Illness Characteristics of COVID-19 in Children Infected with the SARS-CoV-2 Delta Variant

SARS-CoV-2 variants of concern in children and adolescents with COVID-19: a systematic review

OBJECTIVES

Infections by SARS-CoV-2 variants of concern (VOCs) might affect children and adolescents differently than earlier viral lineages. We aimed to address five questions about SARS-CoV-2 VOC infections in children and adolescents: (1) symptoms and severity, (2) risk factors for severe disease, (3) the risk of infection, (4) the risk of transmission and (5) long-term consequences following a VOC infection.

DESIGN

Systematic review.

DATA SOURCES

The COVID-19 Open Access Project database was searched up to 1 March 2022 and PubMed was searched up to 9 May 2022.

ELIGIBILITY CRITERIA

We included observational studies about Alpha, Beta, Gamma, Delta and Omicron VOCs among ≤18-year-olds. We included studies in English, German, French, Greek, Italian, Spanish and Turkish.

DATA EXTRACTION AND SYNTHESIS

Two reviewers extracted and verified the data and assessed the risk of bias. We descriptively synthesised the data and assessed the risks of bias at the outcome level.

RESULTS

We included 53 articles. Most children with any VOC infection presented with mild disease, with more severe disease being described with the Delta or the Gamma VOC. Diabetes and obesity were reported as risk factors for severe disease during the whole pandemic period. The risk of becoming infected with a SARS-CoV-2 VOC seemed to increase with age, while in daycare settings the risk of onward transmission of VOCs was higher for younger than older children or partially vaccinated adults. Long-term symptoms following an infection with a VOC were described in <5% of children and adolescents.

CONCLUSION

Overall patterns of SARS-CoV-2 VOC infections in children and adolescents are similar to those of earlier lineages. Comparisons between different pandemic periods, countries and age groups should be improved with complete reporting of relevant contextual factors, including VOCs, vaccination status of study participants and the risk of exposure of the population to SARS-CoV-2.

PROSPERO REGISTRATION NUMBER

CRD42022295207.

Roadmap to recovery: Implemented and attitude toward school reopening strategies during the COVID-19 pandemic, a scoping review

The novel coronavirus disease 2019 (COVID-19) has had various financial and life impacts on the world's population. Schools' regular activity and function during the pandemic require balancing the repercussions of suspending in-person education versus health threats. Furthermore, children are one of the prominent victims of the restricted quarantine strategies' effects, which may make them vulnerable to various mental health problems. In this study, we reviewed previously reported strategies and roadmaps regarding the reopening of schools during the COVID-19 pandemic. The following databases were searched from October to December 2021, via multi-step search strategies for "COVID-19," "coronavirus," "school reopening," "roadmaps," "reopening," and "reopening strategies": Google Scholar, PubMed, Scopus, and Web of Science. A total of five papers with roadmaps focusing on reopening schools were included in this study. Fundamental issues and principles of these reviewed roadmaps were: 1) protecting the high-risk students and staff physically and mentally, 2) accelerating the vaccination of essential workers, staff, parents, and students, and 3) improving the COVID-19 testing capacity. Roadmaps for the reopening of the schools should describe some phases and steps for their strategies. Current roadmaps have not mentioned any phases and timelines for this process. Describing some health metrics in the roadmaps for progressing to the next step or returning to the previous ones is also necessary for all roadmaps and should be considered in further studies.

Peter SD, Fisher JC, Levy-Lambert D, Reichl A, Ignacio RC, Slater BJ, Tsao K, Berman L. Postoperative respiratory complications in SARS-CoV-2 positive pediatric patients across 20 United States hospitals: A Cohort Study

INTRODUCTION

Data examining rates of postoperative complications among SARS-CoV-2 positive children are limited. The purpose of this study was to evaluate the impact of symptomatic and asymptomatic SARS-CoV-2 positive status on postoperative respiratory outcomes for children.

METHODS

This retrospective cohort study included SARS-CoV-2 positive pediatric patients across 20 hospitals who underwent general anesthesia from March to October 2020. The primary outcome was frequency of postoperative respiratory complications, including: high-flow nasal cannula/non invasive ventilation, reintubation, pneumonia, Extracorporeal Membrane Oxygenation (ECMO), and 30-day respiratory-related readmissions or emergency department (ED) visits. Univariate analyses were used to evaluate associations between patient and procedure characteristics and stratified analyses by symptoms were performed examining incidence of complications.

