Since December 2020, 3 vaccines have been approved in the United States to prevent severe disease and death caused by COVID-19. The 2-dose vaccinations developed by Pfizer and Moderna are approved for use in individuals aged 18 years and older and under emergency use authorization (EAU) for children aged 6 months to 17 years [1]. The 1-dose Johnson & Johnson vaccination is under EAU for adults aged 18 years and older [1]. Although highly effective [2], vaccination rates remain suboptimal, especially among populations who could benefit most. For example, African Americans comprise 12.4% of the US population but only 10.1% of those who have initiated the series and 10.3% of those who have completed the series as of March 10, 2022 [3]. Despite African Americans being almost 2 times more likely than White Americans to die from COVID-19 [4], vaccine hesitancy remains a major hindrance to reduced vaccine uptake among African Americans [5-9].

Emerging studies demonstrate that vaccine hesitancy is deeply rooted in several overlapping areas: (1) mistrust in health care, government, and research [10-13]; (2) structural racism [14]; and (3) lack of understanding of science related to vaccine-specific issues (eg, efficacy, safety, speed of development) [13,15]. Lack of information, misinformation, and disinformation further drive vaccine hesitancy [10], with social or mass media as the primary source [15]. Because effective communication is necessary to help African Americans make informed decisions about COVID-19 vaccines [16], studies have begun to explore the communication strategies necessary to increase COVID-19 vaccination [17-19]. Trusted messengers are key to COVID-19 information being well received and used [20].

The specific messages related to COVID-19 vaccination are as important as the messenger. Information sources have been developed and disseminated widely to educate communities on messages to use to educate communities on COVID-19 vaccination to increase uptake [21,22]. A few emerging studies have tested messages, including persuasive messaging [18], video-based messages [23], and behavioral nudges [24], on vaccination intention or uptake. African Americans suggest the need for messages that are accurate, targeted, culturally appropriate, and community based [25,26]. However, perceptions of the messages remain unknown, and none (to the best of our knowledge) have actively engaged the African American community to develop or refine the messages on COVID-19 vaccination to ensure cultural appropriateness. Such engagement is key because African Americans’ values and decision-making about the COVID-19 vaccine are strongly shaped by culture [10,12,13,27], and targeting will maximize “fit” of information to an individual’s unique characteristics [28,29].

Message development was guided by 2 psychosocial decision-making models: (1) theory of reasoned action (TRA) [30] and (2) the Health Belief Model (HBM) [31]. These health behavior theoretical models are commonly used to understand vaccination decision-making. The TRA predicts that behavioral intentions to vaccinate against COVID-19 are based on attitudes and subjective norms. The HBM predicts that the likelihood of vaccinating against COVID-19 is based on perceived susceptibility of the individual to SARS-CoV-2, perceived severity of COVID-19, and whether perceived benefits of vaccination outweigh perceived barriers.

Kreuter et al [32] proposed 5 strategies to achieve cultural appropriateness that were used to guide message development. Peripheral strategies increase communication appeal through the title, fonts, colors, and images. Evidential strategies provide data on impact of a health issue in a certain group. Sociocultural strategies address health issues from the social and cultural values of a group. Linguistic strategies fit the program to the native language of a certain group. Lastly, constituent-involving strategies ensure community members’ inclusion in program planning. Developing culturally appropriate, theory-based messages that can be used in communication and behavioral interventions may address concerns about COVID-19 vaccination among African Americans. Because social marketing campaigns have been effective in changing knowledge, attitudes, intentions, and behavior at the community level [33-35], they can disseminate theory-based, culturally appropriate messages and potentially increase COVID-19 vaccination acceptability and uptake. Social marketing is “the application of proven concepts and techniques drawn from the commercial sector to promote changes in diverse socially important behaviors such as drug use, sexual behavior…This marketing approach has an immense potential to affect major social problems if we can only learn how to harness its power” [34].

We describe the development of a theory-based, culturally appropriate library of motivational messages for a social marketing campaign to promote COVID-19 vaccination among African Americans who are vaccine hesitant. Message development occurred in 4 phases: (1) adaptation of a message library based on the literature, (2) review by clinical and research experts for content validation, (3) input and review by a community advisory panel (CAP), and (4) message pretesting via a qualitative study with African American community members to evaluate the accuracy, relevance, and persuasiveness of the messages. The long-term goal is for these messages to be used within interventions aimed at increasing COVID-19 vaccination among African Americans.

