Preventive Behaviors and Mental Health-Related Symptoms Among Immunocompromised Adults During the COVID-19 Pandemic: An Analysis of the COVID Impact Survey

Abstract

The COVID-19 pandemic has disrupted the continuity of care of U.S. adults living with chronic diseases, including immunocompromised adults. Disruption in care may be a barrier to identifying COVID-19 associated sequelae, such as mental health symptoms, among the immunocompromised. Our objectives were to evaluate COVID-19-related preventive behaviors, with a focus on canceling doctor's appointments as a proxy for continuity of care, and to compare COVID-19-related mental health symptoms among the immunocompromised with the general population. We used nationally-representative data of 10,760 U.S. adults from the publicly-available COVID-19 Household Impact Survey. We defined immunocompromised as adults with a self-reported diagnosis of “a compromised immune system” (n = 854, 7.6%). We adherence to self-reported COVID-19 preventive behaviors among immunocompromised adults to others using χ²-tests. We focused on continuity of care and estimated determinants of canceling doctor's appointments among the immunocompromised using multivariable Poisson regression to estimate adjusted prevalence ratios (aPRs) with 95% confidence intervals (95% CIs). We evaluated associations of mental health symptoms with being immunocompromised using multinomial logistic regression and estimated conditional odds ratios (cOR) with 95% CIs. Immunocompromised adults were more likely to adhere to recommended COVID-19 preventive behaviors, including washing or sanitizing hands (96.3% vs. 89.8%, χ² <0.001), maintaining social distance (91.9% vs. 83.7%, χ² <0.001), and canceling a doctor's appointment (47.1% vs. 29.7%, χ² <0.001). Hispanic immunocompromised adults (aPR: 1.47, 95% CI: 1.12–1.92) and immunocompromised women (aPR: 1.25, 95% CI: 1.00–1.56) were more likely to cancel doctor's appointments compared to non-Hispanic White immunocompromised adults and men, respectively. Immunocompromised adults reported higher odds of feeling nervous/anxious/on edge (cOR: 1.89, 95% CI: 1.44–2.51), depressed (cOR: 2.81, 95% CI: 2.17–3.64), lonely (cOR: 2.28, 95% CI: 1.74–2.98), and hopeless (cOR: 2.86, 95% CI: 2.21–3.69) 3–7 days in the last week. Immunocompromised adults were more likely to cancel their doctor's appointments and report COVID19-related mental health symptoms. The continuity of care of immunocompromised adults should be prioritized through alternative interventions, such as telehealth.

Background

As of February 3, 2021, the United States has recorded over 26 million confirmed cases and almost 450,000 deaths due to COVID-19, the syndrome associated with the novel 2019 coronavirus SARS-CoV-2.¹ The pandemic has reshaped daily life due to the widespread enactment of COVID-preventive policies such as quarantine, social distancing at a minimum of 6 ft from others, closures of nonessential businesses, and promotion of COVID-preventive behaviors such as handwashing and wearing of face masks.^2–4 Furthermore, the COVID-19 pandemic has impacted the interaction between patients and health care providers.^5,6 The initial surge of patients with COVID-19 prompted hospitals and clinics to cancel nonessential visits to the doctor's office and delay elective surgical procedures to mitigate the risk of exposure, in accordance with CDC guidelines.^7–10 The use of telemedicine as a substitute for these encounters has since increased substantially, from 11% of U.S. patients using telehealth in 2019 to 46% using telehealth in 2020.^11–13 While there is considerable variation among certain demographics, most patients are seeing their doctors less frequently during the COVID-19 pandemic. This is particularly true of patients with chronic medical conditions^14,15 who must see doctors more frequently compared to the general population; these populations have also been found to be more likely to adhere to COVID-preventive measures such as frequent handwashing and social distancing.^14,16

One understudied high-risk group whose health behaviors are of particular concern during the pandemic is immunocompromised adults or those with a chronic condition leading to decreased immune function either through primary (e.g., inherited disorders of immune function) or secondary (e.g., HIV/AIDS, cancer, post-transplant) mechanisms. Immunocompromised patients are more susceptible to infectious disease and therefore confront an increased magnitude of risk not only of contracting SARS-COV-2 but also of suffering increasing morbidity and mortality due to the complications of COVID-19.^17–22 Awareness of this high-risk status may motivate immunocompromised patients to follow preventive guidelines, whether that be social distancing or cancellation of doctors' appointments. However, immunocompromised patients face a dilemma as they are more frequent users of the health care system; due to their immunodeficient status, they must undergo more frequent medical monitoring in the form of doctor visits and routine laboratory monitoring.