RESULTS

Of 266 SARS-CoV-2 positive patients, 163 (61.7%) were male, and the median age was 10 years (interquartile range 4-14). The majority of procedures were emergent or urgent (n = 214, 80.5%). The most common procedures were appendectomies (n = 78, 29.3%) and fracture repairs (n = 40,15.0%). 13 patients (4.9%) had preoperative symptoms including cough or dyspnea. 26 patients (9.8%) had postoperative respiratory complications, including 15 requiring high-flow oxygen, 8 with pneumonia, 4 requiring non invasive ventilation, 3 respiratory ED visits, and 2 respiratory readmissions. Respiratory complications were more common among symptomatic patients than asymptomatic patients (30.8% vs. 8.7%, p = 0.01). Higher ASA class and comorbidities were also associated with postoperative respiratory complications.

CONCLUSIONS

Postoperative respiratory complications are less common in asymptomatic versus symptomatic SARS-COV-2 positive children. Relaxation of COVID-19-related restrictions for time-sensitive, non urgent procedures in selected asymptomatic patients may be reasonably considered. Additionally, further research is needed to evaluate the costs and benefits of routine testing for asymptomatic patients.

LEVEL OF EVIDENCE

Iii, Respiratory complications.

Public perceptions and interactions with UK COVID-19 Test, Trace and Isolate policies, and implications for pandemic infectious disease modelling

Background The efforts to contain SARS-CoV-2 and reduce the impact of the COVID-19 pandemic have been supported by Test, Trace and Isolate (TTI) systems in many settings, including the United Kingdom. Mathematical models of transmission and TTI interventions, used to inform design and policy choices, make assumptions about the public’s behaviour in the context of a rapidly unfolding and changeable emergency. This study investigates public perceptions and interactions with UK TTI policy in July 2021, assesses them against how TTI processes are conceptualised and represented in models, and then interprets the findings with modellers who have been contributing evidence to TTI policy. Methods 20 members of the public recruited via social media were interviewed for one hour about their perceptions and interactions with the UK TTI system. Thematic analysis identified key themes, which were then presented back to a workshop of pandemic infectious disease modellers who assessed these findings against assumptions made in TTI intervention modelling. Workshop members co-drafted this report. Results Themes included education about SARS-CoV-2, perceived risks, trust, mental health and practical concerns. Findings covered testing practices, including the uses of and trust in different types of testing, and the challenges of testing and isolating faced by different demographic groups. This information was judged as consequential to the modelling process, from guiding the selection of research questions, influencing choice of model structure, informing parameter ranges and validating or challenging assumptions, to highlighting where model assumptions are reasonable or where their poor reflection of practice might lead to uninformative results. Conclusions We conclude that deeper engagement with members of the public should be integrated at regular stages of public health intervention modelling.

Effects of vaccination, new SARS-CoV-2 variants and reinfections on post-COVID-19 complications

Post-COVID-19 complications involve a variety of long-lasting health complications emerging in various body systems. Since the prevalence of post-COVID-19 complications ranges from 8-47% in COVID-19 survivors, it represents a formidable challenge to COVID-19 survivors and the health care system. Post-COVID-19 complications have already been studied in the connection to risk factors linked to their higher probability of occurrence and higher severity, potential mechanisms underlying the pathogenesis of post-COVID-19 complications, and their functional and structural correlates. Vaccination status has been recently revealed to represent efficient prevention from long-term and severe post-COVID-19 complications. However, the exact mechanisms responsible for vaccine-induced protection against severe and long-lasting post-COVID-19 complications remain elusive. Also, to the best of our knowledge, the effects of new SARS-CoV-2 variants and SARS-CoV-2 reinfections on post-COVID-19 complications and their underlying pathogenesis remain to be investigated. This hypothesis article will be dedicated to the potential effects of vaccination status, SARS-CoV-2 reinfections, and new SARS-CoV-2 variants on post-COVID-19 complications and their underlying mechanisms Also, potential prevention strategies against post-COVID complications will be discussed.

