High variability in transmission of SARS-CoV-2 within households and implications for control
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) poses a high risk of transmission in close-contact indoor settings, which may include households. Prior studies have found a wide range of household secondary attack rates and may contain biases due to simplifying assumptions about transmission variability and test accuracy.
Methods
We compiled serological SARS-CoV-2 antibody test data and prior SARS-CoV-2 test reporting from members of 9,224 Utah households. We paired these data with a probabilistic model of household importation and transmission. We calculated a maximum likelihood estimate of the importation probability, mean and variability of household transmission probability, and sensitivity and specificity of test data. Given our household transmission estimates, we estimated the threshold of non-household transmission required for epidemic growth in the population.
Results
We estimated that individuals in our study households had a 0.41% (95% CI 0.32%– 0.51%) chance of acquiring SARS-CoV-2 infection outside their household. Our household secondary attack rate estimate was 36% (27%– 48%), substantially higher than the crude estimate of 16% unadjusted for imperfect serological test specificity and other factors. We found evidence for high variability in individual transmissibility, with higher probability of no transmissions or many transmissions compared to standard models. With household transmission at our estimates, the average number of non-household transmissions per case must be kept below 0.41 (0.33–0.52) to avoid continued growth of the pandemic in Utah.
Conclusions
Our findings suggest that crude estimates of household secondary attack rate based on serology data without accounting for false positive tests may underestimate the true average transmissibility, even when test specificity is high. Our finding of potential high variability (overdispersion) in transmissibility of infected individuals is consistent with characterizing SARS-CoV-2 transmission being largely driven by superspreading from a minority of infected individuals. Mitigation efforts targeting large households and other locations where many people congregate indoors might curb continued spread of the virus.
Article activity feed
-
-
-
SciScore for 10.1101/2021.01.29.20248797: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
No key resources detected.
Results from OddPub: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:This study has several limitations. Our estimate of high household transmission variability may not be robust to alternate assumptions for the way community acquisition risk varies by household. For example, some households could have been comprised of families with both parents working essential jobs during Spring/Summer 2020, with children attending in-person day care or camps, thus placing the entire household at much higher risk of …
SciScore for 10.1101/2021.01.29.20248797: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
No key resources detected.
Results from OddPub: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:This study has several limitations. Our estimate of high household transmission variability may not be robust to alternate assumptions for the way community acquisition risk varies by household. For example, some households could have been comprised of families with both parents working essential jobs during Spring/Summer 2020, with children attending in-person day care or camps, thus placing the entire household at much higher risk of community acquisition compared to households working / caring for children at home. Also, households could have high collective community acquisition probability via attending multi-household gatherings of extended family or other social groups. In these ways, households conceivably could vary considerably in their infection numbers for reasons that don’t involve within-household transmission. We tested the implications of this alternate possibility for household variability in our model by allowing variability in community acquisition by household using an additional dispersion parameter to the MLE model (Supplementary Results). Interestingly, the MLE for the transmission dispersion parameter dh still occurred at high variability in household transmission under this alternate model. Furthermore, the improvement in likelihood was not substantial, such that the more complicated model would not be favored by the likelihood ratio test nor the Bayesian information criterion. However, larger uncertainty ranges under the alternate model suggest that ...
Results from TrialIdentifier: No clinical trial numbers were referenced.
Results from Barzooka: We did not find any issues relating to the usage of bar graphs.
Results from JetFighter: We did not find any issues relating to colormaps.
Results from rtransparent:- Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
- Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
- No protocol registration statement was detected.
-