We conducted a formative study to design and develop a theory-based, culturally appropriate message library that could be used in behavioral and communication interventions to increase COVID-19 vaccine uptake among African Americans. In the existing COVID-19 library, 1 message subset for adults and 1 for parents and caregivers of adolescents was developed and iteratively adapted by coauthors JCE (a behavioral scientist with a background in biology and vaccine hesitancy) and JD (a basic scientist with a background in infectious disease) on the basis of emerging literature and feedback from over 30 educational sessions provided to communities on COVID-19 and the vaccine. A community-academic partnership was formed to iteratively adapt the message library over a 1-year period to ensure that it was theory based, culturally appropriate, and up to date. Our adaptation process occurred through 6 phases, with community input at each phase. See Figure 1 for a depiction of the 1-year process to yield the final message library of adult and parent message sets.

A community-academic partnership was developed between 2 academic partners, Meharry Medical College and St Jude Children’s Research Hospital, and 1 community partner, the Congregational Health and Education Network (CHEN) in 2021. This partnership yielded an interdisciplinary team of experts in basic science, epidemiology, behavioral science, communication, and community engagement. The purpose of this partnership was (1) to develop messages and products for a social marketing campaign using community engagement principles and (2) to implement and evaluate the social marketing campaign. CHEN, a 501c(3), is a collaboration between Nashville General Hospital at Meharry Medical College and other academic institutions, along with faith-based organizations serving African American and Latino congregants [36]. A focus for CHEN during the COVID-19 pandemic was to improve vaccine health literacy and uptake to reduce health inequities in vaccination and related outcomes. The community partner (CHEN) ensured that the messaging represented the top vaccination concerns and was culturally appropriate. The academic partners offered guidance, when needed, to ensure that the partnership was equitable, and bidirectional engagement occurred in all research phases. They also ensured scientific accuracy of messages.

Members of the research team conducted a literature search to identify reasons for COVID-19 vaccine hesitancy and acceptance, along with potential strategies to improve coverage among African Americans. See Multimedia Appendix 1 for the methods and results of the search. This search was conducted from May 2021 until January 2022 to ensure that the library had the latest updates before the launch of the social marketing campaign. Using the literature, the TRA [30] and the HBM [31], and the experiences and skills of the research team, we adapted the existing message library to ensure accuracy and relevance.

An 8-member CAP was formed to provide feedback on concerns about the vaccine and potential strategies to increase vaccination. This panel meets quarterly and comprises African Americans in the Nashville, Tennessee, metropolitan statistical area. Organizations represented included the Matthew Walker Comprehensive Health Center, Health Leads, the National Association for the Advancement of Colored People, the Community Partners’ Network, and the Nashville Health Disparities Coalition. Additional members included a young adult, a physician, and a parent. A subgroup of the CAP (n=4, 50%) reviewed the content and provided feedback on the messages. The feedback from the meetings and content review from the subgroup were used to iteratively develop the messages and images for the library.

Five experts were identified to review the content for accuracy and relevance. These reviews were conducted strategically due to the ever-changing nature of COVID-19 pandemic updates. Specifically, 2 experts reviewed the content prior to, 1 during, and 3 after community review. Selection criteria for these content reviewers were experience in SARS-CoV-2/COVID-19 research, including vaccination, expertise in the use of psychosocial theory, and willingness and ability to review the items.

Adapted from Lawshe [43], we used qualitative and quantitative methods in a 2-phase content review process. Specifically, each message subset was quantitatively evaluated for relevance by using a 3-point Likert scale: “essential,” “useful but not essential,” and “not necessary.” Then, experts reviewed the messages for perceived accuracy and clarity by providing written comments and edits for improvement. Using these methods, we iteratively refined the messages.

This research was approved by Meharry Medical College’s Institutional Review Board (Protocol 21-03-1076). All participants provided oral informed consent.