Air pollution might affect the clinical course of COVID-19 in pediatric patients

Air pollution, to which children are more susceptible than adults, can promote airway inflammation, potentially exaggerating the effects of respiratory viral infection. This study examined the association between the clinical manifestation of COVID-19 in unvaccinated pediatric patients hospitalized in Poland (n = 766) and levels of particulate matter 2.5 (PM_2.5) and benzo(a)pyrene (B(a)P) within a week before hospitalization. Children aged ≤ 12 years exposed to mean and max 24 h B(a)P levels > 1 ng/m³ revealed higher odds of cough, dyspnea, fever, and increased concentrations of inflammatory markers (C-reactive protein, interleukin-6, procalcitonin, white blood cell count). In older patients (13-17 years), elevated mean 24 h B(a)P levels increased odds of dyspnea, fever, and diarrhea, and higher concentrations of C-reactive protein and procalcitonin. Exposure to max 24 h PM_2.5 levels > 20 µg/m³ was associated with higher odds of cough, increased concentrations of C-reactive protein (group ≤12 years), and increased procalcitonin concentration (groups ≤12 years and 13-17 years). In both age groups, length of stay was extended in patients exposed to elevated levels of max 24 h PM_2.5, mean and max 24 h B(a)P. This study suggests that worse air quality, particularly reflected in increased B(a)P levels, might affect the clinical course of COVID-19 in pediatric patients and adds to the disease burden during a pandemic.

COVID-19 due to the B.1.617.2 (Delta) variant compared to B.1.1.7 (Alpha) variant of SARS-CoV-2: a prospective observational cohort study

The Delta (B.1.617.2) variant was the predominant UK circulating SARS-CoV-2 strain between May and December 2021. How Delta infection compares with previous variants is unknown. This prospective observational cohort study assessed symptomatic adults participating in the app-based COVID Symptom Study who tested positive for SARS-CoV-2 from May 26 to July 1, 2021 (Delta overwhelmingly the predominant circulating UK variant), compared (1:1, age- and sex-matched) with individuals presenting from December 28, 2020 to May 6, 2021 (Alpha (B.1.1.7) the predominant variant). We assessed illness (symptoms, duration, presentation to hospital) during Alpha- and Delta-predominant timeframes; and transmission, reinfection, and vaccine effectiveness during the Delta-predominant period. 3581 individuals (aged 18 to 100 years) from each timeframe were assessed. The seven most frequent symptoms were common to both variants. Within the first 28 days of illness, some symptoms were more common with Delta versus Alpha infection (including fever, sore throat, and headache) and some vice versa (dyspnoea). Symptom burden in the first week was higher with Delta versus Alpha infection; however, the odds of any given symptom lasting ≥ 7 days was either lower or unchanged. Illness duration ≥ 28 days was lower with Delta versus Alpha infection, though unchanged in unvaccinated individuals. Hospitalisation for COVID-19 was unchanged. The Delta variant appeared more (1.49) transmissible than Alpha. Re-infections were low in all UK regions. Vaccination markedly reduced the risk of Delta infection (by 69-84%). We conclude that COVID-19 from Delta or Alpha infections is similar. The Delta variant is more transmissible than Alpha; however, current vaccines showed good efficacy against disease. This research framework can be useful for future comparisons with new emerging variants.

Illness Characteristics of COVID-19 in Children Infected with the SARS-CoV-2 Delta Variant

BACKGROUND

The Delta (B.1.617.2) SARS-CoV-2 variant was the predominant UK circulating strain between May and November 2021. We investigated whether COVID-19 from Delta infection differed from infection with previous variants in children.

METHODS

Through the prospective COVID Symptom Study, 109,626 UK school-aged children were proxy-reported between 28 December 2020 and 8 July 2021. We selected all symptomatic children who tested positive for SARS-CoV-2 and were proxy-reported at least weekly, within two timeframes: 28 December 2020 to 6 May 2021 (Alpha (B.1.1.7), the main UK circulating variant) and 26 May to 8 July 2021 (Delta, the main UK circulating variant), with all children unvaccinated (as per national policy at the time). We assessed illness profiles (symptom prevalence, duration, and burden), hospital presentation, and presence of long (≥28 day) illness, and calculated odds ratios for symptoms presenting within the first 28 days of illness.

RESULTS

694 (276 younger (5-11 years), 418 older (12-17 years)) symptomatic children tested positive for SARS-CoV-2 with Alpha infection and 706 (227 younger and 479 older) children with Delta infection. Median illness duration was short with either variant (overall cohort: 5 days (IQR 2-9.75) with Alpha, 5 days (IQR 2-9) with Delta). The seven most prevalent symptoms were common to both variants. Symptom burden over the first 28 days was slightly greater with Delta compared with Alpha infection (in younger children, 3 (IQR 2-5) symptoms with Alpha, 4 (IQR 2-7) with Delta; in older children, 5 (IQR 3-8) symptoms with Alpha, 6 (IQR 3-9) with Delta infection ). The odds of presenting several symptoms were higher with Delta than Alpha infection, including headache and fever. Few children presented to hospital, and long illness duration was uncommon, with either variant.

