Emergency Department Survey of Vaccination Knowledge, Vaccination Coverage, and Willingness to Receive Vaccines in an Emergency Department Among Underserved Populations — Eight U.S. Cities, April–December, 2024

Abstract

Current models of vaccination coverage screening and surveillance might miss underserved populations whose only health care access occurs in emergency departments (EDs). During April-December 2024, a survey of non-critically ill adult patients evaluated in 10 EDs in eight U.S. cities across five states was conducted to ascertain patients' vaccination knowledge, self-reported vaccination coverage, and willingness to receive vaccines in an ED. Among 4,326 patients approached by the research team, 3,285 (75.9%) agreed to participate. Non-Hispanic Black or African American (Black), non-Hispanic White, and Hispanic or Latino (Hispanic) persons each accounted for approximately 30% of participants; 17.9% spoke Spanish as their primary language; 7.8% had unstable or marginal housing; and 21.0% lacked a source of primary health care. Approximately one half (49.4%) had not heard of one or more CDC-recommended vaccines for their age group, and 85.9% had not received one or more of the recommended vaccines. Factors associated with not being up to date with recommended vaccinations included non-Hispanic Black race and ethnicity (adjusted odds ratio [aOR] = 1.93; 95% CI = 1.32-2.85), lack of primary health care (aOR = 2.91; 95% CI = 1.74-5.13), and lack of health insurance (aOR = 3.01; 95% CI = 1.27-8.82). Among 2,821 participants who were not up to date with recommended vaccines, 46.4% said that they would accept one or more missing vaccines if they could be provided during their ED visit, and 86.7% of these persons said they would accept all missing vaccines. The primary reasons for missed vaccine doses were that the participant was unaware of or had not been offered the vaccines. EDs could be explored as additional sites to offer vaccination screening, recommendations, counseling, and referrals to increase vaccination coverage among underserved populations.

Introduction

Review of patient vaccination status, coupled with counseling regarding needed vaccines, is a fundamental component of primary health care. During the previous 2 decades, the list of CDC-recommended vaccinations has grown, with recommendations for varicella zoster (shingles), respiratory syncytial virus (RSV), and COVID-19 vaccines varying by age group and risk factors adding to vaccination schedule complexity (1).

Approximately one third of the U.S. adult population receives no primary health care; these persons rarely, if ever, receive a comprehensive review of their vaccination status, vaccination counseling, and opportunities to receive recommended vaccines (2,3). Emergency departments (EDs), the only health care access point for millions of persons in the United States, might be positioned to address gaps in public health interventions, including vaccination messaging and delivery (4). In two ED-based cluster randomized controlled trials, provision of COVID-19 and influenza vaccination messaging platforms resulted in twofold to threefold increases in vaccine acceptance and receipt rates in underserved populations (5,6). Expanding this ED-based vaccination surveillance concept to examine all CDC-recommended adult vaccines, this study sought to ascertain among non–critically ill adult ED patients, 1) knowledge and receipt of CDC-recommended vaccines for their age group; 2) reasons for previously not receiving and currently not accepting recommended vaccines; and 3) willingness to accept missing vaccines if they were offered in an ED.

Methods

Pilot Testing and Institutional Review Board Review

Initial survey templates were reviewed with a patient community advisory board, and the survey instrument was pilot tested with six ED patients to assess comprehension and response consistency. This research was reviewed and approved by a central institutional review board (IRB) with reliance at five of the ED sites and by individual IRBs at the other five ED sites.*

Study Sites and Sample Size

This study was conducted during April 18–December 31, 2024, at 10 EDs in eight U.S. cities in five states (Chicago, Illinois; Detroit, Michigan; Durham, North Carolina; Fresno, California; Los Angeles, California; Philadelphia, Pennsylvania; San Francisco, California; and Sylmar, California). Together, these EDs reported an estimated 637,392 ED visits in 2024. A priori sample size calculations determined that 2,401 patients (across all sites) needed to be enrolled for the lower and upper bounds of the 95% CIs around the point estimates of the primary outcomes (knowledge and receipt of age-recommended vaccines) to be within a 2% margin of error. To balance enrollment across sites, the study stopped enrollment at any one of the 10 EDs if that ED reached 400–450 enrollees. Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines were followed (7).

