COVID-19 vaccination, infection, and symptoms among cancer survivors in the United States

Abstract

The experiences of cancer survivors with the COVID-19 pandemic in the United States during 2021 and 2022, when vaccinations became widely available, are largely undocumented. Using nationally representative survey data in 2021 and 2022, we found that compared with adults without a cancer history, cancer survivors were more likely to have at least 2 COVID-19 vaccines (2021: 66.6% vs 62.3%, P = .003; 2022: 77.0% vs 72.4%, P < .001) and as likely to have a COVID-19 infection history (2021: 14.1% vs 14.2%, P = .93; 2022: 39.9% vs 39.3%, P = .55) but, once infected, were more likely to develop moderate to severe symptoms (2021: 62.5% vs 54.2%, P = .02; 54.5% vs 61.3%; P = .13). Among cancer survivors, younger age, lower educational attainment, lack of health insurance, and more comorbidities were statistically significantly associated with lower vaccination rates (P < .001). Among infected cancer survivors, being female and younger were associated with higher likelihood of developing moderate to severe symptoms (P < .001). Our findings suggest tailored efforts to prevent and control COVID-19 infections for cancer survivors.

Cancer survivors were prioritized for vaccination during the COVID-19 pandemic on the basis of vulnerability to infection because of their weakened immune systems associated with cancer and/or its treatment (1). However, the experiences of cancer survivors in the United States during the COVID-19 pandemic in 2021 and 2022, after wide availability of vaccines, are largely undocumented. With ongoing recommendations for vaccination (2), resurgent COVID-19 infections (3), and the potential for lasting symptom burden following infection, we examined COVID-19 vaccinations, infections, and symptoms among cancer survivors to inform future efforts to protect this vulnerable population using data from the National Health Interview Survey (NHIS).

The NHIS is an annual cross-sectional household interview survey nationally representative of noninstitutionalized civilian population in the United States. The NHIS has included questions on COVID-19 infection and symptoms since September 2020 and COVID-19 vaccination questions since April 2021. The final sample adult response rate was 50.9% in 2021 with a total size of 29 482 and 47.7% in 2022 with a total size of 27 651 (4,5).

Adults aged 18 years and older with and without a cancer history were identified from the 2021-2022 NHIS, accordingly. Age-adjusted percentages and 95% confidence intervals for COVID-19 vaccinations, infection history, and symptom severity were calculated by cancer history. We further calculated the crude percentage of COVID-19–related outcomes stratifying by sex, age group, race and ethnicity, educational attainment, health insurance coverage, year since cancer diagnosis, and cancer site among cancer survivors. Data from 2021 and 2022 were analyzed separately to estimate prevalence of age-adjusted COVID-19–related outcomes because of changes in NHIS survey questions, prevalent virus strains, and vaccine dissemination over time. We combined survey years when stratifying by socioeconomic factors among cancer survivors because we found the variations by all factors were generally similar across the 2 years. All statistical analyses used SAS 9.4 software (SAS Institute, Cary, NC, USA) and weights to account for complex survey design and nonresponse. We conducted χ² tests to compare groups with 2-sided significance of .05.

In 2021, cancer survivors (n = 3428) were more likely to have 2 or more COVID-19 vaccines (66.6% vs 62.3%, P = .003) and as likely to have COVID-19 infection history (14.1% vs 14.2%, P = .93) compared with adults without a cancer history (n = 26 023), after adjusting for age (Table 1). Once infected, cancer survivors were more likely to develop moderate to severe symptoms (62.5% vs 54.2%, P = .02) compared with adults without a cancer history. A similar pattern was observed in 2022. Moreover, the 2022 data showed that once-infected cancer survivors were more likely to have long COVID symptoms than individuals without a cancer history (20.6% vs 17.3%, P = .04).

Table 1.

Age-adjusted COVID-19–related outcomes among cancer survivors and individuals without a cancer history