Adherence to COVID-19 recommended preventive behaviors may lead to unintended consequences, such as social isolation leading to poor mental health outcomes (Fig. 1).²³ A decrease in interactions with the health care system may lead to delayed detection of COVID-19-related poor mental health symptoms. Poor mental health among the immunocompromised may lead to several negative outcomes, such as low quality of life, poor adherence to medication and treatment related to their chronic condition, and ultimately shorter survival.^24,25 Patterns in adherence to COVID-19 preventive behaviors among the immunosuppressed are currently unknown, although as suggested in prior work,¹⁶ preventive measures might drive them to avoid doctor visits despite their increased health needs.

FIG. 1.

Preventive behaviors and continuity of care among immunosuppressed populations: a conceptual model.

In this study, our primary aim was to identify potential differences in adherence to various COVID-19 preventive measures among those with underlying immunocompromised status compared to the general population, with a specific focus on canceling doctor's appointments to evaluate continuity of care. As a secondary aim we sought to characterize if immunocompromised patients were experiencing elevated COVID-19-related mental health symptoms relative to the general population. Our hypotheses were that immunocompromised adults are more likely to adhere to COVID-19 preventive behaviors and experience COVID-19-related mental health symptoms compared to other adults.

Methods

COVID-19 Impact Survey

Data from the COVID-19 Household Impact Survey are publicly available for download (https://www.covid-impact.org/), and as such, ethical approval for this study was not required. For this analysis, we used data from the publicly available COVID-19 Household Impact Survey.²⁶ Data collection was conducted by the nonpartisan research organization NORC at the University of Chicago for the Data Foundation. The purpose of COVID-19 Household Impact Survey is to provide national and regional statistics about physical health, mental health, economic security, and social dynamics in the United States during the pandemic.²⁶ The survey is designed to provide periodic estimates and time trends of the U.S. adult household population nationwide. Data for the COVID-19 Household Impact Survey were collected through AmeriSpeak^®, which is a probability-based panel designed to be representative of the U.S. household population funded and operated by NORC at the University of Chicago. The AmeriSpeak panel randomly selected U.S. households using area probability and address-based sampling. These sampled households were then contacted by U.S. mail, telephone, and field interviewers. The panel provides sample coverage of ∼97% of the U.S. household population. Exclusion criteria included the following populations: people with P.O. Box only addresses, addresses not listed in the USPS Delivery Sequence File, and some newly constructed dwellings. Interviews were conducted in both English and Spanish. Further details regarding the sampling approach have been previously published.^14,16,27

Data from week 1 (April 20–26, 2020), week 2 (May 4–10, 2020), and week 3 (May 30–June 8, 2020) were available and merged for this analysis. The number of participants invited and percentage of interviews completed by week are as follows: 11,133 invited with 19.7% interviews completed (week 1); 8,570 invited with 26.1% interviews completed (week 2); and 10,373 invited with 19.7% interviews completed (week 3). The analytic sample includes 10,760 adults. The final analytic data were weighted to reflect the U.S. population of adults aged 18 years and above. The demographic weighting variables were obtained from the 2020 Current Population Survey. Panelists were offered a $5 monetary incentive for completing the survey.

