Remember to P.A.C.K. for Racially Inclusive Content Strategy: Resource Guide: A starting point for public health practitioners

Dr. Sara Doan (she/her) & Cristy Kennedy (she/her), Kennesaw State University

Date Posted: March 2023

This document serves as an overview of our research project “Ethics of Inclusive Digital Rhetorics: Communicating Public Health on Social Media,” which later informed the formation of our infographic (above) “Remember to P.A.C.K. for Racially Inclusive Content Strategy.” Here, we document the interconnected issues inspiring our study, our methods, findings, and their relevance in the formation of P.A.C.K:
• Provide inclusive charts;
• Avoid poor timing;
• Communicate to individuals; and
• Know your potential biases.
P.A.C.K. and this resource guide are a starting point for more inclusive health communication on Twitter, allowing readers to better apply P.A.C.K. to their own content strategy practices.

Positionality Statement
Sara Doan, Ph.D. is an Assistant Professor of Technical Communication at Kennesaw State University. She applies technical communication and user experience lenses to projects of visual design and health equity. Cristy Kennedy is an interactive design major at Kennesaw State University. Cristy has worked as health communication design intern for COVIDBlack. This project combines Cristy’s expertise as an African American content strategist with Sara’s experience analyzing public health artifacts.

Interconnected Issues
Virtual content strategies associated with the COVID-19 pandemic caused state and national government agencies to emphasize digital health communication (Hope, 2021), prompting users to turn to social media platforms for medical information (Doan, 2021; Walwema, 2021). This surge of social media usage left marginalized groups digitally underserved, highlighting the need for an ethical and universal strategy for sharing health information (Baldwinson, 2018). New and universal content strategies following ethical guidelines would encourage preventive health behaviors in an inclusive and objective manner, minimizing misinformation and rhetorical barriers (Chu et al., 2020).

Our infographic engages these interconnected issues and applies the results found in our 2021 analysis (Kennedy & Doan, 2021), which focused on answering this question: “How do the Twitter content strategies of large public health organizations such as the CDC differ from those of organizations serving African American Audiences?” This infographic links specific characteristics of health organizations to different levels of public outreach, examining ways public health rhetorically excludes minorities through unanticipated digital redlining. Our infographic combats existing inequalities by providing content strategy tips to create inclusive messaging that relies on Kairos, organization size, and localization.

Drawing on Cecilia Shelton’s techné of marginality (2019) and with Guiseppe Getto, Jack Labriola, & Sheryl, Ruszkiewicz’s (2020) definition of content strategy, our study examines how Organizations Serving African American Audiences (OSAAA) use Twitter to clarify misinformation about COVID-19 that circulates within their communities and serve as a model for government organizations. We designed our methodology with an approach that would add to existing studies regarding COVID-19 health messaging strategies (Walwema, 2021) and COVID-19 misinformation on Twitter (Shahi et al., 2021), with a specific focus on inclusiveness in COVID-19 health messaging.

Although this research provided a foundation for understanding the complexities of COVID-19 content strategy (Agley, 2021; Hope, 2021; Shahi et al., 2021; Sleigh et al., 2021; Walwema, 2021), we wanted to interrogate inclusiveness for African Americans in public health tweets. Therefore, we designed a study that could provide a starting point for designing inclusive messaging to address this systemic issue.

Research Questions
These research questions guided our approach:
• Time: How has the time associated with each user’s tweet frequency affected the relevance and impact of information regarding COVID-19?
• Organization Size and Resources: How does the included content differ according to the proposed target audience? Is it relevant?
• Race and Racism: How has measured effectiveness of tweet output differed amongst audiences based on the included categories?

Platforms and Selected Accounts
To choose the social media platform that would yield the greatest results and provide the clearest solution to this issue, we selected Twitter as a research site due its range of equal visibility and content filtration system. Visibility matters when circulating content, as lack of visibility could lead to the news-find-me perception in which users only view health information associated with the users they follow (Agley, 2021). We used NVivo 12 to extract qualitative tweets and tweet objects from OSAAA and public health organizations to compare their content strategies. Qualitative coding was the most effective analysis tool for content strategy due to its ability to directly analyze the user’s tone, messaging, and frequency.

Twitter accounts chosen for analysis and data extraction were selected for comparison, featuring thirteen Twitter accounts that balanced between OSAAA and state health organizations. OSAAA were found and selected by searching the keywords “Black,” “African American,” and “COVID-19” in Twitter’s engine, while state and public health organizations were selected based on which accounts had the highest frequency of tweets. The selected accounts emphasized information for those living in southern states such as Alabama, South Carolina, and Georgia, which allowed us to examine digital communication within states with populations that are disproportionately underserved throughout digital media and news outlets.

Data Collection and Analysis
Our data collection consisted of extracting tweets from the associated Twitter accounts, removing irrelevant content, and then creating several categorized sections for major themes found throughout the data. The dataset initially consisted of 16,325 tweets that were narrowed down to 4,455 after data filtration. The COVID-19 related tweets sourced from these health organizations were all circulated between the months of March and November of 2020 to ensure uniform distribution and accuracy. We used a text search query to filter out content unrelated to COVID-19.

