Prevalence of Underlying Medical Conditions Associated With Severe COVID-19 Illness in Adult Cancer Survivors in the United States.

Cancer, and other underlying medical conditions including chronic obstructive pulmonary disease, heart diseases, diabetes, chronic kidney disease, and obesity, are associated with increased risk of severe coronavirus disease 2019 (COVID-19) illness. We identified 6411 cancer survivors and 77 748 adults without a cancer history from the 2016-2018 National Health Interview Survey and examined the prevalence and sociodemographic factors associated with these conditions in the United States. Most survivors reported having 1 or more of the conditions (56.4%, 95% confidence interval [CI] = 54.8% to 57.9%, vs 41.6%, 95% CI = 40.9% to 42.2%, in adults without a cancer history), and nearly one-quarter (22.9%, 95% CI = 21.6% to 24.3%) reported 2 or more, representing 8.7 million and 3.5 million cancer survivors, respectively. These conditions were more prevalent in survivors of kidney, liver, and uterine cancers as well as Black survivors and those with low socioeconomic status and public insurance. Findings highlight the need to protect survivors against COVID-19 transmission in health-care facilities and to prioritize cancer patients, survivors, caregivers, and their health-care providers in vaccine allocation.

Cancer, and other underlying medical conditions including chronic obstructive pulmonary disease (COPD), heart diseases, diabetes, chronic kidney disease (CKD), and obesity, are associated with increased risk of severe coronavirus disease 2019 (COVID-19)–associated illness (ie, illness requiring hospitalization, intensive care unit admission, mechanical ventilation) or death (1). A recent study reported that about 40% of adults in the United States have any of these conditions (2). Cancer survivors are prone to developing other conditions because of shared risk factors and long-term effects from treatment (3), thus they may be especially vulnerable to severe COVID-19–associated illness. This study investigates the prevalence and factors associated with these underlying medical conditions among cancer survivors in the United States to inform efforts to prevent and control severe COVID-19–associated illness, including risk-stratified vaccine distribution.

We identified 6411 adult cancer survivors and 77 748 adults without a cancer history from the 2016-2018 National Health Interview Survey (NHIS), a national cross-sectional survey of the civilian, noninstitutionalized population (4). The survey collects information on cancer, COPD, heart diseases, diabetes, CKD, and obesity (body mass index ≥ 30 kg/m2), the 6 underlying medical conditions with the strongest evidence of association with severe COVID-19 illness (1). NHIS data are de-identified and publicly available; institutional review board approval was not required for this study.

Respondents with only nonmelanoma skin cancer (n = 2828) or only reporting cancer diagnosis before age 18 years or not reporting age at diagnosis (n = 349) were excluded. Prevalence of underlying medical conditions among cancer survivors and adults without a history of cancer as well as national population estimates of underlying medical conditions among cancer survivors was calculated using NHIS sample weights. Prevalence differences by cancer history were compared using χ2 test for each medical condition. Age-adjusted prevalence and 95% confidence intervals (CI) of any underlying medical conditions in subpopulations defined by sociodemographic characteristics (ie, age, sex, race and ethnicity, educational attainment, insurance coverage, poverty status, and geographic region) were calculated with logistic regression for cancer survivors and adults without a cancer history. The associations of sociodemographic characteristics with presence of any underlying medical conditions were compared between cancer survivors and adults without a cancer history using the Wald F test for the interaction term of cancer history by the characteristic variable in the logistic regression models. All analyses accounted for complex NHIS design and survey of nonrespondents using SAS, version 9.4 (4). Statistical significance was determined using 2-sided tests with a 5% threshold.

Most cancer survivors reported having at least 1 of the underlying medical conditions (56.4%, 95% CI = 54.8% to 57.9%) and nearly one-quarter (22.9%, 95% CI = 21.6% to 24.3%) reported more than 1 condition, representing 8.7 million and 3.5 million cancer survivors in 2018, respectively. In contrast, the prevalence of any and 2 or more conditions among adults without a cancer history was 41.6% (95% CI = 40.9% to 42.2%) and 10.8% (93% CI = 10.5% to 11.2%), respectively (Figure 1). The most common condition among cancer survivors was obesity (30.8%), followed by heart diseases (25.1%), diabetes (17.0%), COPD (9.2%), and CKD (5.6%). Compared with adults without a cancer history, cancer survivors had statistically significantly higher prevalence of all selected medical conditions (P < .001), except obesity (P = .13) (Figure 1).

Figure 1.

Nationwide estimates of selected underlying medical conditions associated with increased risk for severe coronavirus disease 2019–associated illness among adults with (n = 6411) and without (n = 77 748) a cancer history, National Health Interview Surveys 2016-2018. χ2 test was used to compare the prevalence of each medical condition between cancer survivors and adults without a history of cancer. Two-sided P = <.001 for comparison for all medical conditions except obesity (P = .13). CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; Obesity = body mass index ≥30 kg/m2.