The CAP members’ feedback evolved throughout the message library development process. At the beginning of each meeting, the researchers gained insight into the community’s thinking about the COVID-19 vaccine to determine trends in hesitancy. Top concerns included vaccine safety, the speed of vaccine development, mistrust in research and health care, politicization of the vaccine, and conspiracy theories (eg, tracking chip in the vaccine). We then asked about the presentation of messaging. Members suggested that messages should be concise yet comprehensible across reading levels. Members further indicated the need to discuss immediate and long-term benefits and the risks of vaccination so that the community can make an informed decision. A few members further suggested the need to use numbers and images to explain these concerns more clearly.

Select members of the CAP were asked to conduct a detailed review of the messages and images to identify ways to make the content more relatable and comprehensible. Some even provided preferred sources of content (eg, NPR (National Public Radio), which is media organization that seeks to create a more informed public via air, online, or in-person) to help develop the material. Collectively, we used the feedback from the quarterly advisory panel meetings and subgroup review of messages to update the library.

In total, 6 reviewers validated the content. Of these 6, 2 (33%) were White Americans, and 4 (67%) were African Americans; 4 (67%) were female, and 2 (33%) were male. In addition, 1 (17%) reviewer provided feedback for only adult concerns. Reviewers’ areas of expertise included vaccine development, immunology, vaccine-preventable disease and immunizations, and nurse safety. Messages were iteratively updated based on feedback of experts. Most message subsets were classified as essential or useful but not essential. Content edits and additions to the message library reflected the updates on the coronavirus and the vaccine and strategies to ensure comprehensibility and accuracy. Reordering, rephrasing, and adding (eg, analogies) of content were conducted to increase clarity. Because updates are ongoing for COVID-19, content was deemed evidential once primary concepts (eg, process of vaccine development or purpose of boosters) were validated, with the intent to continue to update the library with expert review.

To finalize the library messages, a readability assessment was conducted. We used 3 primary reading grade level assessment tools and an online consensus tool to assess the reading grade level of the adult and parent message sets. First, the Flesch Reading Ease Score [45] was calculated in Microsoft Word. Higher scores indicate easier readability by the user. Second, the Flesch-Kincaid Grade Level [46] was calculated in Microsoft Word to determine a US school grade level. The Flesch Reading Ease and Flesch-Kincaid Grade Levels are calculated by considering the average sentence length (total number of words divided by total number of sentences) and the average syllables per word (total number of syllables divided by the total number of words) using different underlying formulas. Third, the Simple Measure of Gobbledygook (SMOG) index was hand-scored, in addition to being automatically calculated to determine a US [47] school grade level. The SMOG index is calculated using the number of words with multiple syllables in three 10-sentence samples at the beginning, middle, and end of the text. These 3 readability assessment tools all use word difficulty and sentence length as the main factors in determining how easy or how hard the material is to read. Finally, an online readability calculator [48] was used to determine readability consensus. The online calculator applied 7 commonly used readability formulas to provide a consensus rating. See Table 2 for readability assessment results.

The readability assessment revealed acceptable reading grade level scores for all readability formulas applied. The usual recommended reading grade level is fifth-sixth grade to optimize comprehension, according to the American Medical Association and the United States Department of Health and Human Services. However, in combination with the iterative development process and ongoing review by the community partner (CHEN), the CAP, and experts, the messages reflected use of plain language when polysyllabic or complex words were unavoidable.

After completion of the readability assessment, members of the research team conducted a final review to ensure comprehensibility and accuracy. Minor edits were made. See Table 3 for an example of a concern, along with an example of a message for the concern after the adaptation process.

The final message library had 2 message sets, 1 for adults and 1 for parents. There were 18 message subsets for adults and 17 message subsets for parents. These subsets were placed into 3 categories: vaccine development, vaccine safety, and vaccine effectiveness. Each message subset begins with expressing empathy toward the individuals’ concern. Then the facts are provided around each concern, positive or negative. Lastly, the message subset ends with a positive statement related to COVID-19 vaccination that addresses concerns. All message subsets were reviewed by community leaders and members (constituent-involving strategy). We briefly describe each subset for each group next. Each message subset was presented using 3 modes: content, image, and video. See Table 4 to identify each concern, along with key message attributes and the associated strategy to achieve cultural appropriateness.

