Characterizing transmission is critical to control the spread of an emerging infectious disease. The reproductive number is the widely adopted measure of infectiousness. However, it only measures the average number of secondary cases infected by an infected person, not the heterogeneity in the number of transmissions. Variation of individual infectiousness is particularly highlighted by superspreading events (SSEs), in which a minority of cases are responsible for a majority of transmission events. Such phenomena, illustrated by the ‘80/20 rule’ (i.e., 20% of cases responsible for 80% transmission; Adam et al., 2020; Sun et al., 2021), have been observed in emerging infectious disease outbreaks (Lloyd-Smith et al., 2005), including severe acute respiratory syndrome (SARS) (Shen et al., 2004), Middle East respiratory syndrome (Cauchemez et al., 2016; Cowling et al., 2015), and most recently the COVID-19 pandemic (Adam et al., 2020; Lau et al., 2020; Sun et al., 2021; Wong and Collins, 2020; Zhao et al., 2021). In these outbreaks, the proportion of cases attributed to 80% transmission, and the dispersion parameter that is estimated by fitting the negative binomial distribution to the number of secondary cases (Lloyd-Smith et al., 2005), are considered as measures of transmission heterogeneity.

However, the number of contacts per index cases is often not reported in SSE studies, and hence not incorporated in the analyses. In addition, SSE studies usually analyze clusters from different settings, in which the baseline transmission risk and density of exposure could be different (Thompson, 2021). Finally, studies of transmission heterogeneity that focus on SSEs described in the literature may suffer from publication bias, with larger clusters having higher probability of being observed and reported (Zhao et al., 2021). Therefore, the observed heterogeneity in the number of secondary cases could be the result of large number of contacts in SSE settings, or confounding from these factors (Althouse et al., 2020; Bagdasarian and Fisher, 2020), instead of variation in individual infectiousness.

Households are one of the most important settings for SARS-CoV-2 transmission, with 4- to 10-fold higher transmission risk than other places (Thompson, 2021). Hence, household transmission studies provide an ideal setting to quantify variations in individual infectiousness. In a household transmission study, an index case is identified, and their household contacts are followed up for 1–2 weeks, during which there is high transmission potential (Tsang et al., 2016). Therefore, the number of contacts is known while transmission risks and reporting biases can be controlled. We aim to characterize the variation of individual infectiousness by analyzing data from household transmission studies.

We conducted a systematic review to gather information on the number of secondary cases with the number of household contacts for each household, in the form of number of households with X cases among households of size Y. In total, we identified 17 studies, comprising 13,098 index cases and 31,359 household contacts (Appendix 1—figure 1, Appendix 1—table 1; Bernardes-Souza et al., 2021; Carazo et al., 2022; Dattner et al., 2021; Han et al., 2022; Hart et al., 2022; Hsu et al., 2021; Hubiche et al., 2021; Koureas et al., 2021; Laxminarayan et al., 2020; Layan et al., 2022; Lyngse et al., 2022; Méndez-Echevarría et al., 2021; Posfay-Barbe et al., 2020; Reukers et al., 2021; Wilkinson et al., 2021). Most studies covered a period from January to November 2020, which was dominated by ancestral strains, except for Layan et al. (from December 2020 to April 2021) and Hsu et al. (from January 2020 to February 2021), which covered both ancestral strains and the alpha variant.

We then developed a statistical model to quantify the degree and the impact of variation of infectiousness of cases on transmission dynamics. The individual-based household transmission model describes the probability of infection of household contacts as depending on the time since infection in other infected persons in the household, so that infections from outside the household (community infections), or infections via other household contacts rather than the index case (tertiary infections) are allowed (Cauchemez et al., 2009; Tsang et al., 2014; Tsang et al., 2021). Therefore, the model could estimate the per-contact hazard of infection, which implicitly controls for number of household contacts in households. We extend this model by adding a random effect (${\delta }_i$) on the individual infectiousness of cases. Here, the relative infectiousness of case i compared with case j is exp(${\delta }_i$)/exp(${\delta }_j$). The parameter for variation in individual infectiousness (hereafter denoted as infectiousness variation, ${\sigma }_{var}$) is the standard deviation (SD) of the random effect characterizing individual infectiousness, so that ${\delta }_i$ follows a normal distribution with mean equal to 0 and SD equal to ${\sigma }_{var}$ .

We separately fit the models to 14 studies with more than 150 contacts (Carazo et al., 2022; Dattner et al., 2021; Dutta et al., 2020; Han et al., 2022; Hart et al., 2022; Hubiche et al., 2021; Koureas et al., 2021; Laxminarayan et al., 2020; Layan et al., 2022; Lyngse et al., 2022; Méndez-Echevarría et al., 2021; Reukers et al., 2021; Tsang et al., 2022; Wilkinson et al., 2021; Figure 1, Appendix 1—table 2). For 12 studies out of 14, models with infectiousness variation perform substantially better (range of ∆DIC: 5.8–268) (Figure 2, Appendix 1—table 2). From these 12 studies, the estimated infectiousness variation (${\sigma }_{var})$ ranged from 1.03 to 2.83. This suggests that, the 20% most infectious cases are 2.4- to 10-fold more infectious than the average case. Based on the two largest studies with 6782 and 3727 households (Carazo et al., 2022; Lyngse et al., 2022), the estimated infectiousness variation is 1.48 (95% credible interval [CrI]: 1.29, 1.7) and 1.41 (95% CrI: 1.19, 1.72), suggesting that, among all cases, the 20% most infectious are 3.5-fold (95% CrI: 3.0- to 4.2-fold) and 3.3-fold (95% CrI: 2.7- to 4.3-fold) more infectious than the average case. The estimated daily probability of infection from outside the household and estimated person-to-person transmission probability within households range from 0.0003 to 0.017, and from 0.06 to 0.51, respectively. The estimates of parameters for the relationship between number of contacts and transmission (larger value indicates stronger inverse association) range from 0.43 to 0.92, except for the study by Layan et al., 2022, where it is equal to 0.2. For all studies, the predicted final size distribution is consistent with the observed data and the model fit is judged adequate (Appendix 1—Tables 3–7, Appendix 1—figure 2). Simulation studies demonstrated that there is no important systematic bias, with 88–100% (depending on the parameter) of the 95% credible intervals covering the simulation value (Appendix 1—table 8). This suggests that the algorithm could estimate adequately the posterior distribution. We conduct a sensitivity analysis that assumes the individual infectiousness of cases follows a Gamma distribution, but the fit worsens substantially (Appendix 1—table 9).

Figure 1

Download asset Open asset

Figure 2

Download asset Open asset

We conduct random effects meta-analyses on estimates of individual infectiousness from the 14 identified studies. The pooled estimate of infectiousness variation is 1.33 (95% confidence interval [CI]: 0.95, 1.70), suggesting that the 20% most infectious cases are 3.1-fold (95% CI: 2.2- to 4.2-fold) more infectious than the average case (Figure 3). Based on this fitted distribution, we estimate that 5.9% (95% CI: 1.4%, 11.1%) and 14.9% (95% CI: 7.2%, 20.7%) of cases could be at least 8- and 4-fold more infectious than average cases, respectively.

Figure 3

Download asset Open asset

We further explore if the secondary attack rate (SAR, the proportion of infected contacts), and the SD of the distribution of number of secondary cases (${\sigma }_{sec})$ (Figure 4; Appendix 1—table 10) may be correlated with the infectiousness variation. In meta-regression, we find the infectiousness variation is associated with SAR (Appendix 1—table 11). We estimate that doubling SAR are associated with 0.55 (95% CI: 0.21, 0.89) unit decrease in infectiousness variation, with R-squared equal to 67% respectively. In addition, we find that higher infectiousness variation is associated with only using PCR to ascertain secondary cases. Regarding other statistics, we find that ${\sigma }_{sec}$ is positively associated with mean and SD of number of contacts. Other than these associations, we find no association between these statistics and implementation of lockdown, ascertainment method of index and secondary cases, and the circulating virus of SARS-CoV-2 in the study period.

Figure 4

Download asset Open asset

In this study, we characterize the impact of variation of individual infectiousness on heterogeneity of transmission of COVID-19 in households. We demonstrate that it can be estimated from household data using a modeling approach. The pooled estimate of infectiousness variation from 14 studies suggests that the 20% most infectious cases have 3.1-fold (95% CI: 2.2- to 4.2-fold) higher infectiousness compared with average index cases. This implies there is substantial variation in individual infectiousness of cases in households. Given that we focused our analysis on households with known number of contacts in studies conducted in the early outbreaks caused by ancestral strains, the estimated infectiousness variations are corrected for the variations caused by number of contacts and transmission risks in different settings, difference in pre-existing immunity among contacts (almost everyone is naïve and unvaccinated), and the difference in transmissibility among variants. Hence, this estimated infectiousness variation measures the variation caused by difference in individuals, which may be contributed by both biological factors and host behaviors, but not other potential confounding factors mentioned above.

Regarding host behaviors, multiple contact patterns, particularly by age, could contribute to the variations in infectiousness of cases. For example, mother-child contacts are usually more intense than father-child contacts, and adult cases are more capable of self-isolation within a household compared with children. Furthermore, contact pattern studies suggested that school-age children and young adults tended to mix with people of the same age (Mossong et al., 2008; Mousa et al., 2021). Also, the duration of contact could vary, which could also contribute to the variations in infectiousness of cases (Toth et al., 2015). It should be noted that contacts have different features, including their number, frequency, and duration that may contribute to variations in infectiousness. Previous studies also suggest that the relative importance of contact characteristics on transmission may be different among viruses (De Cao et al., 2014). Individual behaviors may also be influenced by control measures and recommendation from public health agencies. For example, lockdown and stay-at-home orders may increase the time spending at home. Mask-wearing when in contact with other household members, using separate bedrooms and bathrooms, avoiding having meals together may reduce the transmission in households (Ng et al., 2021). Also, social disparity such as occupation may increase or reduce the risk of transmission in households, including the availability of personal protective equipment (PPE), or being healthcare worker (Yang et al., 2021). These factors may have impact of the transmission risk and hence SAR in households. In addition, heterogeneity in these different factors may have contributed infectiousness variation.

