A Decade of Tweets: Visualizing Racial Sentiments Towards Minoritized Groups in the United States Between 2011 and 2021.

BACKGROUND: Research has demonstrated the negative impact of racism on health, yet the measurement of racial sentiment remains challenging. This article provides practical guidance on using social media data for measuring public sentiment. METHODS: We describe the main steps of such research, including data collection, data cleaning, binary sentiment analysis, and visualization of findings. We randomly sampled 55,844,310 publicly available tweets from 1 January 2011 to 31 December 2021 using Twitter's Application Programming Interface. We restricted analyses to US tweets in English using one or more 90 race-related keywords. We used a Support Vector Machine, a supervised machine learning model, for sentiment analysis. RESULTS: The proportion of tweets referencing racially minoritized groups that were negative increased at the county, state, and national levels, with a 16.5% increase at the national level from 2011 to 2021. Tweets referencing Black and Middle Eastern people consistently had the highest proportion of negative sentiment compared with all other groups. Stratifying temporal trends by racial and ethnic groups revealed unique patterns reflecting historical events specific to each group, such as the killing of George Floyd regarding sentiment of posts referencing Black people, discussions of the border crisis near the 2018 midterm elections and anti-Latinx sentiment, and the emergence of COVID-19 and anti-Asian sentiment. CONCLUSIONS: This study demonstrates the utility of social media data as a quantitative means to measure racial sentiment over time and place. This approach can be extended to a range of public health topics to investigate how changes in social and cultural norms impact behaviors and policy.A supplemental digital video is available at http://links.lww.com/EDE/C91.

Examining Twitter-Derived Negative Racial Sentiment as Indicators of Cultural Racism: Observational Associations With Preterm Birth and Low Birth Weight Among a Multiracial Sample of Mothers, 2011-2021.

BACKGROUND: Large racial and ethnic disparities in adverse birth outcomes persist. Increasing evidence points to the potential role of racism in creating and perpetuating these disparities. Valid measures of area-level racial attitudes and bias remain elusive, but capture an important and underexplored form of racism that may help explain these disparities. Cultural values and attitudes expressed through social media reflect and shape public norms and subsequent behaviors. Few studies have quantified attitudes toward different racial groups using social media with the aim of examining associations with birth outcomes. OBJECTIVE: We used Twitter data to measure state-level racial sentiments and investigate associations with preterm birth (PTB) and low birth weight (LBW) in a multiracial or ethnic sample of mothers in the United States. METHODS: A random 1% sample of publicly available tweets from January 1, 2011, to December 31, 2021, was collected using Twitter's Academic Application Programming Interface (N=56,400,097). Analyses were on English-language tweets from the United States that used one or more race-related keywords. We assessed the sentiment of each tweet using support vector machine, a supervised machine learning model. We used 5-fold cross-validation to assess model performance and achieved high accuracy for negative sentiment classification (91%) and a high F1 score (84%). For each year, the state-level racial sentiment was merged with birth data during that year (~3 million births per year). We estimated incidence ratios for LBW and PTB using log binomial regression models, among all mothers, Black mothers, racially minoritized mothers (Asian, Black, or Latina mothers), and White mothers. Models were controlled for individual-level maternal characteristics and state-level demographics. RESULTS: Mothers living in states in the highest tertile of negative racial sentiment for tweets referencing racial and ethnic minoritized groups had an 8% higher (95% CI 3%-13%) incidence of LBW and 5% higher (95% CI 0%-11%) incidence of PTB compared to mothers living in the lowest tertile. Negative racial sentiment referencing racially minoritized groups was associated with adverse birth outcomes in the total population, among minoritized mothers, and White mothers. Black mothers living in states in the highest tertile of negative Black sentiment had 6% (95% CI 1%-11%) and 7% (95% CI 2%-13%) higher incidence of LBW and PTB, respectively, compared to mothers living in the lowest tertile. Negative Latinx sentiment was associated with a 6% (95% CI 1%-11%) and 3% (95% CI 0%-6%) higher incidence of LBW and PTB among Latina mothers, respectively. CONCLUSIONS: Twitter-derived negative state-level racial sentiment toward racially minoritized groups was associated with a higher risk of adverse birth outcomes among the total population and racially minoritized groups. Policies and supports establishing an inclusive environment accepting of all races and cultures may decrease the overall risk of adverse birth outcomes and reduce racial birth outcome disparities.

