https://orcid.org/0000-0002-9087-0944
Abstract
Emergency department visits associated with Coronavirus Disease 2019 (COVID-19) continue to indicate racial and ethnic inequities. We describe the sociodemographic characteristics of individuals receiving COVID-19 vaccination in the emergency department and compare with an outpatient clinic population and emergency department (ED) patients who were eligible but not vaccinated.
We conducted a retrospective analysis of electronic health record data at an urban academic ED from May to July 2021. The primary aim was to characterize the ED-vaccinated population, compared with ED patients who were eligible but unvaccinated and the physically adjacent outpatient vaccination clinic population.
A total of 627 COVID-19 vaccinations were administered in the ED. Overall, 49% of ED patients during that time had already received at least one vaccine dose prior to ED arrival. Hispanic, non-Hispanic Black patients, and patients on non-commercial insurance had higher odds of being vaccinated in the ED as compared with outpatient clinic setting. Among eligible ED patients, men and patients who were uninsured/self-pay were more likely to accept ED vaccination.
This ED COVID-19 vaccination campaign demonstrated a higher likelihood to vaccinate individuals from racial/ethnic minority groups, those with high social vulnerability, and non-commercial insurance, when compared with a co-located outpatient vaccination clinic.
Introduction
Coronavirus Disease 2019 (COVID-19) infection disproportionately affects racial and ethnic minorities, and vaccination rates in these communities remain below the national average.1 COVID-19 infection rates continue to fluctuate, and individuals who remain unvaccinated are at high risk for severe illness, hospitalization and death.2 As of October 2021, non-Hispanic White individuals have a 1.2 times higher vaccination rate than non-Hispanic Black individuals and a 1.1 times higher rate than Hispanic individuals.2
It is estimated that one-fifth of the US population utilizes emergency departments (ED) as their primary source of health care, and there are 130 million ED visits per year.3,4 Previous ED-based influenza and pneumococcal vaccination campaigns successfully reached patients who have higher barriers to vaccination.5 As COVID-19 vaccinations became available, a survey conducted from December 2020 to March 2021 in 15 US EDs found that the 61% of respondents reported they would accept a COVID-19 vaccination as part of their ED care.6 However, vaccine hesitancy remains common among ED patients, especially among those of younger age, female sex, Black race, or Hispanic ethnicity.6
Goals of this investigation
An UCHealth University of Colorado Hospital ED (Aurora, CO) began offering the single-dose Ad26.COV2.S vaccine (J&J/Janssen) vaccination to patients receiving care in the ED in May 2021. The primary aim of this study was to characterize the population receiving vaccination in the ED setting. Secondary and tertiary goals were to compare this population to the adjacent outpatient vaccination clinic and evaluate the operational effects of offering vaccination in the ED setting for broad deployment.
Methods
Study setting and population
This was a retrospective secondary analysis of electronic health record data at an urban academic ED. ED-based J&J/Janssen vaccination began on 4 May 2021, and data were analyzed for the first 2 months (4 May–5 July 2021) of the program. Only patients who were being discharged from the ED were eligible for the ED vaccination program. Vaccinations were offered to admitted patients at discharge; however, these patients were not included in this analysis. This was an opt-in program, in which providers were prompted to evaluate COVID-19 vaccination history and consider offering ED vaccination by an electronic health record alert based on each patient’s recorded vaccination status. This study was approved as exempt research by the Colorado Multiple Institutional Review Board.
Study protocol
We analyzed sociodemographic characteristics for the following groups: (i) patients newly vaccinated during the ED visit, (ii) patients who were eligible (unvaccinated) but did not receive vaccination during the ED visit and (iii) individuals who received their vaccination at the adjacent UCHealth outpatient vaccination clinic during the same time period. Vaccination status at the time of ED visit was confirmed by the Colorado Immunization Information System. We also collected ED operational process metrics, including ED length of stay, visit day, time of day, placement in observation status and discharge disposition. A pre-existing dedicated ED pharmacy team prospectively recorded the number of wasted doses per day and documented allergic reactions as part of the clinical record.
Data analysis
Covariates of interest included age, gender, race, ethnicity, primary language, insurance status and Social Vulnerability Index (SVI). The Centers for Disease Control and Prevention’s SVI is a validated measure calculated from 15 social factors in an individual’s residential census tract.7 SVI is used to determine the risk of needing public support and resources, resulting in percentile ranking (lowest to highest vulnerability).7 We analyzed factors independently associated with ED vaccination using multivariable logistic regression. We calculated 95% confidence intervals for operational metrics using either bootstrapping for skewed continuous data or two proportion z-tests with continuity correction as appropriate for categorical data. All analyses were performed using R statistical software version 4.0.2 (Vienna, Austria).
