Transmission potential of influenza A/H7N9, February to May 2013, China
 Gerardo Chowell^{1, 2}Email author,
 Lone Simonsen^{1, 3},
 Sherry Towers^{2},
 Mark A Miller^{1} and
 Cécile Viboud^{1}
DOI: 10.1186/1741701511214
© Chowell et al.; licensee BioMed Central Ltd. 2013
Received: 29 April 2013
Accepted: 30 August 2013
Published: 2 October 2013
Abstract
Background
On 31 March 2013, the first human infections with the novel influenza A/H7N9 virus were reported in Eastern China. The outbreak expanded rapidly in geographic scope and size, with a total of 132 laboratoryconfirmed cases reported by 3 June 2013, in 10 Chinese provinces and Taiwan. The incidence of A/H7N9 cases has stalled in recent weeks, presumably as a consequence of live bird market closures in the most heavily affected areas. Here we compare the transmission potential of influenza A/H7N9 with that of other emerging pathogens and evaluate the impact of intervention measures in an effort to guide pandemic preparedness.
Methods
We used a Bayesian approach combined with a SEIR (SusceptibleExposedInfectiousRemoved) transmission model fitted to daily case data to assess the reproduction number (R) of A/H7N9 by province and to evaluate the impact of live bird market closures in April and May 2013. Simulation studies helped quantify the performance of our approach in the context of an emerging pathogen, where humantohuman transmission is limited and most cases arise from spillover events. We also used alternative approaches to estimate R based on individuallevel information on prior exposure and compared the transmission potential of influenza A/H7N9 with that of other recent zoonoses.
Results
Estimates of R for the A/H7N9 outbreak were below the epidemic threshold required for sustained humantohuman transmission and remained near 0.1 throughout the study period, with broad 95% credible intervals by the Bayesian method (0.01 to 0.49). The Bayesian estimation approach was dominated by the prior distribution, however, due to relatively little information contained in the case data. We observe a statistically significant deceleration in growth rate after 6 April 2013, which is consistent with a reduction in A/H7N9 transmission associated with the preemptive closure of live bird markets. Although confidence intervals are broad, the estimated transmission potential of A/H7N9 appears lower than that of recent zoonotic threats, including avian influenza A/H5N1, swine influenza H3N2sw and Nipah virus.
Conclusion
Although uncertainty remains high in R estimates for H7N9 due to limited epidemiological information, all available evidence points to a low transmission potential. Continued monitoring of the transmission potential of A/H7N9 is critical in the coming months as intervention measures may be relaxed and seasonal factors could promote disease transmission in colder months.
Keywords
Influenza A/H7N9 Transmissibility Reproduction number Spillover Animal reservoir Emerging infection Influenza A/H5N1 Swine influenza Transmission potential China Realtime estimationBackground
An outbreak of novel A/H7N9 influenza virus infections rapidly unfolded in Eastern China, with the first laboratoryconfirmed case identified in Shanghai on 31 March 2013 and a total of 132 laboratoryconfirmed cases and 38 fatalities reported as of 3 June 2013 [1, 2]. Although the number of new A/H7N9 cases has stalled since early May 2013, several features of this virus have heightened concerns for its pandemic potential and prompted an intense public health response from the Chinese authorities and international health organizations. Foremost, the rapid progression of new cases in urban centers in April 2013 and the severity of the disease have been worrisome. Although the exact route of transmission remains unclear, current evidence points to frequent spillovers from a yettobeconfirmed avian reservoir, suspected to involve poultry [3–6]. Although genetic analyses of the novel virus have revealed potential signs of adaptation to mammalian hosts [7], to date, sustained humantohuman transmission has not been established through contact tracing analysis [3, 4] but cannot be ruled out. About 23% [4] of the A/H7N9 patients report having no prior exposure to live animals, underscoring the potential role of transmission by the environment, aerosols and undocumented contacts with infected individuals. Further, recent experimental studies indicate that the A/H7N9 virus is able to spread efficiently among ferrets via direct contact, although airborne transmission is less efficient [8].
A particular cause for concern is the fact that poultry infected with the A/H7N9 virus seem to exhibit relatively mild symptoms [9], which may extend the infectious period in this host. This is in stark contrast to highly pathogenic A/H5N1 influenza viruses, which typically kill poultry within a few days. Silent and undetected A/H7N9 infections in poultry increase the likelihood of zoonotic infections which, in turn, enhance the potential for acquisition of sustained humantohuman transmission properties.
Preliminary studies suggest a low incidence of A/H7N9 infection in chickens and pigeons in affected areas [1, 5]. Nevertheless, live bird markets were preemptively closed and sick birds culled since 6 April 2013 in Shanghai and 16 April 2013 in Zhejiang, which may have slowed down the progression of the outbreak [10]. A quantification of the rate of viral transmission to humans and the effectiveness of intervention measures would be particularly useful to guide public health responses and provide a comprehensive risk assessment of the A/H7N9 threat.
The reproduction number, R, is a key epidemiological tool for assessing the transmission potential of an emerging infection and monitoring the likelihood of largescale outbreaks. Estimates of R >1 signal the potential for an emerging pathogen to generate a major epidemic while R <1 indicates that transmission chains cannot be sustained in the population.
In the case of an emerging infection, obtaining near real time estimates of R is essential to guide intervention strategies. Bayesian estimation approaches [11–13] are naturally wellsuited for situations where epidemiological data are gradually accumulating, due to their flexibility to incorporate prior information. In these approaches, prior information is sequentially updated as more complete outbreak data become available, providing posterior distributions of the epidemiological parameters of interest [11–13]. In contrast, more traditional 'epidemic curve fitting’ approaches have been typically used to provide retrospective estimates of R once the outbreak is over [14–19]. Alternative estimation approaches are based on detailed individuallevel information on prior exposure to suspected animal reservoirs and/or contact with infected patients [20–22].
