key: cord-0705501-56uzzrh8
authors: Molteni, E.; Sudre, C. H.; Canas, L. S.; Bhopal, S. S.; Hughes, R. C.; Antonelli, M. S.; Murray, B.; Klaser, K.; Kerfoot, E.; Chen, L.; Deng, J.; Hu, C.; Selvachandran, S.; Read, K.; Pujol, J. C.; Hammers, A.; Spector, T.; Ourselin, S.; Steves, C. J.; Modat, M.; Absoud, M.; Duncan, E. L.
title: Illness duration and symptom profile in a large cohort of symptomatic UK school-aged children tested for SARS-CoV-2
date: 2021-05-08
journal: nan
DOI: 10.1101/2021.05.05.21256649
sha: 76d542e38288389dace322132d575a96f02cb405
doc_id: 705501
cord_uid: 56uzzrh8

Background In children, SARS-CoV-2 is usually asymptomatic or causes a mild illness of short duration. Persistent illness has been reported; however, its prevalence and characteristics are unclear. We aimed to determine illness duration and characteristics in symptomatic UK school-aged children tested for SARS-CoV-2 using data from the COVID Symptom Study, the largest citizen participatory epidemiological study to date. Methods Data from 258,790 children aged 5-17 years were reported by an adult proxy between 24 March 2020 and 22 February 2021. Illness duration and symptom profiles were analysed for all children testing positive for SARS-CoV-2 for whom illness duration could be determined, considered overall and within younger (5-11 years) and older (12-17 years) age groups. Data from symptomatic children testing negative for SARS-CoV-2, matched 1:1 for age, gender, and week of testing, were also assessed. Findings 1,734 children (588 younger children, 1,146 older children) had a positive SARS-CoV-2 test result and calculable duration of illness with the study time frame. The commonest symptoms were headache (62.2%) and fatigue (55.0%). Median illness duration was six days (vs. three days in children testing negative); and was positively associated with age (rs 0.19, p<1.e-4) with median duration seven days in older vs. five days in younger children. Seventy-seven (4.4%) children had illness duration =>28 days (LC28); LC28 was more common in older compared with younger children (59 (5.1%) vs. 18 (3.1%), p=0.046). The commonest symptoms experienced by children with LC28 were fatigue (84.4%), headache and anosmia (both 77.9%); however, by day 28 the median symptom burden was two. Only 25 (1.8%) of 1,379 children experienced symptoms for [≥]56 days. Few children (15 children, 0.9%) in the negatively-tested cohort experienced prolonged symptom duration; however, these children experienced greater symptom burden (both throughout their illness and at day 28) than children positive for SARS-CoV-2. Interpretation Some children with COVID-19 experience prolonged illness duration; reassuringly, symptom burden in these children did not increase with time, and most recovered by day 56. Some children who tested negative for SARS-CoV-2 also had persistent and burdensome illness. Thus, a holistic approach for all children with persistent illness during the pandemic is required.

To date, the COVID-19 pandemic has resulted in >120 million cases of infection and 2.5 124 million deaths globally, 1 with widespread health, economic, and social chaos. In adults, 125 severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) causes a 126 predominantly respiratory illness 2 of median duration 11 days. 3 In contrast, children with 127 SARS-CoV-2 infection are often asymptomatic (43%-68% 4 ) or have relatively mild 128 symptoms, most commonly cough and fever; 4-9 and life-threatening illness or death is rare. In September 2020, coinciding with full re-opening of primary and secondary schools in the 150 UK, governance for CSS data usage was extended to allow analysis of data from children 151 (i.e., individuals aged <18 years). The UK subsequently experienced further waves of the 152 pandemic (up to 30,000 new cases per day in November 2020 and 60,000 new cases per 153 day in January 2021), 1,17 with 29 December 2020 the peak date for new specimens testing 154 positive over the entire pandemic to date. 17 During this time, there was widespread testing 155 availability for individuals with key symptoms of fever, cough and anosmia, in contrast to the 156 very limited access during the first wave, when testing was mostly restricted to individuals 157 presenting to hospital. 18,19 158 159 Stay-at-home directives and school closures during these later waves resulted in unusually 160 low incidences of commonly circulating viruses such as influenza, adenovirus, and 161 respiratory syncytial virus over the 2020-21 UK winter, both generally and in school-aged 162 children specifically. 20 However, overlap in symptomatology meant many individuals (both 163 adults and children) with respiratory illnesses other than COVID-19 were tested for SARS-164

