The NOVA system is the most widely accepted method of classifying food based on the nature, extent and purpose of food processing(Reference Monteiro, Cannon and Moubarac1,Reference Moubarac, Parra and Cannon2) . According to the NOVA classification system, ultra-processed foods (UPF) are not simply modified foods. Rather, they are formulations made using ingredients not typically found in a household pantry such as industrial food substances and additives(Reference Monteiro, Cannon and Moubarac1). UPF are often attractively packaged and extensively marketed as convenient replacements for less processed or whole foods(Reference Machado, Martínez Steele and Levy3). Typically high in saturated fat, sugar and salt and low in micronutrients and fibre, UPF are often, but not always, energy dense and nutrient poor(Reference Monteiro, Cannon and Moubarac1). They are not recommended in national dietary guidelines, with consumption actively discouraged by some (i.e. Brazil(4)). However, recent evidence has shown high consumption across Western countries of up to 60 % energy intake. Emerging evidence indicates that UPF consumption is associated with poor mental and physical health, including CVD, cerebrovascular disease, depression and all-cause mortality(Reference Marino, Puppo and Del Bo5–Reference Suksatan, Moradi and Naeini10). These risks remain when data are adjusted for diet quality or pattern, suggesting that UPF intake increases the risk of chronic disease regardless of an otherwise healthy diet(Reference Dicken and Batterham11).
Sports nutrition guidelines recommend that athletes ‘fuel for the work required’(Reference Impey, Hammond and Shepherd12,Reference Impey, Hearris and Hammond13) , with periodised nutrition plans individually tailored for athletes’ training, competition and physique goals(Reference Burke, Hawley and Jeukendrup14). Fuelling recommendations use a food-first approach, recognising that a carefully planned whole-food diet provides appropriate fuel and hydration for performance and recovery(Reference Thomas, Erdman and Burke15,Reference Tiller, Roberts and Beasley16) . Additionally, whole foods provide athletes with all nutrients and bioactive compounds needed to promote and support health(Reference Close, Kasper and Walsh17–Reference Koelman, Egea Rodrigues and Aleksandrova19).
Sports foods are a type of sports supplement, defined as ‘specialised products used to provide a convenient source of nutrients when it is impractical to consume everyday foods’(Reference Burke20,Reference Garthe and Maughan21) . They contain nutrients tailored to support exercise-related goals and packaged conveniently as sports drinks, gels, confectionery, electrolyte supplements, isolated protein supplements, and mixed macronutrient supplements such as sports bars, powders, and liquid meals (see Table 1). Sports foods are effective in specific situations to achieve hydration, fuelling, recovery, training adaptation and electrolyte balance(Reference Peeling, Castell and Derave22,Reference Maughan, Burke and Dvorak23) . They also provide safe alternatives where there are food intolerances, allergies, preferences, limited energy budgets, limited availability of foods, food hygiene or contamination risks(Reference Close, Kasper and Walsh17,Reference Garthe and Ramsbottom24) . However, sports foods are intended for sport-specific use, not to replace an everyday diet. They do not provide all the nutrition needed for good health(Reference Peeling, Castell and Derave22). Furthermore, due to the nature of their production and formulation, commercial sports foods are typically classified according to the NOVA system as UPF.
Sports nutrition recommendations and studies of dietary intake in athletes tend to focus on energy, macronutrients and select micronutrients as they relate to performance(Reference Thomas, Erdman and Burke15,Reference Kerksick, Wilborn and Roberts25) . Less is understood about dietary patterns and diet quality in athletes, and few studies have focused on athletes’ intake of sports foods. Those that have assessed sports food intake have predominantly done so as part of a broad investigation of sports supplement use. Garthe and Maughan(Reference Garthe and Maughan21) reviewed the published literature in 2018, mostly of studies now more than 10 years old, and found supplement use by 40–100 % of athletes. They cited sports supplements, vitamin and mineral supplements, and herbs as the most used supplements. It is unclear what proportion of these sports supplements were sports foods. Reasons cited for consuming sports supplements were direct performance benefits, health benefits, following advice or practice of esteemed peers or coaches, a ‘just in case’ approach, convenience or financial sponsorship(Reference Garthe and Maughan21). A review of dietary supplement use (including sports foods) by masters athletes found that approximately 60 % of masters athletes used supplements, predominantly for health reasons(Reference Harnett, Climstein and Walsh26). Other recent studies have found high rates (49–97 %) of sports supplement use among young elite athletes(Reference Jovanov, Đorđić and Obradović27–Reference Waller, Kerr and Binnie30). While evidence remains limited, Maughan et al. (Reference Maughan, Burke and Dvorak23) concluded that athlete supplement use varies according to sports/activities, is greater in men, is influenced by others and increases with age and level of training/performance.
