"I Been Taking Adderall Mixing it With Lean, Hope I Don't Wake Up Out My Sleep": Harnessing Twitter to Understand Nonmedical Prescription Stimulant Use among Black Women and Men Subscribers.
Joni-Leigh Webster, Sahithi Lakamana, Yao Ge, Abeed Sarker
{"title":"\"I Been Taking Adderall Mixing it With Lean, Hope I Don't Wake Up Out My Sleep\": Harnessing Twitter to Understand Nonmedical Prescription Stimulant Use among Black Women and Men Subscribers.","authors":"Joni-Leigh Webster, Sahithi Lakamana, Yao Ge, Abeed Sarker","doi":"10.1101/2024.12.03.24318408","DOIUrl":null,"url":null,"abstract":"<p><p>Black women and men outpace other races for stimulant-involved overdose mortality despite lower lifetime use. Growth in mortality from prescription stimulant medications is increasing in tandem with prescribing patterns for these medications. We used Twitter to explore nonmedical prescription stimulant use (NMPSU) among Black women and men using emotion and sentiment analysis, and topic modeling. We applied the NRC Lexicon and VADER dictionary, and LDA topic modeling to examine feelings and themes in conversations about NMPSU by gender. We paid attention to the ability of natural language processing techniques to detect differences in emotion and sentiment among Black Twitter subscribers given increased mortality from stimulants. We found that, although emotion and sentiment outcomes match the directionality of emotions and sentiment observed (i.e., Black Twitter subscribers use more positive language in tweets), this belies limitations of NRC and VADER dictionaries to distinguish feelings for Black people. Even still, LDA topic models showcased the relevance of hip-hop, dependence on NMPSU, and recreational use as consequential to Black Twitter subscribers' discussions. However, gender shaped the relevance of these topics for each group. Greater attention needs to be paid to how Black women and men use social media to discuss important topics like drug use. Natural language processing methods and social media research should include larger proportions of Black, Hispanic/Latinx, and American Indian populations in development of emotion and sentiment lexicons, otherwise outcomes regarding NMPSU will not be generalizable to populations writ large due to cultural differences in communication about drug use online.</p>","PeriodicalId":94281,"journal":{"name":"medRxiv : the preprint server for health sciences","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11643189/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"medRxiv : the preprint server for health sciences","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1101/2024.12.03.24318408","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Black women and men outpace other races for stimulant-involved overdose mortality despite lower lifetime use. Growth in mortality from prescription stimulant medications is increasing in tandem with prescribing patterns for these medications. We used Twitter to explore nonmedical prescription stimulant use (NMPSU) among Black women and men using emotion and sentiment analysis, and topic modeling. We applied the NRC Lexicon and VADER dictionary, and LDA topic modeling to examine feelings and themes in conversations about NMPSU by gender. We paid attention to the ability of natural language processing techniques to detect differences in emotion and sentiment among Black Twitter subscribers given increased mortality from stimulants. We found that, although emotion and sentiment outcomes match the directionality of emotions and sentiment observed (i.e., Black Twitter subscribers use more positive language in tweets), this belies limitations of NRC and VADER dictionaries to distinguish feelings for Black people. Even still, LDA topic models showcased the relevance of hip-hop, dependence on NMPSU, and recreational use as consequential to Black Twitter subscribers' discussions. However, gender shaped the relevance of these topics for each group. Greater attention needs to be paid to how Black women and men use social media to discuss important topics like drug use. Natural language processing methods and social media research should include larger proportions of Black, Hispanic/Latinx, and American Indian populations in development of emotion and sentiment lexicons, otherwise outcomes regarding NMPSU will not be generalizable to populations writ large due to cultural differences in communication about drug use online.
"我一直在服用阿德拉,并将其与精益混合,希望我不会从睡梦中醒来":利用 Twitter 了解黑人妇女和男子非医疗处方兴奋剂的使用情况》(Harnessing Twitter to Understand Nonmedical Prescription Stimulant Use among Black Women and Men Subscribers.