Biological factors may also contribute to such variations. For example, the SAR for index cases with fever and cough are 1.4- and 1.3-fold higher than index cases without fever and cough, respectively (Madewell et al., 2021). Viral shedding is used as a proxy measure of infectiousness, and consistently it also has substantial variations, as suggested by 11 systematic reviews (Appendix 1—Tables 12–14). Regarding the magnitude of viral shedding, studies report high variations of temporal viral shedding patterns among individuals (He et al., 2020; Jones et al., 2021; Sun et al., 2022). In addition, the duration of viral shedding can be highly heterogeneous, with pooled estimates of mean durations ranging from 11.1 to 30.3 days, and almost all reviews reporting high heterogeneity of estimates (Cevik et al., 2021; Chen et al., 2021b; Díaz et al., 2022; Fontana et al., 2021; Li et al., 2020b; Okita et al., 2022; Qutub et al., 2022; Rahmani et al., 2022; Xu et al., 2020; Yan et al., 2021; Zhang et al., 2021). Such heterogeneities still exist in subgroup analyses by age and severity (Cevik et al., 2021; Fontana et al., 2021; Okita et al., 2022; Qutub et al., 2022; Rahmani et al., 2022; Xu et al., 2020; Yan et al., 2021). The infectious period, proxied by duration of replicant competent virus isolation, is also heterogeneous (Rahmani et al., 2022; Appendix 1—table 14). One review also suggests that heterogeneity in viral shedding is an intrinsic virological factor facilitating higher dispersion parameter for SARS-CoV-2 if we compare it with the corresponding patterns in SARS-CoV-1 and pandemic influenza A(H1N1)pdm09 (Chen et al., 2021a).

The observed variation in individual infectiousness is consistent with past analyses of the dispersion parameter in a negative binomial distribution fitted to number of secondary cases per index case of COVID-19 (Sun et al., 2021; Wong et al., 2015). However, in these studies, the number of contacts was not considered. Therefore, the observed variations may not apply directly to households with limited number of contacts. It should be noted that the SD of the distribution of number of secondary cases (${\sigma }_{sec}$) is only weakly correlated with infectiousness variation. This is because ${\sigma }_{sec}$ is highly correlated with the mean and SD of the number of contacts, suggesting that it may depend on distribution of number of contacts, and hence may not be comparable among studies. Infectiousness variation is also correlated with the SAR. When SAR is higher, it is expected that more contacts are infected and therefore the observed number of secondary cases is less heterogeneous (Adam et al., 2022). We find higher infectiousness variation is associated with only using PCR to confirm secondary cases. One potential reason is that using other methods may lower the sensitivity of detecting infection, resulting in lower estimates of SAR and hence higher estimates of infectiousness variation. Further investigations are needed to explore roles of other potential factors affecting infectiousness variation, such as contact frequency among different regions (Mousa et al., 2021).

An important limitation of our study is that we do not have individual-level data. Therefore, we are unable to determine the impact of demographic factors like age and sex on infectiousness variation. Also, we cannot disentangle the host behaviors from biological factors. Previous analysis suggested no evidence of the impact of age on infectiousness of cases (Lau et al., 2020; Sun et al., 2021). In addition, we could not include factors affecting susceptibility to infection. Our estimates of infectiousness variation should be interpreted in light of these limitations: they capture heterogeneity in infectiousness due to demographic, host, and biological factors. However, in one study that included susceptibility component in the estimation of individual infectiousness, substantial heterogeneity remained with 20% of cases estimated to contribute to 80% of transmission (Tsang et al., 2022). Second, the recruitment methods among studies are different. This may affect the comparability of the results, although all index cases are laboratory-confirmed in all studies (Tsang et al., 2021). Third, we assumed that risks of infection from community for all households are the same, but there were different factors that may affecting this, including occupations, such as healthcare workers, social economic status that related to assess to PPEs (Yang et al., 2021). Finally, most of our identified studies were conducted in the period of circulation of ancestral strains, and therefore the identified infectiousness variation may not be directly applicable to other variants.

In conclusion, we developed a modeling approach to estimate variation in individual infectiousness from household data. Result indicates that there is substantial variation in individual infectiousness, which is important for epidemic management.

The data and computer code (in R language) for conducting the data analysis can be downloaded from https://github.com/timktsang/covid19_transmission_heterogeneity (copy archived at swh:1:rev:634390fa66a4bfb998da691e7cd81cf45e38c6db).

1. 1. Adam DC
   2. Wu P
   3. Wong JY
   4. Lau EHY
   5. Tsang TK
   6. Cauchemez S
   7. Leung GM
   8. Cowling BJ

   (2020) Clustering and superspreading potential of SARS-cov-2 infections in hong kong

   Nature Medicine 26:1714–1719.

   https://doi.org/10.1038/s41591-020-1092-0

   • PubMed
   • Google Scholar
2. Preprint

   1. Adam D
   2. Gostic K
   3. Tsang T
   4. Wu P
   5. Lim WW
   6. Yeung A
   7. Wong J
   8. Lau E
   9. Du Z
   10. Chen D
   11. Ho LM
   12. Martín-Sánchez M
   13. Cauchemez S
   14. Cobey S
   15. Leung G
   16. Cowling B

   (2022) Time-Varying Transmission Heterogeneity of SARS and COVID-19 in Hong Kong

   Research Square.

   https://doi.org/10.21203/rs.3.rs-1407962/v1

   • Google Scholar
3. 1. Althouse BM
   2. Wenger EA
   3. Miller JC
   4. Scarpino SV
   5. Allard A
   6. Hébert-Dufresne L
   7. Hu H

   (2020) Superspreading events in the transmission dynamics of SARS-cov-2: opportunities for interventions and control

   PLOS Biology 18:e3000897.

   https://doi.org/10.1371/journal.pbio.3000897

   • PubMed
   • Google Scholar
4. 1. Bagdasarian N
   2. Fisher D

   (2020) Heterogenous COVID-19 transmission dynamics within singapore: a clearer picture of future national responses

   BMC Medicine 18:164.

   https://doi.org/10.1186/s12916-020-01625-7

   • PubMed
   • Google Scholar
5. 1. Bernardes-Souza B
   2. Júnior SRC
   3. Santos CA
   4. Neto R
   5. Bottega FDC
   6. Godoy DC
   7. Freitas BL
   8. Silva DLG
   9. Brinker TJ
   10. Nascimento RA
   11. Tupinambás U
   12. Reis AB
   13. Coura-Vital W

   (2021) Logistics workers are a key factor for SARS-cov-2 spread in brazilian small towns: case-control study

   JMIR Public Health and Surveillance 7:e30406.

   https://doi.org/10.2196/30406

   • PubMed
   • Google Scholar
6. 1. Carazo S
   2. Laliberté D
   3. Villeneuve J
   4. Martin R
   5. Deshaies P
   6. Denis G
   7. Adib G
   8. Tissot F
   9. Dionne M
   10. De Serres G

   (2022) Characterization and evolution of infection control practices among severe acute respiratory coronavirus virus 2 (SARS-cov-2)-infected healthcare workers in acute-care hospitals and long-term care facilities in québec, canada, spring 2020

   Infection Control and Hospital Epidemiology 43:481–489.

   https://doi.org/10.1017/ice.2021.160

   • PubMed
   • Google Scholar
7. 1. Cauchemez S
   2. Donnelly CA
   3. Reed C
   4. Ghani AC
   5. Fraser C
   6. Kent CK
   7. Finelli L
   8. Ferguson NM

   (2009) Household transmission of 2009 pandemic influenza A (H1N1) virus in the united states

   The New England Journal of Medicine 361:2619–2627.

   https://doi.org/10.1056/NEJMoa0905498

   • PubMed
   • Google Scholar
8. 1. Cauchemez S
   2. Ferguson NM
   3. Fox A
   4. Mai LQ
   5. Thanh LT
   6. Thai PQ
   7. Thoang DD
   8. Duong TN
   9. Minh Hoa LN
   10. Tran Hien N
   11. Horby P

   (2014) Determinants of influenza transmission in south east asia: insights from a household cohort study in vietnam

   PLOS Pathogens 10:e1004310.

   https://doi.org/10.1371/journal.ppat.1004310

   • PubMed
   • Google Scholar
9. 1. Cauchemez S
   2. Nouvellet P
   3. Cori A
   4. Jombart T
   5. Garske T
   6. Clapham H
   7. Moore S
   8. Mills HL
   9. Salje H
   10. Collins C
   11. Rodriquez-Barraquer I
   12. Riley S
   13. Truelove S
   14. Algarni H
   15. Alhakeem R
   16. AlHarbi K
   17. Turkistani A
   18. Aguas RJ
   19. Cummings DAT
   20. Van Kerkhove MD
   21. Donnelly CA
   22. Lessler J
   23. Fraser C
   24. Al-Barrak A
   25. Ferguson NM

   (2016) Unraveling the drivers of MERS-cov transmission

   PNAS 113:9081–9086.

   https://doi.org/10.1073/pnas.1519235113

   • PubMed
   • Google Scholar
10. 1. Celeux G
    2. Forbes F
    3. Robert CP
    4. Titterington DM

    (2006) Deviance information criteria for missing data models

    Bayesian Analysis 1:651–673.

    https://doi.org/10.1214/06-BA122

    • Google Scholar
11. 1. Cevik M
    2. Tate M
    3. Lloyd O
    4. Maraolo AE
    5. Schafers J
    6. Ho A

    (2021) SARS-cov-2, SARS-cov, and MERS-cov viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis

    The Lancet. Microbe 2:e13–e22.

