EUVIMEDEUVIMED entdecken
Open Access

Problematic Social Media Use Profiles and Adolescent Substance Use Across Countries: A Latent Class Analysis With Multilevel Models

Wei Zhong iD, Naiqi Tian iD

Journal of Adolescence · 2026-08 · 98 (6) · 2032-2045

DOI: 10.1002/jad.70201 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Adolescence, online erschienen am 08.06.2026. Verfasst von Wei Zhong und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Sociology, Xiamen University, Xiamen, Fujian, China. Hauptthemen laut MEDLINE: Social Media, Substance-Related Disorders, Adolescent Behavior. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 70 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

While substantial evidence suggests a significant association between problematic social media use (PSMU) and adolescent substance use, the complex nature of PSMU and its cross‐national variations remain unclear.

Objectives

This study aimed to identify distinct patterns of PSMU, compare their prevalence across countries, and examine the associations between these patterns and adolescent substance use.

Methods

This study analyzed a representative adolescent sample ( N = 157,717) from 39 countries, utilizing data from the 2017–2018 Health Behaviour in School‐aged Children (HBSC) survey. Latent class analysis (LCA) was conducted to identify different patterns of PSMU. Multilevel logistic regression models were then used to investigate the associations between these identified PSMU patterns and adolescent substance use, specifically focusing on smoking, drunkenness, and cannabis use.

Results

Five distinct patterns of PSMU were identified: Non‐problematic users (52.60%), Preoccupation‐driven users (8.42%), conflictual‐compulsive users (11.37%), regulation‐failure users (25.16%), and highly‐problematic users (2.45%). The distribution of PSMU patterns varied across countries and regions, suggesting cross‐national heterogeneity rather than a clear geographical or economic gradient. Consistent patterns of association were found between adolescent PSMU and substance use. Highly‐Problematic Users showed the strongest associations with smoking, drunkenness, and cannabis use, followed by conflictual‐compulsive users and preoccupation‐driven users, while regulation‐failure users exhibited the weakest association.

Conclusions

These findings underscore the importance of developing tailored interventions for adolescent PSMU that are based on distinct patterns. Furthermore, special attention should be given to the diverse associations of these PSMU patterns with adolescent substance use.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Social MediaSubstance-Related DisordersAdolescent BehaviorHumansAdolescentFemaleLatent Class AnalysisMaleMultilevel AnalysisSmokingPrevalence

Stichwörter der Autor:innen

latent class analysismultilevel analysisproblematic social media usesubstance use

Autor:innen und Institutionen

  • Department of Sociology, Xiamen University, Xiamen, Fujian, China.
  • Department of Social Work, The Chinese University of Hong Kong, Hong Kong, China.

ORCID-Profile: Wei Zhong, Naiqi Tian

Zitieren

APA

Zhong, W., & Tian, N. (2026). Problematic Social Media Use Profiles and Adolescent Substance Use Across Countries: A Latent Class Analysis With Multilevel Models. Journal of Adolescence, 98(6), 2032-2045. https://doi.org/10.1002/jad.70201

Vancouver

Zhong W, Tian N. Problematic Social Media Use Profiles and Adolescent Substance Use Across Countries: A Latent Class Analysis With Multilevel Models. Journal of Adolescence. 2026;98(6):2032-2045. doi:10.1002/jad.70201

BibTeX
@article{Zhong2026,
  author = {Zhong, Wei and Tian, Naiqi},
  title = {Problematic Social Media Use Profiles and Adolescent Substance Use Across Countries: A Latent Class Analysis With Multilevel Models},
  journal = {Journal of Adolescence},
  year = {2026},
  volume = {98},
  number = {6},
  pages = {2032--2045},
  doi = {10.1002/jad.70201}
}
Kostenlos starten →