COVID-19 Vaccine Perceptions Among Patients With Chronic Disease in a Large Gastroenterology and Hepatology Practice.

INTRODUCTION: To assess beliefs about safety, effectiveness, and delivery of the coronavirus disease 2019 (COVID-19) vaccine among chronic Gastroenterology and Hepatology patients at an academic health system.
METHODS: We asked about vaccine beliefs, vaccine concerns, and preferred location to receive the COVID-19 vaccine.
RESULTS: A total of 1,215 patients responded (response rate: 37%). Most patients believed that vaccines are safe, effective, and that they would take the COVID-19 vaccine at a medical office or pharmacy. However, we identified important sociodemographic factors associated with vaccine hesitancy. DISCUSSION: Patients have high level of trust in the COVID-19 vaccine and are likely to follow their specialist physician recommendations.

INTRODUCTION

On December 11, 2020, the US Food and Drug Administration issued its first emergency use authorization of a messenger RNA vaccine for coronavirus disease 2019 (COVID-19) (1). Currently, the Center for Disease Control Advisory Committee on Immunization Practices recommends vaccination for people at high risk of severe COVID-19 illness due to underlying medical conditions, which include those with liver disease and on immunosuppressive medications (2). Gastroenterology and Hepatology (GI/HEP) practices provide continuity care and may play a role in promotion and delivery of vaccination to these populations. Studies have shown variable vaccine hesitancy in the general population, but little is known about patient perceptions regarding the COVID-19 vaccine among patients with chronic gastrointestinal and liver conditions (3–5). We surveyed beliefs about safety and effectiveness of the COVID-19 vaccine, willingness to accept it, concerns, and optimal location of vaccine administration among a captive sample of patients with inflammatory bowel disease (IBD) and cirrhosis, and liver transplant recipients who receive care within a large health system in Southeastern Pennsylvania.

METHODS

From December 13, 2020, to December 23, 2020, we surveyed patients who had at least 1 outpatient visit at one of 5 GI/HEP tertiary-care or community-based practices in the University of Pennsylvania Health System within the past 6 months (see Survey, Supplemental Digital Content 1, http://links.lww.com/AJG/B988), and who had access to the online electronic health record (EHR) portal (91% of potential patient population are activated). The survey was sent through the EHR portal based on methodology previously developed by our team (6). The survey assessed 3 domains: (i) general vaccine beliefs and behavior, (ii) COVID-19 vaccine concerns, and (iii) optimal location to receive the COVID-19 vaccine with 5-point Likert scale responses (e.g., 1—strongly disagree to 5—strongly agree or 1—least likely, 5—most likely). Items were adapted from a previous survey assessing acceptance of a COVID-19 vaccine and iteratively refined by clinical experts on the study team (4). Sociodemographics, clinical care site, and medical condition were obtained from the EHR. Median household income was estimated based on patient zip code and obtained from the American Community Survey 2019 in inflation-adjusted dollars. Fisher exact tests for categorical and Wilcoxon rank-sum tests for continuous variables compared participant characteristics and responses. For some descriptive and univariable analyses, responses were grouped, e.g., strongly/somewhat agree vs strongly/somewhat disagree/neutral. A multivariable logistic regression model was conducted for vaccine hesitancy (defined as answering strongly/somewhat disagree or neutral to the statement “If a COVID-19 vaccine is available to me, I will take it”) adjusting for sociodemographic factors and self-report of taking the annual influenza vaccine. Analyses were performed with Stata 15.1 (College Station, TX). The study was reviewed by the Institutional Review Board at the University of Pennsylvania and deemed to be quality improvement.

RESULTS

Among 3,113 eligible participants, 1,215 responded with an overall response rate of 37%. A total of 770 (64%) had IBD, 304 (25%) had cirrhosis, and 141 (12%) were liver transplant recipients. Table, Supplemental Digital Content 2, http://links.lww.com/AJG/B989 shows patient characteristics stratified by respondents versus nonrespondents. Respondents were slightly older, more likely to be White, and less likely to have low income. Table 1 shows baseline demographic patient characteristics and responses to key survey items stratified by medical condition (complete responses to all survey items are in Table, Supplemental Digital Content 3, http://links.lww.com/AJG/B990). The mean age was 54 years (interquartile range: 36–66); 87% of respondents were White, 6% were Black, and 7% identified as other; 50% were female. Responses were obtained from tertiary care (73%) and community-based sites (27%). A total of 8% of patients reside in zip codes with less than $45,000 median income.

Table 1.