CONCLUSIONS

COVID-19 in UK school-aged children due to SARS-CoV-2 Delta strain B.1.617.2 resembles illness due to the Alpha variant B.1.1.7., with short duration and similar symptom burden.

Illness Characteristics of COVID-19 in Children Infected with the SARS-CoV-2 Delta Variant

BACKGROUND

The Delta (B.1.617.2) SARS-CoV-2 variant was the predominant UK circulating strain between May and November 2021. We investigated whether COVID-19 from Delta infection differed from infection with previous variants in children.

METHODS

Through the prospective COVID Symptom Study, 109,626 UK school-aged children were proxy-reported between 28 December 2020 and 8 July 2021. We selected all symptomatic children who tested positive for SARS-CoV-2 and were proxy-reported at least weekly, within two timeframes: 28 December 2020 to 6 May 2021 (Alpha (B.1.1.7), the main UK circulating variant) and 26 May to 8 July 2021 (Delta, the main UK circulating variant), with all children unvaccinated (as per national policy at the time). We assessed illness profiles (symptom prevalence, duration, and burden), hospital presentation, and presence of long (≥28 day) illness, and calculated odds ratios for symptoms presenting within the first 28 days of illness.

RESULTS

694 (276 younger (5-11 years), 418 older (12-17 years)) symptomatic children tested positive for SARS-CoV-2 with Alpha infection and 706 (227 younger and 479 older) children with Delta infection. Median illness duration was short with either variant (overall cohort: 5 days (IQR 2-9.75) with Alpha, 5 days (IQR 2-9) with Delta). The seven most prevalent symptoms were common to both variants. Symptom burden over the first 28 days was slightly greater with Delta compared with Alpha infection (in younger children, 3 (IQR 2-5) symptoms with Alpha, 4 (IQR 2-7) with Delta; in older children, 5 (IQR 3-8) symptoms with Alpha, 6 (IQR 3-9) with Delta infection ). The odds of presenting several symptoms were higher with Delta than Alpha infection, including headache and fever. Few children presented to hospital, and long illness duration was uncommon, with either variant.

CONCLUSIONS

COVID-19 in UK school-aged children due to SARS-CoV-2 Delta strain B.1.617.2 resembles illness due to the Alpha variant B.1.1.7., with short duration and similar symptom burden.

Circulation of SARS-CoV-2 Variants among Children from November 2020 to January 2022 in Trieste (Italy)

INTRODUCTION

The ongoing coronavirus disease 19 (COVID-19) outbreak involves the pediatric population, but to date, few reports have investigated the circulation of variants among children.

MATERIAL AND METHODS

In this retrospective study, non-hospitalized pediatric patients with SARS-CoV-2-positive nasopharyngeal swabs (NPS) were enrolled at the Institute for Maternal and Child Health-IRCCS Burlo Garofolo, Trieste (Italy), from November 2020 to January 2022. SARS-CoV-2 variants were identified by in vitro viral isolation, amplification, automatic sequencing of the receptor binding domain (RBD) of the SARS-CoV-2 spike coding gene, and subsequent next-generation sequencing. The growth curves of the isolated strains were defined in vitro by infecting Vero-E6 cells and quantifying the viral load in the supernatants up to 72 h post-infection by qRT-PCR. The neutralization activity of sera obtained from a COVID-19 vaccinated subject, recovered (2020) patient, vaccinated and recovered (2021) patient, and seronegative subject was assessed by microneutralization assay against the different variants.