    https://doi.org/10.1016/S2666-5247(20)30172-5

    • PubMed
    • Google Scholar
12. 1. Chen PZ
    2. Bobrovitz N
    3. Premji Z
    4. Koopmans M
    5. Fisman DN
    6. Gu FX

    (2021a) Heterogeneity in transmissibility and shedding SARS-cov-2 via droplets and aerosols

    eLife 10:e65774.

    https://doi.org/10.7554/eLife.65774

    • PubMed
    • Google Scholar
13. 1. Chen C
    2. Zhu C
    3. Yan D
    4. Liu H
    5. Li D
    6. Zhou Y
    7. Fu X
    8. Wu J
    9. Ding C
    10. Tian G
    11. Lan L
    12. Liu X
    13. Huang C
    14. Hecht R
    15. Li L
    16. Yang S

    (2021b) The epidemiological and radiographical characteristics of asymptomatic infections with the novel coronavirus (COVID-19): a systematic review and meta-analysis

    International Journal of Infectious Diseases 104:458–464.

    https://doi.org/10.1016/j.ijid.2021.01.017

    • PubMed
    • Google Scholar
14. 1. Cochran WG

    (1954) The combination of estimates from different experiments

    Biometrics 10:101.

    https://doi.org/10.2307/3001666

    • Google Scholar
15. 1. Cowling BJ
    2. Park M
    3. Fang VJ
    4. Wu P
    5. Leung GM
    6. Wu JT

    (2015) Preliminary epidemiological assessment of MERS-cov outbreak in south korea, may to june 2015

    Euro Surveillance 20:7–13.

    https://doi.org/10.2807/1560-7917.es2015.20.25.21163

    • PubMed
    • Google Scholar
16. 1. Dattner I
    2. Goldberg Y
    3. Katriel G
    4. Yaari R
    5. Gal N
    6. Miron Y
    7. Ziv A
    8. Sheffer R
    9. Hamo Y
    10. Huppert A

    (2021) The role of children in the spread of COVID-19: using household data from bnei BRAK, israel, to estimate the relative susceptibility and infectivity of children

    PLOS Computational Biology 17:e1008559.

    https://doi.org/10.1371/journal.pcbi.1008559

    • PubMed
    • Google Scholar
17. 1. De Cao E
    2. Zagheni E
    3. Manfredi P
    4. Melegaro A

    (2014) The relative importance of frequency of contacts and duration of exposure for the spread of directly transmitted infections

    Biostatistics 15:470–483.

    https://doi.org/10.1093/biostatistics/kxu008

    • PubMed
    • Google Scholar
18. 1. Díaz LA
    2. García-Salum T
    3. Fuentes-López E
    4. Reyes D
    5. Ortiz J
    6. Chahuan J
    7. Levican J
    8. Almonacid LI
    9. Valenzuela GH
    10. Serrano E
    11. Budnik S
    12. Gandara V
    13. Gallardo A
    14. Seydewitz MF
    15. Ferrés M
    16. Cofré C
    17. Álvarez M
    18. Pavez C
    19. Candia R
    20. Monrroy H
    21. Espino A
    22. Rada G
    23. Ortiz L
    24. Valderrama S
    25. Salinas E
    26. Toro A
    27. Ortega M
    28. Pizarro M
    29. Medina RA
    30. Riquelme A

    (2022) High prevalence of SARS-cov-2 detection and prolonged viral shedding in stools: a systematic review and cohort study

    Gastroenterologia y Hepatologia 45:593–604.

    https://doi.org/10.1016/j.gastrohep.2021.12.009

    • PubMed
    • Google Scholar
19. 1. Dutta S
    2. Kaur RJ
    3. Bhardwaj P
    4. Charan J
    5. Bist SKS
    6. Detha MD
    7. Kanchan T
    8. Sharma P
    9. Misra S

    (2020) Household transmission of COVID-19: a cross-sectional study

    Infection and Drug Resistance 13:4637–4642.

    https://doi.org/10.2147/IDR.S285446

    • PubMed
    • Google Scholar
20. 1. Fontana LM
    2. Villamagna AH
    3. Sikka MK
    4. McGregor JC

    (2021) Understanding viral shedding of severe acute respiratory coronavirus virus 2 (SARS-cov-2): review of current literature

    Infection Control and Hospital Epidemiology 42:659–668.

    https://doi.org/10.1017/ice.2020.1273

    • PubMed
    • Google Scholar
21. 1. Han WGH
    2. Swart A
    3. Bonačić Marinović A
    4. Eggink D
    5. Reimerink J
    6. Wijsman LA
    7. van der Veer B
    8. van den Brink S
    9. van den Brandt A-M
    10. van Tol S
    11. Godeke G-J
    12. Brouwer F
    13. Hoogerwerf M
    14. Reukers DFM
    15. Rots N
    16. Reusken C
    17. Meijer A
    18. Dutch F. F. X.-COVID-19 Research Group

    (2022) SARS-cov-2 RNA and antibody dynamics in a dutch household study with dense sampling frame

    Scientific Reports 12:7937.

    https://doi.org/10.1038/s41598-022-11480-6

    • PubMed
    • Google Scholar
22. 1. Hart WS
    2. Abbott S
    3. Endo A
    4. Hellewell J
    5. Miller E
    6. Andrews N
    7. Maini PK
    8. Funk S
    9. Thompson RN

    (2022) Inference of the SARS-cov-2 generation time using UK household data

    eLife 11:e70767.

    https://doi.org/10.7554/eLife.70767

    • PubMed
    • Google Scholar
23. 1. He X
    2. Lau EHY
    3. Wu P
    4. Deng X
    5. Wang J
    6. Hao X
    7. Lau YC
    8. Wong JY
    9. Guan Y
    10. Tan X
    11. Mo X
    12. Chen Y
    13. Liao B
    14. Chen W
    15. Hu F
    16. Zhang Q
    17. Zhong M
    18. Wu Y
    19. Zhao L
    20. Zhang F
    21. Cowling BJ
    22. Li F
    23. Leung GM

    (2020) Temporal dynamics in viral shedding and transmissibility of COVID-19

    Nature Medicine 26:672–675.

    https://doi.org/10.1038/s41591-020-0869-5

    • PubMed
    • Google Scholar
24. 1. Hedges LV
    2. Vevea JL

    (1998) Fixed- and random-effects models in meta-analysis

    Psychological Methods 3:486–504.

    https://doi.org/10.1037/1082-989X.3.4.486

    • Google Scholar
25. 1. Higgins JPT
    2. Thompson SG
    3. Deeks JJ
    4. Altman DG

    (2003) Measuring inconsistency in meta-analyses

    BMJ 327:557–560.

    https://doi.org/10.1136/bmj.327.7414.557

    • PubMed
    • Google Scholar
26. 1. Hsu CY
    2. Wang JT
    3. Huang KC
    4. Fan ACH
    5. Yeh YP
    6. Chen SLS

    (2021) Household transmission but without the community-acquired outbreak of COVID-19 in taiwan

    Journal of the Formosan Medical Association 120 Suppl 1:S38–S45.

    https://doi.org/10.1016/j.jfma.2021.04.021

    • PubMed
    • Google Scholar
27. 1. Hubiche T
    2. Phan A
    3. Leducq S
    4. Rapp J
    5. Fertitta L
    6. Aubert H
    7. Barbarot S
    8. Chiaverini C
    9. Giraudeau B
    10. Lasek A
    11. Mallet S
    12. Labarelle A
    13. Piram M
    14. McCuaig C
    15. Martin L
    16. Monitor L
    17. Nicol I
    18. Bissuel M
    19. Bellissen A
    20. Jullien D
    21. Lesort C
    22. Vabres P
    23. Maruani A
    24. Research Group of the Société Française de Dermatologie Pédiatrique (French Society of Pediatric Dermatology)

    (2021) Acute acral eruptions in children during the COVID-19 pandemic: characteristics of 103 children and their family clusters

    Annales de Dermatologie et de Venereologie 148:94–100.

    https://doi.org/10.1016/j.annder.2020.11.005

    • PubMed
    • Google Scholar
28. 1. Jing QL
    2. Liu MJ
    3. Zhang ZB
    4. Fang LQ
    5. Yuan J
    6. Zhang AR
    7. Dean NE
    8. Luo L
    9. Ma MM
    10. Longini I
    11. Kenah E
    12. Lu Y
    13. Ma Y
    14. Jalali N
    15. Yang ZC
    16. Yang Y

    (2020) Household secondary attack rate of COVID-19 and associated determinants in guangzhou, china: a retrospective cohort study

    The Lancet. Infectious Diseases 20:1141–1150.

    https://doi.org/10.1016/S1473-3099(20)30471-0

    • PubMed
    • Google Scholar
29. 1. Jones TC
    2. Biele G
    3. Mühlemann B
    4. Veith T
    5. Schneider J
    6. Beheim-Schwarzbach J
    7. Bleicker T
    8. Tesch J
    9. Schmidt ML
    10. Sander LE
    11. Kurth F
    12. Menzel P
    13. Schwarzer R
    14. Zuchowski M
    15. Hofmann J
    16. Krumbholz A
    17. Stein A
    18. Edelmann A
    19. Corman VM
    20. Drosten C

    (2021) Estimating infectiousness throughout SARS-cov-2 infection course

    Science 373:eabi5273.

    https://doi.org/10.1126/science.abi5273

    • PubMed
    • Google Scholar
30. 1. Koureas M
    2. Speletas M
    3. Bogogiannidou Z
    4. Babalis D
    5. Pinakas V
    6. Pinaka O
    7. Komnos A
    8. Tsoutsa S
    9. Papadamou G
    10. Kyritsi MA
    11. Vontas A
    12. Nakoulas V
    13. Sapoynas S
    14. Kanellopoulos N
    15. Kalompatsios D
    16. Papadouli V
    17. Dadouli K
    18. Soteriades S
    19. Mina P
    20. Mouchtouri VA
    21. Anagnostopoulos L
    22. Stamoulis KE
    23. Agorastos K
    24. Petinaki EA
    25. Prezerakos P
    26. Tsiodras S
    27. Hadjichristodoulou C