Main measures

We defined our study population of interest based on one questionnaire item: “Has a doctor or other health care provider ever told you that you have any of the following: Diabetes; High blood pressure or hypertension; Heart disease, heart attack, or stroke; Asthma; Chronic lung disease or COPD; Bronchitis or emphysema; Allergies; a Mental health condition; Cystic fibrosis; Liver disease or end-stage liver disease; Cancer; a Compromised immune system; or Overweight or obesity.” We defined immunocompromised adults as those participants with a self-reported diagnosis of “a compromised immune system.”

We defined COVID-19 preventive behaviors using participants' response (yes/no) to the following questions: “Which of the following measures, if any, are you taking in response to the coronavirus?” Participants were able to select all that applied from a list of 19 options. We focused on the following options: canceled a doctor appointment; visited a doctor or hospital; canceled or postponed work activities; canceled or postponed school activities; canceled or postponed dentist or other appointments; avoided some or all restaurants; worked from home; studied from home; canceled or postponed pleasure, social, or recreational activities; avoided public or crowded places; avoided contact with high-risk people; washed or sanitized hands; kept 6 ft distance from those outside my household; stayed home because I felt unwell; and wiped packages entering my home.

In addition, we evaluated COVID-19-related self-reported mental health symptoms experienced in the last 7 days. To evaluate mental health symptoms, we used participants' responses to the following questions: “In the past 7 days, how often have you: Felt nervous, anxious, or on edge; Felt depressed; Felt lonely; Felt hopeless about the future; Had physical reactions such as sweating, trouble breathing, nausea, or a pounding heart when thinking about your experience with the coronavirus pandemic?” Participants were able to choose from the following list of options for each mental health symptom: Not at all or <1, 1–2, 3–4, and 5–7 days.

Covariates

The following covariates were included in the analyses: age (18–29, 30–44, 25–59, 60+), sex (male, female), marital status (married/living with a partner, widowed/divorced/separated, never married), race/ethnicity [non-Hispanic (NH) White, NH-Black, Hispanic, NH-Asian, NH-Other], education (no high school diploma, HS graduate or equivalent, some college, baccalaureate degree, or above), employment status (employed/unemployed), household income (<$50,000, $50,000–<$100,000, ≥$100,000), population density (rural, suburban, urban), census region (Northeast, Midwest, South, West), and insurance status. These covariates were chosen a priori based on review of the published literature on demographic factors influencing COVID-19 preventive behaviors among adults with chronic diseases.^16,28–30

Data analysis

Survey weights were applied to reflect population-based estimates. Descriptive statistics was summarized, by immunocompromised status, in percentages among all respondents. Chi-squared (χ²) tests were used for bivariate comparison of mental health symptoms experienced in the last 7 days and COVID-19 preventive behaviors among the immunocompromised compared to immunocompetent adults. To estimate determinants of immunocompromised adults canceling doctor's appointments, we calculated prevalence ratios with Poisson regression using robust estimation of standard errors.^26–28 Potential variables for inclusion in the model were assessed using available sociodemographic variables and age-adjusted Poisson regression analysis. Due to the exploratory nature of this analysis using a predictive framework, an arbitrary p value of <.10 was used as criteria to include the variable in the multivariable Poisson regression model. For multivariable Poisson regression models, adjusted prevalence ratios (aPRs) and confidence intervals (95% CIs) for each independent variable were calculated. We conducted sensitivity analyses to evaluate the association of canceled school or childcare among women with immunocompromised conditions who canceled their doctor's appointments using Poisson regression.

Finally, we used multinomial logistic regression to calculate conditional odds ratios (cOR) with 95% CI to compare mental health symptoms experienced in the last 7 days among the immunosuppressed to the immune competent, after adjustment for age, sex, race/ethnicity, insurance status, cancer survivor status, and having a comorbid condition. We adjusted for cancer survivorship status as a cancer diagnosis may lead to immunosuppression during treatment and is also associated with poor mental health symptoms during the pandemic as we showed in prior work.³¹ Comorbid conditions included were diabetes; high blood pressure or hypertension; heart disease, heart attack, or stroke; asthma; chronic lung disease or COPD; bronchitis or emphysema; allergies; cystic fibrosis; liver disease or end-stage liver disease; or overweight or obesity. We also conducted a sensitivity analysis to evaluate mental health symptoms among the immunosuppressed who did not have an existing mental health condition based on self-report. Based on the exploratory nature of this analysis, we did not include an adjustment for multiple comparisons.^32,33 All statistical analyses were conducted using Stata IC 15 (StataCorp LLC, College Station, TX). Sampling weights were applied to provide results that were nationally representative of the U.S. adult population.