RESULTS

In total, 32 SARS-CoV-2-positive children, 16 (50%) females, with a median age of 1.4 years (range: 1 day-13 years), were enrolled. The D614G amino acid substitution was detected in all isolated and amplified viral strains. Of the 32 isolates, 4 (12.5%) carried a nonsynonymous nucleotide mutation leading to the N439K (3/4), lineage B.1.258 (∆H69/∆V70), and S477N (1/4) substitution. In 7/32 (21.8%) isolates, amino acid substitutions allowed the identification of a delta variant, lineage B.1.617.2-AY.43, and in 1/32 (3.1%), the Omicron strain (B.1.1.529.BA1) was identified. The growth curves of the B.1, B.1.258 (∆H69/∆V70), B.1.617.2-AY.43, and B.1.1.529.BA1 variants did not show any significant differences. A reduction in the serum neutralizing activity against B.1.258 (∆H69/∆V70) only in a vaccinated subject (1.7-fold difference), against B.1.617.2-AY.43 in a vaccinated subject and in recovered patients (12.7 and ≥2.5-fold differences, respectively), and against B.1.1.529.BA1 variant (57.6- and 1.4-fold differences in vaccinated and in vaccinated and recovered patients) were observed compared to the B.1 variant.

CONCLUSIONS

SARS-CoV-2 variants carrying the B.1.258 (∆H69/∆V70) and S477N substitutions were reported here in a pediatric population for the first time. Although the growth rates of the isolated strains (B.1.258, B.1.617.2-AY.43, B.1.1.529.BA1) did not differ from the B.1 variant, neutralizing activity of the sera from vaccinated subjects significantly decreased against these variants. Attention should be devoted to the pediatric population to prevent the spread of new SARS-CoV-2 variants in an unvaccinated and predominantly naive population.

Predictive value of isolated symptoms for diagnosis of SARS-CoV-2 infection in children tested during peak circulation of the delta variant

Background

Coronavirus disease 2019 (COVID-19) testing policies for symptomatic children attending US schools or daycare vary, and whether isolated symptoms should prompt testing is unclear. We evaluated children presenting for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing to determine if the likelihood of having a positive SARS-CoV-2 test differed between participants with 1 symptom vs ≥2 symptoms, and to examine the predictive capability of isolated symptoms.

Methods

Participants aged <18 years presenting for clinical SARS-CoV-2 molecular testing in 6 sites in urban/suburban/rural Georgia (July–October, 2021; Delta variant predominant) were queried about individual symptoms. Participants were classified into 3 groups: asymptomatic, 1 symptom only, or ≥2 symptoms. SARS-CoV-2 test results and clinical characteristics of the 3 groups were compared. Sensitivity, specificity, positive predictive values (PPVs), and negative predictive values (NPVs) for isolated symptoms were calculated by fitting a saturated Poisson model.

Results

Of 602 participants, 21.8% tested positive and 48.7% had a known or suspected close contact. Children reporting 1 symptom (n = 82; odds ratio [OR], 6.00 [95% confidence interval {CI}, 2.70–13.33]) and children reporting ≥2 symptoms (n = 365; OR, 5.25 [95% CI, 2.66–10.38]) were significantly more likely to have a positive COVID-19 test than asymptomatic children (n = 155), but they were not significantly different from each other (OR, 0.88 [95% CI, .52–1.49]). Sensitivity and PPV were highest for isolated fever (33% and 57%, respectively), cough (25% and 32%), and sore throat (21% and 45%); headache had low sensitivity (8%) but higher PPV (33%). Sensitivity and PPV of isolated congestion/rhinorrhea were 8% and 9%, respectively.

Conclusions

With high Delta variant prevalence, children with isolated symptoms were as likely as those with multiple symptoms to test positive for COVID-19. Isolated fever, cough, sore throat, or headache, but not congestion/rhinorrhea, offered the highest predictive value.

COVID-19 Infection in Children and Infants: Current Status on Therapies and Vaccines

Since the beginning in December 2019, the SARS-CoV-2 outbreak appeared to affect mostly the adult population, sparing the vast majority of children who only showed mild symptoms. The purpose of this investigation is to assess the status on the mechanisms that give children and infants this variation in epidemiology compared to the adult population and its impact on therapies and vaccines that are aimed towards them. A literature review, including in vitro studies, reviews, published guidelines and clinical trials was performed. Clinical trials concerned topics that allowed a descriptive synthesis to be produced. Four underlying mechanisms were found that may play a key role in providing COVID-19 protection in babies. No guidelines are available yet for therapy due to insufficient data; support therapy remains the most used. Only two vaccines are approved by the World Health Organization to be used in children from 12 years of age, and there are currently no efficacy or safety data for children below the age of 12 years. The COVID-19 clinical frame infection is milder in children and adolescents. This section of the population can act as vectors and reservoirs and play a key role in the transmission of the infection; therefore, vaccines are paramount. More evidence is required to guide safely the vaccination campaign.

