April 4-6, 2024 • Hyatt Regency • Lexington, KY
Innovations in Health Communication
Innovations in Health Communication
Abstract: Opioid-Related News Articles and Stigmatizing Messages: A Content Analysis of National U.S. News Coverage from 2017-2022
In tandem with rising opioid overdose deaths, opioid-related news coverage has increased (Hswen et al., 2020). Recognizing that such coverage has been stigmatizing, the Associated Press (AP) updated their guidelines in 2017 to recommend journalists avoid terms that may create stigma like “addict,” “junky,” and “abuser.” Despite this updated guideline, national news media still utilize these stigmatizing labels that describe people in terms of only their opioid use (Bessette et al., 2022; McGinty et al., 2019). However, two notable gaps persist in our understanding of stigmatizing news content about opioids. First, research has primarily focused on stigma labels without attention to other stigma message types. As the model of stigma communication (MSC; Smith, 2007; Smith et al., 2019) argues, people may also communicate stigma marks that evoke disgusting imagery (e.g., “needle track marks”), stigma responsibility messages that communicate control of the stigmatized condition (e.g., “it’s the person’s fault they got addicted”), and stigma peril messages that describe the harm people with a stigmatized condition pose to others (e.g., “people who use opioids are a moral stain on society”). These four message features—labels, marks, responsibility, and peril—can prompt stigmatizing responses (Ledford et al., 2021), but research has not documented the presence of these messages in the news. Second, since the AP’s updated guidelines, research has not analyzed the use of opioid-related stigma messages over time. This study subsequently aims to chart a timeline of stigmatizing news coverage of opioids from 2017-2022.
Proportional stratified random sampling was used to extract a sample of opioid-related newspaper coverage from June 1, 2017 (after AP’s update) to December 31, 2022. The search was conducted among three of the five largest U.S. newspapers by circulation: The New York Times, The Washington Post, and USA Today. Each newspaper was searched using a string of Boolean search terms that identified the many types of opioids, and 11, 216 articles were retrieved. The population was then stratified proportionally both by year of coverage and newspaper as the researchers selected a 25% sample, totaling 2, 804 articles for coding.
After achieving intercoder reliability between two coders (Krippendorf’s α > .7 for all variables), preliminary, descriptive analysis of the first 10 percent of the sample (280 articles) suggested that labels (e.g., “addict,” “abuser,” “junkie”) were the most prominent stigma type, appearing in 59 of the coded articles (21%), followed by mark messages (n = 11, 4%), responsibility messages (n = 5, 2%), and peril messages (n = 4, 1%). While preliminary, this analysis suggests that national news coverage has still not met the AP’s update substance abuse reporting criteria and that additional types of stigma messages get circulated in the news. By KCHC 2024, all 2,804 articles will be coded and segmented by year to offer more substantive analysis of trends. In addition, researchers will also code the data to test whether news articles that include narratives are less likely to include stigmatizing message features given the power of narratives to reduce stigma (Heley et al., 2020).
Proportional stratified random sampling was used to extract a sample of opioid-related newspaper coverage from June 1, 2017 (after AP’s update) to December 31, 2022. The search was conducted among three of the five largest U.S. newspapers by circulation: The New York Times, The Washington Post, and USA Today. Each newspaper was searched using a string of Boolean search terms that identified the many types of opioids, and 11, 216 articles were retrieved. The population was then stratified proportionally both by year of coverage and newspaper as the researchers selected a 25% sample, totaling 2, 804 articles for coding.
After achieving intercoder reliability between two coders (Krippendorf’s α > .7 for all variables), preliminary, descriptive analysis of the first 10 percent of the sample (280 articles) suggested that labels (e.g., “addict,” “abuser,” “junkie”) were the most prominent stigma type, appearing in 59 of the coded articles (21%), followed by mark messages (n = 11, 4%), responsibility messages (n = 5, 2%), and peril messages (n = 4, 1%). While preliminary, this analysis suggests that national news coverage has still not met the AP’s update substance abuse reporting criteria and that additional types of stigma messages get circulated in the news. By KCHC 2024, all 2,804 articles will be coded and segmented by year to offer more substantive analysis of trends. In addition, researchers will also code the data to test whether news articles that include narratives are less likely to include stigmatizing message features given the power of narratives to reduce stigma (Heley et al., 2020).