In this report, we estimate the transmission potential of the influenza A/H7N9 virus by relying on daily official notifications of laboratoryconfirmed cases in mainland China. In particular, we focus on assessing whether the progression of the outbreak is consistent with unsustained humantohuman transmission dynamics in line with R <1 and whether intervention measures may have reduced transmission. Further, we compare R estimates for A/H7N9 with those for other zoonotic pathogens that have recently caused pandemic concern.
Methods
Data sources
Ethics
The dataset of laboratoryconfirmed cases of avian influenza A H7N9 infection was part of a continuing public health investigation of an emerging outbreak and was, therefore, exempt from institutional review board assessment.
Estimation of the reproduction number R
We adopted a sequential Bayesian framework combined with a susceptibleexposedinfectiousremoved (SEIR) transmission model to estimate R for influenza A/H7N9 [11, 12, 23]. Here, the theoretical R value is a fixed (unknown) quantity, and R estimates are updated in a sequential Bayesian framework as data accumulate over time. This approach was previously applied to study the dynamics of the A/H5N1 influenza outbreaks in Asia [23], the 1918 influenza pandemic in San Francisco, USA [12], and the 2009 A/H1N1 influenza pandemic in China [24]. In this model, the population is assumed to be wellmixed. Susceptible individuals (S) come in contact with infectious individuals (I) and progress to the exposed stage (E) with an average latency period of k^{1} days. Exposed individuals (E) then progress to the infectious stage (I), with an average infectious period of γ^{1} days. Both the latent and infectious periods are assumed to be exponentially distributed.
This model assumes that all A/H7N9 cases originate from humantohuman transmission and, hence, provides an upper bound on the transmissibility of A/H7N9. We also conducted simulation studies to assess the performance of this approach in the situation of an emerging pathogen, where most human cases are due to spillover events originating from exposure to an animal reservoir or the environment, and humantohuman transmission is limited [See Additional file 1].
In this approach, both the latent and infectious periods (1/γ, 1/κ) are fixed and, hence, the only parameter to be estimated is R. We made two different assumptions for the latent and infectious periods to illustrate a short infection process consistent with seasonal influenza [25] (k^{1} = 1.5 days and γ^{1} = 1.5 days, so that the generation interval is 3.0 days) and a longer infection process in line with descriptions of the prolonged course of A/H7N9 infections in humans (k^{1} = 3 days, γ^{1} = 3 days, so that the generation interval is 6.0 days) [4, 5, 26].
Bayesian inference of the reproduction number R
We formulate the model in discrete time probabilistic form to account for the discrete nature of the influenza case data and estimate the distribution of R using Bayes’ theorem.
where the denominator is a normalization factor. Hence, Equation (5) defines the sequential Bayesian estimation scheme, where the posterior probability distribution of R can be used as a prior to generate a posterior distribution at the next time step.
We have to set an initial prior on R to initialize the sequential approach at t = 0, which can reflect any a priori knowledge of the disease. Based on preliminary R estimates derived from the exposure history of A/H7N9 patients (see below), we assumed normal distributions centered around 0.2 (SD = 0.2) and 0.5 (SD = 0.2) as initial priors for R; both distributions were lefttruncated at 0. We also consider a more extreme prior center at R = 1 in Additional file 1.
To compute numerically the posterior of R at each daily iteration, we use Equation 5, relying on the posterior from the previous day as the new prior, following [11, 12]. The posterior R distribution was evaluated using 1,000 discrete bins between 0 and 1.5.
Simulation studies
We carried out simulation studies to evaluate the performances of the Bayesian sequential estimation method in the context of an emerging pathogen. Specifically, we simulated A/H7N9 influenza outbreaks using a modified SEIR transmission process including different levels of humantohuman transmission (as measured by R) together with spillover events originating from a hypothetical reservoir. We varied the true R in the range 0.1 to 2.0 and modeled spillover events as a constant daily rate of new infections arising from exposure to the reservoir (α, in the range 1 to 10 infections per day). We used the model to simulate daily outbreak data, applied the Bayesian estimation method to these data, and confronted the estimated R with the true R [see Additional file 1].
These simulations were designed to gauge the level of error associated with neglecting transmission from environmental or animal sources in our main Bayesian estimation approach, and also to assess the sensitivity of R estimates to prior distribution assumptions, under different epidemiological scenarios.
Variance on case series of A/H7N9 influenza
The SEIR transmission model imposes a requirement on the mean of A/H7N9 cases, but variance can be modeled in a more flexible manner. Because we are dealing with disease count data, the most general choice is the Poisson distribution, where the mean equals the variance. As sensitivity analysis we considered a Negative Binomial distribution which allows for greater variance and better accounts for overdispersed data, and assumed the variance to be twice the mean.
Estimating the impact of live bird market closures
To estimate the impact of live bird market closures in the most affected provinces of Shanghai and Zhejiang, we fit an exponential curve with intrinsic growth rate r to the daily case time series in the preintervention period (before 6 April). We used a Negative Binomial log likelihood fit to account for overdispersion in case counts. The 95% confidence intervals on the growth rate were determined from the range of values of r that yield log L = log L_max  s^2/2 where s = 1.96, and L_max is the value of the likelihood at the bestfit value of r[27]. Using the exponential model fit up to 6 April, we forecasted the expected number of A/H7N9 cases in subsequent weeks. We confronted the progression of reported cases past 6 April against that predicted by the preintervention model as an indication of the effectiveness of control measures.
