Clustering of adolescent health behaviours and inequalities across Europe

Background

Health behaviours in adolescence tend to cluster, yet little is known about how these patterns vary across European settings. Identifying behavioural profiles may provide more actionable insights than single-risk approaches. This study aimed to identify clusters of health behaviours and examine their distribution across countries and socioeconomic groups.

Methods

Cross-sectional baseline data from the SUNRISE study were analysed, including 1,979 adolescents with complete data. Behavioural variables included tobacco/vaping, alcohol use, cannabis use, physical activity and dietary habits. Cluster analysis (k-means) was performed to identify behavioural profiles. Associations between clusters, country and family affluence (FAS) were examined using multinomial logistic regression adjusted for age, gender and migrant status.

Results

Four distinct behavioural profiles were identified: low-risk (32%), inactive (28%), substance-use dominant (15%) and multiple-risk (25%). The multiple-risk cluster showed co-occurrence of substance use, physical inactivity and poor diet, with a mean of 3.1 risk behaviours (SD 0.8). In contrast, the low-risk cluster had a mean of 0.6 behaviours (SD 0.5). Cluster membership differed significantly across countries (p < 0.001), with the proportion of adolescents in the multiple-risk cluster ranging from <15% in some regions to > 35% in others. Lower family affluence was associated with higher odds of belonging to the multiple-risk cluster (adjusted OR≈1.8, 95% CI 1.4–2.3). Adolescents in the inactive cluster reported high prevalence of insufficient physical activity (>90%) but low substance use, highlighting heterogeneity across profiles.

Conclusions

Adolescent health behaviours cluster into distinct and socially patterned profiles that vary substantially across countries. Targeting behavioural clusters rather than individual risks may improve the effectiveness of prevention strategies and reduce health inequalities. (GANº 101136829)

Key messages

• Adolescent health behaviours cluster into distinct profiles across Europe, with one in four showing multiple co-occurring risks.

• Lower socioeconomic status is linked to higher odds of multiple-risk behaviours, highlighting inequalities in adolescent health.