Real-time monitoring the transmission potential of COVID-19 in Singapore, March 2020

The ongoing Coronavirus Disease 2019 (COVID-19) pandemic started with a cluster of pneumonia cases of unknown etiology in Wuhan, China, back in December 2019 [1, 2]. The initial cases have been linked to a wet market in the city of Wuhan, pointing to an animal source of the epidemic [3]. Subsequently, rapid human-to-human transmission of the disease was confirmed in January 2020, and the etiological agent was identified as severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) due to its genetic similarity to the SARS-CoV (severe acute respiratory syndrome coronavirus) discovered in 2003 [4, 5]. The total global case tally has reached 750,890 infections including 36,405 deaths and involving 199 countries as of March 31, 2020 [6]. As the virus continues to spread in the human population, obtaining an accurate “real-time” picture of the epidemic’s trajectory is complicated by several factors including reporting delays and changes in the case definition [7, 8]. Although the COVID-19 case incidence in China has substantially declined, active transmission is now occurring in multiple countries around the world [2]. Epidemiological data from these countries can help to monitor transmission potential of SARS-CoV-2 in near real time.

Outside of China, Singapore, where the first symptomatic imported case (66 years old Chinese male) was reported on January 23, 2020, has been able to maintain relatively low COVID-19 incidence levels through active case finding and strict social distancing measures. Up until March 31, 2020, Singapore has reported 926 laboratory confirmed cases, including 24 reported case importations from Wuhan China and 501 non-Wuhan-related case importations [9]. Imported cases include six individuals who were evacuated from China between January 30 and February 9, 2020, and multiple citizens and long-term Singapore pass holders returning from Asia, Europe, and North America in late March 2020 [9‐11]. Moreover, Singapore has reported 3 deaths as of March 31, 2020 [9]. On February 4, 2020, the Ministry of Health of Singapore reported its first local cluster of COVID-19, which was linked to the Yong Thai Hang shop [12]. A total of 18 clusters with 2 or more COVID-19 cases have been reported thus far. Table 1 summarizes the characteristics of the 6 largest clusters in Singapore.

Although large-scale community transmission has not been reported in Singapore, the novel coronavirus can rapidly spread in confined and crowded places, as illustrated by large clusters of COVID-19 cases linked to the Grace Assembly of God Church, the Life Church and Missions Singapore, Wizlearn Technologies, and the SAFRA Jurong cluster [13, 14]. In China, substantial hospital-based transmission of SARS-CoV-2 has been reported, with approximately 3400 cases involving healthcare workers [15]. This pattern aligns well with past outbreaks of severe acute respiratory syndrome (SARS) and Middle Eastern respiratory syndrome (MERS) [16], including substantial nosocomial transmission during the 2003 SARS outbreak in Singapore [17]. To minimize the risk of hospital-based transmission of SARS-CoV-2, the Ministry of Health of Singapore has restricted the movement of patients and staff across hospitals [18]. Also, because multiple unlinked COVID-19 cases have been reported in the community [19] and the recognition that a substantial proportion of asymptomatic cases may be spreading the virus [20‐22], strict social distancing measures have been put in place including advising the public against large social gatherings in order to mitigate the risk of community transmission [23]. These social distancing measures reduce the risk of onward transmission not only within Singapore, but also beyond the borders of this highly connected nation [24]. A recent influx of imported cases from Asia, Europe, and North America into Singapore has triggered travel bans and restrictions for travelers and citizens [9].

The reproduction number is a key threshold quantity to assess the transmission potential of an emerging disease such as COVID-19 [25, 26]. It quantifies the average number of secondary cases generated per case. If the reproduction number is below 1.0, infections occur in isolated clusters as self-limited chains of transmission, and persistence of the disease would require continued undetected importations. On the other hand, reproduction numbers above 1.0 indicate sustained community transmission [16, 26]. Using epidemiological data and mathematical modeling tools, we are monitoring the effective reproduction number, R_t, of SARS-CoV-2 transmission in Singapore in real-time, and here we report the evolution of R_t by March 17, 2020. Specifically, we characterize the growth profile and the effective reproduction number during the first transmission wave from the daily case series of imported and autochthonous cases by date of symptoms onset after adjusting for reporting delays, and we also derive an estimate of the overall reproduction number based on the characteristics of the clusters of COVID-19 in Singapore.