    (2021) Transmission dynamics of SARS-cov-2 during an outbreak in a roma community in thessaly, greece-control measures and lessons learned

    International Journal of Environmental Research and Public Health 18:2878.

    https://doi.org/10.3390/ijerph18062878

    • PubMed
    • Google Scholar
31. 1. Langan D
    2. Higgins JPT
    3. Jackson D
    4. Bowden J
    5. Veroniki AA
    6. Kontopantelis E
    7. Viechtbauer W
    8. Simmonds M

    (2019) A comparison of heterogeneity variance estimators in simulated random-effects meta-analyses

    Research Synthesis Methods 10:83–98.

    https://doi.org/10.1002/jrsm.1316

    • PubMed
    • Google Scholar
32. 1. Lau MSY
    2. Grenfell B
    3. Thomas M
    4. Bryan M
    5. Nelson K
    6. Lopman B

    (2020) Characterizing superspreading events and age-specific infectiousness of SARS-cov-2 transmission in georgia, USA

    PNAS 117:22430–22435.

    https://doi.org/10.1073/pnas.2011802117

    • PubMed
    • Google Scholar
33. 1. Laxminarayan R
    2. Wahl B
    3. Dudala SR
    4. Gopal K
    5. Mohan B C
    6. Neelima S
    7. Jawahar Reddy KS
    8. Radhakrishnan J
    9. Lewnard JA

    (2020) Epidemiology and transmission dynamics of COVID-19 in two indian states

    Science 370:691–697.

    https://doi.org/10.1126/science.abd7672

    • PubMed
    • Google Scholar
34. 1. Layan M
    2. Gilboa M
    3. Gonen T
    4. Goldenfeld M
    5. Meltzer L
    6. Andronico A
    7. Hozé N
    8. Cauchemez S
    9. Regev-Yochay G

    (2022) Impact of bnt162b2 vaccination and isolation on SARS-cov-2 transmission in israeli households: an observational study

    American Journal of Epidemiology 191:1224–1234.

    https://doi.org/10.1093/aje/kwac042

    • PubMed
    • Google Scholar
35. 1. Li Q
    2. Guan X
    3. Wu P
    4. Wang X
    5. Zhou L
    6. Tong Y
    7. Ren R
    8. Leung KSM
    9. Lau EHY
    10. Wong JY
    11. Xing X
    12. Xiang N
    13. Wu Y
    14. Li C
    15. Chen Q
    16. Li D
    17. Liu T
    18. Zhao J
    19. Liu M
    20. Tu W
    21. Chen C
    22. Jin L
    23. Yang R
    24. Wang Q
    25. Zhou S
    26. Wang R
    27. Liu H
    28. Luo Y
    29. Liu Y
    30. Shao G
    31. Li H
    32. Tao Z
    33. Yang Y
    34. Deng Z
    35. Liu B
    36. Ma Z
    37. Zhang Y
    38. Shi G
    39. Lam TTY
    40. Wu JT
    41. Gao GF
    42. Cowling BJ
    43. Yang B
    44. Leung GM
    45. Feng Z

    (2020a) Early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia

    The New England Journal of Medicine 382:1199–1207.

    https://doi.org/10.1056/NEJMoa2001316

    • PubMed
    • Google Scholar
36. 1. Li B
    2. Zhang S
    3. Zhang R
    4. Chen X
    5. Wang Y
    6. Zhu C

    (2020b) Epidemiological and clinical characteristics of COVID-19 in children: A systematic review and meta-analysis

    Frontiers in Pediatrics 8:591132.

    https://doi.org/10.3389/fped.2020.591132

    • PubMed
    • Google Scholar
37. Book

    1. Liu JS

    (2001)

    Monte Carlo Strategies in Scientific Computing

    Springer-Verlag.

    • Google Scholar
38. 1. Lloyd-Smith JO
    2. Schreiber SJ
    3. Kopp PE
    4. Getz WM

    (2005) Superspreading and the effect of individual variation on disease emergence

    Nature 438:355–359.

    https://doi.org/10.1038/nature04153

    • PubMed
    • Google Scholar
39. 1. Lyngse FP
    2. Kirkeby C
    3. Halasa T
    4. Andreasen V
    5. Skov RL
    6. Møller FT
    7. Krause TG
    8. Mølbak K

    (2022) Nationwide study on SARS-cov-2 transmission within households from lockdown to reopening, denmark, 27 february 2020 to 1 august 2020

    Euro Surveillance 27:2001800.

    https://doi.org/10.2807/1560-7917.ES.2022.27.6.2001800

    • PubMed
    • Google Scholar
40. 1. Madewell ZJ
    2. Yang Y
    3. Longini IM
    4. Halloran ME
    5. Dean NE

    (2021) Factors associated with household transmission of SARS-cov-2: an updated systematic review and meta-analysis

    JAMA Network Open 4:e2122240.

    https://doi.org/10.1001/jamanetworkopen.2021.22240

    • PubMed
    • Google Scholar
41. 1. Méndez-Echevarría A
    2. Sainz T
    3. de Felipe B
    4. Alcolea S
    5. Olbrich P
    6. Goycochea-Valdivia WA
    7. Escosa-García L
    8. Cobo L
    9. Calvo C
    10. Neth O

    (2021) High rates of SARS-cov-2 family transmission in children of healthcare workers during the first pandemic wave in madrid, spain: serologic study

    The Pediatric Infectious Disease Journal 40:e185–e188.

    https://doi.org/10.1097/INF.0000000000003088

    • PubMed
    • Google Scholar
42. 1. Moher D
    2. Liberati A
    3. Tetzlaff J
    4. Altman DG
    5. Group P

    (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

    BMJ 339:b2535.

    https://doi.org/10.1136/bmj.b2535

    • PubMed
    • Google Scholar
43. 1. Mossong J
    2. Hens N
    3. Jit M
    4. Beutels P
    5. Auranen K
    6. Mikolajczyk R
    7. Massari M
    8. Salmaso S
    9. Tomba GS
    10. Wallinga J
    11. Heijne J
    12. Sadkowska-Todys M
    13. Rosinska M
    14. Edmunds WJ

    (2008) Social contacts and mixing patterns relevant to the spread of infectious diseases

    PLOS Medicine 5:e74.

    https://doi.org/10.1371/journal.pmed.0050074

    • PubMed
    • Google Scholar
44. 1. Mousa A
    2. Winskill P
    3. Watson OJ
    4. Ratmann O
    5. Monod M
    6. Ajelli M
    7. Diallo A
    8. Dodd PJ
    9. Grijalva CG
    10. Kiti MC
    11. Krishnan A
    12. Kumar R
    13. Kumar S
    14. Kwok KO
    15. Lanata CF
    16. de Waroux OLP
    17. Leung K
    18. Mahikul W
    19. Melegaro A
    20. Morrow CD
    21. Mossong J
    22. Neal EF
    23. Nokes DJ
    24. Pan-Ngum W
    25. Potter GE
    26. Russell FM
    27. Saha S
    28. Sugimoto JD
    29. Wei WI
    30. Wood RR
    31. Wu J
    32. Zhang J
    33. Walker P
    34. Whittaker C

    (2021) Social contact patterns and implications for infectious disease transmission - a systematic review and meta-analysis of contact surveys

    eLife 10:e70294.

    https://doi.org/10.7554/eLife.70294

    • PubMed
    • Google Scholar
45. 1. Ng OT
    2. Marimuthu K
    3. Koh V
    4. Pang J
    5. Linn KZ
    6. Sun J
    7. De Wang L
    8. Chia WN
    9. Tiu C
    10. Chan M
    11. Ling LM
    12. Vasoo S
    13. Abdad MY
    14. Chia PY
    15. Lee TH
    16. Lin RJ
    17. Sadarangani SP
    18. Chen MI-C
    19. Said Z
    20. Kurupatham L
    21. Pung R
    22. Wang L-F
    23. Cook AR
    24. Leo Y-S
    25. Lee VJ

    (2021) SARS-cov-2 seroprevalence and transmission risk factors among high-risk close contacts: a retrospective cohort study

    The Lancet. Infectious Diseases 21:333–343.

    https://doi.org/10.1016/S1473-3099(20)30833-1

    • PubMed
    • Google Scholar
46. 1. Okita Y
    2. Morita T
    3. Kumanogoh A

    (2022) Duration of SARS-cov-2 RNA positivity from various specimens and clinical characteristics in patients with COVID-19: a systematic review and meta-analysis

    Inflammation and Regeneration 42:16.

    https://doi.org/10.1186/s41232-022-00205-x

    • PubMed
    • Google Scholar
47. 1. Posfay-Barbe KM
    2. Wagner N
    3. Gauthey M
    4. Moussaoui D
    5. Loevy N
    6. Diana A
    7. L’Huillier AG

    (2020) COVID-19 in children and the dynamics of infection in families

    Pediatrics 146:e20201576.

    https://doi.org/10.1542/peds.2020-1576

    • PubMed
    • Google Scholar
48. 1. Qutub M
    2. Aldabbagh Y
    3. Mehdawi F
    4. Alraddadi A
    5. Alhomsy M
    6. Alnahdi A
    7. Fakeeh M
    8. Maghrabi A
    9. Alwagdani M
    10. Bahabri N

    (2022) Duration of viable SARS-cov-2 shedding from respiratory tract in different human hosts and its impact on isolation discontinuation polices revision; a narrative review

    Clinical Infection in Practice 13:100140.

    https://doi.org/10.1016/j.clinpr.2022.100140

    • PubMed
    • Google Scholar
49. 1. Rahmani A
    2. Dini G
    3. Leso V
    4. Montecucco A
    5. Kusznir Vitturi B
    6. Iavicoli I
    7. Durando P

    (2022) Duration of SARS-cov-2 shedding and infectivity in the working age population: a systematic review and meta-analysis

    La Medicina Del Lavoro 113:e2022014.

    https://doi.org/10.23749/mdl.v113i2.12724

    • PubMed
    • Google Scholar
50. 1. Reukers DFM
    2. van Boven M
    3. Meijer A
    4. Rots N
    5. Reusken C
    6. Roof I
    7. van Gageldonk-Lafeber AB
    8. van der Hoek W
    9. van den Hof S

    (2021) High infection attack rates of SARS-cov-2 in dutch households revealed by dense sampling

    Infectious Diseases 74:52–58.

    https://doi.org/10.1101/2021.01.26.21250512

    • Google Scholar
51. 1. Shen Z
    2. Ning F
    3. Zhou W
    4. He X
    5. Lin C
    6. Chin DP
    7. Zhu Z
    8. Schuchat A

    (2004) Superspreading SARS events, Beijing, 2003

    Emerging Infectious Diseases 10:256–260.

    https://doi.org/10.3201/eid1002.030732

    • PubMed
    • Google Scholar
52. Book

    1. Spiegelhalter DJ
    2. Thomas A
    3. Best NG

    (2000)

    WinBUGS Version 1.3 User Manual

    Cambridge: Medical Research Council Biostatistics Unit.

