaks
A total of 216 articles involving 302 outbreak events were included in this analysis. The attack rates varied from 0.06% to 88.89% between the different outbreaks. The beta-binomial model showed that norovirus outbreaks had lower attack rates in children (RR: 0. 13 2.41) than in South China. There was no significant difference in the other factors, e.g., the seasons and routes of transmission (Appendix S2 pp. [29][30].

Discussion
In recent years, China has made progress in controlling the diarrheal diseases but the disease burden of norovirus infection remains high [12]. Therefore, we conducted an exhaustive literature collection of the published articles, explored the epidemiological and clinical characteristics and dynamics of norovirus infection caused by different genotypes in China, as well as the patterns of norovirus outbreaks in different populations, seasons, regions and places.
Our research showed that the nation-wide pooled positive detection rate (11.54%) in the etiological surveillance of norovirus was similar to that of the active surveillance of patients with acute diarrhea managed by the Chinese Centers for Disease Control and Prevention (12.47%) [17], which was lower than the global overall detection rate (18%) and that in developing countries (19%) [20]. Our results show that older adults have the highest attack rates during outbreaks and reports of norovirus outbreaks among elderly people in communities and nursing homes have become more frequent in recent years [21,22]. Most current global initiatives to reduce the burden of diarrhea focus on children under five years of age, so the burden of disease in older people is a growing public health challenge that will have increasingly negative consequences if the issue is ignored. Therefore, due attention is required.
Norovirus infection is also known as winter vomiting disease [23]. Our study shows that norovirus outbreaks occurred mainly from November to March, with the fewest in July and August. Our findings were consistent with the data of the Public Health Emergency Event Surveillance System and CaliciNet China, which showed that most norovirus outbreaks occur between October and March, while the outbreaks were less common in July and August [13,14]. The outbreak peaks were different in the North and South, but the lowest outbreak occurred in July and August both in the North and South. The monthly difference in norovirus prevalence between northern and southern China may be associated with temperature and rainfall [24,25]. In China, cooler temperatures from November to March play an important role in the spread of the virus. In general, there are more outbreaks in the cold season than in the warm season, and the global seasonal pattern of norovirus indicates that its epidemic peak is in winter, suggesting that there is some correlation between temperature and norovirus epidemics. At the same time, studies have also suggested that there is a positive correlation between rainfall and seasonality [7]. Different environments and settings also play a decisive role in the spread of norovirus. Nurseries, primary schools, secondary schools and universities were the main outbreak settings, which is consistent with the data from the Public Health Emergency Event Surveillance System and CaliciNet China, which show that norovirus outbreaks occur predominantly in nurseries and schools [13][14][15]. The government requires school staff to check and screen children attending nurseries and schools for fever, vomiting and diarrhea every morning and to report any infectious disease immediately [26]. This may be why nurseries and schools in China have seen the most outbreaks of norovirus. The peak of the outbreak was the same in nurseries, primary schools and secondary schools, but the outbreak peak in universities was different from the other three major sites. In the major settings, the fewest outbreaks occurred in July and August. The attack rate at universities is lower than elsewhere, which is supported by the beta-binomial model. Universities had the lowest attack rate of norovirus outbreaks, probably because university students and staff are healthy adults with occasional close contact, in contrast to nurseries, health care settings, etc., where sick individuals are either treated or are very young or very old and are taken care of, thus, have more close contacts.
GII.4 and the recombinant genotypes were the dominant genotypes in both the etiological surveillance and outbreak investigations, consistent with the surveillance results in several regions of China and in the global population [20,27,28]. GII.3 was mainly observed in etiological surveillance, and most outbreaks events were caused by GII. 17. In some studies, non-GII.4 strains had replaced GII.4 as the predominant causes of norovirus outbreaks since 2014, while in this study, the phenomenon started in 2015, with the difference likely due to publication delays [15,29]. In addition, due to the publication delay, GII.17 was the predominant genotype in the norovirus outbreaks in China since 2015, while other studies have shown that GII.17 has been the major cause of norovirus outbreaks in China since 2014 [29,30]. In addition, consistent with the results of the Japanese study, a new norovirus variant, GII.17 [P17], has been prevalent in Japan since December 2014 and became the dominant genotype of norovirus outbreaks in March 2015 [31]. GII.2 [P16] was the recombination genotype that caused the most outbreaks of norovirus, which was consistent with CaliciNet China's result [14]. In nurseries, norovirus outbreaks were dominated by GII.2, while recombinant genotypes dominated in the norovirus outbreaks in primary schools. In addition, GII.17 and GII.4 were the dominant genotypes in secondary schools and universities, respectively.
The main symptoms of norovirus infection are vomiting and diarrhea, which usually lasts a relatively short time. The clinical presentation of norovirus infection varies by age group and genotype. Diarrhea was the most common symptom among adult patients and vomiting was the most common symptom among patients in other age groups. For infection with different genotypes, the most common symptom was diarrhea in GII.4infected patients and vomiting in GII.17-infected patients, the same as in the Swedish study [32]. In addition, it has been shown that infection with the GII.4 strain leads to more severe consequences than infection with a non-GII.4 strain [23].
There are some limitations to this study. First, our study suffers from publication bias. Most of these studies have been reported from South China, which may affect the comparability of norovirus infection characteristics between North and South China. Second, since our data are from different authors, and most outbreaks do not share the same variables, this limits the number of studies analyzed. We also excluded articles that did not mention specific information for each outbreak in the summary outbreak analysis, resulting in a reduction in the number of outbreaks included in the analysis. Third, the dual nomenclature of ORF1 and VP1 sequences can be used to identify recombinant noroviruses [5], but most of the norovirus genotypes in the literature we have included are determined with partial VP1 sequences, which prevents us from fully studying the recombinant noroviruses. This study contributes to the attainment of an enhanced understanding of norovirus infection in China, demonstrates the need to consider the wider surveillance of norovirus, thereby helping to inform future research and surveillance efforts, and provides an overview of norovirus epidemiology for future vaccine policy decisions.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v15061336/s1. Four Supplementary Appendices, as part of this study, are publicly available and are provided with this paper, including Appendix S1, Appendix S2, Appendix S3 and Appendix S4. Institutional Review Board Statement: Ethical review and approval were waived for this study due to all data were collected from publicly available sources and were de-identified.
Informed Consent Statement: Patient consent was waived as the data were collected from publicly available sources and de-identified.