Results

Table 1 summarizes the demographic characteristics of the study population overall and stratified by immunocompromised status. Overall, 7.6% of adults reported to have a compromised immune system. Sixty-five percent of immunocompromised adults were 45 years of age or above. Immunocompromised adults were mostly female (61.8%), married or living with a partner (57.8%), NH White (74.1%), and with at least some college education (61.8%). Most immunocompromised adults lived in urban areas (68.9%) and had either employer-sponsored (43.2%) or Medicare insurance (45.1%). The most common comorbid conditions included having allergies (63.3%), hypertension (44.0%), overweight/obese (42.2%), and a mental health condition (32.7%).

Table 1.

Characteristics of COVID Impact Survey Respondents (n = 10,760), a Nationally Representative Survey of the United States Among Adults With and Without a Compromised Immune System (April–June 2020)

	Adults with a compromised immune system (n = 753)		Other adults (n = 10,007)
	Col %	95% CI	Col %	95% CI
Age
18–29	13.2	9.7–17.7	21.0	19.7–22.4
30–44	21.5	18.2–25.3	25.5	24.4–26.7
45–59	29.7	25.6–34.1	24.1	23.0–25.3
60+	35.6	31.3–40.1	29.3	28.1–30.6
Sex
Male	38.2	33.7–42.9	49.3	47.9–50.7
Female	61.8	57.1–66.3	50.7	49.3–52.1
Marital status
Married/living with partner	57.8	53.2–62.3	57.7	56.3–59.1
Widowed/divorced/separated	22.6	19.1–26.5	18.1	17.1–19.1
Never married	19.6	16.2–23.5	24.2	22.9–25.5
Race/ethnicity
White, NH	74.1	70.0–77.9	61.8	60.4–63.2
Black, NH	8.7	6.4–11.7	11.9	11.1–12.9
Hispanic	10.8	8.2–13.9	17.0	15.9–18.3
Other, NH	5.7	4.2–7.9	8.6	7.8–9.5
Employed in the past 7 days	32.7	28.5–37.2	51.2	49.8–42.6
Education
No HS Diploma	8.9	6.0–13.0	9.7	8.6–10.8
HS Graduate	29.3	24.8–34.2	27.7	26.4–29.1
Some college	33.0	29.1–37.1	27.4	26.3–28.4
Baccalaureate or above	28.8	24.9–33.2	35.2	34.0–36.5
Household income
<$50,000	54.5	49.7–59.2	44.9	43.5–46.3
$50,000–<$100,000	30.0	25.8–34.5	32.4	31.1–33.6
≥$100,000	15.5	12.3–19.5	22.8	21.6–24.0
Population density
Rural	7.8	5.9–10.2	9.3	8.5–10.1
Suburban	23.3	19.6–27.6	18.4	17.4–19.4
Urban	68.9	64.4–73.0	72.3	71.1–73.5
Insurance type or health coverage plans
Purchased plan	19.4	15.6–23.9	17.0	15.9–18.1
Employer sponsored	43.2	38.5–48.0	52.7	51.3–54.1
TRICARE	4.4	2.7–7.1	4.8	4.3–5.3
Medicaid	36.9	32.5–41.6	22.3	21.2–23.5
Medicare	45.1	40.5–49.8	23.8	22.7–25.0
Dually eligible (Medicare and Medicaid)	23.9	20.3–27.9	8.6	7.9–9.4
VA	3.7	2.2–6.2	4.5	4.0–5.0
Indian Health Service	0.8	0.3–2.2	1.1	0.8–1.5
No insurance	4.5	2.8–7.1	9.1	8.2–9.9
Comorbid conditions
Diabetes	21.4	17.7–25.8	11.1	10.3–12.0
Hypertension	44.0	39.4–48.7	32.2	30.9–33.4
Heart disease	14.6	11.7–18.1	6.3	5.7–7.0
Asthma	27.0	22.9–31.5	13.0	12.1–13.9
COPD	12.1	9.5–15.2	4.0	3.5–4.6
Bronchitis or emphysema	22.3	18.7–26.3	10.4	9.6–11.2
Allergies	63.3	58.7–67.6	41.0	39.7–42.4
Mental health condition	32.7	28.4–37.4	14.1	13.2–15.1
Cancer	26.0	22.1–30.5	6.5	5.9–7.1
Overweight/obese	42.2	37.6–46.8	32.5	31.3–33.8
Liver disease or end-stage liver disease	5.4	3.8–7.6	0.9	0.7–1.2
Cystic fibrosis	0.9	0.4–2.1	0.4	0.3–0.7