Results
Characteristics of study subjects
From 4 May through 5 July 2021, 627 COVID-19 vaccinations were administered in the ED. Overall, 49% (6895/14 039) of ED patients during that time had already received at least one vaccine dose prior to ED arrival. Of the vaccination-eligible population, 9% (627/7144) were vaccinated during their ED visit. Median age of ED-vaccinated patients was 37 years [27.0, 49.0]; 61% were male, 34% non-Hispanic White, 32% Hispanic, 27% non-Hispanic Black and median SVI was 0.81 (Table 1).
Demographic information of patients who were seen in the ED, stratified by vaccination status, or vaccinated at outpatient clinic between 4 May and July 5 2021 (N = 31 760)
| ED vaccinated | ED eligible not vaccinated | Already vaccinated | UCHealth outpatient clinic vaccinated | Overall | |
|---|---|---|---|---|---|
| (N = 627) | (N = 6517) | (N = 6895) | (N = 17 721) | (N = 31 760) | |
| Age | |||||
| Mean (SD) | 38.8 (14.8) | 38.9 (16.3) | 49.3 (18.6) | 40.2 (14.2) | 41.9 (16.2) |
| Median [Q1, Q3] | 37.0 [27.0, 49.0] | 35.0 [26.0, 49.0] | 49.0 [33.0, 64.0] | 39.0 [28.0, 50.0] | 39.0 [29.0, 53.0] |
| Age group | |||||
| 18–44 | 419 (66.8%) | 4498 (69.0%) | 3017 (43.8%) | 11 073 (62.5%) | 19 007 (59.8%) |
| 45–64 | 183 (29.2%) | 1454 (22.3%) | 2291 (33.2%) | 5740 (32.4%) | 9668 (30.4%) |
| ≥65 | 25 (4.0%) | 565 (8.7%) | 1587 (23.0%) | 908 (5.1%) | 3085 (9.7%) |
| Gender | |||||
| Female | 244 (38.9%) | 3447 (52.9%) | 3932 (57.0%) | 8947 (50.5%) | 16 570 (52.2%) |
| Male | 383 (61.1%) | 3070 (47.1%) | 2963 (43.0%) | 8774 (49.5%) | 15190 (47.8%) |
| Race/ethnicity | |||||
| American Indian and Alaska Native | 8 (1.3%) | 55 (0.8%) | 58 (0.8%) | 78 (0.4%) | 199 (0.6%) |
| Asian/Pacific Islander | 5 (0.8%) | 155 (2.4%) | 319 (4.6%) | 970 (5.5%) | 1449 (4.6%) |
| Hispanic | 199 (31.7%) | 1724 (26.5%) | 1703 (24.7%) | 2522 (14.2%) | 6148 (19.4%) |
| Non-Hispanic Black | 171 (27.3%) | 1899 (29.1%) | 1384 (20.1%) | 1658 (9.4%) | 5112 (16.1%) |
| Non-Hispanic White | 214 (34.1%) | 2260 (34.7%) | 3075 (44.6%) | 8341 (47.1%) | 13 890 (43.7%) |
| Other | 19 (3.0%) | 216 (3.3%) | 254 (3.7%) | 641 (3.6%) | 1130 (3.6%) |
| Unknown | 11 (1.8%) | 208 (3.2%) | 102 (1.5%) | 3511 (19.8%) | 3832 (12.1%) |
| Primary language | |||||
| English | 539 (86.0%) | 5635 (86.5%) | 5856 (84.9%) | 15 054 (85.0%) | 27 084 (85.3%) |
| Other | 19 (3.0%) | 343 (5.3%) | 399 (5.8%) | 1605 (9.1%) | 2366 (7.4%) |
| Spanish | 69 (11.0%) | 539 (8.3%) | 640 (9.3%) | 1062 (6.0%) | 2310 (7.3%) |
| Insurance | |||||
| Medicaid | 232 (37.0%) | 1978 (30.4%) | 1551 (22.5%) | 2224 (12.6%) | 5985 (18.8%) |
| Medicare | 53 (8.5%) | 660 (10.1%) | 1883 (27.3%) | 1203 (6.8%) | 3799 (12.0%) |
| Other | 119 (19.0%) | 1570 (24.1%) | 867 (12.6%) | 81 (0.5%) | 2637 (8.3%) |
| Commercial | 67 (10.7%) | 979 (15.0%) | 1620 (23.5%) | 10 387 (58.6%) | 13 053 (41.1%) |
| Uninsured/self-pay | 156 (24.9%) | 1330 (20.4%) | 974 (14.1%) | 3826 (21.6%) | 6286 (19.8%) |
| SVI | |||||
