Household Transmission of SARS-CoV-2 from Children and Adolescents (New England Journal of Medicine letter, September 2, 2021)

Summary

A letter to the editor of The New England Journal of Medicine, dated September 2, 2021, titled Household Transmission of SARS-CoV-2 from Children and Adolescents, by Victoria T. Chu and colleagues at the Centers for Disease Control and Prevention and the Georgia Department of Public Health. The authors describe a retrospective cohort study of overnight camp attendees and their household contacts, with 224 index patients and 526 household contacts. They report that transmission occurred in 35 of 194 households (18%) and that the secondary attack rate in those households was 45% (95% CI, 36 to 54). A table gives unadjusted and adjusted odds ratios for a secondary case.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Captured: 2026-06-25
Source site: www.nejm.org

Archived web capture. Preserved source — captured from www.nejm.org; archived 2026-06-25.

Cite this document

“Household Transmission of SARS-CoV-2 from Children and Adolescents (New England Journal of Medicine letter, September 2, 2021).” Original at nejm.org. Archived at Pandemic Darlings on 2026-06-25, https://pandemicdarlings.com/source-documents/household-transmission-of-sars-cov-2-from-children-and-adolescents-new-3a16a20f38/, SHA-256 893b57eef4191b419428fa3aa86fee1dc83585e732d985796da1aa0f4dbd07e3.

Full text

T h e n e w e ng l a n d j o u r na l o f m e dic i n e

Household Transmission of SARS-CoV-2 from Children
and Adolescents
To the Editor: Severe acute respiratory syn- at an overnight camp,3 we conducted a retro-
drome coronavirus 2 (SARS-CoV-2) infection in spective cohort study involving camp attendees
children is often asymptomatic or results in only and their household contacts to assess second-
mild disease.1 Data on the extent of transmis- ary transmission and factors associated with
sion of SARS-CoV-2 from children and adoles- household transmission (additional details are
cents in the household setting, including trans- provided in the Methods section in the Supple-
mission to older persons who are at increased mentary Appendix, available with the full text of
risk for severe disease, are limited.2 After an this letter at NEJM.org).
outbreak of coronavirus disease 2019 (Covid-19) We interviewed 224 index patients who were

Table 1. Unadjusted and Adjusted Odds Ratio for a Secondary Case of SARS-CoV-2 Infection among Household Contacts.*

Variable Univariable Model Multivariable Model
Unadjusted Odds Ratio Adjusted Odds Ratio
(95% CI) (95% CI)
Index patients†
Age — yr
7–10 2.3 (0.7–7.0) 0.7 (0.2–2.9)
11–15 1.1 (0.5–2.8) 0.7 (0.3–1.6)
16–19 1.0 (reference) 1.0 (reference)
Covid-19 symptom status
Symptomatic 5.5 (0.8–40.7) 5.5 (0.8–38.1)
Asymptomatic 1.0 (reference) 1.0 (reference)
Maintained physical distancing
Yes 0.3 (0.1–0.6) 0.4 (0.1–0.9)
No 1.0 (reference) 1.0 (reference)
Always wore a mask around household contacts
Yes 0.2 (0.1–0.6) 0.5 (0.2–1.3)
No 1.0 (reference) 1.0 (reference)
Household contacts†
Contact with index patient‡
Direct contact 8.2 (2.7–24.7) 5.8 (1.8–18.8)
Close contact 5.4 (1.4–20.9) 5.2 (1.2–22.5)
Minimal to no contact 1.0 (reference) 1.0 (reference)

* A generalized-estimating-equation approach with an exchangeable correlation structure was used to account for intra-
household correlation. For the multivariable analysis, key characteristics of the index patients (i.e., age, the presence of
coronavirus disease 2019 [Covid-19] symptoms, isolation from other persons, and mask use) and the level of contact
between the household member and the index patient were selected. The model was fitted after calculation of variance
inflation factors to evaluate the resulting design matrix for multicollinearity; all the variance inflation factor values were
below 2, which indicated minimal correlation among the model variables. Confidence intervals were not adjusted for
multiplicity. Households with more than one index patient were excluded from the univariable and multivariable models.
CI denotes confidence interval, and SARS-CoV-2 severe acute respiratory syndrome coronavirus 2.
† An index patient was defined as a camp attendee for whom evidence of SARS-CoV-2 infection on molecular or antigen
testing was reported and who had the earliest date of onset of Covid-19 in the household. A household contact was de-
fined as a person who stayed in the household for at least 1 night during the period when the index patient was infec-
tious (i.e., from 2 days before the date of onset of Covid-19 until 10 days after the date of onset of Covid-19).
‡ Direct contact included being a caregiver of the index patient, having face-to-face contact with the index patient, being
within 6 ft of the index patient while the patient is coughing or sneezing, or having physical contact with the index
patient. Close contact included being within 6 ft of the index patient for at least 15 minutes and sharing a bedroom,
meals, or a vehicle.

n engl j med 385;10 nejm.org September 2, 2021

The New England Journal of Medicine
Downloaded from nejm.org on August 8, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
Correspondence