Reproduction number estimates based on individuallevel exposure data
As a complementary method to estimate the R for influenza A/H7N9, we used an approach recently developed by Cauchemez et al. for zoonotic infections [20]. In this approach, R = 1p, where p is the estimated proportion of infected patients arising from direct contact with the A/H7N9 reservoir (scenario 1 in [20]). This approach provides a conservative upper bound on R as it assumes that case detection probability is independent of cluster allocation (while in general, once an index case is identified, other infections in the family are more likely to be detected). This is a reasonable approach when humantohuman transmission is low [20].
An alternative approach to estimate R relies on the average size of chains of humantohuman transmission, as R can be estimated by dividing the number of secondary infections occurring within clusters by the number of primary cases with a direct link to the reservoir [21]. Although there is uncertainty in the exposure history of A/H7N9 patients, the nature of the reservoir of this virus, and cluster sizes and frequency, we can use R estimates based on exposure and contact information [22, 23] to set the initial prior distributions for R in our Bayesian estimation scheme.
Finally, we provide a comparative review of the transmission potential of emerging zoonoses using both individuallevel contact tracing and exposure data and transmission model fitting approaches, with a focus on avian influenza A/H5N1, swine influenza A/H3N2v, seasonal and pandemic influenza, Nipah virus and severe acute respiratory syndrome (SARS).
Results
Influenza A/H7N9 epidemic curves
Reproduction number estimates based on the Bayesian sequential approach
Estimates and 95% credible intervals of the reproduction number, R, for the A/H7N9 influenza outbreak in China
Parameters  R estimate (95% CI)  

Zhejiang  Shanghai  
(k^{1} = 3 days and γ^{1} = 3 days)  0.13 (0.01 to 0.46)  0.15 (0.01 to 0.47) 
(k^{1} = 1.5 days and γ^{1} = 1.5 days)  0.11 (0.003 to 0.42)  0.17 (0.01 to 0.49) 
Sensitivity analyses and simulation studies
A sensitivity analysis on the prior distribution for R confirmed that there was high uncertainty in the posterior estimates of R [see Additional file 1: Figure S1]. However, the posterior mean of R and upper 95% credible interval remained below the epidemic threshold (R = 1) as epidemiological data accumulated, no matter the prior. Further, estimates were robust to assumptions regarding variance in case count data [see Additional file 1: Figure S2].
Next, we simulated outbreak data illustrating the spread of an emerging infection, where human cases originate from both humantohuman transmission and direct contact with a hypothetical reservoir. Simulations indicate that the Bayesian estimation approach tends to overestimate R, especially when the true R is low and spillover events are frequent [see Additional file 1: Figure S3]. However, the upper bound of the credible interval of the Bayesian approach was trustworthy, as it remained below 1.0 whenever the true R <0.6. Further, case data from Shanghai and Zhejiang suggest that the reported rate of spillover transmission from the reservoir was in the order of approximately one daily infection in the preintervention period, which is in the lower (and more favorable) range of our simulations.
Importantly, simulations show a substantial change between prior R distribution (centered at 0.2, as in our main analysis) and posterior R distributions, when the true R is above 0.6 [see Additional file 1: Figure S3]. This suggests that if the true R was above 0.6 for A/H7N9, we would have detected a greater change in posterior distribution than we did in the observed outbreak data. Finally, our simulation studies indicate that the proportion of A/H7N9 patients arising from humantohuman transmission is approximately equal to R, when 0.1 ≤ R ≤0.9 [see Additional file 1: Figure S4].
Additional sensitivity analyses considering longer latent and infectious periods did significantly change R estimates (Table 1). Similarly, assuming a Negative Binomial to model overdispersion in A/H7N9 case data did not significantly affect our estimates [see Additional file 1: Table S1, Figure S2].
Impact of intervention measures
Estimates of the reproduction number for A/H7N9 using alternative approaches
As a complementary analysis, we present R estimates for A/H7N9 based on alternative approaches relying on individuallevel information on prior exposure and contacts with infected patients [20, 21].
Comparison of reproduction number estimates for the A/H7N9 influenza viruses, other emerging zoonoses with pandemic potential, and human influenza viruses
Outbreak  R estimate  Source and method 

A/H7N9 outbreak  
Avian influenza A/H7N9 2013, China  0.1 (95% CrI: 0.01 to 0.49)  This study; Bayesian approach from [11] 
Avian influenza A/H7N9 2013, China  0.03 to 0.05  This study; exposurebased approach from [20] 
Avian influenza A/H7N9 2013, China  0.28 (95% CI: 0.11 to 0.45)  Analysis of cluster size distribution from [22] 
Other zoonotic influenza viruses  
Avian influenza H5N1 2003 to 2006, SE Asia and Egypt/Turkey  0.29  
Avian influenza H5N1 – 2004 to 2006; SE Asia and Egypt/Turkey  0.52 to 0.54  [11] Bayesian approach 
Swine influenza H3N2v  2011, USA  0.5 to 0.74  
Human influenza viruses  
1918 A/H1N1 influenza pandemic  1.8 to 5.4  
1957 A/H2N2 influenza pandemic  1.5  [33] growth rate 
1968 A/H3N2 influenza pandemic  1.5  [33] growth rate 
2009 A/H1N1 influenza pandemic  1.2 to 3.1  
Seasonal influenza  1.3  
Other zoonotic viruses  
Nipah virus, Malaysia, 1990s  0.05 to 0.08  
Nipah virus, Bangladesh, 2000s  0.48 to 0.51  Exposurebased and cluster size distribution approaches [20]; to data from [21] 
SARS virus, Singapore, Hong Kong, 2003  2.2 to 3.6  [15, 44] Epidemic model fitted to case series during the preintervention period 
An alternative R estimate is provided by the average distribution of secondary chains of transmission. If we assume that all three A/H7N9 clusters represents one spillover event (primary case) followed by one to two serial transmission events, we obtain R = 4/126 = 0.03. Inclusion of one additional suspected cluster of size two identified by contact tracing [see Additional file 1] results in a slightly higher estimate of R = 5/126 = 0.04. Hence, information on individuallevel exposure and cluster size distribution indicates that R is approximately 0.03 to 0.53, consistent with the broad range of uncertainty obtained in the Bayesian approach.