| Mean (SD) | 0.728 (0.239) | 0.696 (0.263) | 0.624 (0.295) | 0.450 (0.308) | 0.541 (0.315) |
| Median [Q1, Q3] | 0.807 [0.628, 0.932] | 0.763 [0.523, 0.919] | 0.686 [0.398, 0.891] | 0.430 [0.170, 0.734] | 0.560 [0.266, 0.837] |
| Missing | 32 (5.1%) | 780 (12.0%) | 153 (2.2%) | 422 (2.4%) | 1387 (4.4%) |
| ED vaccinated | ED eligible not vaccinated | Already vaccinated | UCHealth outpatient clinic vaccinated | Overall | |
|---|---|---|---|---|---|
| (N = 627) | (N = 6517) | (N = 6895) | (N = 17 721) | (N = 31 760) | |
| Age | |||||
| Mean (SD) | 38.8 (14.8) | 38.9 (16.3) | 49.3 (18.6) | 40.2 (14.2) | 41.9 (16.2) |
| Median [Q1, Q3] | 37.0 [27.0, 49.0] | 35.0 [26.0, 49.0] | 49.0 [33.0, 64.0] | 39.0 [28.0, 50.0] | 39.0 [29.0, 53.0] |
| Age group | |||||
| 18–44 | 419 (66.8%) | 4498 (69.0%) | 3017 (43.8%) | 11 073 (62.5%) | 19 007 (59.8%) |
| 45–64 | 183 (29.2%) | 1454 (22.3%) | 2291 (33.2%) | 5740 (32.4%) | 9668 (30.4%) |
| ≥65 | 25 (4.0%) | 565 (8.7%) | 1587 (23.0%) | 908 (5.1%) | 3085 (9.7%) |
| Gender | |||||
| Female | 244 (38.9%) | 3447 (52.9%) | 3932 (57.0%) | 8947 (50.5%) | 16 570 (52.2%) |
| Male | 383 (61.1%) | 3070 (47.1%) | 2963 (43.0%) | 8774 (49.5%) | 15190 (47.8%) |
| Race/ethnicity | |||||
| American Indian and Alaska Native | 8 (1.3%) | 55 (0.8%) | 58 (0.8%) | 78 (0.4%) | 199 (0.6%) |
| Asian/Pacific Islander | 5 (0.8%) | 155 (2.4%) | 319 (4.6%) | 970 (5.5%) | 1449 (4.6%) |
| Hispanic | 199 (31.7%) | 1724 (26.5%) | 1703 (24.7%) | 2522 (14.2%) | 6148 (19.4%) |
| Non-Hispanic Black | 171 (27.3%) | 1899 (29.1%) | 1384 (20.1%) | 1658 (9.4%) | 5112 (16.1%) |
| Non-Hispanic White | 214 (34.1%) | 2260 (34.7%) | 3075 (44.6%) | 8341 (47.1%) | 13 890 (43.7%) |
| Other | 19 (3.0%) | 216 (3.3%) | 254 (3.7%) | 641 (3.6%) | 1130 (3.6%) |
| Unknown | 11 (1.8%) | 208 (3.2%) | 102 (1.5%) | 3511 (19.8%) | 3832 (12.1%) |
| Primary language | |||||
| English | 539 (86.0%) | 5635 (86.5%) | 5856 (84.9%) | 15 054 (85.0%) | 27 084 (85.3%) |
| Other | 19 (3.0%) | 343 (5.3%) | 399 (5.8%) | 1605 (9.1%) | 2366 (7.4%) |
| Spanish | 69 (11.0%) | 539 (8.3%) | 640 (9.3%) | 1062 (6.0%) | 2310 (7.3%) |
| Insurance | |||||
| Medicaid | 232 (37.0%) | 1978 (30.4%) | 1551 (22.5%) | 2224 (12.6%) | 5985 (18.8%) |
| Medicare | 53 (8.5%) | 660 (10.1%) | 1883 (27.3%) | 1203 (6.8%) | 3799 (12.0%) |
| Other | 119 (19.0%) | 1570 (24.1%) | 867 (12.6%) | 81 (0.5%) | 2637 (8.3%) |
| Commercial | 67 (10.7%) | 979 (15.0%) | 1620 (23.5%) | 10 387 (58.6%) | 13 053 (41.1%) |
| Uninsured/self-pay | 156 (24.9%) | 1330 (20.4%) | 974 (14.1%) | 3826 (21.6%) | 6286 (19.8%) |
| SVI | |||||
| Mean (SD) | 0.728 (0.239) | 0.696 (0.263) | 0.624 (0.295) | 0.450 (0.308) | 0.541 (0.315) |
| Median [Q1, Q3] | 0.807 [0.628, 0.932] | 0.763 [0.523, 0.919] | 0.686 [0.398, 0.891] | 0.430 [0.170, 0.734] | 0.560 [0.266, 0.837] |
| Missing | 32 (5.1%) | 780 (12.0%) | 153 (2.2%) | 422 (2.4%) | 1387 (4.4%) |