CI, confidence interval; NH, non-Hispanic.

Immunocompromised adults were more likely to adhere to recommended COVID-19 preventive behaviors, including washed or sanitized hands (96.3% vs. 89.8%, χ² <0.001), keeping 6 ft distance from those outside their household (91.9% vs. 83.7%, χ² <0.001), avoiding public or crowded places (85.3% vs. 75.4%, χ² <0.001), and avoiding contact with high-risk people (71.1% vs. 57.5%, χ² <0.001) (Table 2). Immunocompromised adults were more likely to cancel a doctor appointment (47.1% vs. 29.7%, χ² <0.001). The prevalence of canceling a doctor's appointment increased from 38.3% to 59.2% between April and early May; and although the prevalence decreased to 41.2%, we continued to observe a significant difference in canceling a doctor's appointment among immunocompromised adults by early June compared to the general population (Fig. 2).

Table 2.

Preventive Behaviors Against COVID-19 of COVID Impact Survey Respondents (n = 10,760), a Nationally Representative Survey of the United States Among Adults With and Without a Compromised Immune System (April–June 2020)

	Adults with a compromised immune system (n = 753)		Other adults (n = 10,007)		p
	Col %	95% CI	Col %	95% CI
Preventive behaviors against COVID-19
Washed or sanitized hands	96.3	94.2–97.7	89.8	88.9–90.6	<.001
Kept six feet distance from those outside my household	91.9	88.9–94.1	83.7	82.6–84.7	<.001
Worn a face mask	86.6	83.2–89.4	84.8	83.9–85.8	.32
Avoided public or crowded places	85.3	81.7–88.2	75.4	74.2–76.6	<.001
Avoided some or all restaurants	79.1	75.0–82.7	69.9	68.6–71.1	<.001
Canceled or postponed pleasure, social, or recreational activities	73.5	69.1–77.4	64.4	63.0–65.7	<.001
Avoided contact with high-risk people	71.1	66.7–75.1	57.5	56.1–58.9	<.001
Wiped packages entering my home	52.7	48.0–57.4	40.7	39.4–42.1	<.001
Canceled a doctor appointment	47.1	42.4–51.8	29.7	28.4–30.9	<.001
Canceled or postponed dentist or other appointment	45.0	40.3–49.7	35.4	34.1–36.7	<.001
Worked from home	26.5	22.6–30.8	32.2	30.9–33.5	.01
Canceled or postponed work activities	28.4	24.3–32.8	27.8	26.6–29.1	.81
Visited a doctor or hospital	16.6	13.2–20.7	9.2	8.4–10.0	<.001
Canceled or postponed school activities	19.1	15.7–23.1	19.9	18.8–21.1	.70
Stayed home because I felt unwell	17.6	14.3–21.6	9.9	9.0–10.8	<.001

FIG. 2.

Proportion of immunocompromised adults who canceled doctor's appointments in response to the COVID-19 pandemic from April to June 2020 (n = 10,760).

