Co-occurrence of habit-forming risk behaviors and their socio-demographic, health status and lifestyle determinants: a population-based cross-sectional study

Junko Kose; Pauline Duquenne; Serge Hercberg; Pilar Galan; Mathilde Touvier; Léopold K Fezeu; Valentina A Andreeva

doi:10.1186/s13690-024-01251-2

Co-occurrence of habit-forming risk behaviors and their socio-demographic, health status and lifestyle determinants: a population-based cross-sectional study

Arch Public Health. 2024 Feb 28;82(1):26. doi: 10.1186/s13690-024-01251-2.

Authors

Junko Kose¹, Pauline Duquenne¹, Serge Hercberg^{1

2}, Pilar Galan¹, Mathilde Touvier¹, Léopold K Fezeu¹, Valentina A Andreeva³

Affiliations

¹ Nutritional Epidemiology Research Group (EREN), Epidemiology and Statistics Research Center, Sorbonne Paris Nord University, INSERM U1153, INRAE U1125, CNAM, University of Paris (CRESS), Bobigny, France.
² Department of Public Health, AP-HP Paris Seine-Saint-Denis Hospital System, Bobigny, France.
³ Nutritional Epidemiology Research Group (EREN), Epidemiology and Statistics Research Center, Sorbonne Paris Nord University, INSERM U1153, INRAE U1125, CNAM, University of Paris (CRESS), Bobigny, France. v.andreeva@eren.smbh.univ-paris13.fr.

Abstract

Background: Although habit-forming risk behaviors frequently co-occur, determinants of concurrent risk behaviors have rarely been investigated. The aim of the present study was to investigate socio-demographic, health status, and lifestyle determinants of single versus concurrent risk behaviors in general-population adults.

Methods: We analyzed data from 32,622 participants (74.5% female; mean age = 57.9 ± 14.2 years) of the NutriNet-Santé cohort who completed the Alcohol Use Disorders Identification Test, the 12-item Cigarette Dependence Scale, the modified Yale Food Addiction Scale 2.0, and the Internet Addiction Test in 2021-2022. Using established cutoffs, participants were first split into 2 groups (presence versus absence) for each risk variable (alcohol use disorders, nicotine dependence, food addiction, Internet addiction) and were then divided into 3 groups (no risk behavior, 1 risk behavior (reference), and ≥ 2 risk behaviors). The association between socio-demographic, health status, and lifestyle exposures and individual/concurrent risk behaviors were investigated with polytomous logistic regression.

Results: Younger age (Odds Ratio (OR) = 2.04; 95% Confidence Interval (CI: 1.62-2.56), current financial difficulties (OR = 1.29; CI: 1.08-1.54), self-perceived poor health (OR = 1.70; CI: 1.32-2.20), overall poor dietary quality (OR = 2.88; CI: 2.06-4.02), being underweight (OR = 1.46; CI: 1.05-2.04), having obesity (OR = 1.62; CI: 1.31-1.99), lack of affection during childhood (OR = 1.41; CI: 1.18-1.69), and a lifetime prevalence or medication use for a mental disorder (OR = 1.46; CI: 1.24-1.73) were positively associated with having ≥ 2 versus 1 risk behavior (all p < 0.05). The comparison of none versus 1 risk behavior revealed the same determinants in addition to having a higher education, being physically active at work, and being overweight.

Conclusions: We investigated determinants of concurrent habit-forming risk behaviors among adults in a large, population-based study. The findings could serve as impetus for future research in this domain and ultimately help guide addiction prevention efforts.

Keywords: Addictive behaviors; Alcohol; Food addiction; General population; Internet addiction; Substance use disorders.

Grants and funding

ANR-21-CE36-0003/Agence Nationale de la Recherche