CoV-2. 165 166

Here we report overall illness duration, individual symptom prevalence and duration, and 167 symptom burden in UK school-aged children testing positive for SARS-CoV-2, whose data 168 were logged through the KCL CSS. We provide similar data for symptomatic children tested 169 for SARS-CoV-2 but who were negative, during the same period. Lastly, we present the 170 prevalence and characteristics of long COVID in children. 171 172

Data were acquired within the COVID Symptom Study (CSS), through a mobile application 174 launched jointly by Zoe Global Ltd. and KCL on 24 March 2020. 16 Briefly, individuals are 175 prompted to report through a smartphone application and provide daily updates of their 176 health status, symptomatology, any SARS-CoV-2 testing, vaccination, and health care 177 access. Symptom assessment includes both direct questions for a list of symptoms (yes/no 178 or other defined options, listed in Supplementary Table 1) , and free-text entry. 16 Adult 179 contributors can also report as a proxy for other persons (children, spouses, elderly relatives, 180 etc.). The relationship between the contributor and proxy-reported individual(s) is not 181 solicited and there is no data linkage between the contributor and proxy-reported 182 individual(s). Children aged 16-17 years can use the app directly as independent 183 contributors or be proxy-reported by an adult. 184 185 UK data from school-aged children (aged 5-17 years) were available from the launch of the 186 app to 22 February 2021, which latter date corresponds to eight weeks after the peak SARS-187

CoV-2 positive specimen date in the UK. 17 Data were considered from launch of the app, 188 and from 1 September 2020 (i.e., return-to-school). The cohort was analysed overall and 189 within two age groups: younger children, aged 5-11 years, corresponding to primary school-190 aged children in the UK; and older children, aged 12-17 years, corresponding to secondary 191 school-aged children in the UK (acknowledging that not all older teenagers attend school (defined as return to asymptomatic state or, if proxy-reporting ceased prior to logging of an 202 asymptomatic report, time of final report). Individuals who had returned to asymptomatic 203 state but for whom another symptomatic report was logged within one week of the last 204 symptomatic report were considered as still being unwell from their initial presentation (i.e., 205 allowing for waxing/waning of illness); calculation of illness duration incorporated these short 206 periods of asymptomatic status. Individuals with reporting gaps longer than one week 207 between symptomatic reports were excluded. Individual symptom prevalence and duration 208

were assessed, with individual symptom duration calculated as time between the first and 209 last report for that symptom. Symptom burden was calculated as the number of different 210 symptoms reported at least once over a defined timeframe (during the first week, the first 28 211 days, at ≥28 days until end of illness, and over the entire duration of illness). Consistent with 212 our previous study in adults, 3 we termed illness with symptoms lasting longer than 28 days 213 as LC28; and longer than 56 days, LC56. Thus, by virtue of data census dates LC28 could 214 only be determined in proxy-reported children whose symptoms commenced on or before 24 215 January 2021, and for LC56 on or before 29 Dec 2020 (noting the peak positive specimen 216 date in the UK was 29 December 2020 17 ). 217 218 Symptom profiles were also assessed in children with COVID-19 who presented for hospital-219 based care (presenting to the accident and emergency department or admitted to hospital), 220 where presentation to hospital followed development of symptoms (i.e., when the positive 221 SARS-CoV-2 test result was not a secondary finding in a hospitalised child). 222 223 Several additional direct symptom questions were added to the app on 4 November 2020 224 (Supplementary Table 2) , some of which overlapped with existing questions (e.g., for 225 cutaneous manifestations, questions on "rash" and "sensitive skin" were added to existing 226 questions on "red welts", "blisters", and "sores"). These additional questions and resultant 227 data are presented in Supplementary Table 2 and Supplementary Figure 1 respectively but  228 were not included in illness duration or symptom burden calculations for the main analysis. 229 230 Free text reporting was also possible, across the entire period. Free text data were divided 231 into themes using frequency of descriptive words; each item within the themes were 232 subsequently independently scrutinised by two clinicians (MA, ELD) to ensure appropriate 233 categorisation, and individuals reporting free text symptoms within those themes were then 234 counted. Free text data are reported here as descriptive statistics (Supplementary Table 3 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (Table 1) . 293

Illness duration was significantly shorter in younger compared with older children (five [IQR 294 2;9] vs. seven days [IQR 3;12]) (Mann-Whitney U test p<1.e-5); and age correlated strongly 295 with illness duration (rs 0.19, p<1.e-4). 296 297 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021.