The evidence to date suggests many athletes consume sports supplements including sports foods. As commercial sports foods are UPF, and UPF consumption is associated with nutrition-related chronic disease, it is timely to consider the impacts of UPF consumption for athletes. National sporting authorities provide sports supplements and sports foods guidelines(31). However, there are no current recommendations for athletes regarding UPF consumption, and no studies to our knowledge investigating the intake of sports foods as a source of UPF in athletes. To support the initiation of research in this space, the aim of this study was to assess athletes’ and exercisers’ intake of and attitudes towards ultra-processed sports foods. This research will identify whether sports foods, as a source of UPF, are commonly consumed by athletes and whether athletes and exercisers are concerned about UPF consumption. This knowledge will inform future interventions and research to promote nutrition-related health for athletes.
Materials and methods
Study design
We used an anonymous, online, cross-sectional survey to assess athletes’ and exercisers’ intake of and attitudes towards sports foods.
Participants and setting
We recruited a convenience sample of athletes and exercisers via social media (Twitter, LinkedIn, Facebook) between October 2021 and February 2022. Eligible participants were Australian residents, aged 18+ years, able to read and write in English, and met minimum physical activity criteria (participate in recreational or competitive sport or engage in regular physical training defined as at least three sessions of moderate or vigorous physical activity per week over the past month(Reference Araújo and Scharhag32,Reference Desbrow, Slater and Cox33) ). Participants were limited to Australian residents because the survey used language aligned with the Australian Institute of Sport Sports Supplement Framework(31) and commercial Australian sports foods.
Survey tool
The survey was adapted from a previously validated survey designed to assess intake and motivations for intake of dietary supplements(Reference Caldwell, McGraw and Thompson34) and administered anonymously online via QuestionPro. Multi-select and free-text questions were used to ask about demographic characteristics (age, gender, income); level, type and frequency of sport participation; use of special diets; frequency of consumption of sports foods over the past 12 months; reasons for choosing or not choosing to consume sports foods; everyday foods used in place of sports foods; and whether participants were concerned about UPF consumption (see Table 2). The adapted survey was developed by academic sports dietitians and piloted for comprehension and face validity with four practising sports dietitians and four athletes competing in state or national level sports. Updates to the survey based on their feedback included a section identifying reasons for not consuming sports foods for non-consumers and ‘accommodate reduced/increased energy budgets or macronutrient targets’ as a reason for consuming or not consuming sports foods.
Data analysis
Responses to multi-select questions were analysed using descriptive statistics with response options presented as a percentage of total responses. Pearson’s χ 2 test was used to assess potential relationships between categorical demographic variables and consumption of sports foods, with Phi and Cramer’s V used to test the strength of association and statistical significance set at P < 0·05. Quantitative data analysis was performed using SPSS version 28.
Responses to free-text questions were analysed thematically by AF and confirmed through review by and discussion with EM. Initial themes were identified following familiarisation with the data. All data were then coded and categorised into themes using a framework method. The brief and distinct nature of most responses enabled quantitative presentation of free-text data (i.e. percentage of responses provided aligned with a theme). Participant responses for everyday foods consumed instead of sports foods were cleaned to remove any responses unrelated to everyday foods and entered into NVivo version 12. Food items were coded according to their level of processing based on the NOVA classification groups (unprocessed or minimally processed foods; processed culinary ingredients; processed foods; UPF; and an additional category for water)(Reference Machado, Martínez Steele and Levy3). Assumptions made when assigning codes were that homemade products used minimally processed ingredients, oats/porridge was made with rolled oats, spreads used were commercially processed, bread was mass produced, juice was shelf-stable, and unprocessed or minimally processed foods were prepared with no additional ingredients.
Ethics
The La Trobe University Human Ethics Committee (HEC21294) and University of South Australia Human Research Ethics Committee (204227) approved this study. All participants provided implied informed consent by clicking ‘I agree’ to proceed to the survey after reading a participant information statement.
The Strengthening the Reporting of Observational Studies in Epidemiology Statement was used to guide the reporting of this study(Reference Von Elm, Altman and Egger35).
Results
Participants
One hundred seventy-eight individuals commenced the survey. One hundred forty participants progressed beyond the demographic questions and were included in the reported results.
