Research

Research Spotlight

Behaviour Change Frameworks and Youth Health Behaviours

What We Looked At

This review examined how behaviour change frameworks (BCFs) have been applied to interventions targeting health behaviours in children and adolescents, with the goal of understanding what this evidence base might offer for reducing young people’s engagement with harmful online content (HOC). The frameworks considered include the COM-B model (Capability, Opportunity, Motivation), the Behaviour Change Wheel (BCW), the Theoretical Domains Framework (TDF), the Theory of Planned Behaviour (TPB), and Intervention Mapping (IM).

The review focused on four domains where BCFs have been applied with young people: sexual health and disease prevention; smoking, vaping, and alcohol use; gambling; and mental health, particularly peer support and help-seeking. It also examined the role of gatekeepers, including parents, educators, and clinicians, as behaviour change targets in their own right. Studies were drawn from diverse international settings.

What the Research Found

Across all areas reviewed, a consistent finding is that providing information alone is insufficient to change behaviour. Young people may understand the risks associated with unprotected sex, smoking, alcohol use, or gambling, yet still engage in those behaviours. This knowledge-behaviour gap (Rozendaal, 2022) reflects the fact that behaviour is shaped not only by what people know, but by their motivations, their social environments, and the practical opportunities available to them. BCFs provide a structured way of identifying which of these factors is most relevant for a given population, so that interventions can be designed accordingly.

In sexual health, school-based programmes that go beyond information provision to address self-efficacy, social norms, and practical skills have shown improvements in sexual health knowledge and, to a lesser extent, behaviour (Niland et al., 2024). Digital interventions have also demonstrated moderate gains in knowledge and some behavioural improvement, though effects tend to diminish over time and do not always translate into reduced STI diagnoses (Mo et al., 2025; Zhu et al., 2026). Personalised, theory-informed approaches appear more effective than generic educational programmes because they address the specific barriers and motivations relevant to individual users (Davidovich and Dall, 2012). Research consistently identifies peer influence, fear of judgement, access to services, and concerns about stigma as important barriers (Flood et al., 2024; Newby et al., 2017).

Substance use research applying COM-B and the TDF found that smoking, vaping, and alcohol use among young people arise from interactions between limited self-regulation, peer influence, access to substances, and social motivations such as belonging and confidence (Lakshmi et al., 2023; Vasiliou et al., 2021; Palmer et al., 2023). Interventions incorporating social support, self-monitoring, goal-setting, and personalised feedback showed the most consistent results, particularly in digital formats (Boumparis et al., 2026; Kiselev et al., 2025). Multicomponent programmes that addressed individual behaviour alongside school policy and the social environment were also more effective than single-focus approaches (Andersen et al., 2023; Ba-Break et al., 2023). Even so, results across the literature are mixed: some studies report significant improvements while others find no lasting effect, and effects often decline at longer follow-up (Lu et al., 2025; O’Logbon et al., 2023). Implementation quality, teacher confidence, and cultural relevance all appear to affect whether an intervention works in practice (Nolan and Gutman, 2023).

Application of BCFs to adolescent gambling remains more limited. Most existing research has used the TPB rather than COM-B or BCW, and the focus has been on attitudes and intentions rather than the wider determinants of behaviour. One longitudinal study using COM-B found that gambling develops through a self-reinforcing cycle in which early curiosity and social acceptance gradually give way to habitual behaviour, suggesting that interventions are most likely to be effective in the early stages before habits are established (Botella-Guijarro et al., 2022). Evidence for effective gambling interventions in this age group is currently weak (St-Pierre et al., 2017).

In the area of mental health, BCFs have been applied to understand what influences young people’s engagement with digital mental health services, including peer support platforms, webchat counselling, and online forums. Studies found that anonymity reduces stigma and encourages disclosure, while accessible platform design strengthens young people’s practical ability to engage with services (Ananya et al., 2024; Lim et al., 2025). Social opportunity, in the form of positive peer interactions and responsive moderators, also emerged as an important facilitator. Evidence for effectiveness is promising but mixed: peer support may enhance engagement and perceived support, though its impact on sustained mental health outcomes remains uncertain (Ali et al., 2015; Birrell et al., 2023). Research on moderators of online forums highlights that those managing sensitive content also require training, supervision, and structured support (Perowne and Gutman, 2022).

On the role of gatekeepers, evidence across multiple settings shows that parents, educators, and clinicians face greater capability and opportunity barriers than motivational ones: they may want to support young people but lack the knowledge, training, or organisational resources to do so effectively (Leather et al., 2022; Lau-Zhu et al., 2022). This has direct relevance to online harms, where adults’ ability to recognise risks, facilitate disclosure, and respond appropriately similarly depends on knowledge, skills, and supportive environments. A systems-based perspective, drawing on Bronfenbrenner’s Bioecological Model (1979), suggests that effective interventions need to address behaviour change across families, schools, and wider services, not only among young people themselves.

What This Means in Practice

The accumulated evidence across these domains points consistently towards theory-informed, multicomponent approaches as the most effective design for behaviour change interventions with young people. Single-strategy programmes, whether information-based or purely digital, tend to produce modest and short-lived effects. Interventions are more likely to succeed when they address capability, opportunity, and motivation simultaneously and when they are tailored to the specific barriers faced by the target population rather than assuming that raising awareness will be sufficient.

For work addressing young people’s engagement with harmful online content, this evidence base suggests several practical directions. Interventions will need to go beyond education to address the social motivations and peer dynamics that make harmful content appealing or difficult to avoid. Digital delivery offers scalability and accessibility, but should be designed to maintain engagement over time and should complement, rather than replace, relational support. Involving young people in the design of interventions and ensuring that content is personally relevant are likely to improve both uptake and effectiveness.

The gatekeeper evidence highlights that working with parents, educators, and clinicians must be part of any effective strategy. Adults need knowledge, confidence, and practical support to respond appropriately when young people encounter harmful content, and interventions targeting only young people without also addressing the systems around them are likely to have limited reach. Organisations providing training or guidance to adults in these roles should draw on BCFs to diagnose which specific barriers are most significant in their context, and to select strategies that target those barriers directly.

It is important to note that much of this evidence base comes from areas such as sexual health and smoking rather than harmful online content specifically, and that many studies involve older adolescents or young adults rather than children. Caution is therefore appropriate when applying these findings to younger age groups or to the particular features of online environments. The evidence nonetheless provides a valuable foundation, and the frameworks themselves offer a structured methodology for developing and evaluating future interventions in this area.

References