The Mediterranean Athlete’s Nutrition: Are Protein Supplements Necessary?

Many factors are key to design the optimal nutrition for athletes, total caloric intake, nutrient timing and periodization, macro- and micronutrient intake [1]. Here, we are addressing macronutrient needs of Mediterranean athletes with a focus on total daily protein intake. There seems to be a cultural shift in the dietary realm in which there is a marked increase on the importance of protein intake. Historically, fat intake was deemed as the cause of obesity and cardiovascular diseases, while now the finger is being pointed at carbohydrates, most likely related to the diabetes incidence in developing countries. There is also a parallel increase in protein supplement companies ultimately leading to marketing campaigns targeting also non-athletes that focus on the necessity of dietary integration without providing warnings on the possible risks from diets too high in protein [2].

The US Dietary Reference Intakes (DRI) specify a daily dietary protein intake for all individuals aged 19 years and older of 0.8 g/kg/day (Institute of Medicine, 2005). More specific to athletes, in 2016 the Academy of Nutrition and Dietetics (AND), Dietitians of Canada (DC), and the American College of Sports Medicine (ACSM) published their joint stance on the correct nutrition supporting athletic performance. They stated that protein ingestion level for athletes should fall between 1.2 and 2.0 g/kg/day, this range being necessary to support metabolic adaptation, repair, remodeling, and for protein turnover [3]. In contrast with previous published recommendations [4,5], this joint stance does not distinguish specific protein recommendation based on sport types or gender. The importance of protein intake for athletes is to preserve the lean body mass. Data suggest that protein intake between 1.8 and 2.4 g/kg/day are necessary to preserve lean body mass in individuals who are in caloric restriction, secondary to either dietary restriction or high caloric expenditure [6]. Moreover, during a heavy resistance training program lasting 8 weeks, Antonio et al. [7] observed positive body composition remodelling by applying high protein intake, up to 3.4 g/kg/day.

Increase in protein recommendations for athletes seems to follow the socio-cultural trend “carbs bad/protein good” as well as the dictates of the Paleo or Carnivore diets [8,9,10]. There is not a good grasp, however, of the athlete’s ability to meet the daily recommended protein intake while living in countries that have a tradition of following the Mediterranean diet. Although not strictly defined, the Mediterranean diet is the traditional diet that is eaten by populations surrounding the Mediterranean Sea. It is based on a high intake of fruits, vegetables, legumes, and unrefined cereals, a moderate to high intake of olive oil and fish (depending on the proximity to the sea), a moderate intake of milk, yogurt, and cheese, and a low consumption of meat or meat products [11]. A recent review of the literature sought to more clearly define the Mediterranean diet and concluded that it is a diet that provides 36.6 ± 4.9% energy from total fat, 42.8 ± 3.3% energy from carbohydrate and 14.9 ± 2.3% from protein [12]. Although proteins are found in other foods (grains, legumes, and milk products), meat and fish have the highest concentration of high-quality protein. People following the Mediterranean diet eat a maximum of one serving per week of red meat while a general Western diet provides up to three servings per week [13,14]. Protein intake is crucial in an athlete’s performance as it is necessary for the physiological remodeling that is desired from training, i.e., increase in lean muscle mass accompanied by a loss of fat mass [6]. The Western diet allows an athlete to easily reach adequate protein intake as demonstrated by previous studies [15]. However, the Western diet is not considered an optimal dietary approach to meet micronutrient and antioxidants needs [16]. On the contrary, the Mediterranean diet is a healthy dietary approach also for athletes, but its ability to provide sufficient proteins to athletes is still not clear. Monitoring protein intake in athletes living in Mediterranean countries allows the evaluation of protein supplementation as necessary or only just hype. In order to assess the athlete’s ability to reach the minimum suggested protein levels in a culture that predominantly follows the Mediterranean diet, we surveyed a wide range of athletes that at least competed at an amateur level, resided in Italy or Spain, and tracked their eating and sport practicing habits for three consecutive days, differentiating food from supplement-originating macronutrients.

Between 2017 and 2019, Exercise and Sport Science students from the University of Bologna, set to recruit amateur athletes willing to agree to journal their food intake for three consecutive days (2 workdays and 1 weekend day). The study was carried out according to the Declaration of Helsinki. The diaries were filled out anonymously, and the nutritionist used in the analysis was blinded to the participants as she was not involved in the recruitment of the subjects. Our Bioethics Committee does not assign a protocol number to completely anonymous surveys such as this one. Study participants had to be greater than 18 years old, at least competeing at amateur level, and practice at least 3 h/week in one sport, and reside in Italy or Spain at the time of the study participation. Protein supplementation was captured but was not an inclusion or exclusion criteria. If subjects met criteria, they were asked to sign a consent form, at which points participants were explained in detail how to fill-out a 3-day food diary, an instrument commonly used to access food intake [17,18] and could ask any clarifying questions (3-day food diary facsimile is available as Supplementary Materials). A total of 166 participants completed the dietary journal, 125 males and 41 females. Subjects were grouped by metabolic activity type: anaerobic, aerobic, and mixed exercise according to the metabolic features of the sport practiced.

