Covid-19 is a leading cause of death in children and young people ages 0-19 years in the United States

Introduction

In the one year period April 1, 2021 - March 31, 2022 there were more than 381,000 deaths from Covid-19 in the US¹ (a rate of 115 / 100,000 population). In children and young people (CYP) aged 0-19 years there were 764 deaths reported in this time period (0.9 / 100,000 population). The overall risk of death from Covid-19 in CYP is thus significantly reduced compared to other age groups in the US. Given deaths in US CYP from all causes are rare (49.4 per 100,000 in 2019 for 0-19 year olds; 25.0 per 100,000 for 1-19 year olds), the mortality burden of Covid-19 is best understood by comparing it to other significant causes of CYP mortality from a recent pre-Covid-19 period. For this purpose, we use the US Centers for Disease Control and Prevention (CDC) WONDER database of mortality statistics, grouped according to underlying cause of death. The groups of causes are defined by the US National Center for Health Statistics (NCHS)’s list of 113 “rankable causes of death”, originally drawn up in 1951 to allow comparisons for public health purposes, and regularly updated². Covid-19 (ICD code U07.1) was added to this list in October 2020³. We consider the leading causes of death in 2019, that is, the ordered list of groups of causes that occur most frequently among NCHS’s rankable causes of death². Leading causes of death are one among various ways of understanding mortality and burden of disease.

Methods

We obtained leading causes of death by age group in 2019 from CDC WONDER⁴, which has been comprehensively described elsewhere², and compared these to Covid-19 mortality in our study period, April 1, 2021-March 31, 2022, obtained from CDC WONDER Provisional Mortality Statistics¹ (the most recent 1 year period in which data is close to complete⁵.) To determine pediatric death rates by cause of death, we used 2021 population size estimates by single year of age from the US Census Bureau⁶. We note that NCHS’s list of rankable causes of death group together many individual ICD codes², while Covid-19 deaths count only those for which a single ICD code (U07.1) was listed as the underlying cause of death; we are therefore comparing the direct effect of a single pathogen (SARS-CoV-2) to groupings of multiple underlying causes of death. We also note that the CDC WONDER Provisional Mortality Statistics could still be revised in the future. They also do not include deaths where Covid-19 was considered to have been a contributing cause of death,¹ and thus differ from other datasets reporting Covid-19 deaths, e.g. NCHS has a separate source reporting counts of deaths involving Covid-19⁷ (a previous version of this preprint incorrectly used these data⁸.)

Results

Like many diseases, pediatric Covid-19 death rates follow a characteristic U-shaped curve in the US (Figure 1). In the study period, Covid-19 death rates in infants aged <1 year were 3.5 deaths per 100,000 population. The Covid-19 death rate was below 1 per 100,000 in children aged 1-14, increasing to 1.8 per 100,000 in 15-19-year olds. Considering the marked, age-specific U-shape in pediatric Covid-19 death rates, we next compared the burden of Covid-19 to other causes of death in five age brackets: <1 year, 1-4 years, 5-9, 10-14, and 15-19 years.

• Download figure
• Open in new tab

Figure 1.

Covid-19 death rates in the US for children and young people, where Covid-19 is listed as the underlying cause of death (ICD code U07.1) on the death certificate¹. Rates are calculated as Covid-19 deaths for the period April 1, 2021-March 31, 2022 per 100,000 population (2021 population estimates).

In 2019, leading causes of CYP deaths comprised perinatal conditions (12.7 / 100,000), accidents (9.1 / 100,000), congenital malformations or deformations (6.5 / 100,000), assault (3.4 / 100,000), suicide (3.4 / 100,000), malignant neoplasms (2.1 / 100,000), diseases of the heart (1.1 / 100,000), and influenza and pneumonia (0.6 / 100,000).

