Table 1

Estimates of the infectiousness of asymptomatic compared with symptomatic individuals, by study type (PCR studies, modelling studies and contact tracing studies)

AuthorLocationNumber of relevant study subjectsMeasure of relative infectiousness
PCR studies
 Letizia et al13South Carolina,
USA
51The viral load at diagnosis was approximately four times as high (4.39, S.gene; 4.3, N.gene; 5.22, ORF 1ab) in the five symptomatic participants compared with the 46 asymptomatic participants. However, some asymptomatic participants had higher viral loads estimated on the basis of the cycle threshold (Ct).
 Uhm et al11South Korea396There was no significant difference in the first follow-up Ct value for the E, RdRP and N genes between symptomatic and asymptomatic individuals.
 Lavezzo et al1Vo’, Italy58There was no significant difference between asymptomatic and symptomatic participants based on viral PCR recovered from the nasopharyngeal swabs.
 Danis et al12Les Contamines-Montjoie, France††The viral load was described as being similar among the six study subjects (one asymptomatic and five symptomatic cases).
 Long et al9Wanzhou, China74The initial Ct values were similar among the asymptomatic and mild symptomatic individuals, whereas the duration of shedding among asymptomatics was longer than for mild symptomatic individuals.
Modelling studies, estimate (95% CI)
 Nabi2Dhaka, BangladeshNA*0.57†
 Li et al10ChinaNA*0.55 (0.46 to 0.62)‡,**
0.50 (0.37 to 0.69)§,**
0.43 (0.31 to 0.61)¶,**
Contact tracing studies
 Park et al14Seoul, South Korea225 Secondary attack rate:
  •  Among asymptomatic primary cases=0/4=0.

  •  Among symptomatic and presymptomatic primary cases=34/221=0.05.


 Relative risk (RR)=0, but Fisher’s exact test (null hypothesis of no difference), p=1.0.
 Therefore, the point estimate for RR (equivalent to relative infectiousness) is equal to 0, but we do not have sufficient evidence to conclude that it is significantly different to 1.
 Cheng et al15Taiwan2735 Secondary attack rate:
  •  Among asymptomatic primary cases=0/91=0.

  •  Among symptomatic primary cases=22/2644=0.008.


 RR=0. Fisher’s exact test, p=1.0.
 Therefore, the point estimate for RR (equivalent to relative infectiousness) is equal to 0, but we do not have sufficient evidence to conclude that it is significantly different to 1.
 Han16South Korea192 (all contacts) and 36 (close contacts only) Secondary attack rate among all contacts:
  •  Among asymptomatic primary cases=0/52=0 (among contacts).

  •  Among symptomatic and presymptomatic primary cases=7/140=0.05.


 Secondary attack rate among close contacts only:
  •  Among asymptomatic primary cases=0/14=0.

  •  Among symptomatic and pre-symptomatic primary cases=7/22=0.32.


 RR=0. Fisher’s exact test, p=0.19 (all contacts), p=<0.001 (close contacts).
 Conclusion:
  •  All contacts: the point estimate for RR (equivalent to relative infectiousness) is equal to 0, but we do not have sufficient evidence to conclude that it is significantly different to 1.

  •  Close contacts only: the point estimate for RR (equivalent to relative infectiousness) is equal to 0, and we have sufficient evidence to conclude that it is significantly different to 1.

 Plucinski et al3Diamond Princess cruise ship, Japan750.78 (0.58 to 1.04)
 Luo et al17Guangzhou, China2610 Adjusted OR for transmission risk, based on the severity of the primary cases‡:
  •  Asymptomatic: 0.37 (0.04 to 3.79).

  •  Mild: 0.56 (0.33 to 0.94).

  •  Moderate: 1 (reference category).

  •  Severe: 1.04 (0.57 to 1.90).

  • The number of relevant study subjects refers to the number of individuals with available PCR results (PCR studies) or the number of close contacts (contact tracing studies).

  • *Population studied at national level.

  • †The average of the following set of values for relative infectiousness: {0.5249, 0.5855, 0.5995, 0.59, 0.5498}.

  • ‡After controlling for age, sex, exposure setting and symptoms in the index (primary) cases.

  • §Model for the period 10–23 January 2020.

  • ¶Model for the period 24 January–3 February 2020.

  • **Model for the period 24 January–8 February 2020.

  • ††Asymptomatics were defined as ‘undocumented cases’.