Identifying Hidden Barriers to PrEP Adherence Among Young Men Who Have Sex with Men: Application of Natural Language Processing.
In the United States, the incidence rate of new HIV diagnoses continues to increase among young men who have sex with men (YMSM). Despite the availability of effective preventive strategies such as pre-exposure prophylaxis (PrEP), the rate of PrEP adherence remains markedly low especially among YMSM...
| Publicado en: | AIDS & Behavior Vol. 30; no. 1; pp. 253 - 263 |
|---|---|
| Autores principales: | , , , , , , , |
| Formato: | research tables/charts Journal Article |
| Publicado: |
Springer Nature
Jan2026
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=190984217&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 190984217 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 10907165 G1T jtl: AIDS & Behavior issn: 10907165 maglogo: N pubinfo: dt: Jan2026 vid: 30 iid: 1 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 190984217 187378900 190984217 190984217 10.1007/s10461-025-04863-z 190984217 ppf: 253 ppct: 10 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Identifying Hidden Barriers to PrEP Adherence Among Young Men Who Have Sex with Men: Application of Natural Language Processing. aug: au: Min, Se Hee Scroggins, Jihye Kim Duncan, Dustin T. Garofalo, Robert Janulis, Patrick Francis Kuhns, Lisa Xiao, Fengdi Schnall, Rebecca affil: https://ror.org/00b30xv10 University of Pennsylvania School of Nursing, 418 Curie Blvd, 19104, Philadelphia, PA, USA sug: subj: HIV Infections Prevention and Control Pre-Exposure Prophylaxis Medication Compliance Men Who Have Sex With Men Psychosocial Factors Natural Language Processing Healthcare Disparities Human Male Adult United States Secondary Analysis Prospective Studies Risk Assessment Insurance, Health Social Determinants of Health Patient Dropouts Machine Learning Support Vector Machine Random Forest Thematic Analysis ROC Curve Statistical Significance Data Analysis Software Descriptive Statistics Funding Source Adult: 19-44 years Male ab: In the United States, the incidence rate of new HIV diagnoses continues to increase among young men who have sex with men (YMSM). Despite the availability of effective preventive strategies such as pre-exposure prophylaxis (PrEP), the rate of PrEP adherence remains markedly low especially among YMSM. Previous studies have primarily relied on structured surveys with predefined responses to identify barriers to PrEP adherence which may not fully capture the complexities behind the barriers. This is a secondary data analysis of data from a prospective cohort study focusing on YMSM vulnerable to HIV. A total of 581 participants provided free-text responses regarding reasons for discontinuing PrEP, which served as the primary outcome for the analysis. Natural language processing was conducted to identify potential barriers to PreP adherence and to uncover any previously unidentified barriers in this population. A total of nine categories were identified, with the most prevalent being lack of sexual activity (n = 128), followed by issues related to monogamy/partnership/long-term relationship (n = 73), specific insurance or coverage issues (n = 52), medication-related concerns (n = 41), side effects/health concerns (n = 39), forgetfulness/inconvenience associated with the medication regimen (n = 33), limited healthcare access (n = 26), personal reasons (n = 9), and financial insecurity (n = 8). The NLP analysis demonstrated moderate performance via support vector machine, random forest, gradient boost, and random forest (F-score = 0.75). Our study provides critical insights into specific barriers faced by high-risk YMSM, emphasizing the need for development of targeted interventions aimed at these barriers to improve PrEP access and utilization. pubtype: Academic Journal doctype: research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|