Risks and Opportunities to Ensure Equity in the Application of Big Data Research in Public Health.

The big data revolution presents an exciting frontier to expand public health research, broadening the scope of research and increasing the precision of answers. Despite these advances, scientists must be vigilant against also advancing potential harms toward marginalized communities. In this review, we provide examples in which big data applications have (unintentionally) perpetuated discriminatory practices, while also highlighting opportunities for big data applications to advance equity in public health. Here, big data is framed in the context of the five Vs (volume, velocity, veracity, variety, and value), and we propose a sixth V, virtuosity, which incorporates equity and justice frameworks. Analytic approaches to improving equity are presented using social computational big data, fairness in machine learning algorithms, medical claims data, and data augmentation as illustrations. Throughout, we emphasize the biasing influence of data absenteeism and positionality and conclude with recommendations for incorporating an equity lens into big data research.

Estimated Travel Time and Spatial Access to Abortion Facilities in the US Before and After the Dobbs v Jackson Women's Health Decision.

Importance: Abortion facility closures resulted in a substantial decrease in access to abortion care in the US. Objectives: To investigate the changes in travel time to the nearest abortion facility after the Dobbs v Jackson Women's Health Organization (referred to hereafter as Dobbs) US Supreme Court decision. Design, Setting, and Participants: Repeated cross-sectional spatial analysis of travel time from each census tract in the contiguous US (n = 82 993) to the nearest abortion facility (n = 1134) listed in the Advancing New Standards in Reproductive Health database. Census tract boundaries and demographics were defined by the 2020 American Community Survey. The spatial analysis compared access during the pre-Dobbs period (January-December 2021) with the post-Dobbs period (September 2022) for the estimated 63 718 431 females aged 15 to 44 years (reproductive age for this analysis) in the US (excluding Alaska and Hawaii). Exposures: The Dobbs ruling and subsequent state laws restricting abortion procedures. The pre-Dobbs period measured abortion access to all facilities providing abortions in 2021. Post-Dobbs abortion access was measured by simulating the closure of all facilities in the 15 states with existing total or 6-week abortion bans in effect as of September 30, 2022. Main Outcomes and Measures: Median and mean changes in surface travel time (eg, car, public transportation) to an abortion facility in the post-Dobbs period compared with the pre-Dobbs period and the total percentage of females of reproductive age living more than 60 minutes from abortion facilities during the pre- and post-Dobbs periods. Results: Of 1134 abortion facilities in the US (at least 1 in every state; 8 in Alaska and Hawaii excluded), 749 were considered active during the pre-Dobbs period and 671 were considered active during a simulated post-Dobbs period. Median (IQR) and mean (SD) travel times to pre-Dobbs abortion facilities were estimated to be 10.9 (4.3-32.4) and 27.8 (42.0) minutes. Travel time to abortion facilities in the post-Dobbs period significantly increased (paired sample t test P <.001) to an estimated median (IQR) of 17.0 (4.9-124.5) minutes and a mean (SD) of and 100.4 (161.5) minutes. In the post-Dobbs period, an estimated 33.3% (sensitivity interval, 32.3%-34.8%) of females of reproductive age lived in a census tract more than 60 minutes from an abortion facility compared with 14.6.% (sensitivity interval, 13.0%-16.9%) of females of reproductive age in the pre-Dobbs period. Conclusions and Relevance: In this repeated cross-sectional spatial analysis, estimated travel time to abortion facilities in the US was significantly greater in the post-Dobbs period after accounting for the closure of abortion facilities in states with total or 6-week abortion bans compared with the pre-Dobbs period, during which all facilities providing abortions in 2021 were considered active.

