Association of Age and Pediatric Household Transmission of SARS-CoV-2 Infection | Infectious Diseases | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Flow Diagram of Study Cohort
Flow Diagram of Study Cohort
Figure 2.  Bubble Plot of Age-to-Age Transmission
Bubble Plot of Age-to-Age Transmission
Table 1.  Characteristics of Pediatric Index Cases by Age Group
Characteristics of Pediatric Index Cases by Age Group
Table 2.  Adjusted Odds Ratios and 95% Confidence Intervals for the Associations Between Index Case Age Group and Odds of Transmitting SARS-CoV-2 to Household Contacts
Adjusted Odds Ratios and 95% Confidence Intervals for the Associations Between Index Case Age Group and Odds of Transmitting SARS-CoV-2 to Household Contacts
1.
Zhang  J, Litvinova  M, Liang  Y,  et al.  Changes in contact patterns shape the dynamics of the COVID-19 outbreak in China.   Science. 2020;368(6498):1481-1486. doi:10.1126/science.abb8001PubMedGoogle ScholarCrossref
2.
Zhu  Y, Bloxham  CJ, Hulme  KD,  et al.  A meta-analysis on the role of children in severe acute respiratory syndrome coronavirus 2 in household transmission clusters.   Clin Infect Dis. 2021;72(12):e1146-e1153. doi:10.1093/cid/ciaa1825PubMedGoogle ScholarCrossref
3.
Chang  T-H, Wu  J-L, Chang  L-Y.  Clinical characteristics and diagnostic challenges of pediatric COVID-19: a systematic review and meta-analysis.   J Formos Med Assoc. 2020;119(5):982-989. doi:10.1016/j.jfma.2020.04.007PubMedGoogle ScholarCrossref
4.
Hyde  Z.  Difference in SARS-CoV-2 attack rate between children and adults may reflect bias.   Clin Infect Dis. 2021;ciab183. doi:10.1093/cid/ciab183PubMedGoogle Scholar
5.
Goldstein  E, Lipsitch  M, Cevik  M.  On the effect of age on the transmission of SARS-CoV-2 in households, schools and the community.   medRxiv. Preprint posted July 28, 2020. doi:10.1101/2020.07.19.20157362Google Scholar
6.
van der Hoek  W, Backer  JA, Bodewes  R,  et al.  De rol van kinderen in de transmissie van SARS-CoV-2.   Ned Tijdschr Geneeskd. 2020;164(25):D5140.PubMedGoogle Scholar
7.
Ludvigsson  JF.  Children are unlikely to be the main drivers of the COVID-19 pandemic: a systematic review.   Acta Paediatr. 2020;109(8):1525-1530. doi:10.1111/apa.15371PubMedGoogle ScholarCrossref
8.
Arnedo-Pena  A, Sabater-Vidal  S, Meseguer-Ferrer  N,  et al  COVID-19 secondary attack rate and risk factors in household contacts in Castellon (Spain): preliminary report.   Rev Enf Emerg. 2020;19(2):64-70.Google Scholar
9.
Jing  Q-L, Liu  M-J, Zhang  Z-B,  et al.  Household secondary attack rate of COVID-19 and associated determinants in Guangzhou, China: a retrospective cohort study.   Lancet Infect Dis. 2020;20(10):1141-1150. doi:10.1016/S1473-3099(20)30471-0PubMedGoogle ScholarCrossref
10.
Dattner  I, Goldberg  Y, Katriel  G,  et al  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.   medRxiv. Preprint posted October 11, 2020. doi:10.1101/2020.06.03.20121145Google Scholar
11.
Sun  K, Wang  W, Gao  L,  et al.  Transmission heterogeneities, kinetics, and controllability of SARS-CoV-2.   Science. 2021;371(6526):eabe2424. doi:10.1126/science.abe2424PubMedGoogle Scholar
12.
Li  F, Li  Y-Y, Liu  M-J,  et al.  Household transmission of SARS-CoV-2 and risk factors for susceptibility and infectivity in Wuhan: a retrospective observational study.   Lancet Infect Dis. 2021;21(5):617-628. doi:10.1016/S1473-3099(20)30981-6PubMedGoogle ScholarCrossref
13.
Xin  H, Jiang  F, Xue  A,  et al.  Risk factors associated with occurrence of COVID-19 among household persons exposed to patients with confirmed COVID-19 in Qingdao Municipal, China.   Transbound Emerg Dis. 2021;68(2):782-788. doi:10.1111/tbed.13743PubMedGoogle ScholarCrossref
14.
Wu  J, Huang  Y, Tu  C,  et al.  Household transmission of SARS-CoV-2, Zhuhai, China, 2020.   Clin Infect Dis. 2020;71(16):2099-2108. doi:10.1093/cid/ciaa557PubMedGoogle ScholarCrossref
15.
Thompson  HA, Mousa  A, Dighe  A,  et al.  SARS-CoV-2 setting-specific transmission rates: a systematic review and meta-analysis.   Clin Infect Dis. Published online February 9, 2021. doi:10.1093/cid/ciab100Google Scholar
16.
Soriano-Arandes  A, Gatell  A, Serrano  P,  et al; COPEDI-CAT research group.  Household SARS-CoV-2 transmission and children: a network prospective study.   Clin Infect Dis. 2021;ciab228. doi:10.1093/cid/ciab228PubMedGoogle Scholar
17.
Bi  Q, Wu  Y, Mei  S,  et al.  Epidemiology and transmission of COVID-19 in 391 cases and 1286 of their close contacts in Shenzhen, China: a retrospective cohort study.   Lancet Infect Dis. 2020;20(8):911-919. doi:10.1016/S1473-3099(20)30287-5PubMedGoogle ScholarCrossref
18.
Wang  Y, Tian  H, Zhang  L,  et al.  Reduction of secondary transmission of SARS-CoV-2 in households by face mask use, disinfection and social distancing: a cohort study in Beijing, China.   BMJ Glob Health. 2020;5(5):e002794. doi:10.1136/bmjgh-2020-002794PubMedGoogle Scholar