7 to 19 years of age and for whom there was contact (adjusted odds ratio with close contact,
evidence of SARS-CoV-2 infection on the basis of 5.2; 95% CI, 1.2 to 22.5; and adjusted odds ratio
molecular or antigen laboratory testing. A total with direct contact, 5.8; 95% CI, 1.8 to 18.8). We
of 198 of these campers (88%) were symptom- excluded missing data from the regression mod-
atic; symptoms developed in 141 of these 198 els, and confidence intervals were not adjusted
children or adolescents (71%) after they returned for multiplicity.
home from camp. This retrospective study showed that the ef-
Of 526 household contacts of these index ficient transmission of SARS-CoV-2 from school-
patients, 377 (72%) were tested for SARS-CoV-2, age children and adolescents to household mem-
and 46 (12%) of those who were tested had bers led to the hospitalization of adults with
positive results. An additional 2 secondary cases secondary cases of Covid-19. In households in
of infection were identified according to clinical which transmission occurred, half the house-
and epidemiologic criteria.4 A total of 38 of the hold contacts were infected. The secondary at-
48 secondary cases (79%) occurred in house- tack rates in this study were probably underesti-
holds where the index patient had become symp- mates because test results were reported by the
tomatic after returning home from camp; the patients themselves and testing was voluntary. In
median serial interval (i.e., the interval between addition, a third of the index patients returned
the onset of symptoms in the index patient and home from camp after the onset of symptoms,
the onset of symptoms in the household con- when they were presumably not as infectious as
tacts infected by that patient) was 5.0 days (95% they were before and during the onset of symp-
confidence interval [CI], 4.0 to 6.5). Transmis- toms,5 and two thirds adopted physical distanc-
sion occurred in 35 of 194 households (18%); in ing because of a known exposure at camp; both
these households, the secondary attack rate was of these factors probably reduced the transmis-
45% (95% CI, 36 to 54) (48 of 107 household sion of SARS-CoV-2 in the household. When
contacts). Among the household contacts who feasible, children and adolescents with a known
became infected and who were at least 18 years exposure to SARS-CoV-2 or a diagnosis of
of age, 4 of 41 (10%) were hospitalized (length Covid-19 should remain at home and maintain
of hospital stay, 5 to 11 days); none of the 7 per- physical distance from household members.
sons with a secondary case of infection who were Victoria T. Chu, M.D., M.P.H.
younger than 18 years were hospitalized. Anna R. Yousaf, M.D.
Of the index patients who responded to our Karen Chang, Ph.D.
question regarding preventive measures, 146 of Noah G. Schwartz, M.D.
217 (67%) reported that they had maintained Clinton J. McDaniel, M.P.H.
physical distancing and 73 of 216 (34%) reported Scott H. Lee, Ph.D.
that they had always worn masks around con- Centers for Disease Control and Prevention
tacts during the infectious period after they re- Atlanta, GA
pgz4@cdc.gov
turned home. In a univariable logistic-regression
model, among the index patients who were 18 Christine M. Szablewski, D.V.M.
Marie Brown, M.P.H.
years of age or younger, the increasing use of
Cherie L. Drenzek, D.V.M.
physical distancing and masks was associated
Georgia Department of Public Health
with the older age of the patient (with age as a Atlanta, GA
continuous variable, odds ratio for physical dis- Emilio Dirlikov, Ph.D.
tancing, 1.4; 95% CI, 1.2 to 1.5; odds ratio for Dale A. Rose, Ph.D.
mask use, 1.4; 95% CI, 1.2 to 1.6). In a multivari- Julie Villanueva, Ph.D.
able regression model, the risk of a secondary Alicia M. Fry, M.D.
case of infection among household contacts was Aron J. Hall, D.V.M.
lower among contacts of index patients who had Hannah L. Kirking, M.D.
practiced physical distancing than among con- Jacqueline E. Tate, Ph.D.
tacts of index patients who did not (adjusted Tatiana M. Lanzieri, M.D.
odds ratio, 0.4; 95% CI, 0.1 to 0.9) (Table 1). Rebekah J. Stewart, M.S.N., M.P.H.
Household members who had close or direct Centers for Disease Control and Prevention
contact with the index patient had a higher risk Atlanta, GA
of infection than those who had minimal to no for the Georgia Camp Investigation Team*