Association of "#covid19" Versus "#chinesevirus" With Anti-Asian Sentiments on Twitter: March 9-23, 2020.

Objectives. To examine the extent to which the phrases, "COVID-19" and "Chinese virus" were associated with anti-Asian sentiments.Methods. Data were collected from Twitter's Application Programming Interface, which included the hashtags "#covid19" or "#chinesevirus." We analyzed tweets from March 9 to 23, 2020, corresponding to the week before and the week after President Donald J. Trump's tweet with the phrase, "Chinese Virus." Our analysis focused on 1 273 141 hashtags.Results. One fifth (19.7%) of the 495 289 hashtags with #covid19 showed anti-Asian sentiment, compared with half (50.4%) of the 777 852 hashtags with #chinesevirus. When comparing the week before March 16, 2020, to the week after, there was a significantly greater increase in anti-Asian hashtags associated with #chinesevirus compared with #covid19 (P < .001).Conclusions. Our data provide new empirical evidence supporting recommendations to use the less-stigmatizing term "COVID-19," instead of "Chinese virus."

Leveraging black-market street buprenorphine pricing to increase capacity to treat opioid addiction, 2010-2018.

Increasing capacity to provide buprenorphine, a treatment for opioid addiction, can help mitigate the opioid epidemic in the United States. This study models black-market pricing of buprenorphine to better understand supply and demand for opioid addiction treatment. A mixed effects linear model was used to quantify the effect of county-level racial composition, health insurance coverage, and drug characteristics on price variation. From November 2010 to June 2018, there were 2481 submissions for street buprenorphine transactions in the StreetRx dataset. The mean price was $3.95/mg (SD = $23.12/mg). Price decreased 3.05% each year and was highest in the summer and spring. Brand name buprenorphine was on average 11.18% more expensive than generic buprenorphine. Buprenorphine/naloxone combinations were on average 19.75% less expensive than pure buprenorphine. Purchases in bulk were on average 10.51% cheaper than purchases not in bulk. Street buprenorphine in film form was on average 14.34% more expensive than in pill/tablet form. Buprenorphine street price was 17.12% higher in spring and 22.26% higher in summer compared to fall. For every percentage point increase in percent white, buprenorphine sold for 0.88% higher price. For every percentage point increase in health insurance coverage, street buprenorphine sold for 0.02% lower price. Findings demonstrate that geographic, demographic, and socioeconomic factors shape the diversion of opioid addiction treatment to the black-market. Buprenorphine street pricing can help estimate public need, gaps in care and emerging public health priorities.

Evaluation of Volume of News Reporting and Opioid-Related Deaths in the United States: Comparative Analysis Study of Geographic and Socioeconomic Differences.

BACKGROUND: News media coverage is a powerful influence on public attitude and government action. The digitization of news media covering the current opioid epidemic has changed the landscape of coverage and may have implications for how to effectively respond to the opioid crisis. OBJECTIVE: This study aims to characterize the relationship between volume of online opioid news reporting and opioid-related deaths in the United States and how these measures differ across geographic and socioeconomic county-level factors. METHODS: Online news reports from February 2018 to April 2019 on opioid-related events in the United States were extracted from Google News. News data were aggregated at the county level and compared against opioid-related death counts. Ordinary least squares regression was used to model opioid-related death rate and opioid news coverage with the inclusion of socioeconomic and geographic explanatory variables. RESULTS: A total of 35,758 relevant news reports were collected representing 1789 counties. Regression analysis revealed that opioid-related death rate was positively associated with news reporting. However, opioid-related death rate and news reporting volume showed opposite correlations with educational attainment and rurality. When controlling for variation in death rate, counties in the Northeast were overrepresented by news coverage. CONCLUSIONS: Our results suggest that regional variation in the volume of opioid-related news reporting does not reflect regional variation in opioid-related death rate. Differences in the amount of media attention may influence perceptions of the severity of opioid epidemic. Future studies should investigate the influence of media reporting on public support and action on opioid issues.