Demographic information of patients who were seen in the ED, stratified by vaccination status, or vaccinated at outpatient clinic between 4 May and July 5 2021 (N = 31 760)
| ED vaccinated | ED eligible not vaccinated | Already vaccinated | UCHealth outpatient clinic vaccinated | Overall | |
|---|---|---|---|---|---|
| (N = 627) | (N = 6517) | (N = 6895) | (N = 17 721) | (N = 31 760) | |
| Age | |||||
| Mean (SD) | 38.8 (14.8) | 38.9 (16.3) | 49.3 (18.6) | 40.2 (14.2) | 41.9 (16.2) |
| Median [Q1, Q3] | 37.0 [27.0, 49.0] | 35.0 [26.0, 49.0] | 49.0 [33.0, 64.0] | 39.0 [28.0, 50.0] | 39.0 [29.0, 53.0] |
| Age group | |||||
| 18–44 | 419 (66.8%) | 4498 (69.0%) | 3017 (43.8%) | 11 073 (62.5%) | 19 007 (59.8%) |
| 45–64 | 183 (29.2%) | 1454 (22.3%) | 2291 (33.2%) | 5740 (32.4%) | 9668 (30.4%) |
| ≥65 | 25 (4.0%) | 565 (8.7%) | 1587 (23.0%) | 908 (5.1%) | 3085 (9.7%) |
| Gender | |||||
| Female | 244 (38.9%) | 3447 (52.9%) | 3932 (57.0%) | 8947 (50.5%) | 16 570 (52.2%) |
| Male | 383 (61.1%) | 3070 (47.1%) | 2963 (43.0%) | 8774 (49.5%) | 15190 (47.8%) |
| Race/ethnicity | |||||
| American Indian and Alaska Native | 8 (1.3%) | 55 (0.8%) | 58 (0.8%) | 78 (0.4%) | 199 (0.6%) |
| Asian/Pacific Islander | 5 (0.8%) | 155 (2.4%) | 319 (4.6%) | 970 (5.5%) | 1449 (4.6%) |
| Hispanic | 199 (31.7%) | 1724 (26.5%) | 1703 (24.7%) | 2522 (14.2%) | 6148 (19.4%) |
| Non-Hispanic Black | 171 (27.3%) | 1899 (29.1%) | 1384 (20.1%) | 1658 (9.4%) | 5112 (16.1%) |
| Non-Hispanic White | 214 (34.1%) | 2260 (34.7%) | 3075 (44.6%) | 8341 (47.1%) | 13 890 (43.7%) |
| Other | 19 (3.0%) | 216 (3.3%) | 254 (3.7%) | 641 (3.6%) | 1130 (3.6%) |
| Unknown | 11 (1.8%) | 208 (3.2%) | 102 (1.5%) | 3511 (19.8%) | 3832 (12.1%) |
| Primary language | |||||
| English | 539 (86.0%) | 5635 (86.5%) | 5856 (84.9%) | 15 054 (85.0%) | 27 084 (85.3%) |
| Other | 19 (3.0%) | 343 (5.3%) | 399 (5.8%) | 1605 (9.1%) | 2366 (7.4%) |
| Spanish | 69 (11.0%) | 539 (8.3%) | 640 (9.3%) | 1062 (6.0%) | 2310 (7.3%) |
| Insurance | |||||
| Medicaid | 232 (37.0%) | 1978 (30.4%) | 1551 (22.5%) | 2224 (12.6%) | 5985 (18.8%) |
| Medicare | 53 (8.5%) | 660 (10.1%) | 1883 (27.3%) | 1203 (6.8%) | 3799 (12.0%) |
| Other | 119 (19.0%) | 1570 (24.1%) | 867 (12.6%) | 81 (0.5%) | 2637 (8.3%) |
| Commercial | 67 (10.7%) | 979 (15.0%) | 1620 (23.5%) | 10 387 (58.6%) | 13 053 (41.1%) |
| Uninsured/self-pay | 156 (24.9%) | 1330 (20.4%) | 974 (14.1%) | 3826 (21.6%) | 6286 (19.8%) |
| SVI | |||||
| Mean (SD) | 0.728 (0.239) | 0.696 (0.263) | 0.624 (0.295) | 0.450 (0.308) | 0.541 (0.315) |
| Median [Q1, Q3] | 0.807 [0.628, 0.932] | 0.763 [0.523, 0.919] | 0.686 [0.398, 0.891] | 0.430 [0.170, 0.734] | 0.560 [0.266, 0.837] |
| Missing | 32 (5.1%) | 780 (12.0%) | 153 (2.2%) | 422 (2.4%) | 1387 (4.4%) |
| ED vaccinated | ED eligible not vaccinated | Already vaccinated | UCHealth outpatient clinic vaccinated | Overall | |
|---|---|---|---|---|---|
| (N = 627) | (N = 6517) | (N = 6895) | (N = 17 721) | (N = 31 760) | |
| Age | |||||
| Mean (SD) | 38.8 (14.8) | 38.9 (16.3) | 49.3 (18.6) | 40.2 (14.2) | 41.9 (16.2) |