We evaluated determinants of canceling doctor's appointments specifically among immunocompromised adults (n=753, Table 3). On multivariable analyses, we found that compared to immunocompromised adults aged 60 years and above, adults aged 30–44 (aPR: 0.76, 95% CI: 0.59–0.98) and 45–49 years (aPR: 0.78, 95% CI: 0.62–0.99) were less likely to cancel doctor's appointments. Hispanic immunocompromised adults had 47% higher prevalence of canceling doctor's appointments compared to NH-White adults (aPR: 1.47, 95% CI: 1.12–1.92). In addition, immunocompromised women were more likely to cancel doctor's appointments (aPR: 1.25, 95% CI: 1.00–1.56) compared to men.

Table 3.

Determinants of Adults with a Compromised Immune System Canceling a Doctor's Appointment among COVID Impact Survey, a Nationally Representative Survey of the United States (n = 753) (April–June 2020)

	%	95% CI	Unadjusted PR (only adjusted for age)	95% CI	Adjusted PR	95% CI
Age
18–29	13.2	9.7–17.7	0.69	0.45–1.07	0.67	0.43–1.04
30–44	21.5	18.2–25.3	0.79	0.61–1.00	0.76	0.59–0.98
45–49	29.7	25.6–34.1	0.78	0.62–0.99	0.78	0.62–0.99
60+	35.6	31.3–40.1	Ref.		Ref.
Sex
Male	38.2	33.7–42.9	Ref.		Ref.
Female	61.8	57.1–66.3	1.29	1.04–1.60	1.25	1.00–1.56
Race/ethnicity	74.1	70.0–77.9
White, NH	8.7	6.4–11.7	Ref.		Ref.
Black, NH	10.8	8.2–13.9	1.20	0.86–1.65	1.20	0.88–1.64
Hispanic	5.7	4.2–7.9	1.51	1.13–2.00	1.47	1.12–1.92
Other, NH	74.1	70.0–77.9	1.27	0.91–1.76	1.25	0.92–1.70
Marital status
Married/living with partner	57.8	53.2–62.3	Ref.		Ref.
Widowed/divorced/separated	22.6	19.1–26.5	1.15	0.93–1.42
Never married	19.6	16.2–23.5	0.73	0.53–0.99
Region
Northeast	19.2	15.6–23.5	Ref.		Ref.
Midwest	25.4	21.7–29.5	0.99	0.76–1.30
South	34.4	29.9–39.2	0.80	0.59–1.07
West	21.0	17.7–24.7	0.91	0.67–1.22
Employment status
Not employed	67.3	62.8–71.5	Ref.		Ref.
Employed/self-employed	32.7	28.5–37.2	0.89	0.70–1.13
Household income
<$50,000	54.5	49.7–59.2	1.09	0.79–1.49
$50,000–<$100,000	30.0	25.8–34.5	1.15	0.82–1.60
≥$100,000	15.5	12.3–19.5	Ref.
Population density
Rural	7.8	5.9–10.2	1.09	0.83–1.44
Suburban	23.3	19.6–27.6	0.95	0.74–1.21
Urban	68.9	64.4–73.0	Ref.

PR, prevalence ratio.

As a sensitivity analysis, we focused on women with a compromised immune system and evaluated the association of reporting personal plans affected by the closure of pre-K through 12th grade or childcare with canceling their doctor's appointments. Immunocompromised women were more likely to cancel a doctor's appointment if they reported that their personal plans were affected by the closure of pre-K through 12th grade or childcare (60.9%) compared with women who did not (40.6%; χ² p = .001). After adjustment for age and having at least one other comorbid condition, we found that women with a compromised immune system who reported that their plans were affected by the closure of pre-K through 12th grade or childcare had a 48% higher prevalence of canceling their doctor's appointments (aPR: 1.48; 95% CI:1.17–1.88) compared with women who did not.

