Incubation Periods of SARS-CoV-2 Wild-Type, Delta, and Omicron Variants–Dominant Periods in Singapore

Abstract

This study in Singapore analysed the incubation periods of the following SARS-CoV-2 variants: Wuhan-Hu-1, Delta, and Omicron. Three pandemic waves were examined: Wuhan-Hu-1 (January 2020–March 2021), Delta (May–October 2021), and Omicron (January–June 2022). Data from the SingHealth COVID-19 registry, covering patients from 23 January 2020 to 10 June 2022, were used to calculate incubation periods during the three time periods. The study found median incubation periods of 11 days for Wuhan-Hu-1, 3 days for Delta, and 3 days for Omicron (p-value: <0.001). This study highlighted the impact of different containment measures and the importance of robust EMR systems for tracking and managing infectious diseases. Key challenges included accurate contact tracing and IT infrastructure capabilities. The findings support the use of EMR data for future infectious disease preparedness in Singapore.

1. Introduction

Since the beginning of the coronavirus disease 2019 (COVID-19) pandemic, Singapore’s national strategy for COVID-19 pandemic control included robust contact tracing and the effective implementation of the Quarantine Order [1]. The timeframe for contact tracing and quarantine, two essential components of outbreak prevention and control, was determined by the incubation period of the SARS-CoV-2 wild-type (Wuhan-Hu-1) strain [2]. As the pandemic progressed, other variants of concern, as determined by the World Health Organization, emerged, including the Delta (May–early December 2021) and the Omicron (January–July 2022) variants [3]. Early containment measures were based on the incubation period of the Wuhan-Hu-1 strain or focused on only one variant of concern [4,5,6,7]. We aimed to determine the trend in COVID-19 incubation periods during three distinct pandemic waves corresponding to key SARS-CoV-2 variants in Singapore and the impact of these trends on hospital bed planning for bed allocation to non-COVID-19 patients requiring acute care.

2. Materials and Methods

2.1. Study Setting and Data Source

SingHealth is one of three national clusters in Singapore comprising public hospitals, community hospitals, national speciality centres, and regional primary care clinics. During the pandemic, the SingHealth COVID-19 registry, a database of all patients with laboratory-confirmed COVID-19 who had encounters at SingHealth institutions, was set up. We conducted a COVID-19 registry-based cohort study between 23 January 2020 and 10 June 2022 by extracting data on patients infected with COVID-19 from the SingHealth COVID-19 registry. This included three SingHealth hospitals and nine regional primary care clinics. The study included all patients who were part of the SingHealth COVID-19 registry and had documented exposure to COVID-19 prior to the onset of COVID-19 infection. This study was approved by the SingHealth Centralised Institutional Review Board (CIRB) [CIRB Ref No.: 2022/2040].

2.2. Identification of SARS-CoV-2 Variants

Three prevalent strains were considered in this study, including Wuhan-Hu-1 (23 January 2020–1 March 2021), the Delta variant (5 May–31 October 2021), and Omicron variant (19 January–10 June 2022). These strains were derived from the metadata collected from genomic sequences shared via GISAID, the global data science initiative, and confirmed by polymerase chain reaction (PCR) tests (Figure 1) [8]. All PCR samples for patients within the registry were collected at designated Ministry of Health (MOH)-approved COVID-19 test providers using nasopharyngeal swabs [9,10]. Supervised self-administered antigen rapid tests (ARTs) had been made available in Singapore since early 2022 and were widely used at COVID-19 testing centres with less reliance on PCR results [11,12,13]. Data on COVID-19 test results in the registry were obtained from both PCR and ART via clinician testing. Data on COVID-19 PCR and ART were uploaded to a national database, which could send automated tags to institution electronic medical records (EMRs) should a patient who had tested positive for COVID-19 present.

2.3. Determining the Various Incubation Periods of Variants

Contact tags were assigned to individuals identified as contacts of COVID-19 and were only applied once patients met the epidemiological criteria for significant contact, as determined by the Singapore MOH (see Supplementary Table S1). Eligible patients were identified through this contact tag that appeared in the EMR. During the Omicron period, when the national strategy shifted from containment to endemicity, and quarantine was discontinued, only patients with preceding international travel received restriction orders.