Year	2021					2022
Cancer status	Cancer survivors (n = 3428; weighted n = 29 441 100)		Individuals without a cancer history (n = 26 023; weighted n = 223 496 423)			Cancer survivors (n = 3218; weighted n = 22 757 396)		Individuals without a cancer history (n = 24 393; weighted n = 232 296 018)
	No.	Weighted percentage (95% CI)	No.	Weighted percentage (95% CI)	P	No.	Weighted percentage (95% CI)	No.	Weighted percentage (95% CI)	P
No. of vaccinations
0	360	25.2 (22.8 to 27.8)	4986	28.6 (27.6 to 29.6)	.01	343	19.2 (17.0 to 21.6)	4549	22.6 (21.8 to 23.5)	.01
1	166	7.9 (6.6 to 9.5)	1711	9.0 (8.5 to 9.6)	.15	73	3.7 (2.8 to 4.9)	1032	5.0 (4.6 to 5.3)	.04
≥2	2006	66.6 (63.8 to 69.3)	12270	62.3 (61.2 to 63.4)	.003	2071	77.0 (74.6 to 79.3)	16002	72.4 (71.5 to 73.3)	<.001
COVID history	336	14.1 (12.4 to 16.0)	3313	14.2 (13.6 to 14.8)	.93	949	39.9 (37.8 to 42.1)	9170	39.3 (38.5 to 40.0)	.55
Symptoms at their worst among those who have had COVID-19
No symptoms	37	7.5 (5.0 to 11.1)	301	8.9 (7.7 to 10.2)	.44	122	8.6 (6.9 to 10.6)	987	10.4 (9.5 to 11.4)	.08
Mild symptoms	92	29.7 (23.8 to 36.3)	1181	36.8 (34.8 to 38.9)	.05	338	37.3 (33.5 to 41.2)	3500	38.2 (37.0 to 39.3)	.66
Moderate to severe symptoms	206	62.5 (55.8 to 68.7)	1827	54.2 (52.0 to 56.4)	.02	487	54.5 (50.5 to 58.3)	4676	51.3 (50.0 to 52.6)	.13
Lost taste or smell among those who have had COVID-19	181	61.0 (54.8 to 66.9)	2192	66.6 (64.7 to 68.5)	.07	NA	NA	NA	NA	NA
Long-COVID symptomsa	NA	NA	NA	NA	NA	188	20.6 (17.4 to 24.2)	1606	17.3 (16.4 to 18.2)	.04

a

Long COVID symptoms last 3 months or longer. Long-term symptoms may include tiredness or fatigue, difficulty thinking, concentrating, forgetfulness or memory problems (sometimes referred to as brain fog), difficulty breathing or shortness of breath, joint or muscle pain, fast-beating or pounding heart (also known as heart palpitations), chest pain, dizziness on standing, depression, anxiety, or mood changes. The percentages presented are age adjusted. Source: The National Health Interview Survey 2021-2022. CI = confidence interval; NA = not applicable.

Characteristics of cancer survivors were generally similar in 2021 and 2022 (Supplementary Table 1, available online). Compared with 2021, cancer survivors in 2022 had higher vaccination rates (83.1% vs 77.0% ≥2 vaccinations, P < .001), more COVID-19 infections (31.5% vs 10.8%, P < .001), and less severe symptoms once infected (51.3% vs 63.3% moderate to severe symptoms, P < .05) (Table 2). Among survivors, younger age, lower educational attainment, lack of health insurance coverage and fewer comorbidities were associated with lower vaccination rates (all P < .001) (Table 2). Sex, age, cancer site, and year since cancer diagnosis were associated with COVID-19 infection history, with the highest COVID-19 infection rate among cancer survivors who were female (22.2%), aged 45-54 years (30.9%), with a blood cancer history (30.4%), and diagnosed with cancer more than 5 years ago (24.5%). Among survivors infected with COVID-19, being female was associated with higher likelihood of developing moderate to severe symptoms (P < .001), losing taste or smell (P < .05), and experiencing long COVID symptoms (P < .05). Younger age was associated with higher likelihood of developing moderate to severe symptoms (P < .001) and losing taste or smell (P < .05). Moreover, being a Hispanic individual (P < .001) and without any private insurance (P < .05) were both associated with higher likelihood of losing taste or smell. We found similar patterns among individuals without a cancer history, except that non-Hispanic Black people were more likely to be unvaccinated than other racial and ethnic groups among individuals without a cancer history, but no differences by race and ethnicity in vaccination rates were observed among cancer survivors (P_interaction = .03; Table 2; Supplementary Table 2, available online).

Table 2.

COVID-19–related outcomes among cancer survivors stratified by sociodemographic factors