Investigation of Geographic and Macrolevel Variations in LGBTQ Patient Experiences: Longitudinal Social Media Analysis.

BACKGROUND: Discrimination in the health care system contributes to worse health outcomes among lesbian, gay, bisexual, transgender, and queer (LGBTQ) patients. OBJECTIVE: The aim of this study is to examine disparities in patient experience among LGBTQ persons using social media data. METHODS: We collected patient experience data from Twitter from February 2013 to February 2017 in the United States. We compared the sentiment of patient experience tweets between Twitter users who self-identified as LGBTQ and non-LGBTQ. The effect of state-level partisan identity on patient experience sentiment and differences between LGBTQ users and non-LGBTQ users were analyzed. RESULTS: We observed lower (more negative) patient experience sentiment among 13,689 LGBTQ users compared to 1,362,395 non-LGBTQ users. Increasing state-level liberal political identification was associated with higher patient experience sentiment among all users but had stronger effects for LGBTQ users. CONCLUSIONS: Our findings highlight that social media data can yield insights about patient experience for LGBTQ persons and suggest that a state-level sociopolitical environment influences patient experience for this group. Efforts are needed to reduce disparities in patient care for LGBTQ persons while taking into context the effect of the political climate on these inequities.

Racial and Ethnic Disparities in Patient Experiences in the United States: 4-Year Content Analysis of Twitter.

BACKGROUND: Racial and ethnic minority groups often face worse patient experiences compared with the general population, which is directly related to poorer health outcomes within these minority populations. Evaluation of patient experience among racial and ethnic minority groups has been difficult due to lack of representation in traditional health care surveys. OBJECTIVE: This study aims to assess the feasibility of Twitter for identifying racial and ethnic disparities in patient experience across the United States from 2013 to 2016. METHODS: In total, 851,973 patient experience tweets with geographic location information from the United States were collected from 2013 to 2016. Patient experience tweets included discussions related to care received in a hospital, urgent care, or any other health institution. Ordinary least squares multiple regression was used to model patient experience sentiment and racial and ethnic groups over the 2013 to 2016 period and in relation to the implementation of the Patient Protection and Affordable Care Act (ACA) in 2014. RESULTS: Racial and ethnic distribution of users on Twitter was highly correlated with population estimates from the United States Census Bureau's 5-year survey from 2016 (r2=0.99; P<.001). From 2013 to 2016, the average patient experience sentiment was highest for White patients, followed by Asian/Pacific Islander, Hispanic/Latino, and American Indian/Alaska Native patients. A reduction in negative patient experience sentiment on Twitter for all racial and ethnic groups was seen from 2013 to 2016. Twitter users who identified as Hispanic/Latino showed the greatest improvement in patient experience, with a 1.5 times greater increase (P<.001) than Twitter users who identified as White. Twitter users who identified as Black had the highest increase in patient experience postimplementation of the ACA (2014-2016) compared with preimplementation of the ACA (2013), and this change was 2.2 times (P<.001) greater than Twitter users who identified as White. CONCLUSIONS: The ACA mandated the implementation of the measurement of patient experience of care delivery. Considering that quality assessment of care is required, Twitter may offer the ability to monitor patient experiences across diverse racial and ethnic groups and inform the evaluation of health policies like the ACA.

Feasibility of using social media to monitor outdoor air pollution in London, England.

