Back

Characteristics of COVID-19 in children and potential risk factors for requiring mechanical ventilation; an analysis of 22,490 cases from the United States

Verma, R.; Ramphul, K.; Lohana, P.; Sombans, S.; Ramphul, Y.; Pekyi-Boateng, P. K. K.

2023-02-08 pediatrics
10.1101/2023.02.06.23285543 medRxiv
Show abstract

The pandemic of Coronavirus disease 2019 (COVID-19) has lasted more than two years and caused millions of deaths. While the characteristics and outcomes have been more widely studied in the adult population, we conducted an in-depth analysis via the 2020 National Inpatient Sample to understand the characteristics and predictors for the use of mechanical ventilation in patients of ages 18 and less in the United States. Twenty-two thousand four hundred ninety hospitalizations involving COVID-19-positive children were found. 52.7% (11850 cases) were females, 37.0% were Hispanics, 38.0% (8555 cases) were in the first percentile 0-25th of Median household income, and 66.9% used Medicaid. In total, 1140 cases (5.1%) needed mechanical ventilation. Among factors such as obesity (aOR 1.662, 95%CI 1.368-2.019, p<0.001), Blacks (vs. White) (aOR 1.472, 95%CI 1.23-1.761, p<0.001), private insurances (aOR 1.241, 95%CI 1.06-1.453, p=0.007) or remaining forms of payment other than Medicaid or private insurances (aOR 1.763, 95%CI 1.428-2.177, p<0.001, vs. Medicaid), ages 6 to 10 years (aOR 1.531, 95%CI 1.259-1.862, p<0.001, vs. ages 0-5) showed higher odds of needing mechanical ventilation. On the contrary, Females (aOR 0.54, 95%CI 0.472-0.617, p<0.001, vs. Males), hospitalized patients in November (aOR 0.542, 95%CI 0.399-0.736, p<0.001) and December (aOR 0.446, 95%CI 0.329-0.606, p<0.001) (vs. April), Hispanics (aOR 0.832, 95%CI 0.699-0.99, p=0.038, vs. White), ages 16-18 years (aOR 0.804, 95%CI 0.673-0.96, p=0.016, vs. 0-5years), and in the 76th-100th median household income percentile (aOR 0.783, 95%CI 0.628-0.976, p=0.03, vs. 0-25th percentile) showed reduced odds. 9.6% of patients on mechanical ventilation died.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.