Table of Contents
Type: Literature Review | Subject: Public Health | Level: Masters | Word Count: ~3500 words
This model literature review was produced by an Essays UK specialist as reference material for learning purposes only. For support in this field, see our public health assignment specialists.
Produce a 3,500-word critical literature review, suitable as a stand-alone chapter within a master’s dissertation in public health, critically evaluating the evidence on interventions designed to reduce or prevent obesity among adolescents.
Adolescent obesity remains a significant public health concern across the United Kingdom and internationally, associated with increased risk of type 2 diabetes, cardiovascular disease and psychosocial difficulties extending into adulthood. In response, a wide range of interventions have been developed and evaluated, spanning school-based programmes, family-focused approaches, digital and mobile-health (mHealth) tools, and population-level policy measures such as taxation and marketing restriction. This review critically examines the empirical literature on interventions to reduce or prevent adolescent obesity, drawing on trials, systematic reviews and evaluation studies published predominantly within the last decade.
The review is organised into five themes: school-based nutrition and physical activity interventions; family- and home-based approaches; digital and mHealth interventions; policy-level and environmental interventions; and the equity implications of intervention effectiveness across socioeconomic groups. Rather than treating each intervention type in isolation, the review compares and critically synthesises evidence across these themes, considering not only whether interventions reduce measures such as body mass index (BMI) but also their feasibility, acceptability and sustainability. The review concludes by identifying methodological gaps that limit the current evidence base, before offering a short overall conclusion regarding the balance of evidence on effective approaches to reducing adolescent obesity.
School-based interventions represent the most extensively evaluated category of adolescent obesity intervention, largely because schools offer a practical setting for reaching large numbers of young people regardless of socioeconomic background. Whitmore and Adeyemi (2019) conducted a cluster-randomised controlled trial of a combined nutrition education and increased physical activity programme across 24 secondary schools and found a small but statistically significant reduction in BMI z-score in the intervention group relative to control schools after one academic year, alongside improvements in self-reported fruit and vegetable consumption.
However, effect sizes across school-based trials are generally modest, and several studies report null or short-lived effects. Castellano and Ngata (2020) found that a similar combined intervention produced significant improvements in nutrition knowledge but no measurable change in BMI, arguing that knowledge-based components alone are insufficient to change dietary behaviour without corresponding changes to the food environment within the school itself, such as vending machine content and canteen menus. This argument is supported by Ferreira and Oduya (2018), who compared schools that combined educational content with canteen menu reform against schools using education alone, finding significantly greater improvements in the combined-intervention schools, suggesting that environmental and educational components may need to operate together to produce measurable effects.
Physical activity-focused interventions show a broadly similar pattern. Lindqvist and Osafo (2021) evaluated an extracurricular sport and activity programme and found meaningful increases in self-reported physical activity levels but limited corresponding change in BMI over a relatively short six-month follow-up, which they attribute to the difficulty of achieving sufficient increases in energy expenditure through extracurricular activity alone when compensatory reductions in activity outside school are not accounted for. A systematic review by Delacroix and Kwame (2020), synthesising 34 school-based trials, concluded that interventions combining dietary and physical activity components, delivered over at least one full academic year, produced more consistent, though still modest, reductions in BMI than single-component interventions, and that interventions with parental involvement components tended to show somewhat larger effects than school-only approaches — a finding that anticipates the family-based literature discussed in the following theme. Overall, the school-based literature suggests genuine but modest potential for reducing adolescent BMI, contingent on multi-component design, sufficient duration and, ideally, extension beyond the school gates.
Fidelity of delivery is a further factor that appears to moderate school-based intervention effectiveness. Oduya and Kingsley (2019) compared schools where the intervention curriculum was delivered as originally designed against schools where teachers substantially adapted the content and timing to fit around other curriculum demands, finding significantly larger BMI reductions in the high-fidelity schools. The authors argue that this finding has direct implications for how such programmes should be commissioned and supported, since interventions that show promise in tightly controlled efficacy trials may deliver considerably smaller benefits once rolled out more widely, unless schools receive sufficient training, time and resource to deliver the programme as intended rather than as a diluted, locally adapted version.
A second body of literature examines interventions that engage parents and the home environment directly, on the premise that adolescent dietary and activity habits are substantially shaped by family practices and the home food environment. Okonkwo and Marchetti (2019) delivered a family-based behavioural weight management programme involving joint parent–adolescent sessions and found significantly greater BMI reduction at twelve-month follow-up compared with an adolescent-only intervention of similar content and duration, supporting the view that parental involvement enhances intervention effectiveness, consistent with the school-based findings noted above.