Outcomes	No. of vaccinations				COVID history		Symptoms at their worst among those who have had COVID-19				Lost taste or smell among those who have had COVID-19		Long COVID
	0	1	≥2	P	Yes	P	No	Mild	Moderate to severe	P	Yes	P	Yes	P
Survey year				<.001		<.001				<.05		NA		NA
2021	16.2	6.75	77.0		10.8		9.0	27.7	63.3		54.5		NA
2022	14.0	2.92	83.1		31.5		11.3	37.4	51.3		NA		23.3
Sex				.08		<.05				<.001		<.05		<.05
Male	13.8	6.7	79.5		19.5		12.8	40.2	47.0		63.9		17.5
Female	18.0	6.8	75.2		22.2		9.3	31.3	59.4		68.7		26.9
Age, y				<.001		<.001				<.001		<.05		.13
18-44	28.1	6.1	65.7		28.8		4.3	36.4	59.4		70.5		33.9
45-54	21.8	7.6	70.6		30.9		12.1	30.9	57.0		71.1		24.4
55-64	15.4	7.2	77.4		23.3		6.5	34.0	59.5		52.8		23.1
65-74	11.9	3.4	84.7		19.0		11.2	37.6	51.3		45.2		21.7
75 and older	10.7	2.8	86.5		15.6		17.6	34.3	48.1		43.6		17.7
Race and ethnicity				.64		.06				.44		<.001		.21
Hispanic	15.9	4.5	79.6		27.0		10.8	32.4	56.7		73.2		29.0
Non-Hispanic Black	14.7	4.4	81.0		19.7		11.3	44.5	44.2		65.0		10.3
Non-Hispanic Othera	13.9	2.4	83.7		17.2		22.7	22.5	54.8		59.4		18.7
Non-Hispanic White	15.1	5.0	79.9		20.9		10.1	34.9	54.9		64.6		23.6
Education				<.001		.05				.24		.26		.12
Less than high school graduate	19.7	5.5	74.9		20.9		12.6	31.5	56.0		68.2		23.5
High school graduate	18.0	4.7	77.2		23.3		10.6	33.9	55.4		64.5		27.2
College graduate or more	7.8	4.2	88.0		19.7		9.0	39.2	51.8		66.1		19.3
Insurance status among adults aged 18-64 years				<.001		.20				.31		<.05		.21
Any private	15.1	7.1	77.8		27.8		8.2	34.6	57.2		66.8		23.7
Public only	29.2	6.2	64.7		23.7		4.2	32.2	63.6		75.1		35.4
Uninsured	45.1	10.2	44.7		22.4		11.4	23.6	65.0		73.1		36.7
Insurance status among adults aged 65 years and older				.06		.77				.16		.34		.74
Medicare and private or private only	8.9	3.1	88.0		17.3		11.0	36.4	52.6		48.7		21.9
Medicare and public (Medicaid)	18.8	2.6	78.7		20.1		24.2	25.9	49.8		52.5		26.8
Medicare Advantage	12.2	3.1	84.6		16.8		13.7	41.3	45.1		46.3		17.5
Medicare only and other	12.3	3.4	84.3		17.3		17.8	30.5	51.7		57.8		18.7
No. of comorbiditiesb				<.001		.38				.23		.08		.23
0	19.2	7.1	73.8		22.1		13.0	38.8	48.2		61.4		17.2
1	14.7	5.3	80.0		22.0		8.4	36.6	55.0		64.8		24.1
≥2	14.0	3.9	82.1		20.4		11.0	32.9	56.1		48.4		25
Year since diagnosis				.61		<.001				.78		.11		.69
≤2	14.9	5.8	79.4		16.8		13.2	30.1	56.7		35.2		19.4
3-5	15.7	5.6	78.6		17.7		8.2	38.3	65.8		65.8		18.3
>5	14.9	4.2	80.9		24.5		10.4	35.5	54.1		53.9		24.4
Cancer sites				<.05		<.05				.20		.45		.07
Female breast	13.7	4.0	82.3		17.9		11.7	29.0	59.3		44.7		19.4
Colorectal	15.6	4.4	80.0		19.4		7.4	34.4	58.1		41.8		27.2
Prostate	11.5	3.7	84.8		19.5		13.9	42.6	43.4		56.9		12.5
Blood cancerc	11.5	4.7	83.8		30.4		8.29	30.6	60.6		59.0		28.8
Other	16.5	5.3	78.1		21.7		10.3	35.5	54.2		41.1		25.3

a

Non-Hispanic other includes non-Hispanic Asian, American Indian and Alaska Native, and other single and multiple races. The percentages presented are crude. Source: The National Health Interview Survey 2021-2022. NA = not applicable.

b

Comorbidities include chronic obstructive pulmonary disease, emphysema, chronic bronchitis, arthritis, cardiovascular conditions (coronary heart disease, angina, heart attack, stroke), asthma, high cholesterol, and hypertension.

c

Blood cancer includes leukemia, lymphoma, and myeloma.

