Alex M Russell, Danny Valdez, Mingxuan Wang, Jon-Patrick Allem, Brandon G Bergman, John F Kelly, Dana M Litt, Philip M Massey
{"title":"Content analysis of substance use disorder recovery discourse on Twitter: From personal recovery narratives to marketing of addiction treatment.","authors":"Alex M Russell, Danny Valdez, Mingxuan Wang, Jon-Patrick Allem, Brandon G Bergman, John F Kelly, Dana M Litt, Philip M Massey","doi":"10.1111/acer.15531","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Substance use disorder (SUD) is a prodigious public health issue characterized by a substantial treatment gap. Despite challenges, millions have resolved a prior significant alcohol or drug problem, increasingly using online supports as a part of their recovery efforts (e.g., virtual mutual-help group meetings, traditional social networking sites [SNS]). However, the content surrounding SUD recovery-related discussion on SNS such as Twitter remains largely unexamined. To fill this gap, we explored public tweets using SUD recovery-related hashtags.</p><p><strong>Methods: </strong>From January 1, 2022, to December 31, 2022, 455,968 public tweets were collected using SUD recovery-related hashtags. Natural language processing was used to identify and remove irrelevant groupings of tweets from the dataset, resulting in a final corpus of 186,460 tweets. A random subsample of 1800 tweets was extracted for content analysis, involving codebook development, manual annotation by trained coders, and inter-rater reliability assessment (average Cohen's κ = 0.77).</p><p><strong>Results: </strong>Nearly half (41.7%) of SUD recovery-related posts were from individuals in or seeking recovery, while 21.3% originated from addiction treatment industry accounts. Common themes included addiction treatment marketing (27.4%; some of which promoted scientifically unsupported products or services), emotional support (15.6%), celebrating a recovery milestone (15.4%), alcohol/drug-related sociopolitical commentary (14.7%), expressions of gratitude (11.5%), and mutual-help group participation (8.7%).</p><p><strong>Conclusions: </strong>SUD recovery-related content on Twitter reflected individuals seeking social support during efforts to initiate or maintain recovery. However, these accounts may be met with marketing material from entities that misrepresent their services or promote products based on unsubstantiated claims. Stricter (or enforcement of existing) regulations may be warranted to protect vulnerable SNS platform users from entities seeking to exploit them for financial gain.</p>","PeriodicalId":72145,"journal":{"name":"Alcohol (Hanover, York County, Pa.)","volume":" ","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Alcohol (Hanover, York County, Pa.)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1111/acer.15531","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"SUBSTANCE ABUSE","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Substance use disorder (SUD) is a prodigious public health issue characterized by a substantial treatment gap. Despite challenges, millions have resolved a prior significant alcohol or drug problem, increasingly using online supports as a part of their recovery efforts (e.g., virtual mutual-help group meetings, traditional social networking sites [SNS]). However, the content surrounding SUD recovery-related discussion on SNS such as Twitter remains largely unexamined. To fill this gap, we explored public tweets using SUD recovery-related hashtags.
Methods: From January 1, 2022, to December 31, 2022, 455,968 public tweets were collected using SUD recovery-related hashtags. Natural language processing was used to identify and remove irrelevant groupings of tweets from the dataset, resulting in a final corpus of 186,460 tweets. A random subsample of 1800 tweets was extracted for content analysis, involving codebook development, manual annotation by trained coders, and inter-rater reliability assessment (average Cohen's κ = 0.77).
Results: Nearly half (41.7%) of SUD recovery-related posts were from individuals in or seeking recovery, while 21.3% originated from addiction treatment industry accounts. Common themes included addiction treatment marketing (27.4%; some of which promoted scientifically unsupported products or services), emotional support (15.6%), celebrating a recovery milestone (15.4%), alcohol/drug-related sociopolitical commentary (14.7%), expressions of gratitude (11.5%), and mutual-help group participation (8.7%).
Conclusions: SUD recovery-related content on Twitter reflected individuals seeking social support during efforts to initiate or maintain recovery. However, these accounts may be met with marketing material from entities that misrepresent their services or promote products based on unsubstantiated claims. Stricter (or enforcement of existing) regulations may be warranted to protect vulnerable SNS platform users from entities seeking to exploit them for financial gain.