| Median [Q1, Q3] | 37.0 [27.0, 49.0] | 35.0 [26.0, 49.0] | 49.0 [33.0, 64.0] | 39.0 [28.0, 50.0] | 39.0 [29.0, 53.0] |
| Age group | |||||
| 18–44 | 419 (66.8%) | 4498 (69.0%) | 3017 (43.8%) | 11 073 (62.5%) | 19 007 (59.8%) |
| 45–64 | 183 (29.2%) | 1454 (22.3%) | 2291 (33.2%) | 5740 (32.4%) | 9668 (30.4%) |
| ≥65 | 25 (4.0%) | 565 (8.7%) | 1587 (23.0%) | 908 (5.1%) | 3085 (9.7%) |
| Gender | |||||
| Female | 244 (38.9%) | 3447 (52.9%) | 3932 (57.0%) | 8947 (50.5%) | 16 570 (52.2%) |
| Male | 383 (61.1%) | 3070 (47.1%) | 2963 (43.0%) | 8774 (49.5%) | 15190 (47.8%) |
| Race/ethnicity | |||||
| American Indian and Alaska Native | 8 (1.3%) | 55 (0.8%) | 58 (0.8%) | 78 (0.4%) | 199 (0.6%) |
| Asian/Pacific Islander | 5 (0.8%) | 155 (2.4%) | 319 (4.6%) | 970 (5.5%) | 1449 (4.6%) |
| Hispanic | 199 (31.7%) | 1724 (26.5%) | 1703 (24.7%) | 2522 (14.2%) | 6148 (19.4%) |
| Non-Hispanic Black | 171 (27.3%) | 1899 (29.1%) | 1384 (20.1%) | 1658 (9.4%) | 5112 (16.1%) |
| Non-Hispanic White | 214 (34.1%) | 2260 (34.7%) | 3075 (44.6%) | 8341 (47.1%) | 13 890 (43.7%) |
| Other | 19 (3.0%) | 216 (3.3%) | 254 (3.7%) | 641 (3.6%) | 1130 (3.6%) |
| Unknown | 11 (1.8%) | 208 (3.2%) | 102 (1.5%) | 3511 (19.8%) | 3832 (12.1%) |
| Primary language | |||||
| English | 539 (86.0%) | 5635 (86.5%) | 5856 (84.9%) | 15 054 (85.0%) | 27 084 (85.3%) |
| Other | 19 (3.0%) | 343 (5.3%) | 399 (5.8%) | 1605 (9.1%) | 2366 (7.4%) |
| Spanish | 69 (11.0%) | 539 (8.3%) | 640 (9.3%) | 1062 (6.0%) | 2310 (7.3%) |
| Insurance | |||||
| Medicaid | 232 (37.0%) | 1978 (30.4%) | 1551 (22.5%) | 2224 (12.6%) | 5985 (18.8%) |
| Medicare | 53 (8.5%) | 660 (10.1%) | 1883 (27.3%) | 1203 (6.8%) | 3799 (12.0%) |
| Other | 119 (19.0%) | 1570 (24.1%) | 867 (12.6%) | 81 (0.5%) | 2637 (8.3%) |
| Commercial | 67 (10.7%) | 979 (15.0%) | 1620 (23.5%) | 10 387 (58.6%) | 13 053 (41.1%) |
| Uninsured/self-pay | 156 (24.9%) | 1330 (20.4%) | 974 (14.1%) | 3826 (21.6%) | 6286 (19.8%) |
| SVI | |||||
| Mean (SD) | 0.728 (0.239) | 0.696 (0.263) | 0.624 (0.295) | 0.450 (0.308) | 0.541 (0.315) |
| Median [Q1, Q3] | 0.807 [0.628, 0.932] | 0.763 [0.523, 0.919] | 0.686 [0.398, 0.891] | 0.430 [0.170, 0.734] | 0.560 [0.266, 0.837] |
| Missing | 32 (5.1%) | 780 (12.0%) | 153 (2.2%) | 422 (2.4%) | 1387 (4.4%) |
ED vaccinated compared with outpatient clinic
When comparing the ED vaccination population to the outpatient vaccination clinic population, Hispanic and non-Hispanic Black patients had higher odds of being vaccinated in the ED relative to non-Hispanic White patients (odds ratio [OR] 1.34; 95% confidence interval [CI] 1.03–1.73 and OR 1.80; 95%Cl 1.41–2.29, respectively), as did patients with higher SVI (OR 1.26; 95%CI 1.21–1.31 for ten percentile point SVI increase) (Fig. 1A). Patients with Medicaid (OR 10.64; 95%CI 7.89–14.35) or who were uninsured/self-pay (OR 5.53; 95%CI 4.02–7.59) were more likely to be vaccinated in the ED relative to commercially insured patients (Fig. 1A).
