Background

Coronavirus disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus (SARS-CoV-2). It has spread over 223 countries and has been declared a pandemic on March 11, 2020 by World Health Organization [1]. The number of COVID-19 patients is rapidly increasing globally. More than 113.5 million confirmed cases and 2.5 million deaths were reported globally by March 2, 2021 [1]. The COVID-19 pandemic has become a worldwide public health issue.

The incubation period, known as the interval between initial infection and onset of disease, is an important index to characterize the spread of infectious disease and formulate quarantine measures. For example, the mean incubation period is often used to calculate the reproduction number, and the maximum incubation period is used to determine the duration of quarantine [2]. For COVID-19, its average incubation period has a wide range, ranging from 2.87 days [3] to 17.6 days [2]. The 14-day quarantine strategy faced a challenge because the incubation period of COVID-19 exceeded 14 days [87]. In this study, among 11 545 patients, more than 10% developed the disease 14 days after infection in the 11 545 patients. Consistent with the results in 218 patients with precise data, the most extended incubation period was 26 days, and 27 patients had an incubation period of more than 14 days. Based on available evidence, the most extended incubation period of COVID-19 was 34 days in Shanghai’s patients, far more than 14 days [4]. The extreme tail end of the incubation period may be affected by the sample size, the knowledge of SARS-CoV-2, and the observation period. Previous studies indicated that 5–10% of patients infected with SARS-CoV-2 had an incubation period longer than 14 days [88]. The above-mentioned evidence indicated that an extended quarantine period was needed to prevent the spread of SARS-CoV-2. When the quarantine interval was longer, more patients will be identified, and the epidemic will be controlled faster. Our findings suggested that the 21-day quarantine strategy will reduce the number of patients without symptoms by 80% compared with the 14-day quarantine strategy.

Take Wuhan as an example, after implementing a strict quarantine strategy of all residents on January 23, the number of new cases with COVID-19 decreased rapidly, with a 50% reduction after 14 days (February 6) and a 75% reduction after 28 days (February 20) [89]. However, the impact of the nucleic acid test on the development of quarantine strategies remained unknown. The nucleic acid test played a vital role in the prevention and control of the COVID-19 pandemic. A previous study reported that 8% of patients had a negative report of nucleic acid test after being quarantined for 14 days [90]. Cai et al. found that over 5% of patients had an incubation period of more than 14 days, which defined the incubation period as the interval from the earliest exposure to laboratory confirmation of COVID-19 or onset of symptoms and signs [91]. We hypothesized that, if only relying on clinical symptoms or nucleic acid test results, then the 14-day quarantine strategy will result in a part of patients not being recognized as confirmed cases. Therefore, adopting a 21-day quarantine strategy was recommended, particularly for places with insufficient detection resources or a high risk of being infected with SARS-CoV-2, such as contacting the patients infected with SARS-CoV-2 or coming from cities or countries where COVID-19 was epidemic.

In addition, the potential patients should be quarantined as soon as possible. We found that about 15% of patients had an incubation period of fewer than 3 days. In India, 25% of patients had an incubation period of fewer than 3 days (25th percentile: 3.0 days) [32]. One-third of patients in Singapore had an incubation period of less than or equal to 3 days [35]. Patients with COVID-19 were infectious before they developed symptoms [92]. Our study found that half of the patients developed symptoms within 7 days after infection. A meta-analysis showed that the mean serial interval of COVID-19 was 5.5 days [93]. Therefore, for COVID-19, the serial interval was shorter than the incubation period. We hypothesized that the spread of SARS-CoV-2 occurred on average 1.5 days before the onset of the disease. If all close contacts are quarantined on the 3rd day after infection, then more than 15% of the people infected with SARS-CoV-2 in close connections may have infected others. When COVID-19 patients and their close and sub-close contacts were quarantined for the first time, further transmission will be terminated.

The incubation period of COVID-19 varied with age. We found that the incubation period was different among age groups. In 218 patients, the incubation period presented a U-shaped curve with increasing age. The middle-aged group (41–60 years) had the shortest incubation period among the other groups, particularly the elderly group (≥ 61 years) and those aged 18–40 years. A similar age-specific distribution of incubation period was reported in the previous study with 136 patients, showing the shortest incubation period in patients aged 45 to 59 years [7]. Another study with 2 555 patients also found a U-shaped curve distribution of incubation period in patients [6]. The mechanism of the effect of age on the COVID-19 incubation period was unclear. Possible explanations include a less intense immune response, a delay in the onset of symptoms, and a shorter exposure time and exposure rate in the elderly and children.

Focusing on the incubation period among different SARS-CoV-2 variants is important. Our finding showed that the incubation period in the mainland of China was longer than that outside the mainland of China (6.5 days vs 4.6 days, P < 0.001). Phylogenetic network analysis revealed that the types of SARS-CoV-2 were different among Chinese, Europeans, and Americans in the early stage of the COVID-19 pandemic [94]. Therefore, we hypothesized that the incubation period of COVID-19 may be different among SARS-CoV-2 variants. Recently, an outbreak of SARS-CoV-2 variant B.1.617.2 occurred in Guangzhou, China. In this outbreak, the mean incubation period was 4.4 days (95% CI 3.9–5.0), which was shorter than that previously reported in China [95]. These evidences indicated that mutations affected the incubation period of COVID-19. However, more research was needed to explore the relationship of incubation period of COVID-19 with SARS-CoV-2 variants.

However, no difference in incubation period was observed between males and females. The previous study suggested that male was more susceptible to COVID-19, which may be due to the high plasma concentration of ACE2 [96]. However, our study did not observe a difference in incubation period by gender. Nie et al. also showed an insignificant difference in incubation period by gender (5 days vs 4 days; P = 0.22) [8]. Yang et al. presented a similar conclusion [97]. The evidence indicated that gender maybe not a factor affecting the incubation period of COVID-19.

Several limitations should be stated in the present study. First, the current meta-analysis did not include studies published in languages other than Chinese or English because of language restrictions. Most of the included studies were from China. Secondly, information bias may also exist. As far as we know, this study had the largest sample size to date, but a part of patients was missing the date of onset. Although no significant difference in incubation periods was observed among the three estimation methods of the date of onset, the estimated date of onset may be biased from the actual date of onset. In addition, most patients had a doubly interval-censored data rather than precise date of infection. For extreme exposure intervals, we limited the exposure interval to 14 days, which might neglect a small part of the extreme incubation period. Third, in the 218 patients with precise data, the proportion of incubation period exceeding 14 days was higher than that in previous studies. This result would obtain an extreme right tail for the incubation period of COVID-19. Finally, the active contact tracing and testing (nucleic acid testing and antibody testing) may truncate the time between exposure to identifying infected persons. Therefore, caution should be taken when formulating quarantine strategies because the estimation of the incubation period did not involve nucleic acid tests and antibody tests.

Conclusions

This study provides evidence on the incubation period of COVID-19 to understand the transmission of disease and formulate preventive measures. The mean incubation period is 6.0 days globally, but it is longer in the mainland of China (6.5 days) than in other regions (4.6 days). The region- or age-specific incubation period should be paid attention to. Moreover, 10% of patients had an incubation period over 14 days in Chinese population, suggesting that the 14-day quarantine period may not be enough.