Patient Enrollment and Survey Administration

During weekdays (typically 9:00 a.m.–5:00 p.m., Monday–Friday), research staff members reviewed ED patient care dashboards (lists of patients in the ED with their age and primary symptoms) to identify adults aged ≥18 years who met inclusion and exclusion criteria^†; they then offered enrollment to eligible patients using a scripted consent form. Age-specific surveys (i.e., for persons aged 18–26,^§ 27–49, 50–64, and ≥65 years) were administered in English, Spanish, or (at two sites) Mandarin (Supplementary Box). Participants were asked to respond to the following questions about each CDC-recommended vaccine for their age group as of April 2024 (7): 1) “Have you heard about [VACCINE]?”; 2) “Have you received [VACCINE]?”; 3) (If no), “Why have you not received [VACCINE]?”; and 4) “Would you accept any of these [MISSING VACCINES] today during your ED visit if they could be provided to you?” Plain language accompanied by explanations was used whenever possible to describe the recommended vaccines (e.g., “shingles” rather than “varicella zoster”). Age thresholds for survey questions and time thresholds for up-to-date status were determined using the CDC Recommended Adult Immunization Schedule (as of April 2024)^¶ (1). For example, age ≥60 years was used to prompt the series of questions about RSV,** pneumococcal, shingles, COVID-19, influenza, and tetanus-containing vaccines; tetanus vaccination was considered up to date only if the most recent dose had been received within the previous 10 years. In addition to vaccination questions, participants were asked questions about their race, ethnicity, health insurance, primary care, and housing.^††

Analysis

Participant characteristics were summarized as raw counts and frequency percentages, and aggregate key survey question responses were summarized as percentages with 95% CIs, excluding nonresponses in proportion denominators. Three multiple logistic regression models were fitted, one each with incomplete vaccination knowledge, vaccine coverage, and willingness to accept one or more missed vaccines as dependent variables; model adjustment factors included age group, sex, city, and characteristics previously associated with low vaccination coverage (Black or African American [Black] race, Hispanic or Latino [Hispanic] ethnicity, non-English primary language, no source of primary health care, and lack of health insurance) (4). Model results are reported as adjusted odds ratios (aORs) with 95% CIs, and two-tailed p-values; p-values <0.05 and aORs with 95% CIs that excluded 1 were considered statistically significant. Analyses were conducted using RStudio (version 2023.12.1; RStudio, Inc.).

Results

Participant Characteristics

Among 4,326 non–critically ill patients approached in the 10 participating EDs, 3,285 (75.9%) agreed to participate. Among these, 54.6% were female; 29.6% were Black, 27.0% were non-Hispanic White, and 31.5% were Hispanic; 7.8% had marginal or unstable housing; 21.0% lacked primary health care; and for 17.9%, Spanish was the primary language (Table 1).

Participant Knowledge About Vaccines

Approximately one half (49.4%; 95% CI = 47.7%–54.1%) of participants had not heard of one or more recommended vaccines (Table 2). In adjusted models, the following characteristics were associated with not having heard of one or more vaccines: male sex (aOR = 2.01; 95% CI = 1.71–2.34), Black race (aOR = 2.01; 95% CI = 1.62–2.48), Hispanic ethnicity (aOR = 1.94; 95% CI = 1.51–2.48), non-English primary language (aOR = 2.42; 95% CI = 1.82–3.22), lack of primary health care (aOR = 1.68; 95% CI = 1.35–2.08), and being uninsured (aOR = 1.56; 95% CI = 1.14–2.16) (Table 3).

Participant Self-Reported Vaccine Receipt

Overall, 85.9% (95% CI = 84.8%–87.2%) of participants reported that they had not received one or more of the recommended vaccines. Associated factors included Black race (aOR = 1.93; 95% CI = 1.32–2.85), lack of primary health care (aOR = 2.91; 95% CI = 1.74–5.13), being uninsured (aOR = 3.01; 95% CI = 1.27–8.82), age ≥27 years; and having an ED visit in Chicago (aOR = 2.58; 95% CI 1.34–5.38), Detroit (aOR = 2.24; 95% CI = 1.05–5.34), or Fresno (aOR = 1.95; 95% CI = 1.09–3.59) (Table 3).