As in adults without cancer history, older age was statistically significantly associated with higher prevalence of medical conditions among cancer survivors. However, even in the youngest age group 18-44 years, nearly half of cancer survivors (47.6%) had at least 1 additional condition associated with severe COVID-19 illness (Table 1). In addition to increasing prevalence with age, medical conditions were more prevalent among male survivors (59.9%), those with less than high school completion (68.0%), non-Hispanic Black (67.2%), low income (71.7%), living in the South (59.2%), and 18- to 64-year-olds with public insurance only (71.2%). The prevalence of any underlying conditions was the highest among survivors of kidney (73.9%), liver (71.8%), and uterine (71.5%) cancers.

Table 1.

Nationwide estimates of prevalence of any of the 5 selected underlying medical conditions associated with increased risk of severe COVID-19-associated illness among adults with and without cancer history—United States, 2016-2018

Characteristic	Cancer survivors (n = 6411)		Adults without a cancer history (n = 77 748)
	Sample No. (weighted %)	Age-adjusted prevalence of any medical conditions (95% CI), %c	Sample No. (weighted %)	Age-adjusted prevalence of any medical conditions (95% CI), %c
Age, y
18-44	495 (9.6)	47.6 (42.1 to 53.1)	32 168 (48.7)	32.6 (31.8 to 33.4)
45-64	2012 (36.1)	51.6 (48.9 to 54.3)	26 315 (33.6)	46.8 (45.9 to 47.8)
65-84	3337 (47.2)	61.2 (59.0 to 63.4)	16 996 (15.9)	56.1 (55.2 to 57.1)
85 or older	567 (7.1)	60.5 (55.4 to 65.6)	2269 (1.8)	57.4 (54.6 to 60.0)
Sex
Male	2452 (40.3)	59.9 (57.4 to 62.3)	35 672 (48.7)	42.5 (41.7 to 43.4)
Female	3959 (59.7)	54.0 (51.9 to 56.1)	42 076 (51.3)	40.7 (39.9 to 41.4)
Education
Less than high school	790 (11.7)	68.0 (63.8 to 71.9)	8849 (11.8)	48.4 (47.0 to 49.8)
Graduated high school or equivalent	1648 (25.6)	60.0 (56.9 to 63.1)	18 911 (24.4)	45.6 (44.5 to 46.6)
More than high school	3973 (62.8)	52.7 (50.7 to 54.7)	49 988 (63.8)	38.8 (38.1 to 39.5)
Race
Non-Hispanic White	5263 (80.1)	55.2 (53.5 to 57.0)	53 909 (63.6)	41.0 (40.3 to 41.7)
Hispanic	377 (7.2)	61.4 (55.2 to 67.1)	9671 (16.5)	43.8 (42.4 to 45.2)
Non-Hispanic Black	535 (8.5)	67.2 (61.9 to 72.1)	8903 (12.4)	49.3 (47.7 to 50.9)
Non-Hispanic otherb	236 (4.2)	47.1 (39.4 to 54.9)	5265 (7.4)	28.7 (26.5 to 30.9)
Insurance
Any private (<65 y)	1663 (31.5)	48.1 (42.1 to 54.1)	40 510 (57.3)	38.6 (37.7 to 39.5)
Public coverage (<65 y)	663 (11.1)	71.2 (65.2 to 76.6)	10 871 (14.7)	51.6 (50.4 to 52.9)
Uninsured (<65 y )	181 (3.1)	60.5 (55.8 to 64.9)	7102 (10.3)	42.6 (41.3 to 43.9)
Medicare with any private (≥65 y )	2098 (29.2)	55.5 (50.2 to 60.6)	9332 (8.4)	41.6 (39.7 to 43.5)
Medicare only (≥65 y)	1806 (25.0)	60.5 (55.8 to 64.9)	9933 (9.3)	42.6 (41.3 to 43.9)
Poverty status (federal poverty level)
Less than 100%	643 (8.2)	71.7 (67.1 to 75.8)	10 151 (11.0)	49.1 (47.7 to 50.4)
100%-199%	1127 (15.2)	67.8 (64.1 to 71.3)	13 359 (16.3)	47.9 (46.8 to 49.1)
200%-399%	1836 (28.0)	59.7 (56.7 to 62.6)	20 501 (26.5)	44.4 (43.5 to 45.4)
400% or above	2318 (40.6)	47.7 (45.1 to 50.3)	29 088 (40.1)	36.2 (35.4 to 37.1)
Unknown	487 (7.9)	50.7 (45.0 to 56.3)	4649 (6.1)	34.8 (33.0 to 36.7)
Region
Northeast	1166 (19.4)	53.1 (49.7 to 56.5)	12 738 (17.9)	37.7 (36.3 to 39.2)
Midwest	1541 (23.7)	58.3 (55.0 to 61.5)	17 896 (21.9)	44.5 (43.3 to 45.6)
South	2241 (35.3)	59.2 (56.4 to 62.0)	28 054 (36.3)	44.0 (42.8 to 45.1)
West	1463 (21.6)	52.5 (49.2 to 55.8)	19 060 (23.8)	38.2 (36.9 to 39.5)
Cancer type
Breast	1669 (25.3)	47.4 (44.2 to 50.6)	—	—
Prostate	1001 (15.6)	56.4 (52.6 to 60.1)	—	—
Melanoma	629 (9.7)	54.2 (49.1 to 59.2)	—	—
Colon	493 (7.5)	65.3 (59.5 to 70.7)	—	—
Cervix	483 (7.6)	65.0 (59.6 to 69.9)	—	—
Uterus	354 (5.4)	71.5 (64.0 to 78.1)	—	—
Lung	288 (4.1)	65.3 (58.4 to 71.6)	—	—
Lymphoma	255 (4.2)	56.5 (48.9 to 63.8)	—	—
Thyroid	250 (4.2)	62.2 (54.2 to 69.6)	—	—
Bladder	213 (3.1)	62.4 (54.1 to 70.0)	—	—
Kidney	183 (2.7)	73.9 (65.5 to 80.9)	—	—
Ovary	171 (2.6)	60.4 (50.2 to 69.8)	—	—
Leukemia	104 (1.8)	58.7 (47.7 to 69.0)	—	—
Pharynx	75 (1.2)	58.6 (44.2 to 71.7)	—	—
Bone	70 (1.0)	60.8 (46.3 to 73.6)	—	—
Liver	60 (0.9)	71.8 (55.0 to 84.1)	—	—
Brain	55 (0.9)	55.4 (39.7 to 70.2)	—	—
Other	767 (12.3)	56.7 (52.3 to 61.0)	—	—