Individual symptom prevalence and duration are shown in Table 2 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) Table 3) . 368

Irritability (three children), emotional difficulties (two children), and behavioural difficulties 369 (one child) were reported rarely; impaired attention, impaired concentration, and anxiety 370 were not reported. 371 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021. ;

374 15,597 symptomatic children (8,761 younger children, 6,836 older children), tested for 375 SARS-CoV-2 infection but with a negative result, with logging at least once weekly and with 376 calculable illness duration within the requisite time frame, were proxy-reported. 377

Demographic details of the randomly selected matched control sample are shown in Table 1 . last report was two [IQR 1;3]; and proxy-reporting usually ceased early in illness (logging 423 cessation rates: 11.3% of children with illness duration <10 days; 5.5% ≥10 days). Thus, for 424 children for whom proxy-reporting ceased prior to logging of a healthy report, we have 425 assumed that proxy-reporting cessation coincided with illness resolution (i.e., that adults 426 stopped proxy-reporting because the child had recovered) and illness duration was 427 calculated assuming last day of logging corresponded to last day of illness. 428

We considered the impact of this assumption. Excluding all children with a positive SARS-430

CoV-2 test for whom a healthy report was not logged, median duration of illness in the 431 remaining 1,551 was unchanged (6 days [IQR 3;11]) with prevalence of LC28 3.5% 432

[55/1551], within the confidence intervals for LC28 using data from the entire cohort. In this large study of UK school-aged children recruited via the CSS study app, we have 450

shown that symptomatic infection with SARS-CoV-2 in children is usually short, with a 451 median duration of illness of six days compared to 11 days in adults, 3 and with low symptom 452 burden (Figure 3 ). Prolonged illness duration can occur but is infrequent (4.4% with 453 symptoms ≥28 days; 1.8% with symptoms ≥56 days), considerably lower than observed in 454 adults using the same disease definitions (prevalence in adults of LC28: 13.3% and of LC56: 455 4.5%). 3 Age is a risk factor for longer symptom duration (both overall, and specifically for 456 illness duration ≥28 days), consistent with our previous findings in adults. Anosmia (here, both anosmia and dysosmia) was also common in our cohort (39.6% overall: 465 22.4% of younger children, 48.3% of older children). Certainly, for older children this 466 symptom was reported more often than was observed in a previous small study of 467 adolescents (aged 10-19 years) with 'mild to moderate' COVID-19, ascertained from a single 468 centre, in whom prevalence of anosmia was 24.1%, 24 noting that anosmia was one of the 469 core symptoms determining access to testing in the UK during our study period. 470

In children with symptoms ≥28 days, symptom burden was low by day 28 (median symptom 472 burden of two). However, fatigue was a symptom at some stage in almost all (84.4%) of 473 these children (Figure 4, Supplementary Figure 6 ). In adults, fatigue has been reported in 474 many studies as the commonest symptom of long COVID, although prevalence varies. In our 475 previous study in adults, 3 fatigue was almost universal in LC28 (97.7% of cases experienced 476 fatigue during their illness). Other studies have also reported persistence of fatigue after 477 acute infection in adults (e.g., 53.1% at 60 days 25 ; and 52.3% at 10 weeks 26 ). 478 479 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Here, a strength of our study was our ability to compare contemporaneous illness profiles of 513 symptomatic children testing positive vs. negative for SARS-CoV-2, matched for age, 514 gender, and week of testing. Children testing positive for SARS-CoV-2 had longer median 515 illness duration (six days vs. three days in negatives); and were more likely to have illness 516 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021. ; https://doi.org/10.1101/2021.05.05.21256649 doi: medRxiv preprint duration ≥28 days (4.4% vs. 0.9%). However, some children testing negative for SARS-CoV-517 2 also had symptoms that lasted >28 days, and these children had a higher symptom count 518 both over the duration of their illness and at day 28 (Supplementary Figures 4, 5) , 519 acknowledging here that our sample size is small. We considered whether some children 520

with long illness duration who tested negative for SARS-CoV-2 might have false negative 521 results. However, there is no evidence that sensitivity and specificity of testing for SARS-522

