Treats, indulgence, and dietary patterns: what the evidence says
Quick summary: The scientific evidence on treats and dietary indulgence points consistently in one direction: overall dietary patterns are the primary determinant of diet-related health outcomes, and occasional discretionary foods — sweets, takeaways, cakes — have a minimal impact when consumed infrequently within an otherwise balanced diet. Problems arise when high-sugar, high-fat, or ultra-processed foods shift from occasional exceptions to habitual defaults, displacing nutrient-dense foods and elevating intake of added sugars, saturated fat, and sodium. The psychological evidence adds a further dimension: rigid dietary restriction tends to undermine long-term adherence, while flexible, pattern-based approaches support sustainability. The evidence does not support banning treats — it supports keeping them in their appropriate place.
How researchers study dietary patterns and discretionary foods
Most of what we know about the health effects of specific food categories comes from one of three research designs: large-scale prospective cohort studies, randomised controlled trials (RCTs), and systematic reviews or meta-analyses that pool data across multiple studies.
Prospective cohort studies follow large populations over years or decades, recording dietary habits — typically via food frequency questionnaires or 24-hour dietary recalls — and tracking health outcomes including cardiovascular disease, diabetes, obesity, cancer incidence, and all-cause mortality. These studies generate the majority of evidence on dietary patterns, but carry inherent limitations: self-reported dietary data is subject to recall bias and measurement error; people who eat more discretionary foods often differ systematically from those who eat fewer (lower socioeconomic status, higher stress, less sleep, less physical activity); and controlling for all relevant confounders is methodologically challenging.
RCTs assign participants to specific dietary interventions and measure outcomes under controlled conditions. They are the gold standard for causal inference, but dietary RCTs face practical constraints: participants cannot be blinded to what they are eating; high-calorie food interventions raise ethical questions; and achieving compliance over long follow-up periods is difficult. Most dietary RCTs are therefore short in duration and limited in scope.
Systematic reviews and meta-analyses sit at the top of the evidence hierarchy. They pool data from multiple studies to generate more precise estimates of effect size and can assess the consistency of findings across different populations and methodologies. The evidence on dietary patterns and discretionary food consumption is supported by several high-quality meta-analyses, as discussed below.
The primacy of overall dietary patterns
The most consistent and robustly evidenced finding in nutrition science over the past two decades is that overall dietary pattern — not individual foods or nutrients — is the primary driver of diet-related health outcomes. This finding is now embedded in major dietary guidelines worldwide and endorsed by the leading nutrition research bodies.
KEY RESEARCH
Freeland-Graves, J., & Nitzke, S. (2013). Position of the Academy of Nutrition and Dietetics: total diet approach to healthy eating. Journal of the Academy of Nutrition and Dietetics, 113(2), 307–317. https://doi.org/10.1016/j.jand.2012.12.013
The Academy of Nutrition and Dietetics' position statement articulates the 'total diet approach': all foods can fit within a healthy dietary pattern when consumed in appropriate portions and frequencies. This position explicitly rejects good/bad food labelling in favour of a focus on overall intake across time. The statement identifies rigid restriction of any food category as potentially counterproductive to long-term dietary health.
KEY RESEARCH
Lichtenstein, A. H., Appel, L. J., Vadiveloo, M., et al. (2021). 2021 Dietary Guidance to Improve Cardiovascular Health: A Scientific Statement from the American Heart Association. Circulation, 144, e472–e487. https://doi.org/10.1161/cir.0000000000001031
The 2021 American Heart Association dietary guidance — one of the most comprehensive pattern-based frameworks available — emphasises a dietary pattern high in fruits, vegetables, whole grains, legumes, nuts, and lean protein, and low in added sugars, sodium, refined grains, and ultra-processed foods. Critically, the guidance frames this as a pattern rather than a list of prohibited items, acknowledging that occasional discretionary choices within a healthy overall pattern have minimal impact on long-term cardiovascular risk.
KEY RESEARCH
Tessier, A., Wang, F., Korat, A., et al. (2025). Optimal dietary patterns for healthy aging. Nature Medicine, 31, 1644–1652. https://doi.org/10.1038/s41591-025-03570-5
This 2025 longitudinal analysis — following over 100,000 participants across more than 30 years — examined which dietary patterns were most strongly associated with healthy ageing, defined as reaching age 70 free of major chronic disease and with intact physical and cognitive function. Patterns characterised by high intake of fruits, vegetables, whole grains, legumes, and low-fat dairy, with limited ultra-processed food and added sugar, were most strongly associated with healthy ageing outcomes. The findings reinforce that it is the habitual, long-term pattern — not any single meal or treat — that drives health trajectories.
