select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
A study from the OpenSAFELY platform published in the BMJ looks at COVID-19 risk for adults living with and without children.
This Roundup accompanied an SMC Briefing.
Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, University of Leicester, said:
“As a virologist, there are a number of limitations with this paper, and it is not presented in a very reader-friendly way.
“Firstly, there is no viral sequencing data/analysis to determine if the higher COVID-19 infections are resulting from child-to-parent transmission at home (so their viral sequences should be very similar) – or whether parents may have transmitted/acquired the virus amongst themselves due to enhanced contact rates whilst performing child support activities, e.g. shopping, drop-off/pick-up activities – particularly for the younger children (when parent-child viral sequences may differ).
“I would ignore the first wave data/analysis (Figures 2 and 4) because the state of testing, sample collection/delivery and result reporting during the first wave was very variable and some of the early assays did not work very well. Also community testing was discontinued very early on due to lack of testing capacity and the focus was more on testing symptomatic hospitalised cases. So many children and adults in the community were not tested at all, hence the first wave non-hospitalised child/adult COVID-19 figures were very much underestimated – which will have impacted on any trend shown.
“The second wave data is more reliable (Figures 3 and 5) as diagnostic assay protocols, sample collection/transport, and result reporting through Pillar 1 and 2 data streams were more well established (though still missing and viral sequencing data to assess epidemiological linkage) – which show more consistent trends that living with children overall increased the risk of SARS-COV-2 infection, hospitalisation and ICU admission.
“These findings are more consistent with what we know about children and other respiratory virus infections – including the studies below from South Korea and the USA, which show SARS-COV-2 viral loads in children which can be just as high and he shed for just as long as adults: https://jamanetwork.com/journals/jamapediatrics/fullarticle/2770149 (useful editorial); https://jamanetwork.com/journals/jamapediatrics/fullarticle/2770150 (full study); https://www.massgeneral.org/news/press-release/Massachusetts-general-hospital-researchers-show-children-are-silent-spreaders-of-virus-that-causes-covid-19 (useful summary); https://www.jpeds.com/article/S0022-3476(20)31023-4/fulltext (full study); and reports on school-related COVID-19 outbreaks in the USA and Israel: https://www.cdc.gov/mmwr/volumes/70/wr/mm7008e4.htm and
https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2020.25.29.2001352.
“These studies are all consistent with what we know about the behaviour of other seasonal respiratory viruses (like RSV, parainfluenza, influenza, etc.) in children.
“I would be very wary about any studies reporting contrasting findings to these as they would go against the typical biology of the viral-host relationship in this age group. It would be very interesting to hear any plausible explanation as to why UK children would behave very differently in terms of COVID-19 – compared to these other US, South Korean and Israeli children in this regard – more likely, such studies or the test data on which they are based, may be flawed in some way.”
Comment on both this study and the Archives of Disease in Childhood Scotland study:
Prof Mark Woolhouse, Professor of Infectious Disease Epidemiology, University of Edinburgh, said:
“Two papers published today shed light on the role of children in the transmission of COVID-19 and the risks to adults living in the same household.
“The paper by Forbes and colleagues1 estimates the impact of living with children on the frequency and severity of COVID-19 infections in adults.
“This is an observational study based on the medical histories of twelve million adults. The results from this type of study must be interpreted with care – there may be several possible explanations for any patterns found – but the work does add to a growing body of evidence about the role of children in the current pandemic.
“The findings of the study are complex and open to multiple interpretations.
“A key outcome is that adults over 65 living with children were at increased risk of dying from COVID-19 during the second, but not the first, wave. As schools in England were open in the second, but not the first, wave, this result is consistent with a link between transmission in schools and severe COVID-19 in older co-residents.
“This study cannot establish whether any link is due to direct effects (school-acquired infections being brought into the household) or indirect effects (increased activity and exposure of adults in the household while children are in school). However, the authors note that only 3% of adults over 65 share households with children, so any public health impact must be relatively small.
“Another key outcome is that adults 65 and under living with children were at decreased risk of dying from COVID-19 during both waves. This is despite (in the second wave) their having slightly increased risks of being infected and hospitalised. One possible explanation suggested by the authors is that adults living with children are generally healthier.
“A second paper by McAllister and colleagues2 reports results from a similar study in Scotland.
“This study found that living with children under 12 protected adults against COVID-19 infection. There was no increased risk to adults even when schools were open in Scotland. The authors suggest that this may be due to the impact of adult immune systems of exposure to the mild coronavirus infections that circulate in young children.