    • Google Scholar
53. 1. Spiegelhalter DJ
    2. Best NG
    3. Carlin BP
    4. van der Linde A

    (2002) Bayesian measures of model complexity and fit

    Journal of the Royal Statistical Society 64:583–639.

    https://doi.org/10.1111/1467-9868.00353

    • Google Scholar
54. 1. Sun K
    2. Wang W
    3. Gao L
    4. Wang Y
    5. Luo K
    6. Ren L
    7. Zhan Z
    8. Chen X
    9. Zhao S
    10. Huang Y
    11. Sun Q
    12. Liu Z
    13. Litvinova M
    14. Vespignani A
    15. Ajelli M
    16. Viboud C
    17. Yu H

    (2021) Transmission heterogeneities, kinetics, and controllability of SARS-cov-2

    Science 371:eabe2424.

    https://doi.org/10.1126/science.abe2424

    • PubMed
    • Google Scholar
55. 1. Sun K
    2. Tempia S
    3. Kleynhans J
    4. von Gottberg A
    5. McMorrow ML
    6. Wolter N
    7. Bhiman JN
    8. Moyes J
    9. du Plessis M
    10. Carrim M
    11. Buys A
    12. Martinson NA
    13. Kahn K
    14. Tollman S
    15. Lebina L
    16. Wafawanaka F
    17. du Toit JD
    18. Gómez-Olivé FX
    19. Mkhencele T
    20. Viboud C
    21. Cohen C

    (2022) SARS-cov-2 transmission, persistence of immunity, and estimates of omicron’s impact in south african population cohorts

    Science Translational Medicine 14:eabo7081.

    https://doi.org/10.1126/scitranslmed.abo7081

    • PubMed
    • Google Scholar
56. 1. Thompson SG
    2. Sharp SJ

    (1999) Explaining heterogeneity in meta-analysis: a comparison of methods

    Statistics in Medicine 18:2693–2708.

    https://doi.org/10.1002/(sici)1097-0258(19991030)18:20<2693::aid-sim235>3.0.co;2-v

    • PubMed
    • Google Scholar
57. 1. Thompson HA

    (2021) SARS-cov-2 setting-specific transmission rates: a systematic review and meta-analysis

    Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America 73:e754–e764.

    https://doi.org/10.1093/cid/ciab100

    • Google Scholar
58. 1. Toth DJA
    2. Leecaster M
    3. Pettey WBP
    4. Gundlapalli AV
    5. Gao H
    6. Rainey JJ
    7. Uzicanin A
    8. Samore MH

    (2015) The role of heterogeneity in contact timing and duration in network models of influenza spread in schools

    Journal of the Royal Society, Interface 12:20150279.

    https://doi.org/10.1098/rsif.2015.0279

    • PubMed
    • Google Scholar
59. 1. Tsang TK
    2. Cauchemez S
    3. Perera RAPM
    4. Freeman G
    5. Fang VJ
    6. Ip DKM
    7. Leung GM
    8. Malik Peiris JS
    9. Cowling BJ

    (2014) Association between antibody titers and protection against influenza virus infection within households

    The Journal of Infectious Diseases 210:684–692.

    https://doi.org/10.1093/infdis/jiu186

    • PubMed
    • Google Scholar
60. 1. Tsang TK
    2. Lau LLH
    3. Cauchemez S
    4. Cowling BJ

    (2016) Household transmission of influenza virus

    Trends in Microbiology 24:123–133.

    https://doi.org/10.1016/j.tim.2015.10.012

    • PubMed
    • Google Scholar
61. 1. Tsang TK
    2. Wang C
    3. Yang B
    4. Cauchemez S
    5. Cowling BJ

    (2021) Using secondary cases to characterize the severity of an emerging or re-emerging infection

    Nature Communications 12:6372.

    https://doi.org/10.1038/s41467-021-26709-7

    • PubMed
    • Google Scholar
62. Software

    1. Tsang T

    (2022) Covid19_transmission_heterogeneity, version swh:1:rev:634390fa66a4bfb998da691e7cd81cf45e38c6db

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:4d4436affc6aa5ddf265a9a9acb638f66eb0cd5d;origin=https://github.com/timktsang/covid19_transmission_heterogeneity;visit=swh:1:snp:6e5eab4d82393ee3420ab8827a701a0129569d57;anchor=swh:1:rev:634390fa66a4bfb998da691e7cd81cf45e38c6db
63. 1. Tsang TK
    2. Fang LQ
    3. Zhang A
    4. Jiang FC
    5. Ruan SM
    6. Liu LZ
    7. Cowling BJ
    8. Liu W
    9. Yang Y

    (2022) Variability in transmission risk of SARS-cov-2 in close contact settings: A contact tracing study in shandong province, china

    Epidemics 39:100553.

    https://doi.org/10.1016/j.epidem.2022.100553

    • Google Scholar
64. 1. Veroniki AA
    2. Jackson D
    3. Bender R
    4. Kuss O
    5. Langan D
    6. Higgins JPT
    7. Knapp G
    8. Salanti G

    (2019) Methods to calculate uncertainty in the estimated overall effect size from a random-effects meta-analysis

    Research Synthesis Methods 10:23–43.

    https://doi.org/10.1002/jrsm.1319

    • PubMed
    • Google Scholar
65. 1. Wilkinson K
    2. Chen X
    3. Shaw S

    (2021) Secondary attack rate of COVID-19 in household contacts in the winnipeg health region, canada

    Canadian Journal of Public Health = Revue Canadienne de Sante Publique 112:12–16.

    https://doi.org/10.17269/s41997-020-00451-x

    • PubMed
    • Google Scholar
66. 1. Wong G
    2. Liu W
    3. Liu Y
    4. Zhou B
    5. Bi Y
    6. Gao GF

    (2015) MERS, SARS, and ebola: the role of super-spreaders in infectious disease

    Cell Host & Microbe 18:398–401.

    https://doi.org/10.1016/j.chom.2015.09.013

    • PubMed
    • Google Scholar
67. 1. Wong F
    2. Collins JJ

    (2020) Evidence that coronavirus superspreading is fat-tailed

    PNAS 117:29416–29418.

    https://doi.org/10.1073/pnas.2018490117

    • PubMed
    • Google Scholar
68. 1. Xu CLH
    2. Raval M
    3. Schnall JA
    4. Kwong JC
    5. Holmes NE

    (2020) Duration of respiratory and gastrointestinal viral shedding in children with SARS-cov-2: A systematic review and synthesis of data

    The Pediatric Infectious Disease Journal 39:e249–e256.

    https://doi.org/10.1097/INF.0000000000002814

    • PubMed
    • Google Scholar
69. 1. Yan D
    2. Zhang X
    3. Chen C
    4. Jiang D
    5. Liu X
    6. Zhou Y
    7. Huang C
    8. Zhou Y
    9. Guan Z
    10. Ding C
    11. Chen L
    12. Lan L
    13. Fu X
    14. Wu J
    15. Li L
    16. Yang S

    (2021) Characteristics of viral shedding time in SARS-cov-2 infections: A systematic review and meta-analysis

    Frontiers in Public Health 9:652842.

    https://doi.org/10.3389/fpubh.2021.652842

    • PubMed
    • Google Scholar
70. 1. Yang B
    2. Wu P
    3. Lau EHY
    4. Wong JY
    5. Ho F
    6. Gao H
    7. Xiao J
    8. Adam DC
    9. Ng TWY
    10. Quan J
    11. Tsang TK
    12. Liao Q
    13. Cowling BJ
    14. Leung GM

    (2021) Changing disparities in coronavirus disease 2019 (COVID-19) burden in the ethnically homogeneous population of hong kong through pandemic waves: an observational study

    Clinical Infectious Diseases 73:2298–2305.

    https://doi.org/10.1093/cid/ciab002

    • PubMed
    • Google Scholar
71. 1. Zhang Y
    2. Cen M
    3. Hu M
    4. Du L
    5. Hu W
    6. Kim JJ
    7. Dai N

    (2021) Prevalence and persistent shedding of fecal SARS-cov-2 RNA in patients with COVID-19 infection: A systematic review and meta-analysis

    Clinical and Translational Gastroenterology 12:e00343.

    https://doi.org/10.14309/ctg.0000000000000343

    • PubMed
    • Google Scholar
72. 1. Zhao S
    2. Shen M
    3. Musa SS
    4. Guo Z
    5. Ran J
    6. Peng Z
    7. Zhao Y
    8. Chong MKC
    9. He D
    10. Wang MH

    (2021) Inferencing superspreading potential using zero-truncated negative binomial model: exemplification with COVID-19

    BMC Medical Research Methodology 21:30.

    https://doi.org/10.1186/s12874-021-01225-w

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "The effect of variation of individual infectiousness on SARS-CoV-2 transmission in households" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Miles Davenport as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) Both reviewers have highlighted a number of areas where additional clarification (particularly for σ_var) which currently does not have sufficient detail/interpretation to warrant publication.