19.
Hu  S, Wang  W, Wang  Y,  et al  Infectivity, susceptibility, and risk factors associated with SARS-CoV-2 transmission under intensive contact tracing in Hunan, China.   medRxiv. Preprint posted November 3, 2020. doi:10.1101/2020.07.23.20160317Google Scholar
20.
Park  YJ, Choe  YJ, Park  O,  et al; COVID-19 National Emergency Response Center, Epidemiology and Case Management Team.  Contact tracing during coronavirus disease outbreak, South Korea, 2020.   Emerg Infect Dis. 2020;26(10):2465-2468. doi:10.3201/eid2610.201315PubMedGoogle ScholarCrossref
21.
Lyngse  FP, Kirkeby  CT, Halasa  T,  et al  COVID-19 transmission within Danish households: A nationwide study from lockdown to reopening.   medRxiv. 2020. doi:10.1101/2020.09.09.20191239Google Scholar
22.
Lyngse  FP, Mølbak  K, Frank  KT, Nielsen  C, Skov  RL, Kirkeby  CT.  Association between SARS-CoV-2 transmission risk, viral load, and age: a nationwide study in Danish households.   medRxiv. 2021. doi:10.1101/2021.02.28.21252608Google Scholar
23.
Grijalva  CG, Rolfes  MA, Zhu  Y,  et al.  Transmission of SARS-COV-2 infections in households: Tennessee and Wisconsin, April-September 2020.   MMWR Morb Mortal Wkly Rep. 2020;69(44):1631-1634. doi:10.15585/mmwr.mm6944e1PubMedGoogle ScholarCrossref
24.
Koh  WC, Naing  L, Chaw  L,  et al.  What do we know about SARS-CoV-2 transmission? a systematic review and meta-analysis of the secondary attack rate and associated risk factors.   PLoS One. 2020;15(10):e0240205. doi:10.1371/journal.pone.0240205PubMedGoogle Scholar
25.
Paul  LA, Daneman  N, Brown  KA,  et al.  Characteristics associated with household transmission of SARS-CoV-2 in Ontario, Canada: a cohort study.   Clin Infect Dis. 2021;ciab186. doi:10.1093/cid/ciab186PubMedGoogle Scholar
26.
Maltezou  HC, Vorou  R, Papadima  K,  et al.  Transmission dynamics of SARS-CoV-2 within families with children in Greece: a study of 23 clusters.   J Med Virol. 2021;93(3):1414-1420. doi:10.1002/jmv.26394PubMedGoogle ScholarCrossref
27.
Posfay-Barbe  KM, Wagner  N, Gauthey  M,  et al.  COVID-19 in children and the dynamics of infection in families.   Pediatrics. 2020;146(2):e20201576. doi:10.1542/peds.2020-1576PubMedGoogle Scholar
28.
Marks  M, Millat-Martinez  P, Ouchi  D,  et al.  Transmission of COVID-19 in 282 clusters in Catalonia, Spain: a cohort study.   Lancet Infect Dis. 2021;21(5):629-636. doi:10.1016/S1473-3099(20)30985-3PubMedGoogle ScholarCrossref
29.
Lee  LYW, Rozmanowski  S, Matthew  P,  et al.  SARS-CoV-2 infectivity by viral load, S gene variants and demographic factors and the utility of lateral flow devices to prevent transmission.   Clin Infect Dis. Published online May 11, 2021. doi:10.1093/cid/ciab421Google Scholar
30.
Heald-Sargent  T, Muller  WJ, Zheng  X, Rippe  J, Patel  AB, Kociolek  LK.  Age-related differences in nasopharyngeal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) levels in patients with mild to moderate coronavirus disease 2019 (COVID-19).   JAMA Pediatr. 2020;174(9):902-903. doi:10.1001/jamapediatrics.2020.3651PubMedGoogle ScholarCrossref
31.
Jones  TC, Mühlemann  B, Veith  T,  et al  An analysis of SARS-CoV-2 viral load by patient age.   medRxiv. Preprint posted June 9, 2020. doi:10.1101/2020.06.08.20125484Google Scholar
32.
Yonker  LM, Neilan  AM, Bartsch  Y,  et al.  Pediatric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2): Clinical presentation, infectivity, and immune responses.   J Pediatr. 2020;227:45-52.e5. doi:10.1016/j.jpeds.2020.08.037PubMedGoogle ScholarCrossref
33.
L’Huillier  AG, Torriani  G, Pigny  F, Kaiser  L, Eckerle  I.  Culture-competent SARS-CoV-2 in nasopharynx of symptomatic neonates, children, and adolescents.   Emerg Infect Dis. 2020;26(10):2494-2497. doi:10.3201/eid2610.202403PubMedGoogle ScholarCrossref
34.
Bullard  J, Funk  D, Dust  K,  et al.  Infectivity of severe acute respiratory syndrome coronavirus 2 in children compared with adults.   CMAJ. 2021;193(17):E601-E606. doi:10.1503/cmaj.210263PubMedGoogle ScholarCrossref
35.
Lewis  NM, Duca  LM, Marcenac  P,  et al.  Characteristics and timing of initial virus shedding in severe acute respiratory syndrome coronavirus 2, Utah, USA.   Emerg Infect Dis. 2021;27(2):352-359. doi:10.3201/eid2702.203517PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    EXPAND ALL
    Intra-Household SARS-CoV-2 Transmission from Children
    Harikrishnan Pandurangan, MDS, FDSRCS(England), PhD | Clinician Researcher, Craniofacial Orthodontist, Teeth"N"Jaws Center, Lake Area, Chennai-34, India.
    The study by Paul et al. is a timely analysis of different pediatric age group infecting the house members and found that 0-3 years have the greatest risk of transmission to caregivers and siblings.