CoV-2 are different in adults compared with children, with sensitivity for PCR SARS-CoV-2 523 tests around 95%. relaxation of personal protection and social distancing measures, it is likely that these 532 illnesses will return to more usual (i.e., higher) levels of circulation in future UK winters. Our 533 data highlight that other illnesses may also have a protracted and burdensome course in 534 children, which will also need consideration in post-pandemic service delivery planning. 535 536 Short and long-term effects of COVID-19 on school performance and learning have been a 537 recent matter of concern. 38 In our cohort, neither attentional problems nor memory 538 complaints nor anxiety were reported. Isolated cases of low mood and/or irritability were 539 consistent with, if not lower than, previously reported statistics in the general school-aged 540 healthy population. 39 Our data do not support anecdotal reports of weakness and seizures 541 as common features in children with COVID-19, whether of short or longer illness duration; 542 and no severe neurological symptoms were reported. However, persistence of symptoms 543 from any illness can be associated with low mood, with adverse long-term outcomes 544 including school refusal and separation anxiety. 40 Although self-reporting adults were able to 545 participate in specific questions regarding mental health, these data could not be proxy-546

reported. This limits our ability to assess mental health issues comprehensively, and any 547 potential impact on school performance and learning, in children during the COVID-19 548 pandemic whether testing positive or negative for SARS-CoV-2. 549 550 In considering how our data relates to other sources, the UK Office for National Statistics 551 (ONS) conducted a round of testing for SARS-CoV-2 (irrespective of symptoms) from 2 to 10 552

December 2020, prior to the UK peak of infection, with deliberate oversampling of schools 553 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021. ; https://doi.org/10.1101/2021.05.05.21256649 doi: medRxiv preprint which had had the highest infection rates at the start of the academic year. 121 schools (41 554 primary, 80 secondary) in 15 local authorities were examined, with a total of 7,089 pupils. 41 555 Primary school pupils and staff had a slightly lower percentage testing positive (0.94% and 556 0.99% respectively) for SARS-CoV-2 than secondary pupils and staff (1.22% and 1.64%). 557 ONS warned that their conclusions might not have general validity; and these figures do not 558 capture time fluctuations as the pandemic progressed. In our dataset, with data captured 559 until 22 February 2021 (thus including the subsequent rise in numbers over Christmas and 560

New Year), 6,043 of 258,790 (2.3%) of proxy-reported children were reported with a positive 561 test (2.3% of younger children, 3.7% of older children), noting that reporting was voluntary, 562 by proxy, and through a specific COVID-19 platform, and that testing was only available for 563 symptomatic individuals. Additionally, the CSS app user base (considered as a whole) is not 564 fully representative of the wider UK population, due to over-representation of female gender, 565 white background, and above-average socioeconomic status. 18 Further, we cannot 566 characterise regional variability, as geographic information was not available for many 567 participants. The ONS has also provided estimates of long COVID in children aged 2-16 years, with the 583 most recent data release (April 2021) suggesting that 9.8% of children aged 2-11 years and 584 13.0% of those aged 12-16 years experience ongoing symptoms five weeks after testing 585 positive for SARS-CoV-2, and 7.4% and 8.2% respectively are still reporting symptoms at 12 586 weeks. 45 These latest figures include a control group (defined as those who were never 587 symptomatic, never tested, never-self isolated, and never a contact of anyone testing 588 positive for SARS-CoV-2) which suggested 'baseline' rates for the same symptoms of 2% 589 amongst 2-to 11-year-olds and 1.7% amongst 12-to 16-year-olds. Earlier, in January 2021, 590 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021. ;