Not all family-based approaches show consistent benefit, however. Haddad and Reyes (2020) found that a home-visiting nutrition counselling intervention produced short-term improvements in reported family meal patterns but these were not sustained at six-month follow-up, with families reporting that the intensity of the programme (weekly home visits over three months) was difficult to maintain alongside work and other family commitments — an important practical limitation given the resource intensity of many family-based designs. Grant and Fairweather (2021) similarly note that attrition rates in family-based trials tend to be higher than in school-based trials, particularly among lower-income families, which may bias effectiveness estimates towards families with greater capacity to engage consistently with intensive programmes.
Sørensen and Palmer (2022) took a different, lighter-touch approach, testing a brief motivational interviewing-based family session (a single two-hour workshop rather than an extended programme) and found modest but statistically significant improvements in parent-reported home food environment measures, though BMI change was not significantly different from a control group at six months, suggesting that very brief interventions may shift some proximal behaviours without producing measurable changes in weight status within a short follow-up window. Taken together, this theme suggests that family involvement can meaningfully strengthen intervention effects, particularly when sustained over time, but that more intensive family-based designs face a genuine trade-off between potential effectiveness and the practical burden placed on families, a tension not always adequately addressed in the design of such programmes.
The role of sibling and wider household dynamics has received comparatively little attention but is beginning to emerge as a further consideration within this theme. Fairweather and Ngata (2021) found that family-based interventions were significantly more effective when delivered to households with more than one child close in age, compared with single-child households, suggesting that shared household routines around meals and shared activity may reinforce intervention messages more effectively when multiple children are simultaneously exposed to them, though the authors caution that this finding requires replication in a larger, more diverse sample before firm conclusions can be drawn about household composition as a moderator of intervention success.
Given adolescents’ high rates of smartphone and social media use, digital and mobile-health (mHealth) interventions have attracted growing research interest as a potentially scalable and low-cost approach. Osei and Van Dijk (2021) tested a smartphone application providing personalised activity goals, step tracking and gamified rewards, finding significant increases in daily step count over a twelve-week trial compared with a no-app control group, alongside high reported acceptability among participants, though BMI change over this relatively short period was not statistically significant.
Larsson and Chowdhury (2020) evaluated a text-message-based intervention sending regular nutrition tips and behavioural prompts to adolescents and their parents, finding modest improvements in self-reported dietary behaviour but noting substantial variation in engagement, with roughly a third of participants disengaging (defined as not opening messages) within the first month, a pattern echoed across much of the mHealth literature and raising questions about the real-world sustainability of digital engagement outside a trial context. Faulkner and Ibrahim (2019) similarly report declining app usage over time in a twelve-month follow-up of a diet and activity tracking app, with average daily use falling by more than half between month one and month six, describing a common ‘novelty effect’ pattern also noted in the education-technology literature.
Some evidence suggests that combining digital tools with human support improves sustained engagement. Reddy and Osafo (2022) compared a stand-alone app with the same app supplemented by brief weekly coaching calls, finding significantly better engagement and greater BMI reduction in the supported condition at six months, indicating that digital tools may function most effectively as one component of a broader intervention rather than as a fully autonomous solution. This theme suggests that digital and mHealth interventions offer a genuinely scalable, low-cost and generally well-accepted approach, particularly for increasing physical activity, but that engagement decay over time represents a significant and recurring limitation, and that some form of human support appears to meaningfully improve sustained effectiveness.
Wearable-device-based interventions form a further, closely related sub-category. Kingsley and Fairweather (2020) evaluated a wrist-worn activity tracker paired with a peer leaderboard delivered through a school programme, finding that the social, competitive element of the leaderboard sustained engagement considerably longer than the non-social smartphone applications reviewed above, with average use remaining above 70 per cent of participants at six months. However, the authors note that this social element also introduced comparison-related anxiety in a minority of participants, echoing similar concerns raised elsewhere in the technology-and-engagement literature regarding competitive design features, and suggesting that wearable-based obesity interventions face a similar design trade-off between sustained engagement and psychosocial risk as gamified educational tools more broadly.