Patient characteristics and vaccine beliefs, behavior, and concerns stratified by medical condition

A total of 89% of respondents completely/somewhat generally agreed that vaccines are safe, 92% that vaccines are effective, and 87% completely/somewhat agreed that the COVID-19 vaccine will help control the pandemic. Greater than 75% would follow their government or doctor's recommendation about receiving the COVID-19 vaccine (91% would follow their GI/liver specialist's recommendation), while 56% would follow their employers' recommendation. A total of 85% agreed strongly/somewhat that they would take the COVID-19 vaccine if available, 7% were neutral, and only 8% somewhat/completely disagreed that they would take the vaccine. Concerns about vaccine safety and effectiveness were reported among 33% and 27%, respectively; 39% were concerned about being prioritized for the vaccine; 20% or less had concerns about vaccine costs or inconvenience of receiving it. Most respondents said they would be very/somewhat likely to receive the vaccine at a medical office or pharmacy; acceptability rates were lower for drive-through sites, places of employment, or community sites (e.g., church and school).

Figure 1 show the percent of patients who strongly/somewhat agreed that the COVID-19 vaccine can control the pandemic, they would take the vaccine if recommended, had concerns about effectiveness, side effects, or concerns about being prioritized for the vaccine—all stratified by demographic and behavioral characteristics. Univariable factors associated with lower vaccine hesitancy were older age and self-report of taking the annual influenza vaccine. Female sex, Black race, and lower household income by zip code were associated with higher hesitancy (Figure 2). In multivariable models, after adjustment for these covariates, sex, income, and receipt of an annual influenza vaccine remained independently associated with vaccine hesitancy (full model data in Table, Supplemental Digital Content 4, http://links.lww.com/AJG/B991).

Figure 1.

COVID-19 vaccine perceptions by patient characteristics among 1,215 respondents in Gastroenterology and Hepatology practices. Bars represent % respondents who answered “strongly/somewhat agree” to each question in the legend. Bivariate comparisons are shown by age, sex, race, level of government trust (high vs low), and self-report of receiving annual influenza vaccine (yes vs no). *Age <65 less likely to take vaccine, more likely to be concerned about effectiveness and side effects; women less likely to take vaccine, more likely to be concerned about effectiveness and side effects; Black participants less likely to take vaccine, more likely to be concerned about side effects; lower government trust less likely to believe vaccine will control pandemic, less likely to take vaccine, more likely to be concerned about effectiveness and side effects; not taking flu vaccine less likely to take vaccine, more likely to be concerned about effectiveness and side effects. P < 0.05 in bivariate comparisons.

Figure 2.

Factors associated with COVID-19 vaccine hesitancy. Model is adjusted for variables shown. Vaccine hesitancy was defined as answering “strongly/somewhat disagree or neutral” to the statement “If a COVID-19 vaccine is available to me, I will take it.” Lower aOR indicates lower vaccine hesitancy. aOR, adjusted odds ratio; CI, confidence interval; IBD, inflammatory bowel disease; LTR, liver transplant recipients.

DISCUSSION

We show early data on patient beliefs and acceptability of COVID-19 vaccination in a large and diverse cohort of patients with chronic gastrointestinal and liver diseases and may be prioritized for early for vaccination efforts. First, we note the high feasibility of assessing patient perceptions via readily available EHR tools, which may be leveraged to rapidly disseminate trusted vaccination information, provide outreach, assist with scheduling, and monitor for side effects. Second, we note that patients have high level of trust in their treating GI/HEP physicians, are more likely to follow their recommendations, and prefer to receive COVID-19 vaccination in a medical setting. Thus, GI/HEP specialists have a responsibility to provide appropriate messaging and counseling to their patients and to possibly help guide them to receive the vaccine. Third, we identify important factors (e.g., female sex, Black race, lower income, and younger age) associated with lower willingness to take the vaccine and higher levels of concern about effectiveness and side effects (female sex) that can guide targeted education and outreach efforts (5). Behavioral factors such as lower levels of willingness to follow government recommendations and not receiving the annual flu vaccine are highly associated with less willingness to take the COVID-19 vaccine and higher concerns about adverse effects.

Despite the large sample size, we acknowledge limitations such as response bias as well as a predominantly White, commercially insured, and tertiary care–based sample that may limit generalizability. We did not have access to data on education, which could impact vaccine acceptance. Our survey was deployed before health-system or individual clinician education efforts to educate patients about the vaccine. Future studies should re-evaluate patient perceptions after targeted education and outreach efforts have been conducted. In summary, we report results in a large, specialty care population that can inform targeted vaccination initiatives. Our findings have relevance beyond the COVID-19 pandemic and highlight the important role GI/HEP specialists have in general vaccination and preventive care counseling for their patients with chronic conditions.

CONFLICTS OF INTEREST

Guarantor of the article: Marina Serper, MD, MS.

Specific author contributions: M.S.: conceptualization: equal; formal analysis: lead; methodology: equal; resources: lead; supervision: lead; and writing—original draft: lead. K.R.R.: conceptualization: supporting; writing—review and editing: supporting. N.A.: conceptualization: supporting; writing—review and editing: supporting. M.B.: writing—review and editing: supporting. S.M.: conceptualization: equal; investigation: equal; methodology: equal; writing—review and editing: equal.

Financial support: M.S. time was supported by grant number K23DK115897, and S.J.M. time was supported by grant number K08CA234326 from the National Institutes of Health.

Potential competing interests: None to report.