2.1

Population

A detailed description of the National Health and Nutrition Examination Survey (NHANES) is provided on the NHANES website . Briefly, NHANES is a series of cross-sectional surveys conducted using a stratified probability sampling design to enroll a representative sample of U.S. civilian, non-institutionalized individuals with deliberate oversampling of certain subgroups to provide more precise estimates in these groups. NHANES has been conducted since 1959 with continuous data collection and data release every two years since 1999. The data collected includes a comprehensive demographic and health survey, physical examination and laboratory studies . The present study used data from the 2015–2016 and 2017–2018 NHANES surveys. NHANES de-identified datasets are publicly available from the CDC and the protocols have been reviewed and approved by the National Center for Health Statistics Ethics Review Board. Local institutional review board review is not required for use of NHANES data.

2.2

Demographics

In NHANES, a standardized questionnaire was used to collect information on age, gender, race/ethnicity, family income and educational attainment. For this study, educational attainment was categorized into three groups: less than a high school education, a high school education and more than a high school education. Poverty income ratio (measure of family income to Department of Health and Human Services (HHS) defined poverty level) was selected as the income variable for this study and categorized into two groups: less than 1.5 and equal to or greater than 1.5.

2.3

Variable definitions

The ten medical conditions identified by the CDC include current cancer, chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), immunocompromised state from solid organ transplant, obesity (BMI ≥30 kg/m ² ), serious heart condition such as heart failure, coronary artery disease or cardiomyopathies, sickle cell disease, current or past smoking, pregnancy and type 2 diabetes . Only conditions for which there was stronger evidence per the CDC of increased risk of severe COVID-19 infection were included in this analysis. The current analysis used age ≥65 as an additional risk criteria based on prior CDC criteria which included it as an increased risk condition, and sensitivity analyses were performed removing this risk factor. Pregnancy was not coded in NHANES. Sickle cell and solid organ transplant were also not coded in NHANES but at a prevalence of 0.03% and 0.25% , respectively, these represent a very small proportion of the US population.

Serious heart conditions was defined as an affirmative response to “ever told you had coronary heart disease” or “ever told you had heart attack” or ever told you had congestive heart failure”. Cerebrovascular disease was not including in the serious heart conditions definition as cerebrovascular disease was specifically excluded as an increased risk condition by the CDC, and was designated one of the 11 conditions where an individual “might be at increased risk” for severe COVID-19 infection. Current cancer was ascertained based on an affirmative response to the following question “ever told you had cancer or malignancy” and a response to the follow-up question “age cancer diagnosed” that confirmed the diagnosis had been made within two years of the survey. Sensitivity analyses were performed assessing 1 year and 5 years intervals since cancer diagnosis. CKD was defined by urine/albumin creatinine ratio ≥ 30 mg/g or estimated glomerular filtration rate 15–59 mL/min/1.73m ² or ESRD, defined by an affirmative response to the question “ever told you had weak/failing kidneys” and “received dialysis in past 12 months” . COPD was defined by an affirmative response to the question “ever told you had emphysema” or “ever told you had COPD” or “ever told you have chronic bronchitis”. Obesity was defined by a calculated body mass index ≥ 30 kg/m ² that was determined from direct measurement of height and weight by study personnel. Smoking was defined as an affirmative response to “smoked ≥ 100 cigarettes in life”. Type 2 diabetes mellitus was defined by a1c ≥ 6.5% or fasting plasma glucose ≥ 126 mg/dL or an affirmative response to “doctor told you have diabetes” .

2.4

Statistical analysis

The NHANES analytic guidelines recommend combining survey cycles to obtain more precise subgroup estimates . We determined the prevalence of the nine risk factors including age ≥ 65 coded in NHANES in the unweighted 2015–2016 and 2017–2018 overall sample of adults aged ≥18 ( N = 10,581) and extrapolated these results to a population of 233.8 million U.S. adults ≥ 18. We then examined the prevalence of these risk factors in subgroups defined by gender, age, race/ethnicity, educational attainment, income and history of heart disease. Using NHANES-supplied sample weights, these prevalence estimates were weighted to obtain nationally representative prevalence estimates. NHANES sample weights take into account the complex NHANES survey design including differential probability of sampling, non-response, and differences between the final sample distribution and target population distribution . Sample weights for the combined 2015–2018 period were obtained by dividing the 2-year weight from the 2015–2016 and 2017–2018 periods by two (the number of two-year cycles) . The weighted prevalence estimates were then used to determine the estimated average population numbers at risk during the 2015–2018 time period per methods provided by the National Center for Health Statistics .

We performed sensitivity analyses excluding those aged ≥65, evaluating the proportion with ≥2 or ≥3 risk factors, and excluding obesity and age ≥65 as increased-risk conditions. Comparisons between the proportions with any increased-risk condition among subgroups were performed using a first-order Rao-Scott chi-square test that accommodates the NHANES sampling methodology.