    Such pediatric age group risk in household transmission may vary globally due to cultural variations. As an Orthodontist treating children mostly with age group of 6-15 years, I noticed in the second wave of this pandemic in India, i.e., from March 2021-June 2021, many household were infected ( full family members). On enquiry, many parents reported that the primary infection were from children
    in the family aged 4-10 years. No studies are published yet from this region. The following points are relevant to the household transmission with that age group.

    1. Children in that age group are very active and parents find it difficult to hold them inside the house.

    2. They were playing with neighbourhood friends and in nearby parks and play grounds. Even when playing indoors they were sharing many contact devices like computers, playstations, mobile phones and cutlery items with their friends and siblings.

    3. In the Indian joint family system, or in nuclear families due to lockdown, most grandparents were staying with their children and grandchildren in the house. These children are affectionately playful with the grandparents which lead to higher infection of the elderly family members from the infected children.

    4. Typically, this age group children were participating in many group session activities for talent enhancement at schools or private organisations and are more prone in carrying infection to the house from outside.

    5. Obviously, it is difficult for these children to practice mask wearing and hand washing regularly and effectively and due to the affectionate closehandling by family members they seem to be at high risk of transmitting SAR-CoV-2 in household environment in our country.

    6. In a urban and sub-urban house structure with single or double rooms only, isolation of infected children is compromised due to space issues and other all family members crowded in a single room.

    Increased awareness of handling infected children, isolation of healthy elderly family members and vaccination of the family members will reduce this intra-household SARS-CoV-2 transmission from pediatric cases.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    August 16, 2021

    Association of Age and Pediatric Household Transmission of SARS-CoV-2 Infection

    Author Affiliations
    • 1Health Protection, Public Health Ontario, Toronto, Ontario, Canada
    • 2Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
    • 3Division of Infectious Diseases, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
    • 4Department of Medicine, University of Toronto, Toronto, Ontario, Canada
    • 5Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada
    • 6Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
    • 7Unity Health Toronto–St Joseph’s Health Centre, Toronto, Ontario, Canada
    • 8Division of Infectious Diseases, The Hospital for Sick Children, Toronto, Ontario, Canada
    • 9Department of Paediatrics, University of Toronto, Toronto, Ontario, Canada
    JAMA Pediatr. Published online August 16, 2021. doi:10.1001/jamapediatrics.2021.2770
    Key Points

    Question  Are there differences in the odds of household transmission of SARS-CoV-2 by younger children compared with older children?

    Findings  In this cohort study of 6280 households with pediatric index cases, the adjusted odds of household transmission by children aged 0 to 3 years was 1.43 compared with children aged 14 to 17 years.

    Meaning  Younger children may have greater risk of transmitting SARS-CoV-2 to caregivers and siblings in the household than older children.

    Abstract

    Importance  As a result of low numbers of pediatric cases early in the COVID-19 pandemic, pediatric household transmission of SARS-CoV-2 remains an understudied topic.

    Objective  To determine whether there are differences in the odds of household transmission by younger children compared with older children.

    Design, Setting, and Participants  This population-based cohort study took place between June 1 and December 31, 2020, in Ontario, Canada. Private households in which the index case individual of laboratory-confirmed SARS-CoV-2 infection was younger than 18 years were included. Individuals were excluded if they resided in apartments missing suite information, in households with multiple index cases, or in households where the age of the index case individual was missing.

    Exposures  Age group of pediatric index cases categorized as 0 to 3, 4 to 8, 9 to 13, and 14 to 17 years.

    Main Outcomes and Measures  Household transmission, defined as households where at least 1 secondary case occurred 1 to 14 days after the pediatric index case.

    Results  A total of 6280 households had pediatric index cases, and 1717 households (27.3%) experienced secondary transmission. The mean (SD) age of pediatric index case individuals was 10.7 (5.1) years and 2863 (45.6%) were female individuals. Children aged 0 to 3 years had the highest odds of transmitting SARS-CoV-2 to household contacts compared with children aged 14 to 17 years (odds ratio, 1.43; 95% CI, 1.17-1.75). This association was similarly observed in sensitivity analyses defining secondary cases as 2 to 14 days or 4 to 14 days after the index case and stratified analyses by presence of symptoms, association with a school/childcare outbreak, or school/childcare reopening. Children aged 4 to 8 years and 9 to 13 years also had increased odds of transmission (aged 4-8 years: odds ratio, 1.40; 95% CI, 1.18-1.67; aged 9-13 years: odds ratio, 1.13; 95% CI, 0.97-1.32).

    Conclusions and Relevance  This study suggests that younger children may be more likely to transmit SARS-CoV-2 infection compared with older children, and the highest odds of transmission was observed for children aged 0 to 3 years. Differential infectivity of pediatric age groups has implications for infection prevention within households, as well as schools/childcare, to minimize risk of household secondary transmission. Additional population-based studies are required to establish the risk of transmission by younger pediatric index cases.

    Introduction

    The role of children in the transmission of SARS-CoV-2 infection requires further study. Early in the pandemic, when countries implemented lockdown measures, close contact was mostly limited to households and testing strategies tended to prioritize health care workers and symptomatic individuals.1-4 As a result, there were relatively few diagnosed pediatric cases of COVID-19,5 and it appeared that the proportion of children involved in the transmission of infection was small compared with adults.2,6,7 Since many jurisdictions relaxed public health measures and reopened educational facilities in fall 2020, the number of pediatric COVID-19 cases has grown, providing the opportunity to better characterize the infectivity of children.