On multivariable analyses, after adjustment for age, gender, race/ethnicity, insurance status, comorbid conditions, cancer diagnosis, and obesity, immunocompromised adults reported higher odds of reporting feeling nervous, anxious, or on edge (cOR: 1.89, 95% CI: 1.44–2.51), depressed (cOR: 2.81, 95% CI: 2.17–3.64), lonely (cOR: 2.28, 95% CI: 1.74–2.98), hopeless (cOR: 2.86, 95% CI: 2.21–3.69), and experiencing a physical reaction when thinking about their experiences during the COVID-19 pandemic (cOR: 3.84, 95% CI: 2.46–5.99) 3–7 days per week in the last 7 days (Fig. 3).

FIG. 3.

Mental health symptoms experienced in the last 7 days among the immunocompromised compared to immune-competent adults from April to June 2020 (n = 10,760).

Sensitivity analyses revealed that compared to immune-competent adults, immunocompromised adults without an existing mental health condition reported higher odds of feeling depressed (cOR: 2.26, 95% CI: 1.63–3.13), hopeless (cOR: 1.85, 95% CI: 1.31–2.62), and experience a physical reaction when thinking about experiences during COVID-19 pandemic (cOR: 2.37, 95% CI: 1.25–4.49) 3–7 days per week in the last 7 days (Supplementary Fig. S1).

Discussion

In this nationally representative study of U.S. adults, we found that immunocompromised adults are adhering to COVID-19 preventive behaviors. In addition, we observed that immunocompromised adults were more likely to cancel doctor's appointments, in particular immunocompromised women and Hispanic immunocompromised adults. Immunocompromised adults were more likely to report experiencing mental health symptoms in the past 7 days, particularly feeling depressed and hopeless, even among those without a prior diagnosed mental health condition. Our findings have important implications for the significance of continuity of care among the immunocompromised in the United States. Providing alternative care options, such as telehealth, to ensure that this vulnerable population can discuss their mental health symptoms with their providers should be prioritized as the COVID-19 pandemic continues to be a global public health concern.

Globally, the burden on the health care system, particularly infectious disease care units, during the COVID-19 pandemic has limited provider's ability to provide quality care to patients with immune-compromising conditions.^34,35 Continuity of care is important to ensure monitoring of symptoms and adherence to treatment among immunocompromised adults, which may include people living with HIV/AIDS, those with lupus, transplant recipients, or adults with cancer undergoing chemotherapy or radiotherapy. Individuals within each of these chronic disease groups have been identified as increasingly vulnerable to morbidity and mortality due to COVID-19³⁶ due to their underlying comorbidities. Professional organizations representing clinicians which treat chronic disease patients have released recommendations to monitor special considerations for chronic disease management during the COVID-19 pandemic.^37–39 These recommendations have included increased social distancing and self-quarantine to reduce likelihood of infection, minimizing hospital and face-to-face encounters through telemedicine visits, and increased attention to potential drug–drug interactions and overlapping toxicities among COVID-19 treatments should infection occur. However, several prior COVID-19 studies have highlighted that the ability to engage in these preventive behaviors is a privilege and one that is not equally afforded to racial/ethnic minorities and vulnerable populations with low socioeconomic backgrounds.¹⁴ This is of particular significance when evaluating the immunocompromised as several clinical subpopulations, including HIV and AIDS, are disproportionately racial/ethnic and sexual minorities in the United States.⁴⁰ Interventions to alleviate disparities in access to preventive behaviors should be prioritized and implemented as the pandemic continues to impact continuity of care. Telehealth options will not be equitably available to all populations due to potential limitations in internet access and access to a private space at home to engage in patient–provider conversations.⁴¹ During the early days of the pandemic, although video visits skyrocketed from 3% to 80%, the proportion of visits by patients 65 or older fell from 41% to 35%, and the proportion of visits of non-English speaking patients and Medicaid-insured patients similarly dropped.⁴² Providers should keep barriers to telehealth in mind when developing a care plan for vulnerable immunocompromised adults in the context of the COVID-19 pandemic. Future research evaluating adherence to treatment and management of chronic conditions in the context of telehealth visits during the COVID-19 pandemic should be prioritized to evaluate the effectiveness of non-in-person clinician visits.