aData are from the National Health Interview Survey 2016-2018. Five selected underlying medical conditions include chronic obstructive pulmonary disease, heart diseases, diabetes, chronic kidney disease, and obesity. Logistic regression was used to calculate the age-adjusted prevalence of any medical conditions. CI = confidence interval; COVID-19 = coronavirus disease 2019; — = the prevalence was adjusted by participants' age with logistic regression.

bNon-Hispanic other includes Asian, non-Hispanic American Indian and Alaska Native only, non-Hispanic Native Hawaiian and Pacific Islander only, and non-Hispanic multiple race.

cWald F test for the interaction term of cancer history by the characteristic variable in the logistic models compares the association of the characteristic with presence of any medical conditions between cancer survivors and adults without a cancer history. Two-sided P = <.001 for all characteristics except age (P = .003).

Compared with patients without a history of cancer, cancer survivors had statistically significantly higher prevalence of any additional condition in every subpopulation defined by sociodemographic characteristic except the similar prevalence in the oldest aged group of 85 years and older. Moreover, the associations of each characteristic with presence of any underlying medical conditions were statistically significantly different between cancer survivors and adults without a cancer history.

More than half (56.4%) of cancer survivors in the United States, approximately 8.7 million in 2018, reported having additional underlying medical conditions associated with severe COVID-19 illness. Prevalence of these conditions among survivors is nearly 40% higher than that in the general population (2). These conditions were more prevalent in survivors of kidney, liver, and uterine cancers as well as Black survivors and those with low socioeconomic status and public insurance, all characteristics historically associated with worse health outcomes. Our results suggest these survivors warrant high priority in COVID-19 transmission prevention and vaccine allocation strategies to minimize the risk of widening of health disparities (5,6).

Limitations of this study include the potential recall error with self-reported data and lack of data on 2 less common conditions identified as severe COVID-related medical conditions: sickle cell disease and history of solid organ transplant (1). NHIS excludes populations living in long-term care facilities or those who are incarcerated, who may be sicker or more vulnerable to severe COVID-19 illness. Moreover, NHIS respondents tend to be heathier than the general US population, so the medical condition burden may be underestimated in our study (7). Although we accounted the complex survey design and survey nonresponse in our analyses, the weighting may not fully account for any unmeasured factors associated with nonresponse. We did not evaluate the burden of underlying medical conditions of severe COVID-19 illness in survivors of other chronic conditions (eg, heart attack, stroke); additional studies are warranted to better inform prevention and vaccine allocation strategies for these populations.

In conclusion, underlying medical conditions related to severe COVID-19 illness are common among cancer survivors, highlighting the need to protect this vulnerable population against transmission in health-care facilities. Now that safe and effective COVID-19 vaccines are available, cancer patients, survivors, caregivers, and their health-care providers should be prioritized in vaccine allocation.

Funding

Not applicable.

Notes

Role of the funder: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Disclosures: Xuesong Han has received grant and research support from AstraZeneca for research outside of the current study.

Author contributions: Conceptualization, investigation, methodology, supervision: CJ, XH, KRY. Data curation, formal analysis, project administration, visualization: CJ, XH. Writing-original draft: All authors. Writing-review & editing: All authors.

Data Availability

The data underlying this article are available in National Center for Health Statistics, National Health Interview Survey, at https://www.cdc.gov/nchs/nhis/index.htm.