The diary was designed by a nutritionist, set to capture food intake description, quantity, and time of meal/snack, as well activity description (type and duration). Athletes were asked to track all intake, whether liquid or solid and whether from foods or supplements. Journal data were then translated into macronutrient content (protein, carbohydrates, and lipids (g)) by the same nutritionist for every journal. The food database provided by the Council for Agricultural Research and Economics—Research Center on Food and Nutrition (CREA AN), that holds a total of 790 different foods and their nutritional characteristics (macronutrient content for every 100 g) was used to translate diary information into macronutrient content [19]. The de-identified information was entered into an excel file (Microsoft Corporation, Redmond, WA, USA) and then transferred to GraphPad Prism version 7.00 (GraphPad Software, San Diego, CA. USA) for statistical analysis.

The data were checked for normal distribution with the use of the D’Agostino and Pearson test (p > 0.05). Since the data passed the normality test, parametric statistics were then used. In order to test if our athletic population was able to intake the recommended protein amounts we used the one-way t test setting the hypothetical mean at 1.6 g/kg/day and significance threshold set at p < 0.05. The hypothetical mean of adequate protein intake of 1.6 g/kg/day was obtained by calculating the mean of the range provided by the joint statement from AND, DC and ACSM (1.2–2.0 mg/kg/day). The data were then subdivided by sport type, and a one-way t test was used to see if there were any differences in protein intake by sport type versus the hypothetical mean. When controlling for repeated measures, Sidak’s multiple comparisons test was used to test for a treatment effect of protein supplementation in subjects who chose to supplement their diet, following a significantly different two-way ANOVA result.

Tukey’s post hoc test was used following other ANOVA results for comparing the six available groups.

Recruited participants were predominantly of Caucasian descent and Italian in nationality; only cross fit athletes were prevalently Spanish citizens. Anthropometric data of study participants (subdivided by sport type and gender) are reported in Table 1. Listed below are the sports that were included in each sport sub-type.

As expected, females weighed significantly less than their male counterparts in all sport types, F (5, 160) = 22.92, p < 0.0001. Using a two-way ANOVA test, we found that body mass index (BMI) was significantly different between the six group (F (5, 160) = 9.233, p < 0.0001). Tukey’s test for multiple comparisons revealed that the only significant gender difference in BMI found among the three sport type groups was between the males and females that were training in anaerobic sports (23.5 vs. 20.7 kg/m², p = 0.034). In addition, two-way ANOVA showed no gender dependent difference in training time (hours/week) (F (1, 160) = 0.2753, p = 0.6005, Supplemental Figure S1).

In this study, we evaluate if the Mediterranean diet, that is recommended among the general population to promote health [20,21,22], may provide the protein quantity necessary for optimal athletic performance. Indeed, the mean of protein daily intake for all study participants was 1.48 ± 0.46 g/kg/day, an amount within the AND, DC and ACSM joint statement range of recommended intake (1.2–2.0 g/kg/day), but significantly below its mean. Our results acquire even more significance considering the findings of a recently published systematic review by Jenner SL et al. [23], who demonstrated that among professional and semiprofessional athletes protein intake is often above the upper limit of the recommended range. A relevant observation is that in our group, 27 percent (n = 44) of the athletes fell below the minimum amount of protein intake of 1.2 g/kg/day. This percentage is rather large considering our population is comprised of competitive athletes.

Our survey data showed that the recruited athletes consumed macronutrients consistent with the Mediterranean diet as described by Davis et al. and Feart et al. [12,24]. Italy being a country with a moderate-to-high degree of Mediterranean diet adherence [25], it was expected that the athletes surveyed follow the Mediterranean Diet, but there was a slight variation in which the athletes sacrificed fat intake with protein. A general basic knowledge of the increase protein requirement in sport nutrition by these athletes was probably what permitted this small adjustment.

Although in the past, expert opinions and guidelines suggested that different sport activities required different protein intake [4,5], more recent guidelines do not maintain these differences and describe that protein intake of those training at higher frequency should fall in the highest end of the range [3]. Our data appear in agreement with this, because even though most of the amateur athletes we surveyed were not training at high frequency, we observed a positive correlation between protein intake and weekly training time.

In addition, we found that caloric distribution of macronutrients of amateur athletes are similar across sports types. Some studies evaluated protein dietary habits among elite athletes participating in different sport activities: in 2015, Baranauskas et al. [26] found that the protein intake was about 1.6 and 1.8 g/kg/day in a group of Lithuanian Olympic endurance athletes. Similarly, a study in Australian professional soccer players showed that protein intake was about 1.8 g/kg/day [27]. These studies suggest that athletes participating in sports relying on different metabolisms have similar protein intake. Moreover, amateur athletes tend to have lower protein intake; for example in a group of College athletes, a group more easily comparable with the one we evaluated in our survey with regard to training hours, a mean protein intake of 1.2 g/kg/day was observed [28].

