A 10-Year Social Media Analysis Exploring Hospital Online Support of Black Lives Matter and the Black Community

IMPORTANCE Tensions around COVID-19 and systemic racism have raised the question: are hospitals advocating for equity for their Black patients? It is imperative for hospitals to be supportive oftheBlackcommunityandacknowledgethemselvesassafespaces,runbycliniciansandstaffwho care about social justice issues that impact the health of the Black community; without the expression of support, Black patients may perceive hospitals as uncaring and unsafe, potentially delaying or avoiding treatment, which can result in serious complications and death for those with COVID-19. OBJECTIVE To explore how hospitals showed public-facing support for the Black community as measured through tweets about social equity or the Black Lives Matter (BLM) movement. DESIGN, SETTING, AND PARTICIPANTS Using a retrospective longitudinal cohort study design, tweets from the top 100 ranked hospitals were collected, starting with the most recent over a 10-year span, from May 3, 2009, to June 26, 2020. The date of the George Floyd killing, May 25, 2020, was investigated as a point of interest. Data were analyzed from June 11 to December 4, 2020. MAIN OUTCOMES AND MEASURES Tweets were manually identified based on 4 categories: BLM, associated with the BLM movement; Black support, expressed support for Black population within the hospital’s community; Black health, pertained to health concerns specific to and the creation of health care for the Black community; or social justice, associated with general social justice terms that were too general to label as Black. If a tweet did not contain any hashtags from these categories, it remained unlabeled.


Introduction
Research has shown that patients who perceive their health care practitioners as concerned and compassionate experience better health outcomes. [1][2][3] To that end, hospitals have increasingly leveraged social media to promote themselves as safe spaces for marginalized populations, such as Black individuals, to receive nondiscriminatory care. [3][4][5][6][7][8] Our study seeks to examine whether hospitals express support for the Black population in relation to social justice issues on the social media platform Twitter.
This approach reflects the dominance of social media in addressing social justice issues online within the United States. In the US, a greater proportion of Black individuals use Twitter relative to White individuals, with the extensive conversations and communities of Black users and Black followers often referred to as Black Twitter. 8,9 The hashtag synonymous with Black social justice issues, #BlackLivesMatter, and its diminutive #BLM, first appeared in 2012, 10 and by 2016, it was the third most used social cause hashtag on the platform, 11 appearing a mean of 17 002 times per day between 2013 and 2018. 12 Although often used following fatal encounters with law enforcement, #BlackLivesMatter also became an important tool to raise awareness around health inequities in Black communities, such as HIV, adequate access to analgesia, and cancer screening. 13,14 In 2020, #BlackLivesMatter took on new urgency owing to the convergence of several public heath crises that disproportionately impacted Black communities, including the COVID-19 pandemic and several high-profile cases of police violence against Black men and women. The Centers for Disease Control and Prevention estimates that cases of COVID-19 are 2.6-fold higher among the Black population than the White population, 15 19,20 and work in essential, public-facing positions with limited social distancing and access to personal protective equipment. 21 The ongoing incidents of police violence have resulted in demonstrations and discussions about the social and health impacts of systemic racism in America. While these events raised awareness, they also appeared to have heightened historic distrust of the medical community among underserved Black communities who expressed concerns about being exploited by researchers testing COVID-19 treatments and vaccines. [22][23][24] It is imperative for hospitals to acknowledge themselves as safe spaces, run by clinicians and staff who care about social justice issues impacting the health of the Black community. Without the promotion of activism for the Black Lives Matter (BLM) movement, Black patients may perceive hospitals as uncaring and unsafe, possibly delaying or avoiding treatment, 25,26 potentially resulting in serious complications and death for those with COVID-19. 27 We explored how hospitals showed public-facing support for the Black community as measured through tweets about social equity and/or the BLM movement. the handle. A total of 90 unique handles were collected from the top 100 hospitals. For hospitals that were part of the same health care system, the same handle represented several hospitals. One hospital did not have a handle linked to their website and was excluded from analysis.
To identify discussions related to the Black population, we reviewed the returned data to collect all hashtags used (including hashtags in retweeted or quoted tweets), conducting a keyword search for Black, AfricanAmerican, and race. Additionally, we conducted searches for specific hashtags related to BLM but not captured by the previous search (eg, BLM and justice to capture hashtags such as racialjustice or justiceforGeorgeFloyd). (2) Black support, expressed support for the Black population within the hospital's community; (3) Black health, pertained to health concerns specific to and the creation of health care for the Black community; and (4) social justice, associated with general social justice terms that were too general to label as Black-focused (eg, "racism," peopleofcolor, and healthjustice).
Both original tweets and retweets were extracted based on these groupings. We pulled all of the hashtags used in the identified tweets of interest, and based on these themes that emerged, we separated hashtags and manually categorized each tweet accordingly. Tweets that contained hashtags from more than 1 category were assigned to multiple categories. The 4 categories and tweet categorization were confirmed by all authors. We again manually reviewed a sample from each hashtag grouping to select the examples of tweets within each category that are provided in Table 1.