    To date, household studies have typically only compared infectivity between young and old individuals, often grouping children with young adults8-12 or dichotomizing age to older adults vs younger adults/children.13,14 These studies have reported mixed results, with some finding that older age (≥20 years) was associated with increased infectivity,6,8-10,15,16 1 study finding younger age (<20 years),12 and others finding no age effect.11,13-15,17-19 Conversely, few household studies have examined differences in infectivity among children,15,16,20-23 likely owing to insufficient sample size.2,24 Two meta-analyses reported no significant differences between younger children and older children for household susceptibility to SARS-CoV-22,15; however, it remains unclear if this holds true for infectivity. These findings warrant a closer look at household transmission of SARS-CoV-2 by children and whether there are any differences in the likelihood of transmission for particular age groups.

    We sought to conduct an age analysis of residents aged 0 to 17 years in Ontario, Canada, who were the index case of SARS-CoV-2 infection in their household between June and December 2020. Pediatric index cases were divided into 4 age groups (0-3, 4-8, 9-13, and 14-17 years) to provide a more granular picture of any age differences. We were also interested in comparing characteristics of index cases by age group, exploring the direction of transmission by age, and assessing whether factors such as symptoms, school/childcare reopening, or school/childcare outbreaks were associated with differences in the odds of transmission from children to their household members.

    Methods
    Study Population

    We derived the study cohort from case data that were reported in provincial disease systems by public health units across Ontario, Canada’s most populous province. All individuals with laboratory-confirmed SARS-CoV-2 infection (via positive nucleic acid amplification test) between June 1 and December 31, 2020, were included. We obtained ethics approval from Public Health Ontario’s Research Ethics Board. Data were deidentified so informed consent was not required.

    Identification of Private Households

    Addresses of case individuals were reviewed and classified as either private households, defined as individual houses or apartments/suites within multiunit dwellings, or congregate settings (eg, homeless shelters or long-term care homes). We excluded any individuals with missing or incomplete address information, individuals residing in congregate settings, and individuals identified as residing in multiunit dwellings but missing suite information. Addresses were then matched between cases using a natural language processing algorithm from Python’s “sklearn” library to identify multicase households. Details of the address matching process have been described previously.25

    Outcomes

    The outcome of interest was secondary household transmission of SARS-CoV-2 infection by a pediatric index case individual (aged 0-17 years). Index cases were defined as the earliest case of a household and were identified by comparing symptom onset dates of cases in the household.2 If symptom onset date was missing, we used specimen collection date as a proxy; no cases were missing both dates. Secondary cases were defined as individuals (adult or pediatric) who had disease onset 1 to 14 days after the index case, per previous studies of household transmission.2,21,22,25 We excluded households with index cases missing age (n = 12) and households with multiple index cases (ie, multiple cases occurring on the earliest case date of the household; n = 4335) because they would present challenges for estimating associations between household transmission and characteristics of the index case.

    Individual-Level and Neighborhood-Level Characteristics of Index Cases

    The main exposure of interest was age group of the index case: 0 to 3, 4 to 8, 9 to 13, and 14 to 17 years. We also a priori selected a group of individual-level and neighborhood-level characteristics of the index case to adjust for in the models. At the individual level, we included gender (which was reported in provincial disease systems), month of disease onset, and testing delay between symptom onset and specimen collection. For the testing delay, we additionally categorized individuals who were asymptomatic, identified as individuals who were missing symptom onset date (thus specimen collection date was used) and were reported as asymptomatic in provincial reportable disease systems. As we only had individual-level household size information reported for 59.6% of index cases, we also included mean family size for the neighborhood of the index case. Mean family size was available from 2016 Canadian census records for aggregate dissemination areas across Ontario, representing areas of approximately 5000 to 15 000 persons.

    For stratified and sensitivity analyses, we additionally included information on the presence of symptoms, whether the index case was linked to a school/childcare outbreak (identified by public health units through their investigations), and whether the index case individual’s disease onset was before or after school/childcare reopening (depending on the age of the index case individual). In Ontario, schools reopened in mid-September 2020 and childcare reopened mid-June 2020.

    Statistical Analysis

    We carried out descriptive analyses to assess the characteristics of pediatric index cases across the 4 age groups and included index case individuals older than 17 years for comparison. Direction of transmission from index case to secondary case by age was also examined. We then applied 3 logistic regression models to obtain odds ratios (OR) and 95% confidence intervals for the associations between index case age group and odds of transmitting SARS-CoV-2 to household contacts: (1) a crude, unadjusted model; (2) a model adjusted for gender and month of disease onset (adjusted model 1); and (3) a model adjusted for gender, month of disease onset, testing delay, and mean family size (adjusted model 2).

    For stratified analyses, these models were refit to subsets of the data broken down by 3 additional index case characteristics: (1) presence of symptoms; (2) association with a school/childcare outbreak; and (3) disease onset before school/childcare reopening.

    For sensitivity analyses, we first updated the definition of secondary cases to increase certainty of the direction of transmission from the index case; we reran the analyses with secondary case individuals who developed symptoms 2 to 14 days after the index case and 4 to 14 days after the index case. Second, we adjusted for individual-level household size instead of neighborhood-level mean family size in the subset of index cases that had this information available. Third, we ran the analysis on a symptomatic cohort, restricting to index case and secondary case individuals who had symptoms reported or a symptom onset date available. Last, we reran the symptomatic cohort analysis with the adjusted definitions for secondary cases (ie, 2-14 days and 4-14 days). Two-sided P values were statistically significant at .05. Analyses were performed using RStudio version 1.2.5033 (RStudio).

    Results

    Between June and December 2020, a total of 6280 private households had a pediatric index case (Figure 1). The mean (SD) age of index case individuals was 10.7 (5.1) years and 2863 (45.6%) were female individuals. Of 6280 households, 1717 (27.3%) experienced secondary household transmission, leading to a median of 2 secondary cases (25th percentile, 1 case; 75th percentile, 2 cases; 90th percentile, 3 cases). This corresponded with an overall crude rate of transmission of 27 341 per 100 000 households with pediatric index cases. Pediatric index cases most frequently transmitted infection to individuals aged 0 to 20 years or 30 to 50 years, with older children tending to transmit to older individuals in those age ranges (Figure 2).