Comparison of transmissibility estimates between influenza A/H7N9 and other zoonotic viruses
Table 2 presents a comparison of R for the A/H7N9 influenza virus, zoonotic influenza viruses, seasonal and pandemic influenza viruses and other viruses of pandemic concern. Estimates are based on a variety of approaches, including transmission model fitting methods and individuallevel exposure history approaches (See Additional file 1 for details).
We compiled R estimates for zoonotic influenza viruses that episodically cause human infections, in particular for avianorigin A/H5N1 and swineorigin A/H3N2v. Estimates in the range 0.52 to 0.54 have been proposed for A/H5N1 in Thailand and Indonesia, based on a Bayesian approach similar to that used here [11]. Using the ratio of secondary infections to primary cases [29], we obtain R approximately 0.29 in this period of relatively intense H5N1 activity.
The H3N2v swineorigin influenza virus has recently become a cause of concern in the US, especially in the context of agricultural fairs in 2011 and 2012. Information on the proportion of patients with direct exposure to swine [30] suggests that R is approximately 0.67. Other approaches making more complex assumptions about surveillance intensity and overdispersion in the distribution of secondary cases indicate that R is approximately 0.5 to 0.74 [20].
In the case of seasonal and pandemic influenza outbreaks, modelfitting approaches reveal that R is 1.3 on average for seasonal outbreaks [41, 42] and 1.2 to 5.4 for pandemic viruses, with the highest estimates associated with the lethal 1918 pandemic [16–18, 31–40] (Table 2).
Nipah virus is another emerging viral zoonosis worth comparing to influenza A/H7N9 (Table 2). Early outbreaks in Malaysia in the late 1990s were associated with low transmission potential, as most cases had direct exposure to swine, with R = 0.05 to 0.08 [43]. In contrast, more recent outbreaks in Bangladesh in 2001 to 2007 were characterized by a higher frequency of humantohuman transmission, with R approximately 0.51 [20, 21]. A similar estimate was obtained by analyzing the cluster size distribution [21].
Table 2 also provides data for the SARS outbreak in 2003, with an estimated R in the range 2.2 to 3.7 based on fitting transmission models to the progression of weekly cases before intervention took place [15, 44]. Hence, taken together, the influenza A/H7N9 virus currently has relatively low estimated transmission potential relative to other zoonotic viruses, although confidence intervals are broad.
Discussion
We have provided near realtime estimates of the transmission potential of the emerging A/H7N9 influenza outbreak in China by applying different methodological approaches to official notifications of laboratoryconfirmed cases. Although there is relatively limited information in the A/H7N9 case data at this point, all available evidence points to R estimates well below 1.0 in Shanghai and Zhejiang provinces, where the majority of cases have been reported. Instead, a deceleration in growth rate in mid April is consistent with the effectiveness of preemptive live bird market closures initiated in early April. Comparison between A/H7N9 and other zoonotic threats suggests a relatively low transmission potential relative to that of other avian or swine influenza viruses and recent Nipah viruses, although further data are necessary to confirm this result.
Our Bayesian SEIR estimation approach assumes that all infections originate from humantohuman transmission and, hence, yields 'worstcase scenario’ R estimates. Our estimation framework was robust to assumptions about the duration of the infectious and latent periods, whether we considered a short serial interval characteristic of seasonal influenza [25] or a prolonged disease course more consistent with early case descriptions [4, 26]. In contrast, the Bayesian approach was very sensitive to assumptions regarding the prior distribution of R, which dominated the inference process. Using assumptions reasonably guided by information on prior patient exposure and the frequency of family clusters, this approach indicates a R well below the epidemic threshold (R = 1.0) in Eastern China. Further, simulation studies suggest that if the true R was above 0.6, we would see a greater shift from prior to posterior distributions than seen in the A/H7N9 data, confirming the low transmission potential of this virus.
Alternative estimation approaches based on individual level contact tracing and prior exposure suggest a range of R of 0.03 to 0.53, in line with a recent modeling study analyzing the cluster size distribution of A/H7N9 cases [22]. These low R estimates are consistent with the results of intense efforts by the Chinese health authorities to monitor contacts of infected cases, which have so far revealed only limited instances of secondary transmission [4]. While the occurrence of three (perhaps four) family clusters of A/H7N9 cases is consistent with short chains of human to human transmission, these clusters do not rule out exposure to common environmental or animal sources. Taken together, information from contact surveys [4] and available R estimates are consistent with a predominance of spillover events from a hypothetical reservoir.
We observed a reduction in the growth rate of H7N9 cases in mid to late April, coinciding with the closure of live bird markets in Shanghai, Zhejiang and large Chinese cities in response to the evolving outbreak. The deceleration in the growth rate was significant in our data as early as 18 April, a period when the effectiveness of these measures was still being debated [45]. Our model is illequipped, however, to predict the progression of the outbreak in the coming weeks if intervention measures are relaxed [46], as information is lacking on the residual prevalence of A/H7N9 in poultry populations in China. Further, we cannot rule out a subsequent rise in A/H7N9 transmission potential in the coming months, as seasonal factors could affect virus prevalence in the (presumed) avian reservoir and promote aviantohuman and possibly humantohuman transmission [47, 48].
