To overcome vaccine hesitancy, messaging and expertise matters

Race concordance and acknowledgment of historical injustice increase ratings of doctors among recent Black flu vaccine users, but only non-expert messengers changed vaccine behavior among those who have never been vaccinated

Reviewed by Penny Sun


Despite the demonstrated benefits of preventive medicine, only 45% of American adults typically get a flu shot during flu season. Vaccine hesitancy is particularly common among Black and white lower socioeconomic status men, who don’t trust doctors and are skeptical of the benefits relative to the perceived risk. For Black Americans, this mistrust is further rooted in historical and current medical racism. Given both the disproportionate impact of COVID-19 on disadvantaged communities and the need for mass vaccination, this study seeks to understand what messages will resonate or backfire among these particularly vaccine-hesitant populations and whether messaging alone is enough to induce vaccine uptake.

Previous research has shown that small interventions like prompting patients to write down the date and time of their vaccine appointment and increasing accessibility through on-site clinics at schools and workplaces are effective at increasing vaccination rates. Drawing on these principles, the authors tested whether messaging could improve vaccine uptake among Black and white men without a college education. Specifically, they tested the effect of three key variables: the race and perceived medical expertise of messengers and acknowledging historical medical injustice within the content. This study is particularly timely due to patterns of vaccine hesitancy and disproportionate risk for and impact of COVID-19 on disadvantaged communities.

Dr. Marcella Alsan, MD, PhD, MPH is a Professor of Public Policy at Harvard Kennedy School. Her research applies microeconomics to study health inequality. Dr. Sarah Eichmeyer, PhD, Msc is an Assistant Professor of Economics at the Ludwig-Maximilians University of Munich. Her research interests are at the intersection of public economics and the economics of digitization, particularly the impacts of digital technologies on political polarization, health, and education. 

Methods and Findings

The researchers surveyed 2,893 non-Hispanic Black and white male adults with no college education who had not yet received the flu vaccine during the 2019-2021 flu seasons. Participants were asked about their baseline attitudes and beliefs about the flu vaccine, then randomly assigned to watch videos with one of 10 messengers reading text about vaccine safety and effectiveness. Messages were identical, except some included a mention of historical racial injustices and others did not. Messengers also varied by race and their perceived medical expertise, signaled by their clothing. 

After viewing the video, participants reported their perception of their messenger’s trustworthiness and qualification, perception of the effectiveness and relevance of the message, level of attention they paid to the statement, their belief in the safety of the flu vaccine before and after the statement, interest in free flu shot coupons,  intent on getting the flu vaccine before and after the message, and intent to get the COVID-19 vaccine. Participants then received a trackable coupon for a free flu shot. Finally, participants described their actual vaccine uptake behavior and vaccine safety beliefs in a follow-up survey at least two weeks later. 

Key Findings:

  • Black participants rated messenger and message effectiveness higher for Black messengers, but this racial concordance did not affect content recall, safety beliefs, or flu vaccination coupon interest. When white “expert” messengers acknowledged historical medical injustice, Black participants also perceived their message to be more effective and had higher rates of COVID-19 vaccine intent, but not flu vaccine uptake. The race of the messenger had no impact on any of the outcome measures for white participants, but only white participants listening to Black messengers exited the survey without completing the study. 
  • Overall, participants who watched videos with “non-expert” messengers rated them substantially less qualified to give medical advice, but remembered more of the message content, were more willing to get the COVID-19 vaccine, and more likely to follow through with actually getting the flu vaccine. White expert messengers had significantly lower flu vaccine uptake and COVID-19 vaccine intent among Black participants, a particularly alarming figure given that 75% of Black patients have a racially discordant doctor.
  • Participants’ past vaccine-seeking behaviors also influenced the relative effectiveness of different message interventions. Only Black participants who got a flu vaccine within the last two years were influenced by race concordance and acknowledgment of historical medical injustice. Black participants who never received a flu vaccine rated “non-expert” messengers as significantly more effective.  They also reported substantially higher flu vaccine intent after seeing “non-expert” messengers than those who had a vaccine in the last two years.


These findings provide a roadmap to design effective campaigns and change vaccine beliefs and behaviors among specific demographics. For audiences who previously used vaccines, race concordant and empathetic “experts” are likely the best messengers. However, for audiences that have historically not been vaccinated, peer figures such as community health workers or citizen ambassadors may be the most effective messengers. Based on these findings, an effective vaccine messaging campaign should include messengers from various racial backgrounds and levels of expertise. 


Thank you for visiting RRAPP

Please help us improve the site by answering three short questions.