    The proportion of index cases in each age group increased with age, with 12% (776 of 6280) aged 0 to 3 years, 20% (1257 of 6280) aged 4 to 8 years, 30% (1881 of 6280) aged 9 to 13 years, and 38% (2376 of 6280) aged 14 to 17 years (Table 1). Compared with index case indivudals in the oldest age group, younger index case individuals had a higher proportion associated with a school/childcare outbreak and shorter testing delays. Index case individuals aged 4 to 8 years and 9 to 13 years had higher proportion with no symptoms reported compared with index case individuals aged 14 to 17 years or aged 0 to 3 years. Across all age groups, more index case individuals had disease onset in the fall/winter (September-December) compared with the summer (June-August), which aligns with the trajectory of the second wave of the pandemic in Ontario.

    Associations With Index Case Characteristics

    Compared with index case individuals aged 14 to 17 years, those aged 0 to 3 years had higher odds of transmitting SARS-CoV-2 to household contacts in all 3 models (crude model: OR, 1.20; 95% CI, 1.01-1.44; adjusted model 1: OR, 1.21; 95% CI, 1.01-1.45; adjusted model 2: OR, 1.43; 95% CI, 1.17-1.75) (Table 2). There were no significant differences in the odds of transmission by the 4- to 8-year and 9- to 13-year age groups, with the exception of adjusted model 2 for index case individuals aged 4 to 8 years (OR, 1.40; 95% CI, 1.18-1.67). Additionally, there were incremental odds of transmitting infection with longer testing delays compared with a 0-day delay (1-day delay: OR, 1.24; 2-day delay: OR, 1.59; 3-day delay: OR, 1.97; 4-day delay: OR, 2.38; ≥5-day delay: OR, 2.98), as well as increased odds with larger mean family size (OR, 1.63 per person increase; 95% CI, 1.43-1.86). No significant differences were observed by gender or by month of disease onset.

    In stratified analyses, we did not observe significant heterogeneity in the odds of transmitting SARS-CoV-2 to household contacts between index case individuals with vs without symptoms reported, index case individuals not associated vs associated with a school/childcare outbreak, or index case individuals with disease onset before vs after school/childcare reopening (eTables 1 and 2 in the Supplement).

    Sensitivity analyses of the crude model and adjusted model 1 resulted in the same direction of association for the 0- to 3-year age group, but confidence intervals widened (eTable 3 in the Supplement). In adjusted model 2, associations were largely unchanged with the 2- to 14-day definition, when controlling for individual-level household size, and in the symptomatic case analysis for the 0- to 3-year age group (adjusted model 2: OR, 1.43; 95% CI, 1.17-1.75; 2-14 days: OR, 1.37; 95% CI, 1.11-1.69; household size: OR, 1.31; 95% CI, 1.02-1.67; symptomatic: OR, 1.32; 95% CI, 1.06-1.64) and 4- to 8-year age group (adjusted model 2: OR, 1.40; 95% CI, 1.18-1.67; 2-14 days: OR, 1.33; 95% CI, 1.11-1.60; household size: OR, 1.31; 95% CI, 1.06-1.62; symptomatic: OR, 1.33; 95% CI, 1.10-1.61). Associations for the 0- to 3-year and 4- to 8-year age groups were also similar to adjusted model 2 in the combined 2- to 14-day definition and symptomatic case analysis (age 0-3 years: OR, 1.25; 95% CI, 1.00-1.57; age 4-8 years: OR, 1.23; 95% CI, 1.00-1.50), as well as for the 0- to 3-year age group with the 4- to 14-day definition (OR, 1.35; 95% CI, 1.07-1.70) and after further restricting to symptomatic cases (OR, 1.26; 95% CI, 0.96-1.64).

    Discussion

    In this study of 6280 pediatric index cases, we observed that children aged 0 to 3 years had greater odds of transmitting SARS-CoV-2 to household contacts compared with children aged 14 to 17 years. This association was observed irrespective of factors such as presence of symptoms, school/childcare reopening, or association with a school/childcare outbreak. We also observed some evidence of greater odds of household transmission by children aged 4 to 8 years after controlling for testing delays and neighborhood-level mean family size (as well as individual-level household size). We identified clustering in our age-to-age transmission plot, likely reflecting the age structure of households with younger individuals living with and transmitting to younger caregivers and siblings.

    To date, there have been challenges with analyzing the role of children in household spread of SARS-CoV-2 owing to low numbers of pediatric index cases. We identified pediatric index case individuals younger than 18 years in approximately 7% of households (6280 of 89 191), which is similar to the proportions observed in international studies from Greece (9% of households with pediatric index cases; the study defined pediatric as age <18 years),26 Switzerland (8% of households with pediatric index cases age <16 years),27 Denmark (5% of households with pediatric index cases age <20 years),21 Hunan, China (5% of households with pediatric index cases age <15 years),19 and Guangzhou, China (5% of households with pediatric index cases age <20 years)9; higher than studies from South Korea (3% of households with pediatric index cases age <20 years)20 and Wuhan, China (1% of households with pediatric index cases age <20 years)12; and lower than a study from the US (14% of households with pediatric index cases age <18 years).23 Meta-analyses examining the age of household index cases additionally reported that 3% to 19% of households had pediatric index cases.2,15 Further, in our study, presumed household transmission by a pediatric index case occurred in 27% of households using the 1- to 14-day definition of secondary transmission (24% and 16% using 2- to 14- and 4- to 14-day definitions, respectively), which is close to an estimate from Catalonia, Spain (28%).16