ONS had reported long COVID symptoms by age; however, updated age-specific estimates 591 for specific symptoms were not included in April 2021 data release. Overall ONS long 592 COVID prevalence estimates were adjusted downwards between January and April 2021 593 (e.g., from 12.9% to 9.8% in 2-to 11-year-olds). 45,46 594 595 There is some disparity between our prevalence data for LC28 in children and those of the 596 ONS. At least part of the reason may be that the ONS required two consecutive 597 asymptomatic visits to define the end of illness. Thus, individuals with relapsing/remitting 598 symptoms in whom asymptomatic periods lasted longer than one week would be captured 599 by ONS but not by our study. Consistent with this, the ONS' sensitivity analysis of the impact 600 of defining symptom discontinuity to a single asymptomatic visit radically lowered their 601 prevalence estimates, especially for ongoing symptoms at 12 weeks, with estimates falling 602 from 13.7% to 0.9% (ref 47 -Table 11 Our study is part of one of the largest citizen science initiatives ever in the UK, with 617 generation of real-time epidemiological data from over 4.5 million users across the UK. We 618 leveraged previously published methodologies assessing illness duration and symptom 619 profiling in adults, including assessment of long COVID. 3 Our data census points allowed us 620 to capture all children with illness duration ≥8 weeks, if they had presented before the date of 621 peak specimen receipt in the UK; and our restriction to individuals whose symptom 622 presentation concorded with test timing allowed an accurate determination of symptom 623 onset. Moreover, by restricting analyses to start from 1 September, we avoided bias due to 624 limited test availability during the first wave of the pandemic. However, despite general 625 availability there were still some blocks to testing -in particular, that individuals were 626 required at least one of a defined list of symptoms (specifically, fever, cough and anosmia), 50 627 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 8, 2021. ; https://doi.org/10.1101/2021.05.05.21256649 doi: medRxiv preprint which list was largely informed by adult symptomatology and might not be sufficiently broad 628 to capture some common paediatric manifestations of COVID-19, (e.g., abdominal pain, 629 reported in 4% of paediatric cases 4 and in 27.8% of our younger children). Related to this, 630 the set of questions asked through the app, although refined between 24 March and 1 631 September, and again on 4 November 2020, was largely informed by research in adults. 632

Although we cannot exclude the possibility that a paediatric version might capture other 633 manifestations of COVID-19, the free text data did not suggest common themes emerging 634 unique to paediatric populations. We did not undertake a formal qualitative analysis of the 635 free-text data given (a) its ad hoc collection rather than a formal qualitative study, and (b) the 636 potential bias arising from introduction of additional direct symptom questions from 4 637

November 2020 (i.e., once a symptom was asked about directly, it was unlikely that to be 638 reported as free text). None-the-less, we have provided the data from free text responses, 639 and from questions asked after 4 November 2020 in the Supplementary Material. We 640 acknowledge that symptoms were reported by proxy rather than directly ascertained; 641 however, this is common in clinical assessment of children, particularly younger children. 642

Additionally, children cannot directly consent to research participation. We also acknowledge 643 the possibility that older children may have reported for themselves using their relatives' 644 phone/log-in function and/or misused proxy-reporting from their own mobile phones; 645 however, we have no means of capturing such activity. 646

Our national cohort of children and young people provides the first systematic description of 648 COVID-19 in children. Our data show that long illness duration after SARS-CoV-2 infection 649 in school-aged children is not common; however, a small proportion of children do have 650 prolonged illness duration and persistent symptoms, validating these children's experiences. 651

Our LC56 data provide reassurance regarding the longer term outcome for these children. 652

The symptom burden in children who tested negative for SARS-CoV-2 but had long illness 653 duration highlights that allocation of appropriate resources will be necessary for any child 654 with prolonged illness, whether due to SARS-CoV-2 infection or other illness. Our study 655 provides timely and critical data to inform discussions around the impact and implications of 656 the pandemic on UK paediatric healthcare resource allocation. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 8, 2021. ; https://doi.org/10.1101/2021.05.05.21256649 doi: medRxiv preprint Table 1 . Characteristics of school-aged children who tested positive for SARS-CoV-2, and the control cohort of children (matched 1:1 for age, gender, and week of testing) who tested negative for SARS-CoV-2. The cohort of children with positive SARS-CoV-2 testing is presented here both as younger and older groups; and for usual (i.e., short) vs. extended illness duration. 

Legend: overall number for the entire cohort of children is given first; numbers within brackets separated by oblique refer to younger children and older children in that order. 'Not valid result' -PCR test result proxy-reported as "failed test" or "still waiting". 'Duration calculable' -illness onset within defined timeframe of testing for SARS-CoV-2, and with defined endpoint (for details, please see Methods). 'Irregular logging' -proxy-reporting with intervals of >7 days between proxy-reports during illness duration). 'Illness onset outside of study bounds' -symptom onset before 1 September 2020 or after 24 January 2021. 'Hosp'

-presenting to hospital (either admitted to hospital or seen in Accident and Emergency ward).