SARS-CoV-2 Saliva Mass Screening in Primary Schools: A 10-Week Sentinel Surveillance Study in Munich, Germany

Representative, actively collected surveillance data on asymptomatic SARS-CoV-2 infections in primary schoolchildren remain scarce. We evaluated the feasibility of a saliva mass screening concept and assessed infectious activity in primary schools. During a 10-week period from 3 March to 21 May 2021, schoolchildren and staff from 17 primary schools in Munich participated in the sentinel surveillance, cohort study. Participants were tested using the Salivette® system, testing was supervised by trained school staff, and samples were processed via reverse transcription quantitative polymerase chain reaction (RT-qPCR). We included 4433 participants: 3752 children (median age, 8 [range, 6-13] years; 1926 girls [51%]) and 681 staff members (median age, 41 [range, 14-71] years; 592 women [87%]). In total, 23,905 samples were processed (4640 from staff), with participants representing 8.3% of all primary schoolchildren in Munich. Only eight cases were detected: Five out of 3752 participating children (0.13%) and three out of 681 staff members (0.44%). There were no secondary cases. In conclusion, supervised Salivette® self-sampling was feasible, reliable, and safe and thus constituted an ideal method for SARS-CoV-2 mass screenings in primary schoolchildren. Our findings suggest that infectious activity among asymptomatic primary schoolchildren and staff was low. Primary schools appear to continue to play a minor role in the spread of SARS-CoV-2 despite high community incidence rates.

Dynamics in COVID-19 symptoms during different waves of the pandemic among children infected with SARS-CoV-2 in the ambulatory setting

The aim of this real-life, big data population-based study was to evaluate differences in symptomatic presentation of children infected with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) between the third and fourth waves of the pandemic in Israel, dominated by the Alpha and Delta variants, respectively. Our cohort included all children and adolescents, members of the second-largest Health Maintenance Organization in Israel that had positive real-time polymerase chain reaction (RT-PCR) test during the third and fourth waves of the pandemic (December 1, 2020, to April 30, 2021, and June 1, 2021, to October 10, 2021, respectively). A total of 32,485 and 44,130 children and adolescents in the third and fourth waves were included in the final analysis. The rate of children with symptomatic disease among patients with documented SARS-CoV-2 infection was higher in the fourth wave compared to the third wave (49.9% vs. 37.5%). The most commonly reported symptom and the only symptom that substantially differed between waves was fever, with 33% of SARS-CoV-2 infected children in the fourth wave vs. 13.6% in the third wave. Preschool children had the lowest prevalence of febrile illness compared to other age groups.

CONCLUSION

Children and adolescents infected during the fourth wave of the pandemic in Israel, a Delta-dominant period, had a significantly higher rate of symptomatic febrile illness than the Alpha-dominant period. This phenomenon occurred across all age groups.

WHAT IS KNOWN

• There are differences in COVID-19 severity among adults and children during different waves of the pandemic. • There is a paucity of data regarding symptomatic characteristics in children in large-scale cohorts aside from hospital settings.

WHAT IS NEW

• In a time period dominated by the Delta variant, there were substantially more children with symptomatic disease and febrile illness compared to a period dominated by the alpha variant. • Preschool children had the lowest rate of febrile illness among all age groups.

Symptom-Based Predictive Model of COVID-19 Disease in Children

BACKGROUND

Testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is neither always accessible nor easy to perform in children. We aimed to propose a machine learning model to assess the need for a SARS-CoV-2 test in children (<16 years old), depending on their clinical symptoms.

METHODS

Epidemiological and clinical data were obtained from the REDCap® registry. Overall, 4434 SARS-CoV-2 tests were performed in symptomatic children between 1 November 2020 and 31 March 2021, 784 were positive (17.68%). We pre-processed the data to be suitable for a machine learning (ML) algorithm, balancing the positive-negative rate and preparing subsets of data by age. We trained several models and chose those with the best performance for each subset.

RESULTS

The use of ML demonstrated an AUROC of 0.65 to predict a COVID-19 diagnosis in children. The absence of high-grade fever was the major predictor of COVID-19 in younger children, whereas loss of taste or smell was the most determinant symptom in older children.

CONCLUSIONS

Although the accuracy of the models was lower than expected, they can be used to provide a diagnosis when epidemiological data on the risk of exposure to COVID-19 is unknown.