    Of the aforementioned studies that included pediatric index cases, only a subset compared associations between pediatric age groups and household transmission with mixed findings. Our results align with 2 Danish studies that found that among children, there were increased odds of transmitting infection with younger age (age 0-5 years: OR, 1.11; 95% CI, 1.01-1.19 vs age 10-15 years: OR, 0.82; 95% CI, 0.78-0.85, compared with the age 30- to 35-year reference group).21,22 One study from Spain also found that the highest OR for transmission was among younger index case individuals 0 to 2 years (OR, 2.27; 95% CI, 0.62-8.35 compared with the age 12- to 15-year reference group), but confidence intervals were wide.16 Other studies instead compared secondary attack rates between pediatric age groups, including a study from South Korea that reported household contacts of index case individuals aged 10 to 19 years had the highest secondary attack rate at 18.6% (95% CI, 14.0%-24.0%) vs 5.3% (95% CI, 1.3%-13.7%) for index case individuals aged 0 to 9 years.20 Conversely, a meta-analysis reported no significant difference in household secondary attack rates for index case individuals aged 10 to 19 years vs index case individuals aged 0 to 9 years.15 Similarly, a study from Tennessee and Wisconsin observed higher household secondary attack rate for younger pediatric index case individuals, but confidence intervals were overlapping; the estimated secondary attack rates were 53% (95% CI, 31%-74%) for index case individuals younger than 12 years and 38% (95% CI, 23%-56%) for index case individuals aged 12 to 17 years.23

    The differences in infectivity for pediatric age groups across studies may be explained by differences in viral shedding, symptom expression, and behavioral factors.4,5,21,22 Viral load is suspected to be an important factor affecting the odds of SARS-CoV-2 transmission.22,28,29 Several studies of age-specific viral shedding of SARS-CoV-2 have reported that viral loads in children are similar or higher than viral loads in adults.22,30-33 In particular, 1 study reported that children younger than 5 years might carry more viral RNA in their nasopharynx than older children and adults.30 Conversely, another study found no significant difference in viral loads for children 10 years or younger vs children aged 11 to 17 years.34 Additionally, biases in testing practices, such as the preferential testing of symptomatic cases and contacts, inherently leads to underdetection of pediatric cases and challenges estimating their rates of transmission.4,5,15 Studies have found younger children are more likely to be asymptomatic,3,16,26 which has been postulated as a reason for lower infectivity because lower secondary attack rates have been reported for asymptomatic index cases compared with symptomatic index cases.2,12,15,24 We found that asymptomatic status and testing delays had strong gradient effects on infectivity, similar to our previous study of household transmission.25 However, even after adjusting for the lower odds of asymptomatic transmission and testing delays in our study, children aged 0 to 3 years and 4 to 8 years remained associated with higher odds of transmitting SARS-CoV-2 to household contacts than children aged 14 to 17 years. A possible explanation for this finding is that younger children are not able to self-isolate from their caregivers when they are sick, irrespective of the timing of testing.21,35

    Limitations and Strengths

    This study has some limitations that should be acknowledged. First, there is the possibility of misclassifying household transmission if secondary case infection was truly acquired outside the household or if the true index case individual of the household was untested. This is particularly relevant for pediatric cases owing to their increased probability of having mild or asymptomatic infection and thus increased probability of infection being missed. We attempted to account for this in sensitivity analyses by modifying the secondary case definition from 1 to 14 days to 2 to 14 and 4 to 14 days and by restricting analysis to symptomatic cases only. Second, this study used multiple evolving data systems for reporting COVID-19 cases in Ontario. As a result, there was some inconsistency regarding how symptoms were reported, which may result in some misclassification of symptomatic cases. We carefully examined the reporting practices over the months covered in the study and selected a combination of variables (symptoms reported, symptom onset date available, and/or asymptomatic flag) we felt reflected the most likely symptom status of cases. Third, we could not reliably calculate secondary attack rates in the study because we did not know the number of noninfected contacts in households for the full cohort. However, after controlling for individual-level household size within the subset of the cohort with this information available, our conclusions were unaltered.

    This study also has several strengths. This is a large, population-based study of all individuals with confirmed SARS-CoV-2 infection in Canada’s most populated province; thus, we had sufficient data to explore transmission within the understudied pediatric group. We were also able to include relevant covariates such as testing delays and household size; our results suggest that early testing of pediatric index cases and reduced household size/crowding may be useful strategies to minimize secondary household transmission by children. Second, the use of a natural language processing algorithm to perform address matching allowed us to reliably identify cases in the same household, rather than relying on contact tracing or other types of epidemiologic linkage that would be difficult to perform for a high volumes of cases. Third, our application of various sensitivity analyses for symptom status and the definition of secondary transmission increased our certainty of the direction of transmission from index case to secondary case, further supporting our findings.

    Conclusions

    As the number of pediatric cases increases worldwide, the role of children in household transmission will continue to grow. We found that younger children may be more likely to transmit SARS-CoV-2 infection compared with older children, and the highest odds of transmission were observed for children aged 0 to 3 years. Differential infectivity of pediatric age groups has implications for infection prevention controls within households and schools/childcare to minimize risk of household secondary transmission. Although children do not appear to transmit infection as frequently as adults, caregivers should be aware of the risk of transmission while caring for sick children in the household setting. As it is challenging and often impossible to socially isolate from sick children, caregivers should apply other infection control measures where feasible, such as use of masks, increased hand washing, and separation from siblings.

    Back to top
    Article Information

    Corresponding Author: Sarah A. Buchan, PhD, Public Health Ontario, 661 University Ave, Floor 17, Toronto, ON M5G 1M1, Canada (sarah.buchan@oahpp.ca).

    Accepted for Publication: May 19, 2021.

    Published Online: August 16, 2021. doi:10.1001/jamapediatrics.2021.2770

    Correction: This article was corrected on September 20 2021, to fix a typo in Table 2.

    Author Contributions: Ms Paul and Dr Buchan had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Paul, Daneman, Schwartz, Science, Brown, Buchan.

    Acquisition, analysis, or interpretation of data: Paul, Schwartz, Science, Brown, Whelan, Chan, Buchan.