We have provided transmissibility estimates for influenza A/H7N9 and other zoonoses using several approaches, which rely on very different assumptions. The Bayesian SEIR modelfitting approach is based on the progression of case incidence; our analyses suggest that currently available A/H7N9 data provide relatively limited information, so that the inference process is heavily dependent on the prior (see also more extreme priors in Additional file 1: Figure S6). This likely stems from the small number of A/H7N9 cases available for study (n = 70 in the two main provinces), in part resulting from the low transmission potential of A/H7N9. Simulations were particularly helpful in showing that if the true R was above 0.6, then we would have most likely identified a shift in the posterior distribution. The lack of observed shift is further evidence that R is low and most likely below 0.6.
In the context of subcritical outbreaks (R <1), alternative methods based on contact tracing and exposure information are attractive, although they depend heavily on prior knowledge of the ecology of the disease. These methods rely on estimates of the proportions of cases arising from humantohuman transmission versus direct exposure to the reservoir [20, 21] and, hence, assume that the reservoir is well known and that onset dates and serial intervals can be accurately determined. Further, methods relying on cluster size distribution are more sensitive to reporting schemes than growth rate methods (for example, if clusters are more likely to be reported once a family member is infected) [22].
Information regarding the reservoir of A/H7N9 and the natural history of this disease is still limited, as would be the case for any emerging zoonosis with limited prior experience. It is intriguing that 23% of A/H7N9 cases do not report any prior contact with poultry (suggesting R is approximately 0.23), and yet clusters are extremely infrequent (suggesting R closer to 0). These conflicting findings could be reconciled with additional information on the prevalence of asymptomatic infections; unfortunately, recent serological information is currently lacking. Overall, all R estimation methods tend to produce high uncertain ranges for A/H7N9. In a similar context, early estimates of the transmissibility of the MERSCoV virus using a related approach were relatively broad, with confidence intervals ranging between 0.5 and 1.1 [49]. A quantitative comparison of the performances of these approaches would be useful in the future as these methods are increasingly applied to characterize the pandemic potential of emerging pathogens (see also [22]).
This study is subject to limitations. First, A/H7N9 incidence could be underreported. However, serological surveys conducted at the end of 2012 in China and Vietnam revealed low levels of prior infections [50, 51]. Moreover, influenzalikeillness surveillance suggests that A/H7N9 infection was an uncommon cause of illness in any age group during March and April 2013 in the most affected areas of China [52]. Our estimates are resilient to underreporting issues as long as the observed case series closely tracks the true course of the outbreak. If case detection had improved over time with increased detection capabilities, this would have artificially quickened the progression of reported cases and, in turn, spuriously overestimated the epidemic growth rate and R. Hence, because of likely increased sampling intensity as the outbreak progressed, we can view our R estimates as upper bounds of the true value.
Second, we have used a simple model to estimate R, relying on a SEIR transmission model typically used for human diseases, while in fact there is likely very little transmission between humans. Our simulations suggest that in the context of frequent spillover events arising from a reservoir, our estimates of R are inflated (consistent with providing worstcase scenarios of the true humantohuman transmission potential of A/H7N9). However, our approach accurately predicts whether an emerging pathogen remains below the critical epidemic threshold (R <1). A more refined approach could integrate more information regarding the hypothetical reservoir and the probability of contacts with humans, and could estimate the relative contribution of each component to overall disease transmission. The yet unresolved nature of the reservoir of A/H7N9 and its ecology hampers the calibration of such models.
Third, our model assumes homogeneous mixing, which may not be valid. We have focused on provincespecific data, which provides a better approximation of wellmixed populations than nationallyaggregated data, especially as most cases arose from large cities (especially Shanghai). Still, there could be residual spatial heterogeneity, which may artificially decrease the estimated R. Overall, our very generic model only requires information on the date of symptoms onset and could be applicable to a variety of emerging infections that include spillovers from a putative reservoir and humantohuman transmission.
Conclusion
In conclusion, we have shown that the available epidemiological data on influenza A/H7N9 are consistent with subcritical transmission potential below R = 0.6 in the first three months of virus circulation in Shanghai and Zhejiang provinces, suggesting infrequent humantohuman transmission events. A decline in the growth rate of influenza A/H7N9 cases in April 2013 highlights the beneficial impact of live bird market closures. The estimated transmission potential of A/H7N9 appears lower than that of other zoonotic threats, although uncertainty remains important due to limited statistical information in the available data. Our proposed approach could be useful to quantify the progression of the outbreak and the impact of control measures in the coming months and help monitor the pandemic potential of this emerging pathogen in near realtime.
Abbreviations
 R:

Reproduction number
 SARS:

Severe acute respiratory syndrome
 SEIR:

Susceptibleexposedinfectiousremoved.
Declarations
Acknowledgments
We are thankful to Drs Hongjie Yu and Liao Qiaohong, China CDC for providing access to official notifications of influenza A/H7N9 cases in China and information on exposure history. We thank Aimee Mead, Fogarty International Center, NIH, for editorial assistance.
This research was conducted in the context of the Multinational Influenza Seasonal Mortality Study (MISMS), an ongoing international collaborative effort to understand influenza epidemiological and evolutionary patterns, led by the Fogarty International Center, National Institutes of Health (http://www.origem.info/misms/index.php). Funding for this project comes in part (LS) from the RAPIDD program of the Science & Technology Directorate, Department of Homeland Security, and from the Office of Global Affairs’ International Influenza Unit in the Office of the Secretary of the Department of Health and Human Services.