Data Analysis
For each category, we explored 4 types of descriptive data. First, how many tweets contained the hashtags that defined the category? Second, how many handles were represented? Third, what were the tweet publication dates? And fourth, what were the most common hashtags?
Analyses were conducted in Python version 2.7.6 (Python Software Foundation). Data were analyzed from June 11 to December 4, 2020.

Results
We collected a total of 281 850 tweets for 90 unique handles for the top 100 hospitals. Each hospital account returned at least 1279 tweets, with 85 handles (94.4%) returning at least 3000 tweets, and 74 handles (82.2%) returning at least 3200 tweets. The earliest tweet collected was published on May 3, 2009, and the latest was on June 26, 2020.

BLM
A total of 274 tweets (0.097%) used a hashtag to support the BLM movement, and a total of 67 handles (74.4%) had tweets in the BLM category. Of those 67 handles, 26 (38.8%) had only 1 tweet in this category. The first tweet was published on April 13, 2018 (5 years after the BLM movement and organization was founded). The date of the George Floyd killing, May 25, 2020, was investigated as a point of interest, and we observed that only 4 tweets that were labeled as BLM predated his killing.
The median date of all tweets labeled within the BLM category was June 5, 2020 (IQR, 3 days). The most common hashtags used among these tweets included #GeorgeFloyd and 4 variations of #BlackLivesMatter, including 2 variations of #whitecoatsforBlacklives.

Summary Results
Results for each category of tweets are summarized in Table 2. Figure 1 shows the number of tweets for each hospital handle for each category. For added context, we plotted each handle's total number of uncategorized tweets on each category's plot. The labeled tweets per hospital ranged from 0 to 46 tweets, and the median date for each category was in 2019, within 1 year of when the data was collected. The uncategorized tweets per hospital ranged from 1279 to 3250 tweets, and the median date was in 2018. Figure 2 shows the cumulative number of tweets across time from May 3, 2009, to June 26, 2020. Again, we added uncategorized tweets to each category's plot for added context.