    Drafting of the manuscript: Paul.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Paul, Schwartz, Brown.

    Administrative, technical, or material support: Paul, Schwartz, Buchan.

    Supervision: Science, Buchan.

    Conflict of Interest Disclosures: Dr Buchan reported grants from Canadian Institutes of Health Research for research on influenza, respiratory syncytial virus, and COVID-19 and grants from Canadian Immunity Task Force for COVID-19 vaccines outside the submitted work. No other disclosures were reported.

    Funding/Support: This study did not have a direct funding source but was supported by Public Health Ontario.

    Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank James Johnson, MPH, and Arezou Saedi, MD, for conducting the address-matching work; Trevor van Ingen, MPH, for providing the neighborhood-level data; and Semra Tibebu, MPH, for cleaning the individual-level household size variable. These individuals made their contributions as part of their roles as paid employees of Public Health Ontario at the time of the study.

    References
    1.
    Zhang  J, Litvinova  M, Liang  Y,  et al.  Changes in contact patterns shape the dynamics of the COVID-19 outbreak in China.   Science. 2020;368(6498):1481-1486. doi:10.1126/science.abb8001PubMedGoogle ScholarCrossref
    2.
    Zhu  Y, Bloxham  CJ, Hulme  KD,  et al.  A meta-analysis on the role of children in severe acute respiratory syndrome coronavirus 2 in household transmission clusters.   Clin Infect Dis. 2021;72(12):e1146-e1153. doi:10.1093/cid/ciaa1825PubMedGoogle ScholarCrossref
    3.
    Chang  T-H, Wu  J-L, Chang  L-Y.  Clinical characteristics and diagnostic challenges of pediatric COVID-19: a systematic review and meta-analysis.   J Formos Med Assoc. 2020;119(5):982-989. doi:10.1016/j.jfma.2020.04.007PubMedGoogle ScholarCrossref
    4.
    Hyde  Z.  Difference in SARS-CoV-2 attack rate between children and adults may reflect bias.   Clin Infect Dis. 2021;ciab183. doi:10.1093/cid/ciab183PubMedGoogle Scholar
    5.
    Goldstein  E, Lipsitch  M, Cevik  M.  On the effect of age on the transmission of SARS-CoV-2 in households, schools and the community.   medRxiv. Preprint posted July 28, 2020. doi:10.1101/2020.07.19.20157362Google Scholar
    6.
    van der Hoek  W, Backer  JA, Bodewes  R,  et al.  De rol van kinderen in de transmissie van SARS-CoV-2.   Ned Tijdschr Geneeskd. 2020;164(25):D5140.PubMedGoogle Scholar
    7.
    Ludvigsson  JF.  Children are unlikely to be the main drivers of the COVID-19 pandemic: a systematic review.   Acta Paediatr. 2020;109(8):1525-1530. doi:10.1111/apa.15371PubMedGoogle ScholarCrossref
    8.
    Arnedo-Pena  A, Sabater-Vidal  S, Meseguer-Ferrer  N,  et al  COVID-19 secondary attack rate and risk factors in household contacts in Castellon (Spain): preliminary report.   Rev Enf Emerg. 2020;19(2):64-70.Google Scholar
    9.
    Jing  Q-L, Liu  M-J, Zhang  Z-B,  et al.  Household secondary attack rate of COVID-19 and associated determinants in Guangzhou, China: a retrospective cohort study.   Lancet Infect Dis. 2020;20(10):1141-1150. doi:10.1016/S1473-3099(20)30471-0PubMedGoogle ScholarCrossref
    10.
    Dattner  I, Goldberg  Y, Katriel  G,  et al  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.   medRxiv. Preprint posted October 11, 2020. doi:10.1101/2020.06.03.20121145Google Scholar
    11.
    Sun  K, Wang  W, Gao  L,  et al.  Transmission heterogeneities, kinetics, and controllability of SARS-CoV-2.   Science. 2021;371(6526):eabe2424. doi:10.1126/science.abe2424PubMedGoogle Scholar
    12.
    Li  F, Li  Y-Y, Liu  M-J,  et al.  Household transmission of SARS-CoV-2 and risk factors for susceptibility and infectivity in Wuhan: a retrospective observational study.   Lancet Infect Dis. 2021;21(5):617-628. doi:10.1016/S1473-3099(20)30981-6PubMedGoogle ScholarCrossref
    13.
    Xin  H, Jiang  F, Xue  A,  et al.  Risk factors associated with occurrence of COVID-19 among household persons exposed to patients with confirmed COVID-19 in Qingdao Municipal, China.   Transbound Emerg Dis. 2021;68(2):782-788. doi:10.1111/tbed.13743PubMedGoogle ScholarCrossref
    14.
    Wu  J, Huang  Y, Tu  C,  et al.  Household transmission of SARS-CoV-2, Zhuhai, China, 2020.   Clin Infect Dis. 2020;71(16):2099-2108. doi:10.1093/cid/ciaa557PubMedGoogle ScholarCrossref
    15.
    Thompson  HA, Mousa  A, Dighe  A,  et al.  SARS-CoV-2 setting-specific transmission rates: a systematic review and meta-analysis.   Clin Infect Dis. Published online February 9, 2021. doi:10.1093/cid/ciab100Google Scholar
    16.
    Soriano-Arandes  A, Gatell  A, Serrano  P,  et al; COPEDI-CAT research group.  Household SARS-CoV-2 transmission and children: a network prospective study.   Clin Infect Dis. 2021;ciab228. doi:10.1093/cid/ciab228PubMedGoogle Scholar
    17.