Demographic characteristics of participants are reported in Table 3. All participants were 18–74 years old; 64 % identified as women. A small number of participants (18 %) competed at a national or international level; the majority competed in individual sports (64 %) and trained 5–9 h per week (49 %). Special diets were reported by 35 % of participants. Forms of vegetarian diets, including vegan and flexitarian, were reported by 22 % of participants.
* Multiple responses permitted.
Sports food consumption
Nearly all participants (95 %) reported consuming any sports foods within the past 12 months (Fig. 1). Sports drinks were the most consumed sports food, reported by 73 % of participants. Liquid meals were the least consumed sports food, reported by 9 % of participants. Protein supplements were the most likely sports food to be consumed frequently, with 40 % of participants reporting consumption at least once per week (15 % daily, 21 % a few times per week, 4 % once per week).
Individual sport athletes were most likely to consume gels (χ2(1) = 23·087, V = 0·406, P < 0·001) and sports drinks (χ2(1) = 3·915, V = 0·167, P = 0·048). Athletes competing at state, national, and international levels were more likely to consume sports drinks than recreational or local athletes (85% v. 70%), but this was not statistically significant (χ2(1) = 3·140, V = 0·150, P = 0·076). Athletes participating in sport more than 10 h per week were more likely to consume sports bars (χ2(1) = 4·715, V = 0·184, P = 0·030), sports drinks (χ2(1) = 7·802, V = 0·236, P = 0·005) and electrolyte supplements (χ2(1) = 5·699, V = 0·202, P = 0·017). There were no statistically significant differences in consumption by age, gender, income or presence of a special diet.
Motivations for sports food consumption
Performance and convenience were the most reported reasons for choosing sports foods (Fig. 2). Cost and taste were the most reported reasons for not choosing to consume sports foods, with considerations for energy/macronutrient targets or lack of performance benefit also frequently selected (Fig. 3). Relative to sports foods, participants reported that everyday foods were less convenient to prepare and to carry during performance and presented a greater risk of spoilage (Fig. 3). However, they reported everyday foods to be more affordable, taste better, present less risk of banned substances and be marginally better at supporting performance/recovery and managing gastrointestinal comfort. Peers (40 %) and coaches/high-performance staff (38 %) were the most frequently reported influencers of decisions regarding sports food consumption.
Everyday food alternatives for sports foods
Participants described a range of foods and beverages used in place of sports foods. More than half (54 %) of the items listed by participants were UPF, such as lollies and muesli bars. Processed foods (processed foods and processed culinary ingredients), such as smoothies, whole foods (unprocessed or minimally processed foods), such as fruit, and water were listed less commonly (18, 19 and 9 %, respectively).
Concern about ultra-processed foods
Approximately half (51 %) of participants reported concern about the health effects of consuming ultra-processed sports foods. They described concerns that these foods were not natural (33 %) or negatively impacted health (30 %). Some had specific concerns related to sugar content (9 %) or impact on gut health (15 %). Nearly two-thirds (61 %) of those not concerned about ultra-processed sports food consumption explained this was because they only used sports foods occasionally, in small amounts, or exclusively for training and performance. Others reported that the benefits related to performance or convenience outweighed the risks. Sixteen percent of those unconcerned reported they were unaware of UPF and its potential impact on health. Concern regarding consumption of ultra-processed sports foods was not related to age, gender, type of sport, level of competition or weekly hours participating in sport.
Discussion
Sports food consumption
In this study, almost all participants reported consuming commercial sports foods that are UPF. No other studies have considered UPF consumption specifically in athletes. In a study assessing sources of carbohydrate used in training by German endurance runners, cyclists and triathletes (n 1081), 87 % of athletes reported using commercial sports nutrition products (which are UPF) exclusively or combined with everyday foods to fuel their training sessions(Reference Reinhard and Galloway37). Most studies do not distinguish between sports foods and other sports supplements but do suggest high intakes of sports supplements among athletes(Reference Garthe and Maughan21). Many (87 %) Australian athletes attending state-based sports institutes consumed sports supplements with sports foods more frequently reported than medical or performance supplements(Reference Waller, Kerr and Binnie30).
Participants in our study consumed sports drinks more commonly than any other sports foods, as have athletes in other studies assessing sports food intake(Reference Waller, Kerr and Binnie30,Reference Reinhard and Galloway37) . The study of German endurance athletes only assessed intake of carbohydrate-rich foods, so their intake of other sports foods, such as protein supplements, is unknown. Protein supplements were consumed more frequently than other sports foods in our study. Few studies have explored frequency of sports food intake (i.e. how often athletes consume sports foods). However, Italian gym users also reported frequent consumption of protein supplements, predominantly in the form of whey protein shakes(Reference Bianco, Mammina and Paoli38).
