Background
Influenza is an important seasonal respiratory illness that is estimated to account for 12 200 hospital admissions and 3500 deaths per year in Canada.[1-3] Emergency department (ED) care for influenza consists mainly of supportive therapies and antiviral medications such as neuraminidase inhibitors, which provide modest benefits by reducing influenza symptoms by less than 1 day.[4] Vaccination is recommended for influenza prevention and is provided free for high-risk groups in British Columbia.[5] In one case-control study, influenza vaccination was shown to reduce mortality by 41%, and among those who had been vaccinated previously mortality was reduced by 75%.[6]

Like influenza, pneumococcal disease is a significant cause of morbidity and mortality in ED patients. Although we know that pneumococcus vaccination reduces the incidence of invasive pneumococcal disease, studies have not been powered to detect a reduction in all-cause mortality.[7] In British Columbia this vaccine is also provided free of charge to high-risk groups.[8] In 2011, influenza and pneumonia combined were the eighth leading cause of death in Canada.[2]

Vaccination programs in the ED have been described previously.[9-12] While the vaccination rates in the ED for pneumococcal disease and influenza have been so low that they are difficult to estimate (less than1% of vaccinations), vaccination in the ED for tetanus has been very effective, with an estimated 27 738 000 vaccinations given in the US between 1992 and 2000.[11] However, the burden of influenza and pneumococcal disease is exponentially greater than that of tetanus, with 114 000 hospitalizations annually for influenza in the US and 49 015 for pneumococcal disease, compared with fewer than 50 hospitalizations for tetanus.[11]

With the significant burden of disease and the capacity to vaccinate any willing, high-risk, previously unvaccinated patients, emergency departments present a significant opportunity to implement a valuable public health intervention. The objective of our study was to determine what proportion of adult patients in the ED qualify for and are willing to be vaccinated against influenza and pneumococcal disease during their visit.

Methods
Patients presenting to the emergency department at Vancouver General Hospital from 1 May to 31 August 2015 were approached to enroll in our study. The convenience sample obtained included adults (19 years and older) who could communicate in English and consented to be screened for demographic characteristics, vaccination status, risk factors for influenza and pneumococcal infection, and contraindications to vaccination. Critically ill patients and patients in severe pain were excluded from the study.

The primary outcome we sought was the proportion of patients presenting to the ED who could be immunized for influenza and pneumococcal disease (i.e., at high risk, unvaccinated, and willing to be vaccinated). Secondary outcomes we sought included the proportion of patients with a contraindication to vaccination, the proportion of patients at high risk for influenza and pneumococcal disease, the characteristics of vaccinated and unvaccinated patients, and the characteristics of patients willing to be vaccinated and unwilling to be vaccinated. Based on previous studies, we estimated that approximately 20% of screened vaccination-eligible patients would be willing to receive immunization in the ED. For this estimated proportion of 0.2 and a desired precision of +/- .05 we calculated a required sample size of 246. Data were reported as descriptive statistics and proportions with confidence intervals.

We received approval for this study from our institutional ethics board and obtained consent from patients to participate in the study.

Results
We screened 413 patients (mean age 55 years) using study inclusion and exclusion criteria (Figure [11]) and collected data on the characteristics of all patients screened (Table 1 [12]). A total of 254 of 358 patients who met the inclusion criteria agreed to participate and completed the Vaccination in Emergency Survey for a response rate of 71%. Our group of primary interest included 52 patients (20%) who were at high risk for influenza complications, unvaccinated, and willing to be vaccinated in the ED, and 39 patients (15%) who were at high risk for pneumococcal disease, unvaccinated, and willing to be vaccinated in the ED.

Looking at the study population overall, 83% of patients were at high risk for complications from influenza and 58% were at high risk for complications from pneumococcal disease. Risk factors reported for influenza (Table 2 [13]) and for pneumococcal disease (Table 3 [14]) show that heart and lung disease were common in both groups, and that many patients had multiple risk factors. Of the patients at high risk for complications, many were unaware that they were at high risk for influenza (56%) and pneumonia (63%). Contraindications to influenza vaccination were present in 0.78% of patients (0.39% had previous severe allergic reaction to vaccine or components and 0.39% had previous Guillain-Barré syndrome). No patient had a contraindication to pneumococcal vaccination (severe allergic reaction to the vaccine or components).