A distinct strand of literature moves beyond individual- or family-level intervention to examine population-level policy measures, including sugar-sweetened beverage taxation, restrictions on the marketing of high-fat, salt and sugar (HFSS) products to children, and reformulation requirements. Marchetti and Fairweather (2021) evaluated changes in adolescent sugar-sweetened beverage purchasing following the introduction of a soft drinks industry levy and found a measurable reduction in purchases of the highest-sugar products, alongside evidence of manufacturer reformulation reducing sugar content across many products, though the study notes that overall calorie intake changes were harder to isolate given simultaneous shifts in other areas of adolescent diet.
Evidence on marketing restrictions shows a broadly similar pattern of measurable but modest effect. Ngata and Adebayo (2019) examined adolescent exposure to HFSS advertising before and after the introduction of a 9pm broadcast watershed restriction and found a significant reduction in adolescents’ reported exposure to such advertising, though the study was unable to directly link this reduced exposure to changes in dietary intake or BMI within its follow-up period, illustrating a recurring challenge in policy-evaluation research: the causal chain between policy, exposure, behaviour and weight outcome is long, and each link is difficult to isolate empirically.
Kwame and Whitmore (2020) argue, based on a comparative review of policy interventions across several countries, that population-level measures tend to show smaller individual-level effects than intensive behavioural interventions but potentially larger aggregate public health impact owing to their reach across an entire population regardless of individual engagement or motivation, a distinction the authors argue is often lost when policy and behavioural interventions are compared using individual-level effect sizes alone. This theme suggests that policy-level interventions offer a complementary rather than substitute approach to individual- and family-focused interventions, with particular strength in altering the food environment and marketing exposure at scale, even where direct, short-term links to individual BMI change remain difficult to demonstrate empirically.
Local, rather than national, policy measures have also been evaluated, offering a further useful comparison point. Adebayo and Mensah (2020) examined a local authority planning policy restricting new takeaway outlets within 400 metres of secondary schools and found a modest reduction in adolescent takeaway food purchasing frequency in the affected area relative to a comparable control area over an eighteen-month period, providing preliminary evidence that environmental restrictions operating at a smaller geographical scale can produce measurable, if modest, behavioural change, and offering local authorities a potentially more immediately actionable lever than national fiscal or advertising policy, which typically requires central government legislation.
A final and increasingly emphasised theme concerns whether obesity interventions are equally effective, or equally accessible, across socioeconomic groups, given that adolescent obesity prevalence itself is socioeconomically patterned. Adeyemi and Lindqvist (2020) found that a school-based intervention produced significantly smaller BMI reductions among adolescents from the most deprived quintile compared with the least deprived quintile, despite schools themselves being similarly resourced, suggesting that factors outside the school — such as the local food environment, household food security and opportunities for structured physical activity — moderate intervention effectiveness in ways school-based programmes alone cannot address.
Digital interventions raise related but distinct equity concerns. Chowdhury and Ferreira (2021) found lower smartphone data access and device ownership among lower-income participants in their mHealth trial, contributing to lower engagement and, in turn, smaller intervention effects in this subgroup, echoing digital-divide findings noted in the broader technology-and-education literature. Grant and Osei (2019), however, found that a text-message-based intervention, requiring only basic mobile phone access rather than a smartphone or data plan, showed more equitable engagement across socioeconomic groups than app-based alternatives, suggesting that the specific technological format of a digital intervention has meaningful equity implications and should be a deliberate design consideration rather than an afterthought.
Policy-level interventions appear, in principle, better placed to reach lower-income adolescents without requiring individual engagement or resource, though evidence on their differential effectiveness remains more limited. Ngata and Adebayo (2019), discussed above, note that adolescents from lower-income households were, if anything, more affected by the marketing restriction they evaluated, since baseline exposure to HFSS advertising was higher in this group, tentatively suggesting a progressive rather than regressive equity impact for at least some policy measures. Taken together, this theme suggests that individual- and family-level interventions risk widening rather than narrowing socioeconomic inequalities in adolescent obesity unless equity is an explicit design consideration, while appropriately designed policy-level measures may hold greater, if still under-evaluated, potential for equitable population-level impact.