Participants had received approximately one half (51.3%) of recommended vaccines (Supplementary Table 1). Influenza vaccine was the most commonly reported missed vaccine among persons aged 18–26 years (64.5%) and 27–49 years (65.6%). Among those aged 50–64 years, shingles vaccine was most commonly missed (63.0% of participants), and among adults aged ≥65 years, the most commonly missed vaccine was RSV vaccine (74.2%). Among vaccines recommended for all age groups, 82.6% of participants reported having received COVID-19 vaccine, 42.9% had received a tetanus-containing vaccine within the previous 10 years, and 41.6% had received an influenza vaccine (Supplementary Table 2).

Participant Willingness to Accept Vaccines if Offered in an ED

Among the 2,821 participants who were not up to date with vaccination, 1,309 (46.4%; 95% CI = 44.6%-48.2%) said they would accept one or more missed vaccines if they were offered in an ED (Table 2); among these persons, 1,135 (86.7%; 95% CI = 84.8%-88.5%) said they would accept all missing vaccines. The only factor associated with more willingness to accept vaccines if offered in an ED was having an ED visit in Sylmar (aOR = 5.25; 95% CI = 3.44-8.22).

The most common reasons participants gave for having missed recommended vaccines were that they had not heard of the vaccine (40.7%) and that they had heard of it but were unaware that the vaccine was recommended for them (9.6%). The most common reasons participants gave for stating that they would not accept vaccines if they were offered to them the same day were that they felt too ill in the ED at that time (27.4%) and that they needed more information about the vaccines and the diseases they prevent (19.3%).

Discussion

Current models of vaccination status screening and surveillance, such as the National Health Interview Survey (NHIS), miss underserved populations whose only health care interactions occur in EDs. In this analysis, high rates of incomplete knowledge about vaccines and low self-reported vaccination coverage were observed across all ED patient age groups, especially among Black and Hispanic participants and participants who lacked a source of primary health care and health insurance. Full vaccination coverage also differed by site (lower in Chicago, Detroit, and Fresno), as did willingness to accept vaccines (higher in Sylmar). Many participants reported that they would accept all missing vaccines if they were offered in an ED. Based on the number of respondents who reported that they would accept all missing vaccines if offered (1,135), this analysis suggests that up-to-date coverage in this population could theoretically more than triple from its current level of 14% to as high as 48%.

Apart from COVID-19 and influenza vaccination coverage, data regarding U.S. adult vaccination coverage are limited. In this ED-based analysis, self-reported up-to-date vaccination coverage rates were lower across all age groups than the 22.8% rate among adults reported from the 2022 NHIS. However, NHIS had a lower response rate (47.7%), used substantially different methods (in-person household interviews), and did not include some vaccines that were included in this analysis (e.g., measles, mumps, and rubella; meningococcal; and hepatitis B vaccines) (8).

Considering the millions of visits to U.S. EDs (155 million in 2022) (9), these findings suggest that asking non–critically ill adults in EDs about their vaccination status, and offering missed vaccinations to those who express interest, might improve vaccination coverage. Although comprehensive provision of all recommended adult vaccines to ED patients is likely impractical, the development and implementation of ED-centered programs to offer recommendations for vaccination and provide vaccine counseling and referrals to vaccine providers are logical next steps. Experience from previous vaccine delivery research suggests that such programs would need to be automated and embedded into the ED workflow so that they do not interfere with or add to the workload of ED operations (4–6).

Limitations

The findings in this report are subject to at least three limitations. First, because no centralized registry for adult vaccination exists, documentation of receipt of vaccines from electronic medical records (EMRs) was not included in the study protocol. Nevertheless, the survey method used is standard for vaccination coverage surveillance, and other investigators have found high concordance between self-reports of vaccination and data reported in EMRs (10). Second, the survey questions related to vaccination receipt are subject to recency and recall biases; participants might have more easily remembered vaccines that they actively sought as senior adults (e.g., shingles vaccines) than those received at younger ages. Finally, reported vaccine acceptance rates might not reflect acceptance rates in other EDs with different populations, and they could have been affected by other factors, including social desirability bias.

Implications for Public Health Practice

This ED-based survey identified substantial gaps in knowledge about and receipt of age-recommended vaccines and suboptimal self-reported coverage. Many patients reported that they would accept vaccines if they were offered to them in an ED. EDs could be explored as additional sites to offer vaccination screening, recommendations, counseling, and referrals to increase vaccine coverage among underserved populations.