The last exposure date was determined by the start date indicated in all three tags (Contact-quarantine, Phone-Surv, Travel-Hist). For positive patients, another tag bearing the positive date of the PCR or ART appeared in the EMR. The incubation period was calculated based on the duration in days from the last exposure date to the date of the first positive COVID-19 test (Supplementary Figure S1). As definite infector and infected relationships could not be established, the calculation of the serial interval was excluded from the analysis.

2.4. Statistical Analysis

Descriptive data were reported as frequency counts and percentages. The incubation period of each variant was reported as the median days and interquartile range (IQR). Differences in the incubation periods among the three variants were analysed using the Wilcoxon Signed Rank test or Kruskal–Wallis test, where appropriate. Statistical analyses were performed using R version 4.0.1 (R Project for Statistical Computing).

3. Results

During the study period, a total of 17,738 patients were identified from the registry who had some documented exposure during contact tracing and subsequently developed COVID-19, separated by the prevailing variant in circulation (Figure 2). The median incubation period was 11 days (IQR: 7–14 days) for the Wuhan-Hu-1 variant, 3 days (IQR: 2–4 days) for the Delta variant, and 3 days (IQR: 0–5 days) for the Omicron variant. Additionally, the distribution of the incubation periods among patients within each period and between periods was not similar (Supplementary Table S1); no differences were observed for incubation periods between genders during the Omicron period.

4. Discussion

A recent meta-analysis comparing the incubation periods between the various strains indicated that the incubation period is between 3 and 5 days for the delta variant and 2–4 days for the omicron variant [4]. These findings corroborate our data on the observed variants. The median incubation period of the SARS-CoV-2 Wuhan-Hu-1 variant is 1–4 days longer compared to other known respiratory viruses, as mentioned in a 2009 systematic review, but it is similar to the median incubation period of other human coronaviruses, such as the Delta and Omicron variants [14].

During the first wave in Singapore, the patients who contracted COVID-19 were dormitory migrant workers living in cramped communal settings, mainly male foreign labourers from developing countries [1]. One observation from the study illustrated that the rate of infection among this group outpaced the infection rate within the community, highlighting differences in policy implementation between Singapore residents and non-residents [8]. This may explain the gender differences observed, where the operations aimed at containing the outbreak among dormitory migrant workers resulted in a different incubation period, while the incubation period for females more closely matched the published literature of 4–6 days [9]. A digital contact tracing system, TraceTogether that was developed by the Singapore GovTech and manufactured by iWOW solutions in Singapore, was used to mitigate the spread of COVID-19 during the Delta and Omicron periods, which may explain why contacts were notified and tested earlier, thereby reducing the overall contact tracing time compared to the manual methods used when the Wuhan-Hu-1 strain was the predominant variant [15,16,17]. As data were obtained through an electronic tag, this did not further stratify the extent of SARS-CoV-2 exposure. More importantly, laboratory testing capacity was reserved for clinically suspected cases rather than screening for contact tracing, which had a higher risk of false positives [18,19]. This affected any lead time gained from more efficient contact tracing.

The strength of this report lies in how easily data could be drawn from routinely captured information within the EMR to help build institutions’ defence against emerging infectious diseases [20]. However, as the disease progressed during the Omicron and Delta waves, cycle threshold (Ct) values from PCR tests became necessary to determine acute infections [21,22]. It is important to note that the effectiveness of the IT systems is dependent on the accuracy of contact tracing during the pandemic, the IT infrastructure’s ability to cope with surges in traffic, the communication pipelines with EMR, and the social–economic resources available to healthcare institutions and the public [23,24].

5. Conclusions

We found that the incubation period for the Delta and Omicron variants was shorter than the Wuhan-Hu 1 variant. We also evaluated whether the incubation periods could be estimated from readily available information in the EMR and identified key factors that could influence these results. Despite these limitations, our data demonstrate the feasibility of such methods for the estimation of incubation periods that may be useful for the management of emerging infectious diseases in Singapore.