A novel household‐based patient outreach pilot program to boost late‐season influenza vaccination rates during the COVID‐19 pandemic

Abstract Background The objective of this study was to test a novel household‐based approach to improve late‐season influenza vaccine uptake during the 2020–2021 season, using Epic's MyChart patient portal messages and/or interactive voice response telephone calls. Methods This study was a non‐blinded, quality improvement program using a block randomized design conducted among patients from Reliant Medical Group clinics residing in a traditional household (≥2 individuals clinically active in the Reliant system living at the same address). Households were randomized 1:1:1 into intervention arms: non‐tailored communication (messaging based on CDC's seasonal influenza vaccination campaign), tailored communication (comprehensive communication including reinforcement of the importance of influenza vaccination for high‐risk individuals), and standard‐of‐care control. Influenza vaccination during the program was captured via medical records, and the odds of vaccination among communication arms versus the control arm were assessed. A survey assessing influenza vaccination drivers was administered using MyChart. Results Influenza vaccination increased by 3.3% during the program period, and no significant differences in vaccination were observed in intervention arms relative to the control arm. Study operationalization faced substantial challenges related to the concurrent COVID‐19 pandemic. Compared with vaccinated survey respondents, unvaccinated respondents less frequently reported receiving a recommendation for influenza vaccination from their healthcare provider (15.8% vs. 42.3%, p < 0.001) or awareness that vaccination could protect themselves and higher risk contacts (82.3% vs. 92.6%, p < 0.001). Conclusions No significant effects of the interventions were observed. Survey results highlighted the importance of healthcare provider recommendations and the need for increased education around the benefits of vaccination.

Influenza vaccination is the most effective method to prevent influenza infections and associated hospitalizations and deaths.
During the 2019-2020 season, an estimated 7.5 million cases, 105,000 influenza-related hospitalizations, and 6300 influenzarelated deaths were averted due to influenza vaccination in the United States. 3 Despite a multitude of efforts to improve the uptake of seasonal influenza vaccines, [4][5][6][7] rates have remained stagnant in the United States over the past 10 years. 8 While rates among children and individuals >65 years have come close to the national target of 70%, those among younger adults have consistently lagged behind. 9 In the wake of the COVID-19 pandemic, public health experts urged the importance of influenza vaccination to mitigate the increased burden of the "twindemic" to the population and healthcare system. The COVID-19 pandemic also marked a turning point in the perception of public health, with an increased emphasis on protection of those around you and thinking in terms of units (i.e., household and family), with regard to virus transmission. This unique situation highlights the importance of both addressing stagnant influenza vaccination rates and ensuring that vaccine uptake is sustained throughout the latter part of the season, a time when uptake traditionally slows, yet influenza can still pose a substantial risk. 10,11 Leveraging the public's recently heightened awareness of infectious disease and vaccination, we sought to explore a novel household-based approach to improve late-season influenza vaccine uptake during the 2020-2021 season, using messages from the

| Program cohorts and randomization
Eligible households were stratified based on risk, which included a high-risk cohort (≥1 household member with a risk factor for severe influenza-or COVID-19-related complications) and a nonhigh-risk cohort (no household members with known risk factors).
High-risk conditions were based on current problem lists (Table S1). Age (i.e., <2 or ≥65 years) alone was not considered as a risk factor in this stratification. Eligible households were random- for all eligible households ( Figure 1).

| Program period
The program period was defined as the period from the initiation of household communication through the end of follow-up and was scheduled to begin on December 1, 2020 ( Figure 1). Due to pandemic-related operational challenges, the start of the program period was delayed until January 13, 2021. Individuals were followed through March 31, 2021 to ascertain influenza vaccination status.

| Communications
Non-tailored and tailored communications were sent to eligible household members via MyChart or IVR beginning January

| Participant survey
To determine the factors that influenced program participants' decision on whether to get vaccinated or not, a survey was administered to all active MyChart participants. Completed surveys were linked to the participant's EHR using an encrypted identifier.

| Ethics approval and patient consent
Prior to data collection, the study was approved by the WCG Institu- 3 | RESULTS

| Randomization and program population characteristics
A total of 36,920 households (94,747 individuals) were initially randomized ( Figure S1). The proportion of households included in the initial randomization from each Reliant clinic was balanced across program arms (Table S3). Given the delay in the program start date, the number of eligible households (i.e., ≥ 1 household member unvaccinated at program start) decreased to 27,658 households (72,208 individuals) ( Figure 2). After excluding ineligible households, the balance between program arms and across clinics remained comparable to that of the initially randomized population (Table S3).    Table 3).

| Participant survey
Among self-reported unvaccinated respondents, 42.1% cited a lack of concern of their risk of influenza infection as the reason for not getting a vaccination. Self-reported vaccinated respondents were more likely to indicate that protecting others around them had a major effect on their

| DISCUSSION
The objective of this study was to develop and evaluate a novel household-based recall-reminder intervention to increase late-season influenza vaccine uptake, leveraging pre-existing technological systems within care delivery organizations. During the program period, influenza vaccination rates among unvaccinated individuals in eligible households increased by only 3.3%, highlighting the degree to which vaccine uptake tapers off in the latter part of the influenza season.
There were no significant differences observed in uptake between the intervention arms relative to the SoC control arm, indicating that the tailored and non-tailored communications did not have an effect.
It is important, however, to interpret these results within the context in which this study was conducted. The concurrent COVID-19 pandemic provided substantial challenges hindering the effectiveness of the intervention. During the nearly 2-month delay between randomization and program start, 11.1% of unvaccinated individuals in the randomized population received an influenza vaccine, rendering more than 25% of households ineligible. Furthermore, the program

DATA AVAILABILITY STATEMENT
The data contained in our database contains proprietary elements owned by Optum and, therefore, cannot be broadly disclosed or made publicly available at this time. The disclosure of this data to third-party clients assumes certain data security and privacy protocols are in place and that the third-party client has executed our standard license agreement which includes restrictive covenants governing the use of the data.

SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.