(A) Comparison of patients vaccinated in the ED to patients vaccinated at the UCHealth outpatient vaccine clinic from 4 May 2021 to 5 July 2021. (B). Comparison of patients vaccinated in the ED to patients who were eligible but did not receive vaccination from 4 May 2021 to 5 July 2021.
ED vaccinated compared with ED eligible but unvaccinated
When comparing the ED-vaccinated population to ED patients who were eligible but did not receive the vaccine, the odds of being vaccinated in the ED were higher for men versus women (OR 1.89; 95%Cl 1.58–2.25), higher SVI (OR 1.05; 95%CI 1.01–1.09 for 10 percentile point SVI increase) and Medicaid (OR 1.54; 95%CI 1.14–2.08) or uninsured/self-pay (OR 1.43; 95%Cl 1.04–1.98) versus commercially insured patients (Fig. 1B). Among ED patients eligible to receive ED vaccination, non-Hispanic Black patients had lower odds of receiving ED vaccination (OR 0.77; 95%Cl 0.61–0.96 versus non-Hispanic White). These results were not altered when additional ED operational metrics were included in the analysis; specifically, ED disposition, time of administration and length of stay (Supplemental Fig. 1).
ED operational metrics
There was no apparent association between ED length of stay and receipt of vaccination as part of ED care (Table 2). We observed that arrival during non-business hours (before 8 AM or after 5 PM or anytime on weekends) was associated with decreased odds of receiving vaccination in the ED (OR 0.82; 95%Cl 0.69–0.99) (Table 2). For the 2-month period analyzed and the 627 ED vaccinations administered, there were 240 wasted doses and no documented allergic reactions or serious adverse events.
ED operational metrics (length of stay, visit day, time of day, placement in observation status and discharge status) for patients who were seen in the ED, stratified by vaccination status between 4 May and 5 July 2021 (N = 7144)
| ED vaccinated | ED eligible not vaccinated | Difference (95% CI) | OR (95% CI) | |
|---|---|---|---|---|
| (N = 627) | (N = 6517) | |||
| ED length of stay (Hours) | ||||
| Median [Min, Max] | 2.5 [0.0167, 24.2] | 2.7 [0.0167, 70.8] | −0.2 (−0.3 to 0.1) | 1.00 (0.96–1.03) |
| ED business hours versus non | ||||
| Non-business hours | 242 (38.6%) | 2243 (34.4%) | 4.2 (0.2 to 8.2) | 0.82 (0.69–0.99) |
| Discharge | ||||
| Discharged-seen in main | 273 (43.5%) | 2626 (40.3%) | 3.2 (−0.8 to 7.3) | Reference |
| Discharged-discharge from intake | 35 (5.6%) | 609 (9.3%) | −3.7 (−5.7 −1.8) | 0.51 (0.35–0.75) |
| Discharged-sent to fast track | 153 (24.4%) | 1068 (16.4%) | 8.0 (4.5 to 11.5) | 1.28 (1.02–1.61) |
| Observation status | 128 (20.4%) | 1004 (15.4%) | 5.0 (1.7 to 8.3) | 1.14 (0.89–1.45) |
| Inpatient | 35 (5.6%) | 1032 (15.8%) | −10.2 (−12.3 to −8.3) | 0.29 (0.19–0.43) |
| Other | 3 (0.5%) | 178 (2.7%) | −2.2 (−3.0 to −1.5) | 0.15 (0.05–0.49) |
| ED vaccinated | ED eligible not vaccinated | Difference (95% CI) | OR (95% CI) | |
|---|---|---|---|---|
| (N = 627) | (N = 6517) | |||
| ED length of stay (Hours) | ||||
| Median [Min, Max] | 2.5 [0.0167, 24.2] | 2.7 [0.0167, 70.8] | −0.2 (−0.3 to 0.1) | 1.00 (0.96–1.03) |