If vaccination were to be offered in the ED, 53% of patients responding to the survey would accept influenza vaccination and 44% would accept pneumococcal vaccination. Among high-risk patients, 55% would accept influenza vaccination and 45% would accept pneumococcal disease vaccination. Many more patients stated that they would consider vaccination if they had the opportunity to talk with a health care provider. The reasons for not wanting to be immunized in the ED for influenza (Table 4 [15]) and for pneumococcal disease (Table 5 [16]) were varied, with “Do not think I am at high risk” being a common response for patients declining influenza vaccination (18%) and pneumococcal disease vaccination (27%).

Conclusions
Our data show that if we offered five patients vaccination for influenza, one of these would be a high-risk patient who had not been reached by other offers of influenza vaccination and was willing to be vaccinated in the ED. This is similar to previously reported numbers.[12] For pneumococcal disease, it would be necessary to offer vaccination to seven patients in order to reach one high-risk, unvaccinated patient who was willing to be vaccinated. Very few patients had contraindications to vaccination, making it feasible to screen a large number of patients rapidly.

Large percentages of the ED patient population in our study were found to be at risk for complications of pneumococcal disease (58%) and influenza (83%), rates higher than those previously reported.[12] Patients seen in Canadian EDs are increasingly complex and have multiple medical comorbidities.[13] In our experience, these patients generally have more comorbidities than would be seen in a family doctor’s office or a community vaccination clinic. The emergency department provides a unique opportunity to reach a particularly high-risk cohort of patients through a vaccination program.

Barriers
We recognize that there are many barriers to influenza and pneumococcus vaccination in the emergency department, including concerns about disruption of ED patient flow, scarcity of time and resources, and personal attitudes of health care staff toward vaccination.[14] To address these concerns, we implemented an influenza vaccination program in our ED that maximized efficiency in immunizing patients while minimizing resources, time, and training required. Physicians screened patients for contraindications to vaccination (previous anaphylactic reaction to the influenza vaccine or components, fever higher than 38 °C, or previous Guillain-Barré syndrome) and wrote an order for influenza vaccine. ED nurses administered influenza vaccine to patients. Since our data showed a very high proportion of our patients at risk for influenza complications, we offered the influenza vaccine to any unvaccinated patients who did not have a contraindication. We used a single vaccine rather than the multiple available products to simplify the vaccination process for ED physicians and nurses. While we did not specifically study the time it took to complete an influenza vaccination in the ED, our screening and administration process was comparable to that used for tetanus vaccination, a process well established in the ED, and it is unlikely that adding influenza vaccination to the patient visit increased ED length of stay or wait times.

Many patients expressed interest in vaccination for influenza and pneumococcal disease but wanted to speak with a health care provider before proceeding. Also, many patients did not realize that they were part of a high-risk group, and some had concerns about side effects and vaccine effectiveness. These and other common reasons for patients declining vaccination represent opportunities for education and informed decision making regarding vaccination.

Study limitations
Our study had some limitations. First, we relied on self-reported data to determine patient vaccination status and the comorbidities that are risk factors for complications of influenza and pneumococcal disease. Patients may not have reported their vaccination status accurately and may have underreported or overreported their comorbidities. Second, the number of patients willing to be vaccinated may be an overestimate since those patients who declined to participate in our study would be more likely to decline vaccination than those who did participate. However, our response rate of 71% was reasonably high and thus the effect of the nonparticipants would likely be small.

Summary
Our study demonstrates that a significant number of high-risk patients presenting to the emergency department would be willing to be vaccinated for influenza and pneumococcal disease while in the ED. In our tertiary care ED we found a very high number of risk factors for complications from influenza and pneumonia and identified many patients with multiple risk factors. Few ED patients have contraindications to influenza and pneumococcus vaccination and many of the reasons for declining vaccination can be addressed through health care provider education and recommendations. ED patients should be considered a target group for vaccination campaigns to prevent serious complications from influenza and pneumococcal disease. 

Acknowledgments
Thank you to Meena Dawar of Vancouver Coastal Health who provided expertise and support from the public health perspective and helped facilitate the implementation of our emergency department vaccination program. 

Competing interests
None declared. 

This article has been peer reviewed.

Dr Taylor is a PGY-5 resident in the Department of Emergency Medicine at the University of British Columbia. Dr Vu is an emergency and critical care physician for Vancouver Coastal Health. He is also a clinical assistant professor in the Faculty of Medicine at the University of British Columbia. Ms Leon Elizalde is an MSc candidate in the School of Population and Public Health at the University of British Columbia. Ms Li-Brubacher is an MSc candidate at the University of British Columbia.