    Bi  Q, Wu  Y, Mei  S,  et al.  Epidemiology and transmission of COVID-19 in 391 cases and 1286 of their close contacts in Shenzhen, China: a retrospective cohort study.   Lancet Infect Dis. 2020;20(8):911-919. doi:10.1016/S1473-3099(20)30287-5PubMedGoogle ScholarCrossref
    18.
    Wang  Y, Tian  H, Zhang  L,  et al.  Reduction of secondary transmission of SARS-CoV-2 in households by face mask use, disinfection and social distancing: a cohort study in Beijing, China.   BMJ Glob Health. 2020;5(5):e002794. doi:10.1136/bmjgh-2020-002794PubMedGoogle Scholar
    19.
    Hu  S, Wang  W, Wang  Y,  et al  Infectivity, susceptibility, and risk factors associated with SARS-CoV-2 transmission under intensive contact tracing in Hunan, China.   medRxiv. Preprint posted November 3, 2020. doi:10.1101/2020.07.23.20160317Google Scholar
    20.
    Park  YJ, Choe  YJ, Park  O,  et al; COVID-19 National Emergency Response Center, Epidemiology and Case Management Team.  Contact tracing during coronavirus disease outbreak, South Korea, 2020.   Emerg Infect Dis. 2020;26(10):2465-2468. doi:10.3201/eid2610.201315PubMedGoogle ScholarCrossref
    21.
    Lyngse  FP, Kirkeby  CT, Halasa  T,  et al  COVID-19 transmission within Danish households: A nationwide study from lockdown to reopening.   medRxiv. 2020. doi:10.1101/2020.09.09.20191239Google Scholar
    22.
    Lyngse  FP, Mølbak  K, Frank  KT, Nielsen  C, Skov  RL, Kirkeby  CT.  Association between SARS-CoV-2 transmission risk, viral load, and age: a nationwide study in Danish households.   medRxiv. 2021. doi:10.1101/2021.02.28.21252608Google Scholar
    23.
    Grijalva  CG, Rolfes  MA, Zhu  Y,  et al.  Transmission of SARS-COV-2 infections in households: Tennessee and Wisconsin, April-September 2020.   MMWR Morb Mortal Wkly Rep. 2020;69(44):1631-1634. doi:10.15585/mmwr.mm6944e1PubMedGoogle ScholarCrossref
    24.
    Koh  WC, Naing  L, Chaw  L,  et al.  What do we know about SARS-CoV-2 transmission? a systematic review and meta-analysis of the secondary attack rate and associated risk factors.   PLoS One. 2020;15(10):e0240205. doi:10.1371/journal.pone.0240205PubMedGoogle Scholar
    25.
    Paul  LA, Daneman  N, Brown  KA,  et al.  Characteristics associated with household transmission of SARS-CoV-2 in Ontario, Canada: a cohort study.   Clin Infect Dis. 2021;ciab186. doi:10.1093/cid/ciab186PubMedGoogle Scholar
    26.
    Maltezou  HC, Vorou  R, Papadima  K,  et al.  Transmission dynamics of SARS-CoV-2 within families with children in Greece: a study of 23 clusters.   J Med Virol. 2021;93(3):1414-1420. doi:10.1002/jmv.26394PubMedGoogle ScholarCrossref
    27.
    Posfay-Barbe  KM, Wagner  N, Gauthey  M,  et al.  COVID-19 in children and the dynamics of infection in families.   Pediatrics. 2020;146(2):e20201576. doi:10.1542/peds.2020-1576PubMedGoogle Scholar
    28.
    Marks  M, Millat-Martinez  P, Ouchi  D,  et al.  Transmission of COVID-19 in 282 clusters in Catalonia, Spain: a cohort study.   Lancet Infect Dis. 2021;21(5):629-636. doi:10.1016/S1473-3099(20)30985-3PubMedGoogle ScholarCrossref
    29.
    Lee  LYW, Rozmanowski  S, Matthew  P,  et al.  SARS-CoV-2 infectivity by viral load, S gene variants and demographic factors and the utility of lateral flow devices to prevent transmission.   Clin Infect Dis. Published online May 11, 2021. doi:10.1093/cid/ciab421Google Scholar
    30.
    Heald-Sargent  T, Muller  WJ, Zheng  X, Rippe  J, Patel  AB, Kociolek  LK.  Age-related differences in nasopharyngeal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) levels in patients with mild to moderate coronavirus disease 2019 (COVID-19).   JAMA Pediatr. 2020;174(9):902-903. doi:10.1001/jamapediatrics.2020.3651PubMedGoogle ScholarCrossref
    31.
    Jones  TC, Mühlemann  B, Veith  T,  et al  An analysis of SARS-CoV-2 viral load by patient age.   medRxiv. Preprint posted June 9, 2020. doi:10.1101/2020.06.08.20125484Google Scholar
    32.
    Yonker  LM, Neilan  AM, Bartsch  Y,  et al.  Pediatric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2): Clinical presentation, infectivity, and immune responses.   J Pediatr. 2020;227:45-52.e5. doi:10.1016/j.jpeds.2020.08.037PubMedGoogle ScholarCrossref
    33.
    L’Huillier  AG, Torriani  G, Pigny  F, Kaiser  L, Eckerle  I.  Culture-competent SARS-CoV-2 in nasopharynx of symptomatic neonates, children, and adolescents.   Emerg Infect Dis. 2020;26(10):2494-2497. doi:10.3201/eid2610.202403PubMedGoogle ScholarCrossref
    34.
    Bullard  J, Funk  D, Dust  K,  et al.  Infectivity of severe acute respiratory syndrome coronavirus 2 in children compared with adults.   CMAJ. 2021;193(17):E601-E606. doi:10.1503/cmaj.210263PubMedGoogle ScholarCrossref
    35.
    Lewis  NM, Duca  LM, Marcenac  P,  et al.  Characteristics and timing of initial virus shedding in severe acute respiratory syndrome coronavirus 2, Utah, USA.   Emerg Infect Dis. 2021;27(2):352-359. doi:10.3201/eid2702.203517PubMedGoogle ScholarCrossref
    ×