Authors’ Affiliations
References
 Butler D: Mapping the H7N9 avian flu outbreaks. Nature. 2013, doi:10.1038/nature.2013.12863Google Scholar
 World Health Organization: Human infection with avian influenza A(H7N9) virus in China Update on May 17th. 2013, Available online: http://www.who.int/csr/don/2013_05_17/en/index.html Google Scholar
 Horby P: H7N9 is a virus worth worrying about. Nature. 2013, 496: 39910.1038/496399a.View ArticlePubMedGoogle Scholar
 Li Q, Zhou L, Zhou M, Chen Z, Li F, Wu H, Xiang N, Chen E, Tang F, Wang D, Meng L, Hong Z, Tu W, Cao Y, Li L, Ding F, Liu B, Wang M, Xie R, Gao R, Li X, Bai T, Zou S, He J, Hu J, Xu Y, Chai C, Wang S, Gao Y, Jin L, et al: Preliminary report: epidemiology of the avian influenza A (H7N9) outbreak in China. N Eng J Med. in press
 Chen Y, Liang W, Yang S, Wu N, Gao H, Sheng J, Yao H, Wo J, Fang Q, Cui D, Li Y, Yao X, Zhang Y, Wu H, Zheng S, Diao H, Xia S, Zhang Y, Chan KH, Tsoi HW, Teng JL, Song W, Wang P, Lau SY, Zheng M, Chan JF, To KK, Chen H, Li L, Yuen KY: Human infections with the emerging avian influenza A H7N9 virus from wet market poultry: clinical analysis and characterisation of viral genome. Lancet. 2013, 381: 19161925. 10.1016/S01406736(13)609034.View ArticlePubMedGoogle Scholar
 CIDRAP News: H7N9 gene study links patient and poultrymarket viruses. Available from: http://www.cidrap.umn.edu/cidrap/content/influenza/avianflu/news/apr2513poultry.html. 25 April 2013
 Gao R, Cao B, Hu Y, Feng Z, Wang D, Hu W, Chen J, Jie Z, Qiu H, Xu K, Xu X, Lu H, Zhu W, Gao Z, Xiang N, Shen Y, He Z, Gu Y, Zhang Z, Yang Y, Zhao X, Zhou L, Li X, Zou S, Zhang Y, Li X, et al: Human infection with a novel avianorigin influenza A (H7N9) virus. N Engl J Med. 2013, 368: 18881897. 10.1056/NEJMoa1304459.View ArticlePubMedGoogle Scholar
 Zhu H, Wang D, Kelvin DJ, Li L, Zheng Z, Yoon SW, Wong SS, Farooqui A, Wang J, Banner D, Chen R, Zheng R, Zhou J, Zhang Y, Hong W, Dong W, Cai Q, Roehrl MH, Huang SS, Kelvin AA, Yao T, Zhou B, Chen X, Leung GM, Poon LL, Webster RG, Webby RJ, Peiris JS, Guan Y, Shu Y: Infectivity, transmission, and pathology of human H7N9 influenza in ferrets and pigs. Science. 2013, 341: 183186. 10.1126/science.1239844.View ArticlePubMedGoogle Scholar
 From SARS to H7N9: will history repeat itself?. Lancet. 2013, 381: 133310.1016/S01406736(13)60865X.
 Shadbolt P: WHO: H7N9 virus 'one of the most lethal so far’. CNN. Available online from: http://www.cnn.com/2013/04/24/world/asia/chinabirdflu/index.html. 26 April 2013
 Bettencourt LM, Ribeiro RM: Real time bayesian estimation of the epidemic potential of emerging infectious diseases. PLoS One. 2008, 3: e218510.1371/journal.pone.0002185.View ArticlePubMedPubMed CentralGoogle Scholar
 Chowell G, Nishiura H, Bettencourt LM: Comparative estimation of the reproduction number for pandemic influenza from daily case notification data. J R Soc Interface. 2007, 4: 155166. 10.1098/rsif.2006.0161.View ArticlePubMedGoogle Scholar
 Birrell PJ, Ketsetzis G, Gay NJ, Cooper BS, Presanis AM, Harris RJ, Charlett A, Zhang XS, White PJ, Pebody RG, De Angelis D: Bayesian modeling to unmask and predict influenza A/H1N1pdm dynamics in London. Proc Natl Acad Sci U S A. 2011, 108: 1823818243. 10.1073/pnas.1103002108.View ArticlePubMedPubMed CentralGoogle Scholar
 Chowell G, Fenimore PW, CastilloGarsow MA, CastilloChavez C: SARS outbreaks in Ontario, Hong Kong and Singapore: the role of diagnosis and isolation as a control mechanism. J Theor Biol. 2003, 224: 18. 10.1016/S00225193(03)002285.View ArticlePubMedGoogle Scholar
 Lipsitch M, Cohen T, Cooper B, Robins JM, Ma S, James L, Gopalakrishna G, Chew SK, Tan CC, Samore MH, Fisman D, Murray M: Transmission dynamics and control of severe acute respiratory syndrome. Science. 2003, 300: 19661970. 10.1126/science.1086616.View ArticlePubMedPubMed CentralGoogle Scholar
 Chowell G, Ammon CE, Hengartner NW, Hyman JM: Estimation of the reproductive number of the Spanish flu epidemic in Geneva, Switzerland. Vaccine. 2006, 24: 67476750. 10.1016/j.vaccine.2006.05.055.View ArticlePubMedGoogle Scholar
 Nishiura H, CastilloChavez C, Safan M, Chowell G: Transmission potential of the new influenza A(H1N1) virus and its agespecificity in Japan. Euro Surveill. 2009, 14.Google Scholar
 Mills CE, Robins JM, Lipsitch M: Transmissibility of 1918 pandemic influenza. Nature. 2004, 432: 904906. 10.1038/nature03063.View ArticlePubMedGoogle Scholar
 Viboud C, Tam T, Fleming D, Handel A, Miller MA, Simonsen L: Transmissibility and mortality impact of epidemic and pandemic influenza, with emphasis on the unusually deadly 1951 epidemic. Vaccine. 2006, 24: 67016707. 10.1016/j.vaccine.2006.05.067.View ArticlePubMedGoogle Scholar