Coding inductively from Joanna Sleigh’s (2021) codebook, and adding codes for race and inclusivity, we analyzed content strategy for public health across 4,455 tweets.

Organizations and Follower Counts as of November 2020
OSAAA: National Organizations: State Organizations:
@BHMinfo – 2,968 @CDCgov – 3,393,197 @ALPublicHealth – 17,935
@BlackInChem – 5,253 @HHSGov – 1,019,538 @GaDPH – 19,935
@BlackInSciComm – 10,680 @PreventCOVID_19 – 269 @scdhec – 26,789
@COVIDBLK – 2,926
@Georgia_NAACP – 13,101
@Morehouse – 4,6224

A total of 4,455 COVID-19 related tweets were coded for the analysis of this qualitative study, leading to a variety of possible categorizations for the results. When evaluating the results of this study, several trends were shown in association with time, size and resources, race, target audience, perceived bias, and engagement. With three types of accounts as this study’s focus, it was expected that certain behaviors would overlap, differ, and contrast.

OSAAA shared the theme of being the accounts with the highest amount of content, as well as the accounts whose Tweet composition were composed of holiday tweets. While OSAAA took into consideration the practices that would best fit the lives of those in marginalized communities, state and national public health organizations ceased to implement such caution.

National public health organizations focused tweets on the most general of audiences, repeatedly mentioning children, pets, and mental health. Despite these practices being derivative, each state faced its own unique struggles during a global pandemic, calling for content that is specialized for the problems faced by their local communities.

The following guidelines are a starting point for creating public health tweets that better meet the needs of African American users:
1. Provide content and charts that include information for all demographics. For example, if you provide a chart of general population infections, provide one with more demographic specific infections as well.
2. Avoid poor timing when uploading or circulating your content. For example, do not stop posting or decrease content during the peak of your targeted crisis.
3. Communicate to your audience as individuals, your audience should not have to do a deep search to find content pertaining to themselves. For example, if you post about food shortages, explain their impacts on states or counties independently.
4. Know your potential biases when uploading content to remain inclusive to all. For example, if you are posting about testing site availability, ask yourself “How does my background or personal perspective change my views on this?” Asking yourself follow-up questions will allow you to think like your audience would and provide them with the proper information.
Continue to educate yourself to avoid microaggressions and provide more inclusive and useful comments.


Agley, Jon. (2021). Expectancy violation and COVID-19 misinformation: A comment on Bogomoletc and Lee’s “Frozen meat against COVID-19 misinformation: An analysis of Steak-umm and positive expectancy violations.” Journal of Business and Technical Communication, 35(4), 496–504.
Baldwinson, Raquel. (2018). Ethics for rhetoric, the rhetoric of ethics, and rhetorical ethics in health and medicine. Rhetoric of Health & Medicine, 1(3–4), 213–238.
Chu, Derek K., Duda, Stephanie., Solo, Karla., Yaacoub, Sally., Schunemann, Holger J., & Akl, Elie A. (2020). Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-COV-2 and COVID-19: A systematic review and meta-analysis. The Lancet, 395(10242), 1973–1987.
Doan, Sara. (2021). Misrepresenting COVID-19: Lying with charts during the second golden age of data design. Journal of Business and Technical Communication, 35(1), 73–79.
Getto, Guiseppe., Labriola, Jack., & Sheryl, Ruszkiewicz. (Eds.). (2020). Content strategy in technical communication (1st ed.). Routledge.
Hope, Lacy. (2021). Protecting pandemic conversations: Tracing Twitter’s evolving content Policies during COVID-19. Journal of Business and Technical Communication, 35(1), 88-93.
Kennedy, Cristy., & Doan, Sara. (2021, February 3). Comparing COVID-19 Messaging between Public Health Accounts and African American Organizations in the Southeast. 10th Annual Symposium on Communicating Complex Information, Old Dominion University.
Shahi, GautamKishore., Dirkson, Anne., & Majchrzak, TimA. (2021). An exploratory study of COVID-19 misinformation on Twitter. Online Social Networks and Media, 22, 100104.
Shelton, Cecilia. (2019). On Edge: A Techné of Marginality [East Carolina University].
Sleigh, Joanna., Amann, Julia., Schneider, Manuel., & Vayena, Effy. (2021). Covid-19 on Twitter: An analysis of risk communication with visuals. BMC Public Health.
Walwema, Josephine. (2021). The WHO health alert: Communicating a global pandemic with WhatsApp. Journal of Business and Technical Communication, 35(1), 35–40.

Author Bios
Sara Doan (she/her) is an assistant professor of technical communication at Kennesaw State University. Her work on data visualizations, preventive health behaviors, and feedback in technical communication courses has appeared in the Journal of Business and Technical Communication, IEEE Transactions on Professional Communication, and Business and Professional Communication Quarterly.

Cristy Kennedy (she/her) is a senior interactive design major at Kennesaw State University. She has experience working as a health communication design intern for COVIDBlack. Her work focuses on using user experience design and content strategy to solve problems regarding healthcare, education, and data.