[image at following page]

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 

Reported Cases and Dearths by Country or Territory

Epidemiologic, clinical, and 688 laboratory findings of the COVID-19 in the current pandemic: Systematic review and 689 meta-analysis

Attributes and predictors of long COVID. 692

Epidemiological and Clinical 694 Characteristics of COVID-19 in Children: A Systematic Review and Meta-Analysis. 695 Frontiers in Pediatrics

Ocular Manifestations and Clinical Characteristics of 697 Children with Laboratory-Confirmed COVID-19 in Wuhan, China

Clinical characteristics of pediatric 700 SARS-CoV-2 infection and coronavirus disease 2019 (COVID-19) in Kuwait

Comparison of Clinical and Epidemiological 703 Characteristics of Asymptomatic and Symptomatic SARS-CoV-2 Infection in Children

Childhood COVID-19: a multicentre retrospective 706 study

Clinical Features and Temporal Changes of RT-PCR 708 and Chest CT in COVID-19 Pediatric Patients

Sick-Samuels AC. Acute covid-19 and 711 multisystem inflammatory syndrome in children

SARS-CoV-2 paediatric inflammatory syndrome. Paediatrics 714 and Child Health

Inflammatory Syndrome in Children -United States

Post-719 COVID-19 Symptom Burden: What is Long-COVID and How Should We Manage It? 720 Lung

Post-acute COVID-19 syndrome

COVID-19 rapid guideline: 724 managing the long-term effects of COVID-19 (NG188). Evidence reviews 2 and 3: 725 prevalence

Real-time tracking of self-reported symptoms 727 to predict potential COVID-19

COVID-19) in the UK -730 Cases by specimen date

Detecting COVID-19 infection hotspots 732 in England using large-scale self-reported data from a mobile application: a 733 prospective, observational study. The Lancet Public Health

Coronavirus (COVID-19) in the UK -Lab-based 736 testing and capacity, by test type

Weekly national Influenza and COVID-19 surveillance report 739 -Week 11 report (up to week 10 data

Signs and symptoms to determine if a patient 744 presenting in primary care or hospital outpatient settings has COVID-19. Cochrane 745 Database of Systematic Reviews

Neurologic and Radiographic Findings Associated with 748 COVID-19 Infection in Children

Post-acute COVID-19 750 outcomes in children with mild and asymptomatic disease

Loss of smell and taste in 754 COVID-19 infection in adolescents

Persistent fatigue following SARS-CoV-2 760 infection is common and independent of severity of initial infection

Prevalence of headache and 763 migraine in children and adolescents: A systematic review of population-based 764 studies. Developmental Medicine and Child Neurology

Prevalence of headache and migraine in schoolchildren

Prevalence of chronic disabling fatigue 769 in children and adolescents

Unidentified chronic fatigue 772 syndrome/myalgic encephalomyelitis (CFS/ME) is a major cause of school absence: 773 Surveillance outcomes from school-based clinics

Natural course of chronic fatigue 776 syndrome/myalgic encephalomyelitis in adolescents. Archives of Disease in 777

Behavioral, virologic, and 779 immunologic factors associated with acquisition and severity of primary epstein-barr 780 virus infection in university students

Baseline characteristics, management, 783 and outcomes of 55,270 children and adolescents diagnosed with COVID-19 and 784 1,952,693 with influenza in France

Steel K, Davies B. 5. Test sensitivity and 788 Specificity

COVID-19 Hospital Activity

Weekly national seasonal respiratory report -Week ending 795 28

Case report and systematic review suggest that children may 801 experience similar long-term effects to adults after clinical COVID-19

803 39 UK Office for National Statistics (ONS) ER. Young people's well-being in the UK

Chronic physical illness and 808 mental health in children

COVID-19 Schools Infection Survey Round

Public Health Scotland. Total Cases by Age and Sex

Public Health Wales TP. Rapid COVID-19 surveillance

Northern Ireland COVID-19 Tracker -Cumulative Number of 821 Individuals Tested Positive by Age

UK Office for National Statistics (ONS) DA. Prevalence of long COVID symptoms 824 and COVID-19 complications

Updated estimates of the prevalence of long 829 COVID symptoms

All data related to Prevalence of ongoing 834 symptoms following coronavirus (COVID-19) infection in the UK

Forgotten children living with long Covid

The 840 Times

Why it's not too late for the UK to pursue a zero-Covid strategy

Knowledge barriers in the symptomatic-846 COVID-19 testing programme in the UK: an observational study. medRxiv 2021. 847 Illness duration (median

BMI: Body Mass Index. IQR: Inter Quartile Range

Symptoms over the duration of illness in children testing positive for SARS-CoV-2. Sample tested positive for SARS-CoV-2 Younger children (aged 5-11 years, n=588) Older children