Overall, current estimates of transmission potential in Singapore, based on two different data sources and different methods, suggest that temporary local transmission potential of SARS-CoV-2 has occurred in Singapore while our most recent estimate of the effective reproduction number is below the epidemic threshold of 1.0 whereas the overall reproduction number derived from the distribution of cluster sizes just barely crosses 1.0 (R = 0.61 (95% CI 0.39, 1.02). Temporary sustained transmission in the beginning of the epidemic can be partly attributed to multiple case importations and initiation of local transmission in the region. While large-scale local transmission has not been reported in Singapore, the fact that asymptomatic and subclinical cases are now well documented for COVID-19 [54] suggests that our estimates could be underestimated [55]. On the other hand, it is not clear if asymptomatic or subclinical cases are as infectious as symptomatic cases. Indeed, we have reported that multiple local cases have yet to be traced to existing transmission chains [9]. Additional data collected during the course of the outbreak will help obtain an improved picture of the transmission dynamics. These findings emphasize the need to strengthen public health interventions including active case contact tracing activities in countries with emerging transmission of SARS-CoV-2 [56]. It is worth noting that imported cases have minor contribution to secondary cases in Singapore, with most of the imported cases dating back to the early phase of the epidemic and between March 10 and 17, 2020. However, there are examples such as the Grant Hyatt Singapore cluster and the Yong Thai Hang cluster that were linked to imported sources, and the original sources had left Singapore before these local clusters emerged [30].

Our R_t estimates for Singapore are substantially lower than mean estimates reported for the COVID-19 epidemic in other parts of the world [57‐65]. This indicates that containment efforts have a significant impact in Singapore (Table 2). However, some differences in the reproduction numbers reported for the epidemic in China may result from different methods, differences in data sources and time periods used to estimate the reproduction number. Similarly, a recent study has shown an average reporting delay of 6.1 days in China [66] which agrees with our mean estimate for cases in Singapore (6.4 days). Moreover, the scaling parameter for growth rate (p) indicates a sub-exponential growth pattern in Singapore, reflecting the effective isolation and control strategies in the region. This is consistent with a sub-exponential growth pattern for Chinese provinces excluding Hubei (p ~ 0.67), as estimated by a recent study [67] whereas an exponential growth pattern was estimated for Hubei (p ~ 1.0) [67].

A previous study on the 2015 MERS outbreak in South Korea reported substantial potential for superspreading transmission despite a subcritical reproduction number [68]. The lower estimate of the dispersion parameter in our study also indicates significant transmission heterogeneity in Singapore. Superspreading events of MERS-CoV and SARS-CoV associated with nosocomial outbreaks are well documented and driven largely by substantial diagnostic delays [16]. Although the average delay from onset of symptoms to diagnosis/reporting for COVID-19 patients in Singapore is at 6.4 days and no superspreading events has been observed yet, the dispersion parameter, k < 1, indicates the probability of observing large clusters and the potential for superspreading such as the SAFRA Jurong cluster [50, 68]. Therefore, public health measures enacted by public health authorities in Singapore that advise the public to avoid mass gatherings and confined places are crucial to prevent disease amplification events. However, the presence of asymptomatic cases in the community represents an ongoing threat although it is not currently known if subclinical cases are less infectious [20, 69]. This highlights the need for rapid testing of suspected cases to quickly isolate those that test positive for the novel coronavirus. To achieve this goal, public health authorities in Singapore are reactivating 900 general practitioner clinics [70]. While new clusters emerge in Singapore, some clusters including the Yong Thai Hang cluster, Seletar Aerospace cluster, Wizlearn Technologies, and the Grand Hyatt cluster have stabilized (no recent additional cases in most clusters). The “Grace Assembly of God and the Life Church and Missions” cluster and the “SAFRA Jurong” cluster continue to be consolidated .

Beyond Singapore, COVID-19 cases are now being reported in 204 countries including identifiable clusters in many parts of the world [2, 71‐74]. Moreover, Singapore has also produced secondary chains of disease transmission beyond its borders [24]. Although Singapore has been detecting and isolating cases with diligence, our findings underscore the need for continued and sustained containment efforts to prevent large-scale community transmission including nosocomial outbreaks. Overall, the current situation in Singapore highlights the need to investigate the imported, unlinked, and asymptomatic cases that could be a potential source of secondary cases and amplified transmission in confined settings. Although Singapore has a world-class health system including a highly efficient contact tracing mechanism in place that has prevented the outbreak from getting out of control [24, 75], continued epidemiological investigations and active case finding efforts are needed to contain the outbreak.

Our study is not exempt from limitations. First, the outbreak is still ongoing and we continue to monitor the transmission potential of COVID-19 in Singapore. Second, onset dates are missing for 12 cases, which were excluded from our analyses. Third, we cannot rule out that additional cases will be added to existing clusters, which may lead to underestimating the reproduction number based on the cluster size distribution. Fourth, some of the cases are associated with generating secondary chains in more than one cluster, which were included in the most relevant cluster.

This is a real-time study to estimate the evolving transmission potential of SARS-CoV-2 in Singapore. Our current findings point to temporary sustained transmission of SARS-CoV-2, with our most recent estimate of the effective reproduction number lying below 1.02. These estimates highlight the significant impact of containment efforts in Singapore while at the same time suggest the need to maintain social distancing and active case finding efforts to stomp out all active or incoming chains of transmission.