Discussion
The findings of this cohort study reflect the low signal of tweets regarding the Black community and generalized social justice from hospitals in the US. It is evident from the lack of signal for labeled tweets relative to uncategorized tweets. This lack of discussion surrounding the Black community is a trend across approximately 10 years of tweets across all hospitals within our data set. Of the small percentage of tweets that had Black community or social justice labels, the earliest tweets are at least 6 years later than the earliest tweet published in the data set. This, along with the median dates within a year of collection, indicate that these discussions are relatively recent.
Most hospitals and hospital systems focus their tweets on somatic health. While some hospitals framed COVID-19 as a public health crisis and noted its disparate impact in Black communities, the signal was low. Hospitals tended to separate messages about Black health and COVID-19. Although while 270 tweets occurred after. Hospital participation in the national response to his killing is evidenced by the most popular hashtag in the group (#GeorgeFloyd). It is not possible to determine whether the increases in tweets related to BLM will continue at these hospitals or reflect a response to a specific event. 33 Institutions may consider ongoing use of social justice hashtags to be outside their scope or overly political, but hospitals can powerfully leverage social justice to demonstrate concern for the health and well-being of the Black members of the communities they serve. 33 Whether intentional or not, these top hospitals are not communicating nor demonstrating outwardly that they care about issues and concerns specific to the Black community. Because of this, Black communities may not feel represented at top hospitals. Erasure 34 from public discourse has been shown to be negatively associated with the quality of care delivered to marginalized populations and with the uptake of care by these populations. 35 A 2021 study by Kiang and Tsai 36 examined the public Total amount of tweets per hospital (log 10 ) in each category (orange dots) and uncategorized tweets (blue dots). The uncategorized hospitals are ordered in the assigned Newsweek ranking 28 from left to right. Hospitals without an orange dot did not publish any tweets labeled within that category. communication statements of 56 leading US medical schools after the George Floyd killing. Kiang and Tsai found that most medical schools failed to make reference to the killing of George Floyd by the police and historical racism against Black people. 36 Social justice has been shown to be a powerful tool in addressing health inequities in Black communities, motivating changes in policy to improve public health outcomes [37][38][39] related to access to care, 40,41 quality of care delivery, 42 and overall health care outcomes. 43 In addition, hospitals that leverage social justice messaging and hashtags may be able to further the reach of their tweets and better engage communities. Indeed, some studies, such as a 2018 study by Edrington,44 have suggested that tweets that contain calls to action related to social and health policy are more likely to be engaged and retweeted.
We recognize that our analysis is based only on hashtags and not on text, creating conservative grouping. However, a critical consideration is that general social media etiquette includes hashtags, especially when a user wants to ensure that their message is clearly stated and grouped with messages like it for increased visibility. Hospitals tend to employ professional communications teams versed in social media norms and likely are well versed in labeling messages for public consumption.
While conservative, we believe these tweets capture the social justice messages that hospitals wanted amplified via signal boosting through hashtags.
To our knowledge, this is the first study that has looked at hospital advocacy of Black social justice issues and BLM in particular. Additionally, this study illustrates a baseline of how hospitals are tweeting. While there were few communications about BLM in this set of tweets, our study could provide guidance for further work concerning how online BLM advocacy may be associated with trust or improved health outcomes in the Black population a hospital serves. By providing hypotheses for future studies, our work may advance research to define the importance of trust and how it relates to promotion of Black community and social justice issues. It also may encourage hospitals to expand their social justice advocacy of historically underserved populations in an effort to increase community trust and improve the health outcomes of affected populations.
Further analysis needs to be conducted to find other potential signals and avenues that communicate effective engagement (eg, examining other social media sites or community-based outreach). We hope that our work inspires future research on how a hospital's advocacy and Social justice D outward-facing discussion centered on health disparities and social injustices could help to advocate for and support the Black population in terms of trust and safety. We hope that future analyses contextualize our findings further by exploring whether community-or minority-based hospitals are more likely to engage in social justice, whether there are differences in levels of hospital engagement by minority group, and sentiment and linguistic properties that extend beyond hashtags. Eventually, it is our aim that this analysis will be expanded to include more US hospitals varying in conventional ranking and community demographics served.

Limitations
This study has some limitations. It is possible that we would have seen more hospitals engaged in these topics if we had expanded our method to have less conservative grouping. We also may not have included all hashtags associated with BLM (eg, #protest, #policebrutality). Again, we argue that if a hospital wanted to publicly engage in these topics, they would have used hashtags central to movements to take advantage of signal boosting. 45,46 We also recognize that we have included only the top 100 ranked hospitals according to Newsweek; thus, our results may not be generalizable to all hospitals in the US. We did not include community-or minority-based hospitals as a comparison to explore if these hospitals that do not rank high in conventional US hospital rankings would rank high in terms of advocacy.

Conclusions
The findings of this cohort study help to shed light on how hospitals use social media and hashtags to address social justice issues impacting the Black communities they serve. It is important for health organizations to realize how they engage with their communities via social media. These brief engagements can communicate an institution's dedication to the health and well-being of the populations they serve. Such engagements are particularly important among underserved and marginalized groups, which historically have not trusted mainstream health organizations and are, in general, heavy users of communications platforms. Health organizations that thoughtfully address social justice issues online may be able to build trust and bolster uptake of their services among these groups and help to improve their health outcomes overall.