In addition to the disruptions in care experienced by immunocompromised adults, the increased mental health symptoms reported also warrant further consideration. The increased mental health symptoms reported among individuals with compromised immune systems may be due to increased fear of contracting COVID-19, given early reports of greater COVID-19 susceptibility among individuals with preexisting health conditions.^43,44 Conversely, the mental health impacts of the COVID-19 pandemic among immunocompromised adults may also be due to increased adherence to COVID-19 preventive behaviors. Previous studies have identified associations between adherence to preventive behaviors, such as social distancing and self-quarantine, with increased depressive symptoms due to factors such as increased social isolation and lack of social support.^44,45 Unique approaches to evaluating mental health symptoms have been used in several settings; for example, in China, online mental health surveys using communication programs, such as Weibo, WeChat, and TikTok, enabled mental health professionals and health authorities to provide mental health services online during the pandemic.^46,47 Remote consultation networks which can be implemented through the internet or telephone were also implemented in China to safely provide mental health services, while maintaining social distancing.⁴⁶ In the United States, telephone-based visits have proven to be a vital tool for caring for vulnerable populations without access to broadband and high-speed internet.⁴⁸ Clinicians managing patients with immunocompromised conditions should have increased awareness of the mental health impacts of the pandemic within members of these groups, conduct mental health screening during face-to-face or telehealth visits, phone-based visits, and provide additional mental health resources during this time. Longitudinal studies of individuals from immunocompromised backgrounds should be conducted to examine onset and persistence of mental health symptoms during and beyond the COVID-19 pandemic.

For this analysis, we were able to utilize nationally representative survey data and, therefore, obtain a representative sample of immunocompromised adults in the United States. However, our results should be interpreted in the context of several limitations. We relied on self-report of chronic conditions and symptoms reported in the 7 days before survey administration, which may lead to measurement error of our outcomes of interest. We were unable to use validated scales such as the General Anxiety Disorder-7 to measure mental health conditions, including depression and anxiety. We were unable to distinguish between types of immunocompromised conditions, and therefore, future research focusing on specific groups such as those living with HIV or organ transplant survivors should be prioritized. In the immunocompromised population, severity of disease measures, such as CD4+ count, is an important measure of significance when evaluating continuity of care; however, these data were unavailable and should be prioritized in future evaluations. Importantly, data regarding immunocompromised adult's mental health symptoms before COVID-19 were unavailable, and therefore, we are unable to conclude if the mental health symptoms are related to the pandemic. We conducted sensitivity analyses to address this concern; however, future research using existing longitudinal cohorts to compare mental health symptoms among the immunocompromised pre- and post-COVID-19 pandemic should be prioritized.

In conclusion, we observed that immunocompromised adults are adhering to COVID-19 preventive behaviors, including canceling doctor's appointments. Certain demographic groups of immunocompromised adults, including women and Hispanic adults, may be more likely to cancel their appointments. The disproportionate impact on women may be related to interruptions in childcare. The continuity of care and engagement in patient–provider interactions of immunocompromised adults should be an urgent priority due to the frequently self-reported mental health symptoms during the COVID-19 pandemic. Identifying mental health symptoms should be prioritized for immunocompromised adults as symptoms such as anxiety or hopelessness can impact their quality of life and adherence to treatment.^24,25,49 Interventions to improve equitable access to mental health care, such as alternatives to telehealth or telephone based care, targeted to vulnerable and at-risk populations should be prioritized.

Authors' Contributions

J.Y.I. conceptualized the article, guided data analysis, interpreted critically, and wrote the article; D.V., A.H., M.C.R. contributed to data interpretation and article writing. All authors have read and approved the submission.

Data Availability

Data are publicly available at the following website: https://www.covid-impact.org/results

Ethics and Consent

Ethical approval was not obtained from the author's respective institutions as the data were made publicly available.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

J.Y.I. is supported by UNC's Cancer Care Quality Training 2T32CA116339-11. M.C.R. is supported by the TRANSPORT—The Translational Program of Health Disparities Research Training 5S21MD012474-02.

Supplementary Material

Supplementary Figure S1