Discretionary foods: what the evidence specifically shows
Discretionary foods — those high in added sugars, saturated fat, sodium, or refined carbohydrates, consumed primarily for enjoyment rather than nutritional necessity — are the category into which most 'treats' fall: confectionery, cakes, biscuits, pastries, desserts, takeaway meals, sugary beverages, and similar products.
The evidence on their health impact operates at two levels: the effect of high habitual consumption, and the effect of occasional consumption within an otherwise healthy diet. These produce very different pictures.
High habitual consumption of added sugars, particularly from sugary beverages, is associated with elevated risk of cardiovascular disease, type 2 diabetes, obesity, and dental caries. The evidence for sugary drinks specifically — where caloric intake is not compensated for by reduced intake elsewhere — is among the most consistent in dietary epidemiology. The risk signal is considerably weaker for solid discretionary foods, and largely disappears in studies that adjust for overall dietary pattern quality.
KEY RESEARCH
Cara, K. C., Goldman, D. M., Kollman, B. K., et al. (2023). Commonalities among dietary recommendations from 2010 to 2021 clinical practice guidelines. Advances in Nutrition, 14, 500–515. https://doi.org/10.1016/j.advnut.2023.03.007
This meta-epidemiological review of 14 major clinical practice guidelines from 2010 to 2021 found near-universal agreement on one point: added sugars should be limited, not eliminated. No guideline advocated complete avoidance of discretionary foods; all framed the recommendation in terms of limiting frequency and portion size within the context of an overall healthy pattern. This reflects the research consensus: the harm from discretionary foods is dose-dependent and pattern-dependent, not binary.
Ultra-processed foods and nutrient displacement
A substantial portion of what might be considered 'treats' — cakes, biscuits, confectionery, fast food — falls within the category of ultra-processed foods (UPFs) as defined by the NOVA classification system. UPFs are industrial formulations made largely from substances extracted from whole foods, with additives including emulsifiers, stabilisers, artificial flavours, and sweeteners, typically absent from minimally processed or home-prepared foods.
The health concern with high UPF consumption is partly about specific nutrients — added sugars, saturated fat, sodium — and partly about what UPFs displace from the diet. Meta-analytic evidence consistently shows that diets high in UPFs tend to be lower in fibre, protein quality, vitamins, and minerals, because UPFs occupy dietary 'space' that might otherwise be taken by fruits, vegetables, legumes, and whole grains.
KEY RESEARCH
Martini, D., Godos, J., Bonaccio, M., et al. (2021). Ultra-processed foods and nutritional dietary profile: a meta-analysis of nationally representative samples. Nutrients, 13, 3390. https://doi.org/10.3390/nu13103390
This meta-analysis of nationally representative dietary surveys found that higher UPF consumption was consistently associated with lower dietary quality scores, lower fibre and micronutrient intake, and higher free sugar, sodium, and saturated fat intake — regardless of country or dietary assessment methodology. The displacement effect is important: it suggests that the harm from high treat consumption is not solely about the treats themselves, but about what they crowd out.
However, the same displacement logic does not apply to occasional consumption. A treat eaten once a week within an otherwise nutrient-dense diet does not meaningfully displace the nutrients provided by the rest of that diet. The dose-response relationship here is non-linear: the marginal harm from the first occasional treat is very small; the harm from each additional treat becomes larger as the proportion of the diet occupied by discretionary foods increases.
The psychological evidence: restriction, flexibility, and sustainability
The health impacts of treats cannot be understood in purely nutritional terms. Eating behaviour is strongly shaped by psychological factors, and the psychological evidence on dietary restriction is both robust and often overlooked in public health messaging.
Several research traditions converge on the same finding: rigid dietary restriction — characterised by all-or-nothing rules, categorical food avoidance, and strong guilt responses to perceived dietary 'failures' — is associated with poorer long-term dietary adherence, increased risk of disordered eating behaviours, and greater susceptibility to overconsumption following any perceived transgression of dietary rules (sometimes called the 'what the hell effect' or 'abstinence violation effect' in the clinical literature).
Flexible dietary restraint — characterised by general adherence to a healthy pattern with conscious accommodation of occasional treats — is associated with better long-term dietary quality, lower rates of binge eating, and greater psychological wellbeing around food. This finding has been replicated across multiple study populations and dietary contexts.
KEY RESEARCH
Dominguez, L. J., Veronese, N., Baiamonte, E., et al. (2022). Healthy aging and dietary patterns. Nutrients, 14, 889. https://doi.org/10.3390/nu14040889
This review of dietary patterns and healthy ageing explicitly identifies the social and psychological dimensions of food as features of healthy dietary patterns, not obstacles to them. The Mediterranean diet — among the best-evidenced patterns for long-term health — is characterised not just by its nutritional composition but by a cultural relationship with food that accommodates shared meals, festive eating, and enjoyment as integral components. Framing these as threats to dietary quality misunderstands what makes such patterns sustainable across decades.