For comparison, in the study period, there were 764 CYP deaths reported whose underlying cause was Covid-19 (0.9 / 100,000), meaning it ranks as the #8 cause of death (Table 2) and accounts for 2.3% of the top 10 causes. Rankings disaggregated by age group are shown in Table 1,demonstrating that Covid-19 ranks consistently in the top 10 causes of death in CYP in all age groups. Ranks are: #7 (<1 year old), #7 (1-4 year olds), #6 (5-9 year olds), #6 (10-14 year olds), and #5 (15-19 year olds). Covid-19 accounted for 0.7% (<1 year old), 2.7% (1-4 year olds), 4.4% (5-9 year olds), 3.9% (10-14 year olds), and 4.2% (15-19 year olds) of the top 10 leading causes of death by age group.

Excluding causes of death unrelated to disease (accidents, assault and suicide), Covid-19 ranked as the #5 cause of death in US CYP aged 0-19 (Table 2). Considering infectious and respiratory diseases only, Covid-19 ranked as the #1 cause of death in US CYP aged 0-19 (Table 3).

Conclusion

Early in the Covid-19 pandemic, comparisons of Covid-19 disease severity between age groups were a vital tool for appropriately allocating limited resources and prioritizing vaccination campaigns. However, long-term public health planning and management needs to be informed by the leading causes of deaths within each age group, a practice in the US dating back at least 7 decades, beginning with the first publication of a leading cause of death ranking in 1952². The most recent comprehensive data pre-pandemic on leading causes of death covers 2019². Comparing to this period, we have shown that Covid-19 was a leading cause of death in CYP aged 0-19 years in the US for the period April 1, 2021-March 31, 2022, ranking #8 among all causes of deaths (comprising 2.3% of the top 10 causes), #5 in disease related causes of deaths (excluding accidents, assault and suicide), and #1 in deaths caused by infectious and respiratory diseases.

These findings need to be considered in the context of several limitations. While analyses of excess deaths have suggested under-reporting bias in Covid-19 deaths⁹, it is important to note that the specific criteria for determining Covid-19 deaths is heterogeneous across the US and may be subject to both under and over reporting depending on the context, in addition to delays in reporting. Some of the cause of death categories vary in specificity, for example, the category of deaths from “influenza and pneumonia” combines a variety of causes, to which SARS-CoV-2 could be a contributing factor^10,11. Relatedly, we only consider Covid-19 as an underlying (and not contributing) cause of death, but Covid-19 amplifies the severe impacts of other diseases and mortality hazards from infection are increased with accompanying comorbidities. It is thus likely that the absolute mortality burden of Covid-19 has been larger in CYP aged 0-19 than our analyses suggest, but the exact magnitude is unknown.

In summary, we find that Covid-19 is now a leading cause of death for CYP aged 0-19 in the US. A future variant or subvariant capable of displacing current Omicron subvariants will, by definition, have a growth advantage over Omicron, and there is no guarantee that its intrinsic severity will be reduced¹². In the context of sustained transmission and circulation of SARS-CoV-2 in the US, the non-trivial risk posed by Covid-19 to CYP warrants utilization of a wide and robust array of tools to limit the extent of infection and severe disease in this age-group, through a combination of safe and efficacious vaccination against the disease¹³, appropriate pharmaceutical treatments, mitigations such as ventilation, and other non-pharmaceutical interventions.

Ethics

According to HRA decision tools (http://www.hra-decisiontools.org.uk/research/), our study is considered Research, and according to the NHS REC review tool (http://www.hra-decisiontools.org.uk/ethics/), we do not need NHS Research Ethics Committee (REC) review, as we only used (1) publicly available, (2) anonymized, and (3) aggregated data outside of clinical settings.

Funding

ES and SF acknowledge funding from the EPSRC (EP/V002910/2), OR from the MRC (MR/V038109/1). SM and SB acknowledge funding from the Novo Nordisk Young Investigator Award (NNF20OC0059309). CW acknowledges funding from the MRC DTP that supports his PhD studies (Award Ref 1975152). This research was funded in whole or in part by EPSRC (EP/V002910/2). For the purpose of Open Access, the author has applied a CC BY public copyright license to any Author Accepted Manuscript (AAM) version arising from this submission.

Code Availability

All data and R scripts for replicating the results in this manuscript are available from: https://github.com/flaxter/covid19pediatric/