| ED business hours versus non | ||||
| Non-business hours | 242 (38.6%) | 2243 (34.4%) | 4.2 (0.2 to 8.2) | 0.82 (0.69–0.99) |
| Discharge | ||||
| Discharged-seen in main | 273 (43.5%) | 2626 (40.3%) | 3.2 (−0.8 to 7.3) | Reference |
| Discharged-discharge from intake | 35 (5.6%) | 609 (9.3%) | −3.7 (−5.7 −1.8) | 0.51 (0.35–0.75) |
| Discharged-sent to fast track | 153 (24.4%) | 1068 (16.4%) | 8.0 (4.5 to 11.5) | 1.28 (1.02–1.61) |
| Observation status | 128 (20.4%) | 1004 (15.4%) | 5.0 (1.7 to 8.3) | 1.14 (0.89–1.45) |
| Inpatient | 35 (5.6%) | 1032 (15.8%) | −10.2 (−12.3 to −8.3) | 0.29 (0.19–0.43) |
| Other | 3 (0.5%) | 178 (2.7%) | −2.2 (−3.0 to −1.5) | 0.15 (0.05–0.49) |
ED operational metrics (length of stay, visit day, time of day, placement in observation status and discharge status) for patients who were seen in the ED, stratified by vaccination status between 4 May and 5 July 2021 (N = 7144)
| ED vaccinated | ED eligible not vaccinated | Difference (95% CI) | OR (95% CI) | |
|---|---|---|---|---|
| (N = 627) | (N = 6517) | |||
| ED length of stay (Hours) | ||||
| Median [Min, Max] | 2.5 [0.0167, 24.2] | 2.7 [0.0167, 70.8] | −0.2 (−0.3 to 0.1) | 1.00 (0.96–1.03) |
| ED business hours versus non | ||||
| Non-business hours | 242 (38.6%) | 2243 (34.4%) | 4.2 (0.2 to 8.2) | 0.82 (0.69–0.99) |
| Discharge | ||||
| Discharged-seen in main | 273 (43.5%) | 2626 (40.3%) | 3.2 (−0.8 to 7.3) | Reference |
| Discharged-discharge from intake | 35 (5.6%) | 609 (9.3%) | −3.7 (−5.7 −1.8) | 0.51 (0.35–0.75) |
| Discharged-sent to fast track | 153 (24.4%) | 1068 (16.4%) | 8.0 (4.5 to 11.5) | 1.28 (1.02–1.61) |
| Observation status | 128 (20.4%) | 1004 (15.4%) | 5.0 (1.7 to 8.3) | 1.14 (0.89–1.45) |
| Inpatient | 35 (5.6%) | 1032 (15.8%) | −10.2 (−12.3 to −8.3) | 0.29 (0.19–0.43) |
| Other | 3 (0.5%) | 178 (2.7%) | −2.2 (−3.0 to −1.5) | 0.15 (0.05–0.49) |
| ED vaccinated | ED eligible not vaccinated | Difference (95% CI) | OR (95% CI) | |
|---|---|---|---|---|
| (N = 627) | (N = 6517) | |||
| ED length of stay (Hours) | ||||
| Median [Min, Max] | 2.5 [0.0167, 24.2] | 2.7 [0.0167, 70.8] | −0.2 (−0.3 to 0.1) | 1.00 (0.96–1.03) |
| ED business hours versus non | ||||
| Non-business hours | 242 (38.6%) | 2243 (34.4%) | 4.2 (0.2 to 8.2) | 0.82 (0.69–0.99) |
| Discharge | ||||
| Discharged-seen in main | 273 (43.5%) | 2626 (40.3%) | 3.2 (−0.8 to 7.3) | Reference |
| Discharged-discharge from intake | 35 (5.6%) | 609 (9.3%) | −3.7 (−5.7 −1.8) | 0.51 (0.35–0.75) |
| Discharged-sent to fast track | 153 (24.4%) | 1068 (16.4%) | 8.0 (4.5 to 11.5) | 1.28 (1.02–1.61) |
| Observation status | 128 (20.4%) | 1004 (15.4%) | 5.0 (1.7 to 8.3) | 1.14 (0.89–1.45) |
| Inpatient | 35 (5.6%) | 1032 (15.8%) | −10.2 (−12.3 to −8.3) | 0.29 (0.19–0.43) |
| Other | 3 (0.5%) | 178 (2.7%) | −2.2 (−3.0 to −1.5) | 0.15 (0.05–0.49) |
Discussion
Main findings of this study
This ED COVID-19 vaccination campaign was more likely to vaccinate individuals from racial/ethnic minority groups, those with high social vulnerability, and non-commercial insurance when compared with a co-located outpatient vaccination clinic.