n engl j med 385;10 nejm.org September 2, 2021

*A complete list of members of the Georgia Camp Investiga- 2. CDC COVID-19 Response Team. Severe outcomes among
tion Team is provided in the Supplementary Appendix, available patients with coronavirus disease 2019 (COVID-19) — United
with the full text of this letter at NEJM.org. States, February 12–March 16, 2020. MMWR Morb Mortal Wkly
Drs. Chu and Yousaf contributed equally to this letter. Rep 2020;69:343-6.
The findings and conclusions in this letter are those of the 3. Szablewski CM, Chang KT, McDaniel CJ, et al. SARS-CoV-2
authors and do not necessarily represent the official position of transmission dynamics in a sleep-away camp. Pediatrics 2021;
the Centers for Disease Control and Prevention (CDC). 147(4):e2020046524.
Supported by the CDC. 4. Coronavirus Disease 2019 (COVID-19): 2020 interim case
Disclosure forms provided by the authors are available at definition, approved August 5, 2020. Atlanta:Centers for Dis-
NEJM.org. ease Control and Prevention, 2020 (https://ndc .services .cdc
This letter was published on July 21, 2021, and updated on April .gov/c ase-definitions/coronavirus-d isease-2019-2020-08-05/).
21, 2022, at NEJM.org. 5. He X, Lau EHY, Wu P, et al. Temporal dynamics in viral shed-
ding and transmissibility of COVID-19. Nat Med 2020;26:672-5.
1. Dong Y, Mo X, Hu Y, et al. Epidemiology of COVID-19 among
children in china. Pediatrics 2020;145(6):e20200702. DOI: 10.1056/NEJMc2031915

Adjuvant Nivolumab in Muscle-Invasive Urothelial Carcinoma
Sindhura Chitikela, M.D.
To the Editor: The benefit of nivolumab main- Lalit Kumar, M.D., D.M.
tenance therapy in the study by Bajorin et al. Ranjit K. Sahoo, M.D., D.M.
(June 3 issue)1 appears to have been driven main- All India Institute of Medical Sciences
ly by the 308 patients who received cisplatin as New Delhi, India
neoadjuvant therapy and reflects cross-sensitivity drranjitmd@gmail.com

(hazard ratio, 0.52; 95% confidence interval [CI] No potential conflict of interest relevant to this letter was
0.38 to 0.71) rather than by the 401 patients who reported.

did not (hazard ratio, 0.92; 95% CI, 0.69 to 1.21). 1. Bajorin DF, Witjes JA, Gschwend JE, et al. Adjuvant nivolumab
This also explains the lack of benefit in the 309 versus placebo in muscle-invasive urothelial carcinoma. N Engl
patients with a creatinine clearance of less than J Med 2021;384:2102-14.
2. Powles T, Park SH, Voog E, et al. Avelumab maintenance
60 ml per minute (hazard ratio, 0.87; 95% CI, therapy for advanced or metastatic urothelial carcinoma. N Engl
0.64 to 1.18) and by those with primary tumors J Med 2020;383:1218-30.
that did not arise from the bladder (hazard ratios, DOI: 10.1056/NEJMc2110897
1.23 and 1.56 in the renal pelvis and ureter, respec-
tively). In another study,2 the 269 patients with ad- To the Editor: Bajorin et al. recently published
vanced urothelial cancer who were treated with the positive results of the phase 3 CheckMate 274
gemcitabine–carboplatin (hazard ratio, 0.66; 95% trial, reporting that nivolumab adjuvant therapy is
CI, 0.47 to 0.91) derived as much benefit from superior to placebo in terms of disease-free sur-
avelumab maintenance as did the 389 patients vival in patients with radically resected muscle-
who were treated with gemcitabine–cisplatin invasive urothelial carcinoma. These surprising
(hazard ratio, 0.69; 95% CI, 0.51 to 0.94), irrespec- results are discordant with those of the Adjuvant
tive of creatinine clearance: at 60 ml or more per Atezolizumab versus Observation in Muscle-
minute, the hazard ratio was 0.68 (95% CI, 0.50 Invasive Urothelial Carcinoma (IMvigor010) trial,1
to 0.92), and at less than 60 ml per minute, the which failed to prove the superiority of atezoliz
hazard ratio was also 0.68 (95% CI, 0.50 to 0.94). umab over observation in the same setting. In
The lack of benefit in patients of European comparing the contrasting results of these simi-
and Asian ancestry requires clarification and lar treatments in the adjuvant phase, some ques-
could be related to the use of cisplatin as neoad- tions arise. Why did the placebo group in the
juvant therapy, the relative proportion of primary CheckMate 274 trial have a relevantly lower rate
tumors that were not located in the bladder, or of disease-free survival than the observation group
the expression of programmed death ligand 1 in the IMvigor010 trial (10.9 vs. 16.6 months)?
(PD-L1). Again, the predictive role of PD-L1 ex- Why is this difference enhanced in the PD-L1-
pression in the patients who were not eligible to positive groups, considering that different tech-
receive cisplatin requires more focus for the pur- niques are used in the evaluation of PD-L1 expres-
pose of identifying those who derived benefit sion, depending on the immunotherapeutic agent
from adjuvant nivolumab. used? Did the different percentages of upper-tract

n engl j med 385;10 nejm.org September 2, 2021

File and source

File: nejmc2031915_9f0476e5da176966_recap1004.pdf
Size: 307,730 bytes
SHA-256: 893b57eef4191b419428fa3aa86fee1dc83585e732d985796da1aa0f4dbd07e3
Our copy: nejmc2031915_9f0476e5da176966_recap1004.pdf
Original: nejm.org