 Cauchemez S, Epperson S, Biggerstaff M, Swerdlow D, Finelli L, Ferguson NM: Using routine surveillance data to estimate the epidemic potential of emerging zoonoses: application to the emergence of US swine origin influenza A H3N2v virus. PLoS Med. 2013, 10: e100139910.1371/journal.pmed.1001399.View ArticlePubMedPubMed CentralGoogle Scholar
 Luby SP, Hossain MJ, Gurley ES, Ahmed BN, Banu S, Khan SU, Homaira N, Rota PA, Rollin PE, Comer JA, Kenah E, Ksiazek TG, Rahman M: Recurrent zoonotic transmission of Nipah virus into humans, Bangladesh, 2001–2007. Emerg Infect Dis. 2009, 15: 12291235. 10.3201/eid1508.081237.View ArticlePubMedPubMed CentralGoogle Scholar
 Nishiura H, Mizumoto K, Ejima K: How to interpret the transmissibility of novel influenza A(H7N9): an analysis of initial epidemiological data of human cases from China. Theor Biol Med Model. 2013, 10: 3010.1186/174246821030.View ArticlePubMedPubMed CentralGoogle Scholar
 Bettencourt LM, Ribeiro RM, Chowell G, Lant T, CastilloChavez C: Towards real time epidemiology: data assimilation, modeling and anomaly detection of health surveillance data streams. Intelligence and security informatics: biosurveillance. Lecture Notes in Comput Sci. 2007, 4506: 7990. 10.1007/9783540726081_8.View ArticleGoogle Scholar
 Yang F, Yuan L, Tan X, Huang C, Feng J: Bayesian estimation of the effective reproduction number for pandemic influenza A H1N1 in Guangdong Province, China. Ann Epidemiol. 2013, 23: 301306.View ArticlePubMedGoogle Scholar
 Ferguson NM, Cummings DA, Cauchemez S, Fraser C, Riley S, Meeyai A, Iamsirithaworn S, Burke DS: Strategies for containing an emerging influenza pandemic in Southeast Asia. Nature. 2005, 437: 209214. 10.1038/nature04017.View ArticlePubMedGoogle Scholar
 Gao HN, Lu HZ, Cao B, Du B, Shang H, Gan JH, Lu SH, Yang YD, Fang Q, Shen YZ, Xi XM, Gu Q, Zhou XM, Qu HP, Yan Z, Li FM, Zhao W, Gao ZC, Wang GF, Ruan LX, Wang WH, Ye J, Cao HF, Li XW, Zhang WH, Fang XC, He J, Liang WF, Xie J, Zeng M, et al: Clinical findings in 111 cases of influenza A (H7N9) virus infection. N Engl J Med. 2013, 368: 22772285. 10.1056/NEJMoa1305584.View ArticlePubMedGoogle Scholar
 Cowan G: Statistical Data Analysis. 1998, Oxford: Oxford University PressGoogle Scholar
 FluTrackers: Laboratoryconfirmed A/H7N9 influenza case series. Available online from: http://www.flutrackers.com/forum/showthread.php?t=202713 (Last accessed on 25 April 2013)
 Pitzer VE, Olsen SJ, Bergstrom CT, Dowell SF, Lipsitch M: Little evidence for genetic susceptibility to influenza A (H5N1) from family clustering data. Emerg Infect Dis. 2007, 13: 10741076. 10.3201/eid1307.061538.View ArticlePubMedPubMed CentralGoogle Scholar
 Lindstrom S, Garten R, Balish A, Shu B, Emery S, Berman L, Barnes N, Sleeman K, Gubareva L, Villanueva J, Klimov A: Human infections with novel reassortant influenza A(H3N2)v viruses, United States, 2011. Emerg Infect Dis. 2012, 18: 834837. 10.3201/eid1805.111922.View ArticlePubMedPubMed CentralGoogle Scholar
 Andreasen V, Viboud C, Simonsen L: Epidemiologic characterization of the 1918 influenza pandemic summer wave in Copenhagen: implications for pandemic control strategies. J Infect Dis. 2008, 197: 270278. 10.1086/524065.View ArticlePubMedPubMed CentralGoogle Scholar
 Chowell G, Viboud C, Simonsen L, Miller MA, AcunaSoto R, Diaz JM, MartinezMartin AF: The 1918–19 influenza pandemic in Boyaca, Colombia. Emerg Infect Dis. 2012, 18: 4856. 10.3201/eid1801.101969.View ArticlePubMedPubMed CentralGoogle Scholar
 Viboud C, Tam T, Fleming D, Miller MA, Simonsen L: 1951 influenza epidemic, England and Wales, Canada, and the United States. Emerg Infect Dis. 2006, 12: 661668. 10.3201/eid1204.050695.View ArticlePubMedPubMed CentralGoogle Scholar
 Boëlle PY, Ansart S, Cori A, Valleron AJ: Transmission parameters of the A/H1N1 (2009) influenza virus pandemic: a review. Influenza Other Respi Viruses. 2011, 5: 306316. 10.1111/j.17502659.2011.00234.x.View ArticleGoogle Scholar
 Fraser C, Donnelly CA, Cauchemez S, Hanage WP, Van Kerkhove MD, Hollingsworth TD, Griffin J, Baggaley RF, Jenkins HE, Lyons EJ, Jombart T, Hinsley WR, Grassly NC, Balloux F, Ghani AC, Ferguson NM, Rambaut A, Pybus OG, LopezGatell H, AlpucheAranda CM, Chapela IB, Zavala EP, Guevara DM, Checchi F, Garcia E, Hugonnet S, Roth C, WHO Rapid Pandemic Assessment Collaboration: Pandemic potential of a strain of influenza A (H1N1): early findings. Science. 2009, 324: 15571561. 10.1126/science.1176062.View ArticlePubMedPubMed CentralGoogle Scholar