Participants competing at higher levels and training for longer durations were most likely to consume sports foods in our study. This suggests athletes may be using sports foods as intended – to provide nutrients and fluid required when needs are unable to be met by everyday foods(Reference Burke20,Reference Garthe and Maughan21) . For example, water is recommended as the beverage of choice for exercise up to 60–90 min with glucose/electrolyte solutions, often in the form of sports drinks, for longer bouts(Reference Kerksick, Wilborn and Roberts25). Contrary to our findings, intakes of Australian athletes attending state-based sports institutes did not differ across performance levels(Reference Waller, Kerr and Binnie30). This may be related to the relatively elite status of athletes in that study compared with our more heterogeneous study population.
Motivation for sports food consumption
Participants in our study stated they consumed sports foods for performance and convenience. Performance may be a particularly important concern for those competing at higher levels and may explain the previously noted higher sports supplement intake among higher-calibre athletes(Reference Maughan, Burke and Dvorak23). Performance was the strongest influence of food choice among athletes attending the 2018 Commonwealth Games and 2017 Universiade competitions(Reference Thurecht and Pelly39), while convenience was the primary reason why German endurance athletes selected sports foods over everyday foods(Reference Reinhard and Galloway37).
Participants in our study cited taste and cost as reasons for not consuming sports foods. Taste, as an element of sensory appeal, was ranked second behind performance as a factor influencing food choice of 2018 Commonwealth Games and 2017 Universiade athletes in Thurect and Pelly’s study(Reference Thurecht and Pelly39) but was considered by only 15 % of German endurance athletes when choosing carbohydrate-rich foods for training sessions(Reference Reinhard and Galloway37). Reinhard and Galloway suggest this discrepancy may be related to Birkenhead and Slater’s assertion that the importance of taste varies by eating occasion(Reference Birkenhead and Slater40). So, taste preferences for everyday foods may not override performance-driven motivations for sports foods. While cost was cited as a reason for not consuming sports foods in our study, the influence of cost on athlete food choice in other studies seems to vary depending on the cohort and the context(Reference Thurecht and Pelly39–Reference Heaney, O’Connor and Naughton41). Athletes with limited disposable income who are responsible for purchasing and providing their own food are most likely to be influenced by cost, and we did not assess these contextual factors. Regardless, as many everyday foods can also be cost-prohibitive, athletes and their support teams may need suggestions for low-cost alternatives to commercial sports foods.
Athletes are commonly concerned about doping when making food choices(Reference Waller, Kerr and Binnie30,Reference Thurecht and Pelly39) . Doping is an important consideration, particularly for athletes competing at levels of competition subject to World Anti-Doping Authority regulations(42). Participants in our study believed that everyday foods presented a lower risk of banned substances than commercial sports foods. However, some protein-fortified everyday foods, such as those prepared by cafés and other food outlets, can present a doping risk for athletes, with batch-tested commercially prepared protein-fortified foods a safer option for athletes looking for convenient high-protein options(Reference O’Bryan, Shaw and Allanson43). Therefore, athletes and their support teams may need further education about doping risks of everyday and sports foods.
In our study, participants reported that peers and coaches/high-performance staff influenced their decisions about sports foods. An earlier review found that coaches, trainers, friends and family were the most commonly used sources of dietary supplement information for athletes(Reference Denham44). In other Australian studies, athletes have reported coaches, sports scientists/physicians, dietitians/nutritionists, peers and the internet as sources of nutrition and sports supplement information(Reference Impey, Hammond and Shepherd12,Reference Shaw, Slater and Burke29,Reference Waller, Kerr and Binnie30,Reference Trakman, Forsyth and Hoye45,Reference Jenner, Belski and Devlin46) . While it is important to understand who influences athletes’ food choices, few (< 10 %) German endurance athletes were influenced by peers, nutritionists or coaches to consume everyday foods over commercially prepared sports foods(Reference Reinhard and Galloway37). Given that food choice, particularly for individual endurance sport athletes, may be influenced more by their personal views than by others, sporting organisations should focus education efforts on athletes themselves.
Everyday food alternatives for sports foods
In this study, more than half of the everyday food items consumed by participants in place of commercial sports foods were UPF. So, athletes may need education on the risk of UPF foods and support to identify non-UPF everyday food options. Furthermore, convenient access to safe and affordable minimally processed options may support athletes to make these changes in practice. Education and support for training and competition venues to provide non-UPF options may lead to enhanced access to these foods where they are most convenient to athletes. Other research has demonstrated that healthy canteen displays at sporting association trade shows can support organisations to plan and implement changes in their canteens(Reference Trakman, Staley and Forsyth47).
