Vaccine-preventable diseases such as influenza, pneumococcal disease, and shingles lead to significant rates of illness, hospitalization, and death among older adults. In the United States, vaccination rates have been mostly unchanged for a decade with lower rates among racial and ethnic minority groups. New and scalable approaches are needed to address this important public health issue. Nudges are changes to the way choices are offered or information is framed that can have outsized effects on behavior. For example, default options are the path of least resistance and the action that takes place if no alternatives are selected. Active choice is a method that prompts a decision-making now, rather than waiting for stakeholders to recognize the need to make the decision on their own. Our groups at the University of Pennsylvania (Penn Medicine) and the University of Washington (UW Medicine) have formed behavioral design teams embedded within the operations of our health systems and have demonstrated how these types of nudges can improve health care value and patient outcomes. Since electronic health records (EHRs) have been adopted by more than 90% of clinicians in the US, this scalable technology platform is an optimal environment to implement and deploy these types of nudges. In this study, we propose to design, test, and implemented personalized nudges to clinicians and patients to target barriers among high-risk subgroups to improve vaccination rates. We will pilot this at two health systems and then implement a pragmatic trial at those health systems and sites in the VA Health System. In the R61 phase, we will focus on the following aims at Penn Medicine and UW Medicine.
Aim 1 : To use EHR data and analytical methods to identify high-risk groups of older adults with suboptimal vaccination rates for influenza, pneumococcal disease, and herpes zoster.
Aim 2 : To assess the feasibility of implementing different types of personalized nudges to clinicians and patients to target the identified groups to improve vaccination rates among older adults.
Aim 3 : To pilot test ways to personalize promising nudges to clinicians and patients for improving vaccination among older adults. In the R33 phase, we will focus on the following aims at Penn Medicine, UW Medicine, and the VA Health System.
Aim 1 : To conduct a 12-month, multisite, cluster randomized, pragmatic trial to evaluate the effectiveness of personalized nudges to clinicians and patients relative to control to improve vaccination rates among older adults.
Aim 2 : To evaluate the effectiveness of the intervention on reducing disparities in vaccination rates related to race/ethnicity and socioeconomic factors.
Aim 3 : To evaluate heterogeneity in treatment effect across clinician, patient, and practice characteristics to further tailor approaches in future intervention design.
Many older adults are at risk of illness, hospitalization, and death from vaccine-preventable diseases. Moreover, there are significant racial and ethnic disparities in care with even lower vaccination rates among African Americans, Hispanics, and Asians. In this study, we will use analytical methods to identify high-risk subgroups, test personalized nudges to address barriers among these groups, and then implement a pragmatic, multisite trial to evaluate the effectiveness of personalized nudges to clinicians and patients to improve vaccination rates among older adults and reduce disparities in care.