2) Questions about parameter identifiability and model validation were not sufficiently addressed – particularly in light of the many individual-level parameters estimated in the analysis.

3) Additional exploration of empirical relationships (such as individual infectiousness and household size, as an example) should be explored. For example, how much of the heterogeneity is driven by biological factors (such as being the primary case).

4) Additional work should be done to help disentangle differences between biological or behavioral factors since this will greatly change the interpretation of inferred parameters. In addition, more contextual information should be provided (socio-economic status and the implications of this as a confounder, a more detailed exploration of interventions that were in place and their implications) should be done.

Reviewer #1 (Recommendations for the authors):

Tsang et al. evaluate the degree of transmission heterogeneity among COVID-19 cases using data compiled from 17 household transmission studies. Transmission heterogeneity and super spreading are important but incompletely understood drivers of epidemic dynamics. While it is clear that substantial heterogeneity exists among infectors, but the relative contribution of behavior (e.g. the number of contacts), timing (e.g. large number of social interactions near the peak of infectiousness), and intrinsic biological differences in infectiousness are not resolved.

Tsang et al. compiled household transmission data from 17 published studies. They independently fit a model to each data set in order to quantify transmission heterogeneity arising from intrinsic biological differences in infectiousness, after controlling for differences in the number of contacts (i.e. household size), and timing (via the inferred time since infection). Repeated application of the same model to multiple datasets is a strength of this study because it allows the authors to assess associations between the inferred parameters and aspects of study design that can influence case ascertainment.

The authors estimate substantial intrinsic transmission heterogeneity via the parameter σ_var, which could be attributed to individual differences in viral loads or differences in the infector's contact intensity with others in the household. They show that intrinsic, within-household variation is associated with other population-level metrics of transmission heterogeneity, including p_80 and p_0.

Although the study focuses on the household size as a confounder of heterogeneity in infectiousness, the ways that the model controls for household size and the interpretation of the inferred heterogeneity parameter, σ_var could be explained more clearly. Furthermore, it seems that basic empirical relationships between individual infectiousness (δ_i) and household size are not explored. Finally, questions about parameter identifiability and model validation could be addressed more extensively, especially in light of the many individual-level parameters estimated in the analysis.

Overall, the study design is sound and the analysis of multiple datasets is a strength. However, the modeling approach and the biological interpretation/significance of the results could be more clearly explained. Also, given the large number of individual-level parameters estimated, it would be ideal to see a direct assessment of parameter identifiability (e.g. correlation plots), and model validation on simulated data (i.e. can the model estimate known parameters from simulated data?). The latter shouldn't be too difficult, given that the authors already have a simulation model for tests of model adequacy.

Specific comments:

1. Please explain assumptions about the individual infectiousness profile f(.) (from section 2 of the Materials and methods). What is the assumed shape of this function? How is it parameterized? How could the inclusion of this function in the model interact with inferred individual times of infection, especially if times of infection are inaccurate? On a related note, the authors might consider introducing the concept of contact timing as a factor that can influence heterogeneity in transmission.

2. The household size is a key conceptual focus of the study, but the main text methods/results don't explain clearly how the model accounts for household size or contact number. My interpretation is that the model accounts for household size in two ways. First. the model estimates the per-contact hazard of infection, which implicitly controls for household size/contact number. Second, the model includes a parameter, β, which represents a dilution effect, wherein the transmission hazard might be diluted in larger households due to less per-individual contact intensity. I'm not sure that the dilution effect is currently explained anywhere in the main text, and it could also be explained more clearly that the model is designed to estimate transmission heterogeneity after adjusting for household size.

2a. It would be helpful if the authors could discuss the distinction between contact number, contact frequency, and contact duration somewhere in the text.

3. Given the focus on household size and the number of contacts, it seems there are missed opportunities to explore how the inferred level of individual infectiousness (δ_i) co-varies with metrics like household size.

4. Please explain in more detail the meta-regression sensitivity analyses.

5. I'm not sure that tables S6-S9 are cross-referenced anywhere in the main text, and their legends aren't sufficiently detailed to fully communicate how to interpret these tables in the context of the broader study.

Reviewer #2 (Recommendations for the authors):

Transmission heterogeneity and superspreading of SARS-CoV-2 have been demonstrated repeatedly through real-world observational studies, with a small fraction of individuals accounting for the majority of onward transmission. However, this observed transmission heterogeneity is likely a superposition of accumulated variations from multiple factors, including but not limited to host behavioral factors such as the variation in contact numbers contact duration, and contact settings; variations in the adoption of preventive measures (mask-wearing, physical distancing, etc.); the effects of NPIs (case isolations and contact quarantines, population-level lockdowns); as well as biological factors including differences in shedding duration and intensity of the primary case, variation in susceptibility among close contacts. Fewer studies, however, attempted to isolate the effects of individual factors contributing to the overall transmission heterogeneity, while controlling for other factors. In the manuscript entitled "The effect of variation of individual infectiousness on SARS-CoV-2 transmission in households", the authors aim at characterizing the variation in individual infectiousness of SARS-CoV-2, controlling for other host factors. To achieve this, the authors performed a meta-analysis of household transmission studies conducted during or not too longer after the initial SARS-CoV-2 wave caused by the ancestral strain across the globe. The authors fitted a household transmission model to the curated data and estimated the variation in infectiousness by introducing random effects of individual infectiousness and its population-level distribution.

The study has several strengths. First, by choosing analyzing data from the household study, the authors were able to control for several key factors contributing to the overall transmission heterogeneity but not variation in individual infectiousness, including the number of contacts as well the setting of transmission (household). The authors also incorporated an additional parameter in the household transmission model to adjust for the impact of household size on household transmission risk, which has been identified as an important risk factor for SARS-CoV-2 transmission within the household across multiple studies. The authors also curated the studies during the early stage of the pandemic so that most household contacts would remain naïve during the study period, and the immune status of the household contact (either due to prior infection or vaccination) would be unlikely to confound the results of the study.

However, the study also has a few limitations. First, it is difficult to disentangle if the observed variation in infectiousness is due to biological factors or behavioral factors. During the study period of interest, public health agencies across the globe were recommending at-home isolation guidelines aiming at reducing transmission within the household, including mask-wearing when in contact with other household members, using separate bedrooms/bathrooms, avoiding having meals together, etc. The differences in the guidelines across nationals/regions as well as the level of compliance with guidelines at the household level would also impact individual household transmission risk. Second, the risk of acquiring infections from the community could be heavily influenced by the socio-economic status, since multiple studies have clearly demonstrated stark disparity of COVID-19 burden, factors such as occupation (essential workers tend to be low-wage jobs) assess to PPE and healthcare were likely to contribute to the observed disparity. Lastly, it is also difficult to entangle if the observed heterogeneity is due to the biological factors of the primary case (i.e., variation in shedding duration/intensity) or the contact (variation in susceptibility). The current formulation of the transmission model only addresses the former not the latter.

For the household transmission model concerning imputation of the timing of infection, it is unclear if the timing of infection is partially missing, or if the timing of infection were unavailable for all studies. If it is the former case, the authors should give a more detailed description of the imputation process in a study-by-study fashion and report the proportion of infection time that was imputed. If it is the latter case, I do not understand the benefit of explicitly modeling the temporal infectiousness profile, thus I would recommend the authors use a simpler chain-binomial model fitting to only the binary outcome (of infected or not) unless the authors make a convincing case otherwise.

Both p80 and p0 are poor statistics to characterize household transmission due to the discrete nature of household contact numbers. For example, it is unintuitive to interpret and compare p80 for a household size of 2 vs 10 (for a household size of 2 conditional on an index case within the household this could only be interpreted as the proportion of households with secondary infections, which is cross-household level statistics, while for a household size of 10, it can be interpreted as a meaningful statistical for each household). Furthermore, the interpretation of p80 as a metric for transmission heterogeneity becomes even murkier if multi-generation transmission within the household is considered, especially for a larger household. For p0, this metric is heavily dependent on household size. For example, one would expect the p0 for a household size of 10 (with on index) to be way smaller than p0 household size of 2 (assuming no dilution effects on transmission with the increase of household size), as in the former case p0 characterize the probability of all 9 uninfected contacts escaping infection from the index, while in the latter p0 characterizes the probability to escape infection by only 1 household contact. Thus, it is meaningless to compare p0 without controlling for household size across studies. For the arguments above, I recommend the authors remove both p80 and p0 from the paper.

Through the formulation of the household transmission model, the authors essentially assume a lognormal distribution of individual infectiousness, which is a long-tailed distribution by nature. I think this is an important point that needs to be highlighted. Has the author tried a normal distribution instead of a log-normal distribution? Would a normal distribution fit better or worse to the data than the log-normal one? I recommend the authors conduct a sensitivity analysis of a normal distribution of the individual infectiousness (i.e., not taking the exponential of the σ term for the hazard of infection function on page 7).

Table S5 is cropped.

Essential revisions:

1) Both reviewers have highlighted a number of areas where additional clarification (particularly for σ_var) which currently does not have sufficient detail/interpretation to warrant publication.

We have modified the discussion to talk about the interpretation and limitation of infectiousness variation. We clarified that the estimated infectiousness variations measures the variation caused by difference in individuals, which may be contributed by both biological factors and host behaviors, but not other environmental factors. Please see Response 7 for more details.

We also added a limitation that our approach could not disentangle the observed heterogeneity from biological factors affecting infectiousness and variations in susceptibility of contacts. Please see Response 14 for more details.

2) Questions about parameter identifiability and model validation were not sufficiently addressed – particularly in light of the many individual-level parameters estimated in the analysis.