When treats become problematic: the dose-response evidence
While occasional treat consumption is well-evidenced as compatible with a healthy diet, the evidence equally supports the view that high habitual consumption of discretionary foods carries real health risk. The critical variable is frequency and proportion of total dietary intake.
Population-level data consistently show that the highest quintiles of added sugar consumption — typically representing intakes well above recommended limits — are associated with elevated cardiovascular disease risk, greater adiposity, and poorer metabolic markers. These associations hold after adjustment for total energy intake, suggesting that sugar composition, not simply calorie excess, contributes to the risk profile.
The most clinically significant signal is for sugary beverages: multiple meta-analyses have found associations between regular consumption and type 2 diabetes risk, independent of overall dietary quality. Liquid calories from sugar-sweetened drinks appear to be less well compensated for than calories from solid food, contributing to positive energy balance over time.
For solid discretionary foods — confectionery, cakes, biscuits — the evidence is less stark and more dependent on overall dietary context. The key threshold appears to be when discretionary foods transition from occasional exceptions to habitual components of the diet: occupying regular meal slots, consumed daily, and present in sufficient quantity to reduce intake of nutritionally necessary foods.
KEY RESEARCH
Neuhouser, M. L. (2019). The importance of healthy dietary patterns in chronic disease prevention. Nutrition Research, 70, 3–8. https://doi.org/10.1016/j.nutres.2018.06.002
Neuhouser's narrative review synthesises evidence across dietary pattern research and concludes that no single food or food group — whether nutritious or discretionary — exerts sufficient influence on health to be evaluated in isolation from the overall pattern in which it sits. The review identifies pattern quality and dietary diversity as the primary levers of chronic disease prevention, and discretionary foods as relevant primarily insofar as they affect overall pattern quality.
Practical implications: where the evidence points
The evidence supports several clear, actionable conclusions — and equally rules out some common responses to the treat question.
First, banning treats is not evidence-based. No major clinical or public health guideline advocates complete avoidance of discretionary foods, because the evidence does not support it — and the psychological evidence suggests it may be counterproductive by increasing dietary rigidity and the risk of overconsumption episodes.
Second, the quality of everyday defaults is the primary modifiable dietary variable for most people. The nutritional evidence consistently identifies habitual intake — of fruits, vegetables, whole grains, legumes, protein sources, and healthy fats — as the strongest dietary predictor of health outcomes. Improving the quality of regular meals has a larger expected health impact than eliminating occasional treats.
Third, portion size and frequency are the meaningful variables for discretionary foods. Within a high-quality overall diet, a moderate portion of a discretionary food consumed occasionally does not move the needle on health outcomes in any clinically meaningful way. The same food consumed daily in large portions, or as a replacement for nutritionally necessary foods, carries real risk.
Fourth, sugary beverages deserve specific attention. The evidence for harm from regular consumption of sugar-sweetened drinks is among the most consistent in dietary research, and is not simply a function of overall dietary quality. Reducing sugary drink consumption — including fruit juices consumed in large quantities — represents a high-value dietary change for many people.
Fifth, the social and cultural dimensions of food are legitimate health considerations, not indulgences to be apologised for. Dietary patterns that are sustainable across decades must accommodate the social functions of food: celebration, connection, shared pleasure. A dietary framework that makes these impossible is unlikely to be maintained.
Bringing it back to The New 5-a-Day
Eating well is Pillar Two of The New 5-a-Day — and the treats question sits at a genuinely interesting junction in the nutritional evidence. On one side: the real and well-documented harms of high discretionary food consumption. On the other: the equally well-documented evidence that occasional indulgence, within a strong dietary pattern, carries minimal risk — and that rigid restriction may carry risks of its own.
The evidence does not resolve this as a tension. It resolves it as a hierarchy. Dietary pattern quality is the primary variable. Treat frequency is a secondary variable, relevant mainly when it undermines the primary one. For most people, in most circumstances, the most effective intervention is not policing treats — it is building a pattern of everyday eating strong enough that treats look after themselves.
That is a less dramatic message than much nutrition communication delivers. It is also the message the evidence consistently supports.
A note on medical advice: This Deep Dive is intended for readers who want to engage with the evidence in more depth. It does not replace professional medical or dietary advice. If you have specific health conditions or concerns related to diet, nutrition, or eating behaviour, please speak to your GP or a registered dietitian.
Want to read further? Full citations for all research referenced in this post are available in the accompanying Reference List.