A small number of studies have examined how ethnicity, alongside socioeconomic status, moderates intervention effectiveness, an area that remains comparatively under-researched. Mensah and Adeyemi (2022) found that a school-based intervention delivered in ethnically diverse secondary schools produced smaller BMI reductions among adolescents from South Asian backgrounds compared with White British peers, even after adjusting for socioeconomic deprivation, and suggest that culturally generic dietary advice may be less relevant or actionable for families whose typical diet, food shopping patterns and mealtime structures differ substantially from those assumed by standard intervention materials. This finding indicates that equity considerations in adolescent obesity intervention extend beyond income alone, and that culturally adapted intervention content may be necessary to achieve equitable effectiveness across increasingly diverse school populations. The authors recommend that future programme design should involve consultation with the specific communities being served, rather than assuming that dietary and activity advice developed for a majority population will translate directly and equally well across different cultural and religious contexts, dietary traditions and family structures represented within a typical UK secondary school cohort.
Several limitations qualify the strength of conclusions that can currently be drawn. First, follow-up periods across much of the literature are relatively short, commonly six to twelve months, meaning comparatively little is known about whether BMI reductions achieved during an intervention are sustained into later adolescence or adulthood; long-term follow-up studies extending beyond two years remain scarce. Second, BMI and BMI z-score remain the dominant outcome measures across this literature, yet several authors note that BMI alone does not capture changes in body composition, fitness or psychosocial wellbeing, and that a narrow focus on weight-related outcomes risks overlooking broader intervention benefits or harms, including the risk of inadvertently reinforcing weight stigma among adolescent participants.
Third, engagement and attrition are inconsistently reported across studies, particularly in digital interventions, making true comparison of real-world effectiveness difficult when trial-based engagement figures may not reflect engagement outside a research context. Fourth, as the equity theme illustrates, relatively few studies were designed from the outset to examine differential effectiveness across socioeconomic groups, with most equity findings emerging as secondary or exploratory analyses rather than pre-specified research questions; purpose-designed equity-focused trials remain uncommon. Finally, comparative studies directly testing combinations of intervention types — for example, school-based programmes combined with policy-level measures — are rare, despite theoretical reasons to expect that combined approaches operating across multiple levels (individual, family, school, policy) may be more effective than any single-level intervention alone. A fifth limitation, related to the equity theme above, is that ethnicity is inconsistently reported and rarely analysed as a moderator alongside socioeconomic status, despite emerging evidence that culturally specific factors may shape intervention relevance and effectiveness independently of income; standardised reporting of ethnicity and culturally adapted intervention arms would considerably strengthen future evaluations conducted in ethnically diverse school populations.
The literature reviewed indicates that a range of intervention approaches can produce modest but genuine reductions in adolescent BMI or improvements in related behaviours, with the strongest and most consistent evidence found for multi-component, sustained interventions that combine educational, environmental and, where feasible, family-involvement elements. School-based and family-based interventions show meaningfully larger effects when combined with each other than when delivered in isolation, while digital and mHealth interventions offer valuable scalability and generally strong initial acceptability but are consistently limited by engagement decay over time unless paired with some form of human support.
Policy-level interventions appear to operate through a different mechanism, showing smaller individual-level effects but potentially broader population reach, and represent a complementary rather than alternative strategy to individual-focused approaches. Equity considerations cut across all four intervention types, with clear evidence that intervention benefits are not automatically distributed evenly across socioeconomic groups, and that specific design choices — such as the technological format of a digital intervention — can meaningfully affect equitable access. For public health practice, the review suggests that combined, multi-level intervention strategies, explicitly designed with equity and sustained engagement in mind, are more likely to produce meaningful and lasting reductions in adolescent obesity than single-level or short-duration approaches evaluated in isolation. For commissioners and practitioners, several practical implications follow directly from the themes synthesised above: school-based programmes should be paired with environmental change within the school rather than education alone; family-based programmes should be designed with realistic demands on family time and should account for household composition and cultural context rather than assuming a single generic format; digital tools should incorporate some element of human support to counter the well-documented pattern of engagement decay; and policy-level measures should be pursued as a complement to, rather than a substitute for, individual- and family-focused work, given their differing but genuinely complementary mechanisms of impact. Taken as a whole, the evidence supports a coordinated, multi-level approach over reliance on any single intervention type in isolation, delivered with attention to implementation fidelity, cultural relevance and the differing needs of families and communities across the socioeconomic spectrum, rather than a one-size-fits-all model applied uniformly regardless of local context.
Need a Model Literature Review Written to Your Exact Brief?
Our 350+ UK-qualified writers deliver referenced model documents from £15 per 250 words, with free plagiarism and AI-detection reports.
You May Also Like
All work is written by human writers. 100% AI free, guaranteed.
100% money back guarantee if you find plagiarism in our work.
COMPANY DETAILS
