|
?:abstract
|
-
OBJECTIVE To assess associations between body mass index (BMI) and severe COVID-19 outcomes: hospitalization, intensive care unit (ICU) admission, and mortality. A secondary aim was to investigate whether associations varied by age. METHODS The cohort comprised COVID-19 positive Veterans Health Administration (VHA) patients (N=9,347). For each outcome, we fit piecewise logistic regression models with restricted cubic splines (knots at BMIs of 23, 30, 39), adjusting for: age, sex, comorbidities, VHA nursing home residence, and race/ethnicity. Supplemental analyses included age-by-BMI interaction terms (alpha=0.05). RESULTS We found evidence of a nonlinear J-curve association between BMI and likelihood of hospitalization and mortality. BMI was associated with increased odds for hospitalization, ICU admission, and mortality among patients with BMIs 30-39, but decreased odds of hospitalization and mortality for patients with BMIs 23-30. Patients under age 75 with BMIs 30-39 had increased odds for mortality with increasing BMI. CONCLUSION Odds for severe outcomes with COVID-19 infection increased with increasing BMI for people with, but not without obesity. This non-linear relationship should be tested in future research. COVID-19 public health messages in VHA, and broadly, should incorporate information about risks associated with all classes of obesity, particularly for those under age 75.
|
![[This page as RDF]](https://research.tib.eu/covid-19/static/rdf-icon.gif){kind=icon}