We also did not find any notable gender-dependent difference in protein intake reported as grams per kilograms per day. There was no significant differences in training time between males and females when subdivided by sport group. In addition, protein intake was normalized to kilograms, which was the only data point that was significantly different between sexes. Our data suggest that our population consumes protein primarily based on weight (kg) and minimally based on hours of training, which is consistent with the guidelines.

When looking closer at the sub-group of athletes who did supplement their protein intake, an unexpected result was observed. Twenty-eight percent of the surveyed population supplemented their daily diet with protein supplements. Interestingly, when those who chose to supplement were separated from those who did not, the daily protein intake excluding intake from supplements was higher in those who chose to supplement, the exact opposite of what one would expect. One would presume that individuals who supplement their diet chose to do so because they are not meeting recommended values; however, these data suggest, that those who already have a sufficient level of daily intake with diet alone are the same individuals who also supplement their diet.

This type of pattern suggests that individuals who are aware of the importance of protein intake, and chose to supplement, are already making food consumption choices that would guarantee them an adequate amount of dietary intake.

Our survey shows, in addition, that in the subgroup not using protein supplements the percent of athlete below the suggested minimum range of 1.2 g/kg/day are high in aerobic, anaerobic and mixed, respectively, 14%, 48%, and 30%. Even considering that our data may not capture exactly the real nutritional status of these athletes due to the low accuracy of a three-day diary, these high values suggest that, aside from athletic coaching, nutritional coaching should be integrated into an athlete’s training living in Italy.

Overall, the Mediterranean diet appears as a dietary regimen able to provide sufficient protein to meet dietary recommendations in amateur athletes. A simple three-day food diary may be a feasible solution to raise awareness of an individual’s protein intake so to direct the necessary diet modification and/or suggest protein supplementation only when effectively needed.

Authors would like to acknowledge and thank Luca Aleotti, Sandro Bartolomei, Alessandro Bettazzi, Lorenzo Duca, Lorenzo Grimaldi, Federico Principi, Samuele Tomasini for recruiting subjects and collecting the survey data. Elena Casiraghi for help in the analysis of data. We would also like to thank the subjects that agreed to participate in the study.

The following are available online at https://www.mdpi.com/2072-6643/12/12/3681/s1↗, Figure S1: Training time (hours/week) of athletes grouped by sport type and then by gender. Two-way ANOVA revealed significant sport type interaction (F (2, 160) = 45.73, p < 0.0001), as expected. However, the two-way ANOVA did not show a gender difference (F (1, 160) = 0.2753, p = 0.6005). Tukey’s post hoc multiple comparisons test showed that there was not gender difference between sexes in any sport group (n.s. = p > 0.05); Figure S2: (a) Percent of carbohydrates composing subjects’ daily caloric intake. (b) Percent of lipids composing subjects’ daily caloric intake. (c) Percent of protein composing subjects’ daily caloric intake. A two-way ANOVA was used to see if there were any sport or gender dependent variation on percent of macronutrients. There was no significant interaction detected by two-way ANOVA for percent carbohydrates (F (2, 160) = 0.5463, p = 0.5802) and percent lipids (F (2, 160) = 2.850, p = 0.0608). The percent protein intake was significant for an interaction, F (2, 160) = 8.304, p = 0.0004), and Tukey’s multiple comparisons test showed a significant difference between genders solely in the anaerobic sport group (p < 0.05); Figure S3: (a) daily protein intake (excluding supplements) subdivided by sport type and sex. Two-way ANOVA showed that protein intake did not vary by gender ((F (1, 160) = 1.378, p = 0.2422). Tukey’s test for multiple comparisons did not show any significant differences between gender in any sport type, n.s. = p > 0.05. (b) daily protein intake (excluding supplements) subdivided by sex of the entire population, Student’s t test did not show any significant differences n.s. = p > 0.05. Figure S4: Correlation between total protein intake (including supplements) and weekly training hours for athletes participating in different sport categories, (a) aerobic, (b) anaerobic, (c) mixed. Correlation analysis revealed a significant (p < 0.05) direct relation between the two variables. Table S1: Percent of population subdivided by sport and then by gender that consumed less than the minimum recommended daily protein intake of 1.2 g/kg/day.

Design the study: A.L., M.C.; Data analysis: C.L.P., S.M., M.C., G.C., A.d.P., M.M., A.L.; Wrote the Manuscript: C.L.P., M.M., A.L.; Revised the manuscript: C.L.P., S.M., S.H., M.M., A.L. All authors have read and agreed to the published version of the manuscript.

This research was funded by University of Bologna RFO2018 funding to S.H, M.M., A.L.

The authors declare no conflict of interest.