We have added the correlation plots for the posterior distribution of parameters, and also rerun a longer MCMC chain (with 500K iterations instead of 100K iterations) to ensure that our estimates of infectiousness variations are valid. Please see Response 7 for more details.

3) Additional exploration of empirical relationships (such as individual infectiousness and household size, as an example) should be explored. For example, how much of the heterogeneity is driven by biological factors (such as being the primary case).

The individual infectiousness parameters are nuisance parameter in our work, since there is no individual level information on the case and contacts. However, our model is designed in a way that covariates affecting susceptibility or infectiousness could be added if individual level information is available. Please see Response 11 for more details.

We also conduct meta-analysis and meta-regression to determine if any factors may be associated with the infectiousness variation. We found that this is not associated with the mean numbers of contacts and household size, and only associated with if the study used PCR to confirm secondary cases only (Table S11).

4) Additional work should be done to help disentangle differences between biological or behavioral factors since this will greatly change the interpretation of inferred parameters. In addition, more contextual information should be provided (socio-economic status and the implications of this as a confounder, a more detailed exploration of interventions that were in place and their implications) should be done.

We have extended our discussion on difference between biological and behavioral factors. In summary, behavioral factors including difference in frequency and duration of contacts caused by different factors, like the role in family, recommendation from public health agencies, among individuals. Biological factors include difference in viral shedding and infectious period among individuals. We also provide more information on other confounding factors, such as social-economic status, recommendation from public health agencies, the implementation of lockdowns that may increase the time spending at home. These may increase of decrease the SAR, which is shown to be negatively correlated with infectiousness variations. Please see Response 7 and 14 for more details.

Reviewer #1 (Recommendations for the authors):

Tsang et al. evaluate the degree of transmission heterogeneity among COVID-19 cases using data compiled from 17 household transmission studies. Transmission heterogeneity and super spreading are important but incompletely understood drivers of epidemic dynamics. While it is clear that substantial heterogeneity exists among infectors, but the relative contribution of behavior (e.g. the number of contacts), timing (e.g. large number of social interactions near the peak of infectiousness), and intrinsic biological differences in infectiousness are not resolved.

Tsang et al. compiled household transmission data from 17 published studies. They independently fit a model to each data set in order to quantify transmission heterogeneity arising from intrinsic biological differences in infectiousness, after controlling for differences in the number of contacts (i.e. household size), and timing (via the inferred time since infection). Repeated application of the same model to multiple datasets is a strength of this study because it allows the authors to assess associations between the inferred parameters and aspects of study design that can influence case ascertainment.

The authors estimate substantial intrinsic transmission heterogeneity via the parameter σ_var, which could be attributed to individual differences in viral loads or differences in the infector's contact intensity with others in the household. They show that intrinsic, within-household variation is associated with other population-level metrics of transmission heterogeneity, including p_80 and p_0.

Although the study focuses on the household size as a confounder of heterogeneity in infectiousness, the ways that the model controls for household size and the interpretation of the inferred heterogeneity parameter, σ_var could be explained more clearly. Furthermore, it seems that basic empirical relationships between individual infectiousness (δ_i) and household size are not explored. Finally, questions about parameter identifiability and model validation could be addressed more extensively, especially in light of the many individual-level parameters estimated in the analysis.

Overall, the study design is sound and the analysis of multiple datasets is a strength. However, the modeling approach and the biological interpretation/significance of the results could be more clearly explained. Also, given the large number of individual-level parameters estimated, it would be ideal to see a direct assessment of parameter identifiability (e.g. correlation plots), and model validation on simulated data (i.e. can the model estimate known parameters from simulated data?). The latter shouldn't be too difficult, given that the authors already have a simulation model for tests of model adequacy.

Thank you for your suggestions. In brief, we summarized that the estimated infectiousness variations measures the variation caused by difference in individuals, which may be affected by both biological factors and host behaviors but not other confounding factors, such as the difference in number of contacts, pre-existing immunity among contacts or difference in transmissibility among variants. We also added more details on the interpretation of the estimated infectiousness variation.

We added ‘Given that we focused our analysis on households with known number of contacts in studies conducted in the early outbreaks caused by ancestral strains, the estimated infectiousness variations are corrected for the variations caused by number of contacts and transmission risks in different settings, difference in pre-existing immunity among contacts (almost everyone is naïve and unvaccinated), and the difference in transmissibility among variants. Hence, this estimated infectiousness variation measures the variation caused by difference in individuals, which may be contributed by both biological factors and host behaviors, but not other confounding factors such as…. ‘ in the Discussion section in the revised manuscript.

We conducted a simulation study, and found there is no important systematic bias, and that 88-100% of 95% credible interval covered the simulation value. We also added the correlation plots in the revised manuscript. The correlation between parameters are moderate. We run long MCMC chain to ensure our results are valid. We conduct a sensitivity by running the MCMC chain with 500,000 iterations and the results are similar to the one with 100,000 iterations. This suggests that 100,000 iterations are enough (our main results).

We added ‘Simulation studies demonstrated there is no important systematic bias, with 88 to 100% of the 95% credible intervals covering the simulation value (Appendix 1 – Table 8). This suggests that the algorithm could estimate adequately the posterior distribution.’ In the Result section of the revised manuscript.

Specific comments:

1. Please explain assumptions about the individual infectiousness profile f(.) (from section 2 of the Materials and methods). What is the assumed shape of this function? How is it parameterized? How could the inclusion of this function in the model interact with inferred individual times of infection, especially if times of infection are inaccurate? On a related note, the authors might consider introducing the concept of contact timing as a factor that can influence heterogeneity in transmission.

We apologize for the confusion. We added a table to summarize the assumed values extracted from the reference papers, and a plot for the assumed individual infectiousness profile function f(.). We also moved the exact formula and more details of the model from appendix to the main text, to enhance the readability.

In the estimation, the infectiousness profile are used to determine the generation time, and it would influence the observed heterogeneity of distribution of number of secondary cases (including all generations). When there is no difference in infectiousness variation among cases, if the generation time is shorter, there will be more generations of cases, and hence the observed distribution of number of secondary cases would be more heterogeneous. Hence, this assumption is setting a ‘baseline’ of observed heterogeneity of observed distribution of number of secondary cases when there is no difference in infectiousness among cases, so that the model could estimate the extra heterogeneity due to variations of individual infectiousness.

We thank for the reviewer’s suggestion about contact timing, which is a good point and we also added in the Discussion. We added ‘the duration of contact could vary, which could also contribute to the variations in infectiousness of cases.’ in the Discussion section of the revised manuscript.

2. The household size is a key conceptual focus of the study, but the main text methods/results don't explain clearly how the model accounts for household size or contact number. My interpretation is that the model accounts for household size in two ways. First. the model estimates the per-contact hazard of infection, which implicitly controls for household size/contact number. Second, the model includes a parameter, β, which represents a dilution effect, wherein the transmission hazard might be diluted in larger households due to less per-individual contact intensity. I'm not sure that the dilution effect is currently explained anywhere in the main text, and it could also be explained more clearly that the model is designed to estimate transmission heterogeneity after adjusting for household size.

We thank the reviewer’s a comprehensive summary on the way that the model accounts for the household size or number of household contacts. We added this information, including the dilution effect, to the revised manuscript as follows:

We added ‘the model could estimate the per-contact hazard of infection, which implicitly controls for number of household contacts in households.’ In the Result section of revised manuscript.

We added ‘In the model, the hazard of infection of individual j at time t from an infected household member i, with infection time t_i in household k, was ${\displaystyle {\lambda }_{i\to j}\left(t\right)=\frac{{\lambda }_h}{X_k^{\beta }}\ast \exp \left({\delta }_i\right)\ast f(t-t_i)}$ where ${\displaystyle {\lambda }_h}$ was the baseline hazard, ${\displaystyle {\delta }_i}$ followed a Normal distribution with mean 0 and standard deviation ${\displaystyle {\sigma }_{\text{var}}}$, which quantified the variation of individual infectiousness (hereafter denoted as infectiousness variation). The relative infectiousness of case i compared with case j was exp(${\displaystyle {\delta }_i}$)/exp(${\displaystyle {\delta }_j}$).

${\displaystyle X_k}$ was the number of household contacts. ${\displaystyle \beta }$ was the parameter describing the relationship between number of household contacts and transmission rate. It ranged from 0 to 1, with 0 indicating that the transmission rate was independent of number of household contacts while 1 indicated that the transmission rate was inversely proportional to number of household contacts (i.e. dilution effect of the contact time per contact which was lower when the number of household contacts is higher).’ to the Method section of the revised manuscript.

2a. It would be helpful if the authors could discuss the distinction between contact number, contact frequency, and contact duration somewhere in the text.

Thank you for your suggestion. We agree this is an important point and added to the revised manuscript. We added ‘It should be noted that there are several components in the contacts, including the numbers, frequencies and duration of contacts that may contribute to the infectiousness variations. Previous studies also suggest that their relative importance in transmission may be different among viruses.’ In the Discussion in the revised manuscript.

3. Given the focus on household size and the number of contacts, it seems there are missed opportunities to explore how the inferred level of individual infectiousness (δ_i) co-varies with metrics like household size.

Thank you for your suggestion. We estimated that the posterior correlation between inferred level of individual infectiousness (δ_i) and household size are not significant ( the credible intervals are all covered one). This suggested the inferred individual infectiousness was not correlated with household size.

4. Please explain in more detail the meta-regression sensitivity analyses.

Sorry for the confusion. Meta-analysis and Meta-regression is conducted to pool the estimates of infectiousness variation, which was estimated separately for 14 identified studies. Furthermore, it allows us to determine the potential association between estimated infectiousness variation and study characteristics, including the following factors: the mean and SD of number of household contacts, implementation of lockdown, ascertainment method of index and secondary cases, only ancestral strains are circulating in study period, and all household contacts were tested. We have clarified in the Method section of the revised manuscript.