Concern about ultra-processed foods
Approximately half of the participants in our study were concerned about the health effects of consuming UPF. Participants unconcerned about the health impacts of UPF reported that use was only occasional, and that performance and convenience benefits outweighed health risks. Education for athletes and their support teams should highlight that negative health consequences of UPF exist regardless of overall diet quality(Reference Dicken and Batterham11) and that any UPF consumption may harm their overall health. They can then make informed decisions about whether to consume ultra-processed sports foods as part of their performance plan.
Strengths and limitations
Our study explored the views of a broad range of athletes across Australia. The sample included a broad distribution of responses from women and men and athletes across diverse sport types and participation levels. Consequently, the results are likely to be generalisable to Australian athletes. Since each country has specific guidance on sports supplements, it is important to conduct supplement-related research within the context of the host country.
We based the survey on a previously validated survey and piloted the adapted version to assess content and face validity before use. The survey forced responses and provided ‘other’ options and free-text boxes throughout the survey to ensure the most relevant options were covered. A combination of question types, including free-text, enabled investigation into both sports food practices and reasons for choices made by athletes. However, we evaluated only the frequency of sports food consumption, not actual amounts consumed or intake as a proportion of total energy intake. Further, we evaluated only sports food intake, not overall diet. As we conducted this study before publication of the Participant Classification Framework(Reference McKay, Stellingwerff and Smith48), participation categories are similar but not fully aligned with the six-stage framework. The five categories presented align with Tiers 1–4, and it is possible, though unlikely, that some athletes in this study may classify as Tier 5 (World Class).
Gibney et al.(Reference Gibney and Forde49) suggest that classification of foods into NOVA categories can be problematic and Astrup and Monteiro(Reference Astrup and Monteiro50) note the lack of consistency in classification of foods by researchers in other studies. So, while in our experience the classification of foods consumed in place of commercial sports foods was straightforward and there was agreement between researchers, it is possible that others may have interpreted and classified the foods differently. To distinguish between categories, we made assumptions for some foods and we have noted these in our methods.
Finally, several publications dispute the NOVA classification system and potential health concerns associated with UPF(Reference Astrup and Monteiro50,Reference Visioli, Marangoni and Fogliano51) . These focus on the idea that the NOVA classification does not add to public health messaging beyond a nutrient or dietary pattern approach. Nonetheless, athletes should consider the impact of added harmful compounds and lost protective elements on health and performance, even when the nutrient composition of a manufactured food product aligns with their sports nutrition plan and dietary targets. At this stage, there is no evidence to demonstrate whether the benefits of exercise can mitigate the adverse health effects of consuming UPF.
Recommendations for future research
Regulations guiding product content, labelling and claims can help athletes and their teams make informed decisions about sports supplements including sports foods(Reference Menal-Puey and Marques-Lopes52). However, they do not address the potential health risk of consuming UPF. Given Dicken and Batterham’s(Reference Dicken and Batterham11) study demonstrating that healthful dietary patterns do not mitigate the chronic health risks of UPF, concerns and unanswered questions remain regarding sports food recommendations for athletes. Future studies are needed to identify whether there is an association between sports food consumption and the development of chronic disease. Further investigations should explore the type, frequency, amount and duration of sports food consumption in various athletic populations and whether any risks are offset or increased by exercise type, frequency and duration. At a practical level, sporting organisations should trial providing education about and access to safe, affordable, convenient, minimally processed food options for athletes and their support teams. Finally, addressing UPF as a strategy to support more sustainable diets for athletes can be explored(Reference Meyer, Reguant-Closa and Nemecek53).
Conclusion
Almost all participants in this study consumed ultra-processed commercial sports foods as convenient options to support performance. Half of the participants were concerned about the potential adverse health effects of consuming UPF. Existing literature suggests that an overall healthy dietary pattern does not mitigate the impact of UPF on health, and more research is needed to understand the impact of exercise. Athletes may consider reducing UPF intake to lower their risk of developing diet-related disease. Level of processing may be used as a simple guide for athletes to adopt health-promoting dietary behaviours that align with their performance nutrition goals. Athletes and their support teams need education and access to safe, affordable, convenient, minimally processed alternatives to commercial sports foods.
Acknowledgements
No funding was received for the conduct of this study or preparation of the manuscript.
E. M. conceptualised the study design. A. F. and E. M. collected, analysed and interpreted the data. A. F. drafted the initial manuscript. A. F. and E. M. critically reviewed the manuscript and approved the final version.
The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.