 Yang Y, Sugimoto JD, Halloran ME, Basta NE, Chao DL, Matrajt L, Potter G, Kenah E, Longini IM: The transmissibility and control of pandemic influenza A (H1N1) virus. Science. 2009, 326: 729733. 10.1126/science.1177373.View ArticlePubMedPubMed CentralGoogle Scholar
 Munayco CV, Gomez J, LagunaTorres VA, Arrasco J, Kochel TJ, Fiestas V, Garcia J, Perez J, Torres I, Condori F, Nishiura H, Chowell G: Epidemiological and transmissibility analysis of influenza A(H1N1)v in a southern hemisphere setting: Peru. Euro Surveill. 2009, 14.Google Scholar
 White LF, Wallinga J, Finelli L, Reed C, Riley S, Lipsitch M, Pagano M: Estimation of the reproductive number and the serial interval in early phase of the 2009 influenza A/H1N1 pandemic in the USA. Influenza Other Respi Viruses. 2009, 3: 267276. 10.1111/j.17502659.2009.00106.x.View ArticlePubMed CentralGoogle Scholar
 Nishiura H, Chowell G, Safan M, CastilloChavez C: Pros and cons of estimating the reproduction number from early epidemic growth rate of influenza A (H1N1) 2009. Theor Biol Med Model. 2010, 7: 110.1186/1742468271.View ArticlePubMedPubMed CentralGoogle Scholar
 Katriel G, Yaari R, Huppert A, Roll U, Stone L: Modelling the initial phase of an epidemic using incidence and infection network data, H1N1 pandemic in Israel as a case study. J R Soc Interface. 2009, 2011: 856867.Google Scholar
 Chowell G, Miller MA, Viboud C: Seasonal influenza in the United States, France, and Australia: transmission and prospects for control. Epidemiol Infect. 2007, 136: 852864.PubMedPubMed CentralGoogle Scholar
 Chowell G, Viboud C, Simonsen L, Miller M, Alonso WJ: The reproduction number of seasonal influenza epidemics in Brazil, 1996–2006. Proc Biol Sci. 2010, 277: 18571866. 10.1098/rspb.2009.1897.View ArticlePubMedPubMed CentralGoogle Scholar
 Parashar UD, Sunn LM, Ong F, Mounts AW, Arif MT, Ksiazek TG, Kamaluddin MA, Mustafa AN, Kaur H, Ding LM, Othman G, Radzi HM, Kitsutani PT, Stockton PC, Arokiasamy J, Gary HE, Anderson LJ: Case–control study of risk factors for human infection with a new zoonotic paramyxovirus, Nipah virus, during a 1998–1999 outbreak of severe encephalitis in Malaysia. J Infect Dis. 2000, 181: 17551759. 10.1086/315457.View ArticlePubMedGoogle Scholar
 Riley S, Fraser C, Donnelly CA, Ghani AC, AbuRaddad LJ, Hedley AJ, Leung GM, Ho LM, Lam TH, Thach TQ, Chau P, Chan KP, Lo SV, Leung PY, Tsang T, Ho W, Lee KH, Lau EM, Ferguson NM, Anderson RM: Transmission dynamics of the etiological agent of SARS in Hong Kong: impact of public health interventions. Science. 2003, 300: 19611966. 10.1126/science.1086478.View ArticlePubMedGoogle Scholar
 World Health Organization: Frequently Asked Questions on human infection caused by the avian influenza A(H7N9) virus Update as of 30. 2013, http://www.who.int/influenza/human_animal_interface/faq_H7N9/en/, April .Google Scholar
 CIDRAP news: NEWS SCAN: H7N9 emergency scalebacks, H7N7 in Germany. 2009, http://www.cidrap.umn.edu/cidrap/content/influenza/avianflu/news/may1713scan.html, H1N1 interventions, low US flu activity.Google Scholar
 Shaman J, Kohn M: Absolute humidity modulates influenza survival, transmission, and seasonality. Proc Natl Acad Sci USA. 2009, 106: 32433248. 10.1073/pnas.0806852106.View ArticlePubMedPubMed CentralGoogle Scholar
 Shaman J, Pitzer VE, Viboud C, Grenfell BT, Lipsitch M: Absolute humidity and the seasonal onset of influenza in the continental United States. PLoS Biol. 2010, 8: e100031610.1371/journal.pbio.1000316.View ArticlePubMedPubMed CentralGoogle Scholar
 Breban R, Riou J, Fontanet A: Interhuman transmissibility of Middle East respiratory syndrome coronavirus: estimation of pandemic risk. Lancet. 2013, 382: 694699. 10.1016/S01406736(13)614920.View ArticlePubMedGoogle Scholar
 Bai T, Zhou J, Shu Y: Serologic study for influenza A (H7N9) among highrisk groups in China. N Engl J Med. 2013, 368: 23392340. 10.1056/NEJMc1305865.View ArticlePubMedGoogle Scholar
 Boni MF, Chau NV, Dong N, Todd S, Nhat NT, de Bruin E, van Beek J, Hien NT, Simmons CP, Farrar J, Koopmans M: Populationlevel antibody estimates to novel influenza A/H7N9. J Infect Dis. 2013, 208: 554558. 10.1093/infdis/jit224.View ArticlePubMedPubMed CentralGoogle Scholar
 Xu C, Havers F, Wang L, Chen T, Shi J, Wang D, et al: Monitoring avian influenza A(H7N9) virus through national influenzalike illness surveillance, China. Emerg Infect Dis. 2013, 19: doi:10.3201/eid1908.130662Google Scholar
 The prepublication history for this paper can be accessed here:http://www.biomedcentral.com/17417015/11/214/prepub
Prepublication history
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.