5. I'm not sure that tables S6-S9 are cross-referenced anywhere in the main text, and their legends aren't sufficiently detailed to fully communicate how to interpret these tables in the context of the broader study.

These tables summarize the systematic review and meta-analysis about viral shedding of COVID-19 cases, which is a proxy of individual infectiousness of cases. They showed heterogeneity in the level of viral shedding of COVID-19 cases, which support our results about variation of individual infectiousness.

Reviewer #2 (Recommendations for the authors):

Transmission heterogeneity and superspreading of SARS-CoV-2 have been demonstrated repeatedly through real-world observational studies, with a small fraction of individuals accounting for the majority of onward transmission. However, this observed transmission heterogeneity is likely a superposition of accumulated variations from multiple factors, including but not limited to host behavioral factors such as the variation in contact numbers contact duration, and contact settings; variations in the adoption of preventive measures (mask-wearing, physical distancing, etc.); the effects of NPIs (case isolations and contact quarantines, population-level lockdowns); as well as biological factors including differences in shedding duration and intensity of the primary case, variation in susceptibility among close contacts. Fewer studies, however, attempted to isolate the effects of individual factors contributing to the overall transmission heterogeneity, while controlling for other factors. In the manuscript entitled "The effect of variation of individual infectiousness on SARS-CoV-2 transmission in households", the authors aim at characterizing the variation in individual infectiousness of SARS-CoV-2, controlling for other host factors. To achieve this, the authors performed a meta-analysis of household transmission studies conducted during or not too longer after the initial SARS-CoV-2 wave caused by the ancestral strain across the globe. The authors fitted a household transmission model to the curated data and estimated the variation in infectiousness by introducing random effects of individual infectiousness and its population-level distribution.

The study has several strengths. First, by choosing analyzing data from the household study, the authors were able to control for several key factors contributing to the overall transmission heterogeneity but not variation in individual infectiousness, including the number of contacts as well the setting of transmission (household). The authors also incorporated an additional parameter in the household transmission model to adjust for the impact of household size on household transmission risk, which has been identified as an important risk factor for SARS-CoV-2 transmission within the household across multiple studies. The authors also curated the studies during the early stage of the pandemic so that most household contacts would remain naïve during the study period, and the immune status of the household contact (either due to prior infection or vaccination) would be unlikely to confound the results of the study.

However, the study also has a few limitations. First, it is difficult to disentangle if the observed variation in infectiousness is due to biological factors or behavioral factors. During the study period of interest, public health agencies across the globe were recommending at-home isolation guidelines aiming at reducing transmission within the household, including mask-wearing when in contact with other household members, using separate bedrooms/bathrooms, avoiding having meals together, etc. The differences in the guidelines across nationals/regions as well as the level of compliance with guidelines at the household level would also impact individual household transmission risk. Second, the risk of acquiring infections from the community could be heavily influenced by the socio-economic status, since multiple studies have clearly demonstrated stark disparity of COVID-19 burden, factors such as occupation (essential workers tend to be low-wage jobs) assess to PPE and healthcare were likely to contribute to the observed disparity. Lastly, it is also difficult to entangle if the observed heterogeneity is due to the biological factors of the primary case (i.e., variation in shedding duration/intensity) or the contact (variation in susceptibility). The current formulation of the transmission model only addresses the former not the latter.

We appreciate the reviewer’s very comprehensive summary of our work and unique insight of the interpretation of our results, and we added many of them in the revised manuscript. We expanded the section of our discussion listing the different limitations, in particular the fact that our approach can not disentangle the observed heterogeneity from biological factors affecting infectiousness and variations in susceptibility of contacts.

We added the followings in the Discussion in the revised manuscript:

“Given that we focused our analysis on households with known number of contacts in studies conducted in the early outbreaks caused by ancestral strains, the estimated infectiousness variations do not include the variations caused by number of contacts and transmission risks in different settings, difference in pre-existing immunity among contacts (almost everyone is naïve and unvaccinated), and the difference in transmissibility among variants. Hence, this estimated infectiousness variation measures the variation caused by difference in individuals, which may be contributed by both biological factors and host behaviors, but not other confounding factors.”

“It should be noted that there are several components in the contacts, including the numbers, frequencies and duration of contacts that may contribute to the infectiousness variations. Previous studies also suggest that their relative importance in transmission may be different among viruses. Individual behaviors may also be influenced by the implemented control measures and recommendation from public health agencies. For example, lockdown and stay-at-home orders may increase the time spending at home. Mask-wearing when in contact with other household members, using separate bedrooms and bathrooms, avoiding having meals together may reduce the transmission in households. Also, social disparity such as occupation may increase or reduce the risk of transmission in households, including the availability of personal protective equipment (PPE), or being healthcare worker. These may have impacted the SAR in households, but also estimated infectiousness variation in this setting.”

“In addition, we could not include factors affecting susceptibility to infection. Our estimates of infectiousness variation should be interpreted in light of these limitations: they capture heterogeneity in infectiousness due to demographic, host and biological factors. However, in one study that included susceptibility component in the estimation of individual infectiousness, substantial heterogeneity remained with 20% of cases estimated to contribute to 80% of transmission.”

“Third, we assumed that risks of infection from community for all households are the same, but there were different factors that may affecting this, including occupations, such as healthcare workers, social economic status that related to assess to personal protective equipment”

For the household transmission model concerning imputation of the timing of infection, it is unclear if the timing of infection is partially missing, or if the timing of infection were unavailable for all studies. If it is the former case, the authors should give a more detailed description of the imputation process in a study-by-study fashion and report the proportion of infection time that was imputed. If it is the latter case, I do not understand the benefit of explicitly modeling the temporal infectiousness profile, thus I would recommend the authors use a simpler chain-binomial model fitting to only the binary outcome (of infected or not) unless the authors make a convincing case otherwise.

Sorry of confusion. The timing of infection were unavailable for all studies. The reason of not using a simple chain-binomial model is that it is impossible to add the parameters characterizing individual infectiousness (δ_i) in the model. When there are different types of susceptible in data, chain-binomial model required adding separate probability parameters for each type of susceptible in the model, with each type of transmission (from household members or from outside households).

We have clarified this by adding ‘Since the data were extracted from publication, the infection time of all cases was unavailable for all studies.’ in the Method section in the revised manuscript. We also added a section in Appendix to explain this.

Both p80 and p0 are poor statistics to characterize household transmission due to the discrete nature of household contact numbers. For example, it is unintuitive to interpret and compare p80 for a household size of 2 vs 10 (for a household size of 2 conditional on an index case within the household this could only be interpreted as the proportion of households with secondary infections, which is cross-household level statistics, while for a household size of 10, it can be interpreted as a meaningful statistical for each household). Furthermore, the interpretation of p80 as a metric for transmission heterogeneity becomes even murkier if multi-generation transmission within the household is considered, especially for a larger household. For p0, this metric is heavily dependent on household size. For example, one would expect the p0 for a household size of 10 (with on index) to be way smaller than p0 household size of 2 (assuming no dilution effects on transmission with the increase of household size), as in the former case p0 characterize the probability of all 9 uninfected contacts escaping infection from the index, while in the latter p0 characterizes the probability to escape infection by only 1 household contact. Thus, it is meaningless to compare p0 without controlling for household size across studies. For the arguments above, I recommend the authors remove both p80 and p0 from the paper.

Thank you. We agree with the reviewer and removed p80 and p0 from the paper.

Through the formulation of the household transmission model, the authors essentially assume a lognormal distribution of individual infectiousness, which is a long-tailed distribution by nature. I think this is an important point that needs to be highlighted. Has the author tried a normal distribution instead of a log-normal distribution? Would a normal distribution fit better or worse to the data than the log-normal one? I recommend the authors conduct a sensitivity analysis of a normal distribution of the individual infectiousness (i.e., not taking the exponential of the σ term for the hazard of infection function on page 7).

Thank you for this suggestion. We conducted a sensitivity analysis for assuming the individual infectiousness followed a Γ distribution, which has a shorter tail compared with lognormal distribution. However, model comparison suggested that assuming γ distribution for individual infectiousness parameter performed substantially worse, compared with the lognormal distribution in the main analysis (Appendix 1 – Table 9).

Table S5 is cropped.

Sorry for the mistake, we revised the table.

Author details

1. Tim K Tsang

   1. WHO Collaborating Centre for Infectious Disease Epidemiology and Control, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China
   2. Laboratory of Data Discovery for Health, Hong Kong, China

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing – review and editing

   For correspondence

   timtsang@connect.hku.hk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5037-6776

2. Xiaotong Huang

   WHO Collaborating Centre for Infectious Disease Epidemiology and Control, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China

   Contribution

   Data curation, Investigation, Visualization

   Competing interests

   No competing interests declared

3. Can Wang

   WHO Collaborating Centre for Infectious Disease Epidemiology and Control, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China

   Contribution

   Data curation, Investigation

   Competing interests

   No competing interests declared

4. Sijie Chen

   WHO Collaborating Centre for Infectious Disease Epidemiology and Control, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China

   Contribution

   Investigation, Visualization

   Competing interests

   No competing interests declared

5. Bingyi Yang

   WHO Collaborating Centre for Infectious Disease Epidemiology and Control, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China

   Contribution

   Writing – review and editing

   Competing interests

   No competing interests declared

6. Simon Cauchemez

   Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, Paris, France

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Writing – review and editing

   Contributed equally with

   Benjamin John Cowling

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9186-4549

7. Benjamin John Cowling

   Laboratory of Data Discovery for Health, Hong Kong, China

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Writing – review and editing

   Contributed equally with

   Simon Cauchemez

   For correspondence

   bcowling@hku.hk

   Competing interests

   consults for AstraZeneca, Fosun Pharma, GlaxoSmithKline, Moderna, Pfizer, Roche and Sanofi Pasteur. The authors report no other potential conflicts of interest

   "This ORCID iD identifies the author of this article:" 0000-0002-6297-7154

• 761

  Page views

• 122

  Downloads

• 2

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.