Curiously, individuals who arrived during business hours were more likely to be vaccinated during their ED visit, although 24/7 availability is purportedly a major benefit of ED-based vaccination. This may have been driven by unmeasured staffing or workflow factors during overnight shifts and should be an area of structured inquiry in future studies. Finally, there was no association between receiving a vaccination and ED length of stay, providing reassurance that this practice can create community benefit without further straining already-overtaxed EDs. The unfortunately low percentage of eligible patients vaccinated during their ED visit (9%) is comparable to the findings of a recent study that showed patients who lacked regular medical care were more commonly vaccine hesitant.8
What is already known on this topic
Globally, COVID-19 vaccine hesitancy has been observed among racial and ethnic minority groups.8,9 Vaccination hesitancy in these communities has been attributed to many factors, including mistrust of government and process of vaccine development due to historical experiences, systemic racism, marginalization and inequity in access to vaccination.10,11 Public health educational campaigns both at a national and local level focused on these communities have shown success in changing beliefs about the COVID-19 vaccine and increasing vaccination rates.12,13
In addition, successful ED vaccination campaigns are not novel but rather have been successfully implemented for hepatitis A, influenza and pneumococcal infections in adult and pediatric EDs.5,14,15 Many of these campaigns have been with public health organization partnership.14
What this study adds
These findings show that it is feasible and safe to offer COVID-19 vaccination in a high-volume ED. The J&J/Janssen vaccine is now ubiquitously available and does not encumber ED staff with the obligation to schedule a follow-up dose, making it logistically easy to provide in ED settings.
Our program relies upon the buy-in of ED providers to inquire about, encourage and provide the vaccine, as well as the willingness of patients to accept the dose. Conversations about vaccine efficacy, side effects and rectification of misinformation are time-consuming, and in a busy ED, providers may not perceive they have time to dispel concerns and answer questions. Future research in this area should focus on vaccine hesitancy in the ED population and providing education to ED providers on how to discuss vaccination recommendations.
Limitations of this study
There are several limitations to this study. This was a single-site study in a state, Colorado, with an already high vaccination rate. Therefore, it is difficult to estimate the generalizability to all regions of the US.
Although the electronic health record alert had an area for providers to document why individuals were refusing vaccination, this was not routinely filled out, so no meaningful data were available for why patients refused vaccination in this ED setting. This information could have provided valuable insight on vaccination hesitancy in our community.
Only the J&J/Janssen vaccine was available for vaccination in the ED setting as it was a single dose with no follow-up visit needed. Therefore, ED patients could have been interested in vaccination but reluctant to get the J&J/Janssen vaccine, causing them to refuse vaccination in the ED. Additional research evaluating these patients could provide additional context to see if patients did receive vaccination after their ED visit and which vaccine they received.
Lastly, we do not know if every eligible patient was offered the vaccine. There was minimal standardization of scripting and workflow across providers, and it is likely that some providers may have approached all patients about vaccination, whereas others may have ignored the clinical decision support entirely and asked none. There could have been bias on chief compliant or other selection bias that was not evaluated in this analysis. The low vaccination rate (9%) for eligible patients illustrates this limitation and could impact the validity of the results.
Conclusion
Our findings show that there is an opportunity to inoculate difficult-to-reach populations in the ED setting who are at higher risk for COVID-19 illness but remain unvaccinated. Although many healthcare systems do not typically embrace the role of the ED in primary prevention, our results show that the ED can play a pivotal role in improving access to a population that is currently underserved and under vaccinated.
Funding
None.
Acknowledgments
We would like to thank Alexander Greig and Rachel Severson for their informatics support of this project.
Sarah K. Wendel, Fellow in Administration, Operations, and Quality
Conner L. Jackson, Research Instructor
Daniel Resnick-Ault, Assistant Medical Director
Gabrielle Jacknin, Pharmacist Supervisor – Emergency Medicine
Richard D. Zane, Chair of Department of Emergency Medicine
Sean S. Michael, Medical Director of UCHealth University of Colorado Emergency Department
Kelly J. Bookman, Vice Chair of Operations
Adit A. Ginde, Vice Chair of Research
