Related Experiment Video
Updated: Jan 14, 2026

Using Cholesky Decomposition to Explore Individual Differences in Longitudinal Relations between Reading Skills
Published on: September 17, 2019
Social determinants of clusters of health behaviours: a longitudinal cohort study using latent- class analysis
Olivia McEvoy1, Richard Layte2
1Department of Sociology, Trinity College Dublin, Dublin, Dublin 2, D02, Ireland. olmcevoy@tcd.ie.
Background:
Much of the burden of non-communicable disease in high-income countries is attributable to harmful engagement with health behaviours. Evidence suggests that these behaviours may be better understood, and more amenable to intervention, if analysed as clusters of behaviours rather than individual behaviours. To date, research on the clustering of health behaviours has focused on the consequences of multi-engagement as opposed to the potential shared antecedent or contemporaneous processes. Moreover, adolescence is a critical period for the development of health behaviours, yet there remain more studies on the clustering of health behaviours in adulthood, when patterns of behaviour are likely already crystallised.
Methods:
Drawing from a nationally representative longitudinal dataset of young people living in the Republic of Ireland, latent- class analysis was used to identified clusters of the health behaviours: diet quality, levels of physical activity, screentime, tobacco smoking and underage alcohol consumption, for the first time in this context. We theorised potential social determinants for engagement with the distinct clusters of behaviours and evaluated these theories empirically using multinominal logistic regression.
Results:
Our findings indicate that, what are typically termed, 'lifestyle' behaviours (diet, physical activity and screentime) tend to co-occur rather than appear in isolation. These lifestyle behaviours appear to be relatively independent of tobacco smoking and underage alcohol consumption, which we term 'substance use.' We demonstrate that the experience of periodic, but mostly persistent poverty (approx. nine years), is associated with membership of clusters characterised by lifestyle behaviours, while reporting oppositional values is distinctly associated with membership of clusters characterised by substance use.
Conclusions:
Our results support the much-established relationship between income poverty and harmful health behaviours related to lifestyle that is evidenced when looking at these behaviours individually and thereby provides further empirical evidence for the theories on 'fundamental causes' of health inequalities and the 'social exclusion framework.' We find an association between oppositional values and substance use behaviours, and furthermore that oppositional values tend to be highest among the income poorest adolescents. We highlight a concurrent influence for substance use in adolescence, which is also relevant for addressing inequalities in health behaviour engagement.
More Related Videos
Related Concept Videos
Longitudinal Research
Longitudinal Studies
Lifestyle Factors and Health
Benefits of Physical Activity
Physical activity, whether through structured exercise or casual activities like walking, biking, or dancing, is a cornerstone of a...
Study Designs in Epidemiology
Observational studies are those where the researcher does not intervene but rather observes natural variations. They include cross-sectional, cohort, and...
Bias in Epidemiological Studies
Factors Affecting Illness
For instance, risk factors are connected to illness,...