Our study aimed to develop and validate a message library for a social marketing campaign to increase COVID-19 vaccination among African Americans. The goal was to provide African American adults and parents with theory-based, culturally appropriate messaging on COVID-19 vaccines to motivate vaccine uptake. We described a multiphase process using community engagement approaches for the message library development with the HBM [31], the TRA [30], and Kreuter’s [32] cultural targeting strategies serving as the conceptual frameworks. Our existing library allowed us to expeditiously adapt the messaging to meet the needs of African Americans. This process can be used by researchers and health care professionals to inform the development of culturally appropriate messages.

Use of formative research to build theory-based, culturally appropriate messaging while applying community engagement principles is critical for communities to play an active role in disease prevention and control measures, such as COVID-19 vaccination [26,49]. This method holds great promise in addressing health disparities, yet is in its infancy [50,51]. Applying Boyer et al’s [52] multilevel approach to stakeholder engagement, we had community member involvement at all phases and varying levels to develop a message library to promote COVID-19 vaccination among African Americans. Engagement approaches included formation of a community-academic partnership, a CAP, and inclusion of community interviewees. Having the community-academic partnership and CAP allowed the community voice to be at the root of the messaging. Using each engagement approach, there was a balance of power to ensure that there was bidirectional communication and a deliberative process to foster respect, and even trust in some instances [53]. Furthermore, this process increased the likelihood of achieving cultural appropriateness of the messages.

Content validation has been recognized as a necessary component of message development and is highly valued [54]. The feedback provided by experts in the content review process was used to evolve the library with accurate and relevant messages. Furthermore, the suggestions for modification enriched the messages. Messages were further tested with a purposeful sample of African Americans for cultural appropriateness (ie, evidential, linguistics, peripheral, and sociocultural strategies). Our results indicated that African American adults and parents viewed the messages positively and indicated that the messages were persuasive, useful, and trustworthy. Feedback yielded distinct strategies to increase relevance, comprehension, and appeal. It is important to understand the target audiences’ response early to determine the likelihood of message effectiveness for the intervention [55].

Using this feedback from a multiphase process, our final message library yielded 18 message subsets for adults, and there were 17 message subsets for parents that were grounded in theory and cultural-targeting strategies. There were 3 preferred modes (ie, messages, images, and videos) for African American adults and parents. Studies demonstrate that multiple modes of communication are effective in increasing health literacy among populations, and plain-language messages, pictures, and videos are commonly cited, particularly in the context of community-level interventions [56]. We believe that this approach will be effective in reaching different characteristics of individuals.

A major strength of this study is the use of theory and culturally appropriate strategies inclusive of community engagement to develop the COVID-19 message library for African American adults and parents. We used different levels of community engagement (ie, community-academic partnership, CAP, and interviewees) to ensure that the messages met the needs of our target population. In addition, we equipped the community with information about COVID-19 vaccines to ease concerns postvaccination or to make an informed decision about getting the vaccine. Furthermore, these individuals can now serve as education resources to their communities.

The next step in our partnership will be to test these messages in a 5-month social marketing campaign in a pilot study. Specifically, these messages are used on a website to provide information on COVID-19 vaccines. Shortened versions of these messages are used to market the website. We will evaluate the impact on attitudes, willingness, and self-reported vaccination status to be reported in a future manuscript. If the intervention demonstrates effectiveness, it could prove that theory-based, culturally appropriate messages in a social marketing campaign can be used as a motivational tool among African Americans.

This study has limitations. Messages may not be generalizable to African Americans outside the southeastern United States. We had a small, purposeful sample, yet findings explained diverse perspectives to ensure messages encompassed multiple viewpoints toward the vaccine. There is potential for selection bias among content experts as they are medical professionals and clinicians from different disciplines and with clinical or research expertise. Furthermore, lack of access (ie, geographical barriers) to the vaccine could prevent uptake regardless of other concerns being addressed.

Vaccine hesitancy continues to negatively impact COVID-19 vaccination among African Americans. Effective interventions are needed to increase vaccine uptake. We believe we have developed validated and pretested theory-based, culturally appropriate messages that can be motivational in different interventions aimed at increasing the COVID-19 vaccination rate among African Americans.