key: cord-1035554-4i8mau9j
authors: Chen, Siyu; Shao, Yong
title: Risk factors for COVID-19 mortality in hospitalised children and adolescents in Brazil
date: 2021-09-15
journal: Lancet Child Adolesc Health
DOI: 10.1016/s2352-4642(21)00266-2
sha: c2c8ef46e9a61d3483bde303a5bcfcf1c16c7095
doc_id: 1035554
cord_uid: 4i8mau9j

nan

www.thelancet.com/child-adolescent Vol 5 October 2021

Surveillance Information System (SIVEP-GRIPE) dataset. Second, the authors should have men tioned that the risk factors included in SIVEP-GRIPE were selfreported (provided by the patients them selves or their families). Therefore, the analysis could be biased by the patients' knowledge regarding their medical condition. Additionally, some variables could be incorrectly coded in the electronic records; for instance, we identified in the current SIVEP-GRIPE platform at least 14 puerperal individuals younger than 10 years, which was probably a data entry error (leading to outliers).

Finally, the author did not mention any effort to test the regression model assumptions (eg, non-linearity relationship and residual analysis). The inclusion of the variables in the final multivariate model was based on a univariate parameter, which could have suppressed other important variables that should be included in the model. There was also no internal validation or crossvalidation. Therefore, we believe that our concerns should affect how the data presented by Oliveira and colleagues 1 should be interpreted.

We declare no competing interests. We read the Article by Eduardo Oliveira and colleagues 1 with great interest and believe the findings from this cohort study are important, given that they directly investigated the risk factors associated with COVID-19 in children and adolescents. In their study, patients were roughly evenly distributed among the three age groups, and risk of death was increased in infants younger than age 2 years and in adolescents aged 12-19 years, relative to children aged 2-11 years. However, the authors did not provide a rationale for the age groupings. The lower age limit of adolescence is generally defined as 10 years, 2 including by the UN and WHO. 3 Additionally, a study of COVID-19 trends between March 1, 2020, and Dec 12, 2020, in young people aged 0-24 years in the USA found that more than 81% of patients were older than 10 years. 4 Therefore, we are interested to know how a different age stratification (<2 years, 2-9 years, and 10-19 years) would affect the study findings, and we believe that comparison between these age groups could provide further insight on the COVID-19 mortality risk in adolescents.

It is important to present the median and mean ages in the three age groups, given that this information will help readers understand how mortality risk is influenced by age within the broad age bands. Having data related to symptoms, comorbidities, admission to intensive care units, and death rate by age groups will also provide a basis for understanding the disparity in death risk among age groups.

The upper-age definition of adolescence has long posed a conundrum and varies across countries. Defining adolescence as age 10-24 years has been proposed to align more closely with adolescents' biological growth and social-role transitions, 2 and some studies on COVID-19 have included patients aged 0-24 years. 4 The study by Oliveira and colleagues 1 included patients younger than 20 years, and the inclusion of patients aged 21-24 years might provide a more comprehensive understanding of COVID-19 in adolescence.

We declare no competing interests. 

The age of adolescence

Global strategy for women's, children's and adolescents' health

Sauber-Schatz EK. COVID-19 trends among persons aged 0-24 years: United States

We thank Jonas Carneiro Cruz and col leagues and Siyu Chen and Yong Shao for their interest in our study in The Lancet Child & Adolescent Health. 1 Here, we further discuss some findings and methodological aspects, specifically the question of the effect of age and comorbidities in the prognosis of paediatric COVID-19.Cruz and colleagues raised concerns about grouping different clinical disorders in a single categorical variable. This issue is interesting from both clinical and methodological points of view. We tested various models that included the variable comorbidities, as dichotomous, categorical, or continuous, and also models including the main chronic pre-existing conditions as separate covariates. In this regard, we did not observe any superiority in clinical contribution among the different