Air pollution is a serious public health concern. Innovative and scalable methods for detecting harmful air pollutants such as PM2.5 are necessary. This study assessed the feasibility of using social media to monitor outdoor air pollution in an urban area by comparing data from Twitter and validating it against established air monitoring stations. Data were collected from London, England from July 29, 2016 to March 17, 2017. Daily mean PM2.5 data was downloaded from the LondonAir platform consisting of 26 air pollution monitoring sites throughout Greater London. Publicly available tweets geo-located to Greater London containing air pollution terms were captured from the Twitter platform. Tweets with media URL links were excluded to minimize influence of news stories. Sentiment of the tweets was examined as negative, positive, or neutral. Cross-correlation analyses were used to compare the relationship between trends of tweets about air pollution and levels of PM2.5 over time. There were 16,448 tweets without a media URL link, with a mean of 498.42 (SD = 491.08) tweets per week. A significant cross-correlation coefficient of 0.803 was observed between PM2.5 data and the non-media air pollution tweets (p < 0.001). The cross-correlation coefficient was highest between PM2.5 data and air pollution tweets with negative sentiment at 0.816 (p < 0.001). Discussions about air pollution on Twitter reflect particle PM2.5 pollution levels in Greater London. This study highlights that social media may offer a supplemental source to support the detection and monitoring of air pollution in a densely populated urban area.

How AI Assistants Could Help Answer Patients' Messages-and Potentially Improve Their Outcomes.

This Medical News article is an interview with John Ayers, PhD, MA, vice chief of innovation in the Division of Infectious Diseases & Global Public Health at the University of California, San Diego, the lead author of a recent study on chatbot responses to patient questions.

How Do Policymakers Regulate AI and Accommodate Innovation in Research and Medicine?

In this Medical News article, JAMA Editor in Chief Kirsten Bibbins-Domingo, PhD, MD, MAS, and Alondra Nelson, PhD, the Harold F. Linder Professor at the Institute for Advanced Study, discuss effective AI regulation frameworks to accommodate innovation.

How to Navigate the Pitfalls of AI Hype in Health Care.

In this Medical News article, Arvind Narayanan, PhD, a professor of computer science at Princeton University, discusses the benefits of using artificial intelligence in research and clinical settings while remaining cautious of hype, biases, and data privacy issues.

AI Developers Should Understand the Risks of Deploying Their Clinical Tools, MIT Expert Says.

This Medical News article is an interview with Marzyeh Ghassemi, a machine learning expert at the Massachusetts Institute of Technology who focuses on health care applications, and JAMA Editor in Chief Kirsten Bibbins-Domingo.

Tapping AI's Strengths-From Operating Room Safety to Wearable Device Interpretation.

In this Medical News interview, Sachin Kheterpal, the University of Michigan Medical School's associate dean for research information technology, joins JAMA Editor in Chief Kirsten Bibbins-Domingo to discuss AI's number-crunching potential for improving patient care.

How Artificial Intelligence Will Enhance Imaging Access and Analysis.

This Medical News article is an interview with Saurabh Jha, a cardiothoracic radiologist and an associate professor of radiology at the University of Pennsylvania, and JAMA Editor in Chief Kirsten Bibbins-Domingo.

US Surgeon General Vivek Murthy: Firearm Violence Is a Public Health Crisis.

This Medical News article is an interview with US Surgeon General, Vivek Murthy, MD, MBA, and JAMA Editor in Chief Kirsten Bibbins-Domingo, PhD, MD, MAS, about a new advisory that declares gun violence a public health crisis.

"Scalable Privilege"-How AI Could Turn Data From the Best Medical Systems Into Better Care for All.

This Medical News article is an interview with JAMA Editor in Chief Kirsten Bibbins-Domingo and physician Atul Butte, the University of California Health System's chief data scientist.

Blind Spots, Shortcuts, and Automation Bias-Researchers Are Aiming to Improve AI Clinical Models.

This Medical News article is an interview with University of Michigan computer scientist Jenna Wiens, whose research interests lie at the intersection of AI and health care, and JAMA Editor in Chief Kirsten Bibbins-Domingo.

How Machine Learning Might Help Improve Cardiac Imaging.

In this Medical News interview, University of California, San Francisco, cardiologist Rima Arnaout, joins JAMA Editor in Chief Kirsten Bibbins-Domingo to discuss the transformative potential of AI on cardiac imaging.