In this large national study, we found that cancer survivors had better COVID-19 vaccine completion than adults without a cancer history. Nonetheless, nearly one-fifth of cancer survivors were not vaccinated in 2022, leaving them at risk of COVID-19 infection and developing severe symptoms during any resurgences. Among survivors, individuals who were female, younger age, Hispanic, and without any private insurance had worse experiences with COVID-19 infection compared with their counterparts. Short-term cancer survivors experienced a lower COVID-19 infection rate than long-term survivors, possibly because of higher prevention and protection measures applied to them given their recent cancer diagnosis and treatment. Although not statistically significant in our results for cancer survivors, other studies have found that in the general population, racial minorities had a higher risk of COVID-19 positivity compared with non-Hispanic White individuals (6). Ongoing efforts to narrow the gap of COVID-19 outcomes between minority groups and the general population are warranted.

The rollout of COVID-19 vaccines in the United States started with health-care personnel and long-term care facility residents; followed by older adults and high-risk populations, including those with underlying medical conditions; then younger adults; and last, children and adolescents, which may partially explain the better vaccination completion among older adults (7-10). We also found that cancer survivors with comorbidities had a higher vaccination completion rate than those without. In contrast, a previous study of patients who were under cancer treatment and had severe COVID-19 symptoms found that having any comorbidity was associated with lower vaccination rates, suggesting that vaccination promotion among individuals with cancer and other health conditions may need to consider their specific priorities in different stages of the cancer continuum (11). Our findings that non-Hispanic Black people were more likely to be unvaccinated than other racial and ethnic groups among individuals without a cancer history, but there were no differences by race and ethnicity in vaccination rates among cancer survivors, suggest that vaccine delivery to all groups was more effective among cancer survivors than the general population. Moreover, we identified the lowest vaccination rate among younger adults aged 16-44 years, uninsured adults, and adults without a high school degree. Our findings highlight the importance of targeting vulnerable populations, including individuals without health insurance and lower education attainment, in future vaccination promotion.

In addition, despite a better vaccination completion compared with individuals without a cancer history, cancer survivors were more likely to experience moderate to severe COVID-19 symptoms once infected, and nearly 1 in 5 experienced long-term COVID symptoms. This may be related to the weakened immune system and high comorbidity burden associated with cancer history and treatments (12). A previous study followed cancer patients diagnosed with COVID-19 in 2020 at a single institution—MD Anderson Cancer Center—until May 2021 and reported that 60% had long COVID-19 symptoms, higher than the prevalence of long COVID-19 symptoms in our sample of cancer survivors in 2022 (13). The discrepancies could be because our samples are longer-term cancer survivors and less likely to be receiving active treatment; vaccination was not available at COVID-19 infection in the previous study, whereas in our study cancer survivors could be infected with COVID-19 after their vaccination, and the virus strains were different. Future research assessing cancer survivors’ COVID-19 symptoms and any lasting health effects are warranted.

This study has limitations. The dates of COVID-19 infection and vaccination were not available in the NHIS; thus, we were unable to examine the relationship of time since infection and vaccination with COVID-19 symptom severity. Moreover, estimates for cancer survivors who lost taste or smell or experienced long COVID symptoms should be interpreted with caution because they were based on small numbers.

With the continuing high infectious rate and seasonal resurgences of COVID-19 infections and ongoing recommendations for vaccination, especially for vulnerable populations, monitoring the impact of COVID-19 infection and the effectiveness of prevention and control strategies continue to be a public health priority (2,3,14,15). Our findings suggest the need for tailored efforts to prevent and control COVID-19 infection for cancer survivors.

Data availability

This study used publicly available deidentified survey data downloadable from https://www.cdc.gov/nchs/nhis/2020nhis.htm.

Author contributions

Nuo Nova Yang, MSPH (Conceptualization; Formal analysis; Investigation; Methodology; Project administration; Writing—original draft; Writing—review & editing), Jingxuan Zhao, MPH (Conceptualization; Investigation; Methodology; Writing—review & editing), Zhiyuan Zheng, PhD (Conceptualization; Investigation; Methodology; Writing—review & editing), K. Robin Yabroff, PhD (Conceptualization; Investigation; Methodology; Writing—review & editing), and Xuesong Han, PhD (Conceptualization; Investigation; Methodology; Project administration; Supervision; Writing—review & editing).

Funding

None.

Conflicts of interest

None.

Acknowledgements

K. Robin Yabroff serves on the Flatiron Health Equity Advisory Board and has received honorarium from the National Comprehensive Cancer Network for workgroup participation. Dr Yabroff, who is a JNCI deputy editor and co-author on this paper, was not involved in the editorial review or decision to publish the manuscript. Part of the findings were presented as online publication at the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting at Chicago, Illinois, June 2-6, 2023.