“Together, these studies suggest two competing effects. The presence of children in the household may increase the risk of exposure to COVID-19 – especially when schools are open – but for various reasons most adults in those households are less likely to develop severe disease, so the net public health impact is very low.
“The two studies published today add to a large body of evidence that has been accumulating since the earliest days of the pandemic suggesting that children play a limited role in COVID-19 transmission. The evidence that closing schools causes short-term and long-term harms to children has always been much stronger than the evidence that closing schools has any substantial effect on the public health burden of COVID-19.”
1 BMJ OpenSAFELY paper: ‘Association between living with children and outcomes from covid-19: OpenSAFELY cohort study of 12 million adults in England’ by Harriet Forbes et al. will be published in the BMJ at 23:30 UK time on Thursday 18 March 2021, which is also when the embargo will lift.
DOI: 10.1136/bmj.n628
2 Archives of Disease in Childhood Scotland paper: ‘Living with Children and Adults’ Risk of COVID-19: Observational Study’ by Rachael Wood et al. will be published in the Archives of Disease in Childhood at 00:01 UK time on Friday 19 March 2021, which is also when the embargo will lift.
Prof Sarah Lewis, Professor of Molecular Epidemiology, University of Bristol, said:
“In this study adults living with children represented larger households. We have seen from other studies that larger household size increases risk of infection and hospitalisation from COVID-19 with a strong dose response relationship. This is due to larger households having more overall contacts outside the home and a greater potential for transmissions in the home. It is therefore not surprising that living with children increases the risk of COVID-19 somewhat, but this does not mean that it is particularly risky to live with children. The increase in risk was small and no obvious dose response relationship was noted. It would have been informative to see the increase in risk per additional adult in the home for comparison.
“The difference between the first and second wave could be explained by the large increase in testing by the second wave. In the first wave COVID-19 was not measured well, and may have resulted in an underestimate of any effect of living with children. There may be some effect of school closures but the peak of infections in the first wave occurred around the time that lockdown and school closures were implemented, so this is unlikely to explain the difference fully.”
Prof Russell Viner, Professor of Adolescent Health, UCL, said:
“The question of whether children and young people going to school bring home significant COVID risk for parents and grandparents is one of the key issues of this pandemic for families and for policy-makers.
“This exceptionally well-done study using large routine datasets doesn’t answer this question finally, but provides an important part of the jigsaw. The authors undertook a comprehensive set of adjustments and sensitivity analyses, very impressively thinking through the many sources of bias in the data they were using. A major limitation is that they focused on adults that lived with children – and only 3% of the over-65s did so – so this paper doesn’t tell us much about the risk of children visiting grandparents.
“In essence they found that living with younger children brought no increased risk of infection in the first wave but did in the second, and that living with teenagers slightly increased risk of infection in both waves. However increases in risk were low.
“This study fits well with a clearer view that is emerging around the role of children and young people in transmitting this virus. Clearly children and young people can and do transmit SARS-CoV-2, the virus that causes COVID-19 disease. Attending school inevitably brings a risk of children and young people bringing virus home into households, particularly those at secondary schools. However these risks appear low across the whole population. Children and young people contribute to transmission but are not the ‘super spreaders’ they can be for other viruses.
“The reopening of schools is based upon a balance between the risks of increasing transmission of the pandemic and the very real risks of not reopening schools. This paper helps to quantify the transmission risk side of the equation. I believe that this thereby supports policies to cautiously reopen schools.”
Prof Keith Neal, Emeritus Professor of the Epidemiology of Infectious Diseases, University of Nottingham, said:
“This paper suggests that children are responsible for only a small part of the spread of COVID-19 especially within households. Having children under the age of 11 was actually protective against infection which possibly reflects the need to deliver child care and not go out to work.
“Work from community infection studies have consistently shown that higher numbers of people in a household correlates with higher risk if infection. This is what would be expected as the more people in a household, the greater chance of a person introducing infection. (50% of Swedish households are single person).
“Comparing a household with children compared to one with the same number of adults without children a greater risk of introduction would be expected simply based on more opportunities of introducing infection into the household.
“What is surprising is how small the effect of children in a household made.
“These results of the effect of children within households is reassuring now schools have reopened.”
‘Association between living with children and outcomes from covid-19: OpenSAFELY cohort study of 12 million adults in England’ by Harriet Forbes et al. was published in the BMJ at 23:30 UK time on Thursday 18 March 2021.
DOI: 10.1136/bmj.n628
Declared interests
Prof Mark Woolhouse: “No CoIs to declare.”
Prof Russell Viner: “No COI.”
None others received.