Abstract Body (Do not enter title and authors here): Background: Persistent racial and ethnic disparities exist in influenza vaccination coverage. Increasing uptake in this population could potentially reduce influenza infection and associated complications in a at-risk population.

Purpose: To explore differences in the effectiveness of behavioral nudges on influenza vaccination rates according to ethnicity.

Methods: We conducted a secondary analysis of the nationwide, pragmatic, registry-based, randomized, implementation trial NUDGE-FLU-CHRONIC, which evaluated the effect of 6 different electronic nudging letter strategies on influenza uptake compared with usual care among persons aged 18-64 years, with a chronic disease associated with an increased risk of adverse influenza-related outcomes. Ethnicity was categorized into two groups, white or non-white. Effect modification by ethnicity was tested on the absolute scale. Absolute differences and relative risks (RR) for each comparison were calculated at a significance level of 0.0071.

Results: Of 299,881 participants (53.2% female, median age 52.0 years [IQR, 39.8-59.0]) randomized, 90.4% were white. Vaccination uptake was significantly lower compared with non-whites (22.0% vs. 37.7%, p<0.001). White residents were older (median [IQR] age: 52.5 [40.5, 59.2] vs. 46.6 [36.0, 55.8]), and had a higher prevalence of heart failure (4.0% vs 2.7%, p<0.001), atrial fibrillation (6.9% vs 3.2%, p<0.001, and hypertension (36.2% vs 28.9, p<0.001), whereas non-whites had a higher prevalence of ischemic heart disease (10.5% vs. 8.9%, p<0.001), and diabetes (24.6% vs. 18.7%, p<0.001). White residents were more likely to be educated, be employed, and had higher income (all p<0.001).

Each letter successfully increased influenza vaccination irrespective of ethnicity; however, the magnitude of effectiveness of each letter was significantly modified by ethnicity (Figure 1; all p<0.001), such that effectiveness was less pronounced among non-whites with the largest difference in effect size observed with the repeated letter and the smallest difference in effect size observed with the cardiovascular gain letter.

Conclusion: Electronic nudging letters consistently led to clinically relevant relative gains in vaccine uptake, irrespective of ethnicity. However, absolute gains from these nudging strategies were somewhat attenuated in non-white residents. These results highlight the need for greater emphasis on vaccine implementation strategies tailored according to ethnicity.