1. Introduction
Low micronutrient intake in athletes can result in deficiencies affecting health and performance, in particular when this occurs for longer periods of time [
1]. Some studies report that many athletes do not meet micronutrient recommendations [
2,
3,
4,
5], whereas others conclude the opposite [
1,
6]. Shifts and variations in food patterns over time [
7,
8], increased availability of nutritional supplements [
9], and changing viewpoints regarding requirements [
10,
11] merit regular monitoring of dietary intake by athletes. At the same time, rapid developments in assessment tools, such as web-based approaches, make it easier to gain insight in the intake of large groups of athletes [
12,
13].
No consensus exists on whether micronutrient requirements are different in athletes as compared to the general population [
14,
15]. In practice, athletes are nowadays often advised to meet the general recommended dietary reference intakes (DRI) for all micronutrients by consuming a diverse diet to ensure nutrient adequacy [
16], paying special attention to optimal intake of iron, vitamin D and calcium, and of antioxidants [
17]. There seems to be less attention on the intake of B-vitamins [
17]. Theoretically, exercise could increase the need for this group of micronutrients [
14]. However, if energy expenditure increases, food intake increases as well, which potentially could result in a higher vitamin B intake. Unfortunately this is not necessarily the case, in particular when athletes make poor dietary choices, resulting in a lower micronutrient intake than expected [
14] because of a low micronutrient density of the diet [
5,
6,
18].
Nutritional supplements are frequently used by athletes [
19], although use can be irregular and varying over time [
20]. Dietary supplements mainly comprise micronutrient supplements such as vitamins and minerals. These supplements may promote athletes’ general health through the prevention and treatment of nutrient deficiencies. On the other hand, sport nutrition products mainly contain macronutrients, such as carbohydrates and protein, but can also contain micronutrients. These sport nutrition products include, but are not limited to, sports drinks, recovery drinks, and sports bars. Finally, the category of ergogenic supplements, such as creatine and caffeine, may contain additional micronutrients although these products are also available as single bioactive substance products [
21]. Therefore, nutritional supplements can be an important source of micronutrient intake.
Even though many athletes use multiple nutritional supplements at a time, this does not necessarily guarantee adequate habitual dietary intake of micronutrients at an individual level [
7]. In addition, case reports have shown that some athletes consume very high doses of certain micronutrients exceeding the upper level (UL) also defined as tolerable upper intake level [
11], possibly resulting in reduced health and performance in the long term [
22]. Examples of frequently reported highly dosed micronutrient supplements used by individual athletes are antioxidants (i.e., vitamin C and E) [
23], vitamin D, iron and magnesium [
21]. Further, practical experience and anecdotal reports suggest a substantial use of high doses of vitamin B6 among athletes.
In the present study we aimed to evaluate the adequacy of micronutrient intake of Dutch elite and sub-elite athletes, using a web-based 24-h recall method with accompanying nutritional supplement questionnaires. In addition, we aimed to assess the effect of nutritional supplements use on micronutrient intake, by making a comparison between both users and non-users of dietary supplements, sport nutrition products or a combination of both. Ultimately this should lead to better identification of athlete groups at risk of inadequate micronutrient intake, either too low or too high.
3. Results
A substantial part of the 759 athletes (
n = 206) provided incomplete dietary information, i.e., less than 2 complete days. As a consequence, a total of 553 athletes, both men (59%) and women (41%), were included in the final analysis of this study. All results described in this paragraph are reported in
Table 1. They recorded an average of 2.83 days by 24-h recalls and accompanying nutritional supplement intake questionnaires per person. For men and women respectively, 157 and 83 participants were classified as endurance athletes, 138 and 104 participants as team sports athletes, and 32 and 39 participants as strength athletes as described in Wardenaar et al. [
24]. Mean age of the men was 23.5 ± 11.5 years and of the women 22.0 ± 7.6 years, and mean exercise time was 93.5 ± 61.3 min per day. Reported energy intake was on average 11.7 ± 3.2 MJ for men and 9.1 ± 2.5 MJ for women. Mean food intake level (FIL) values, as an indicator of the quality of energy reporting, were slightly higher for men (mean ± SD: 1.55 ± 0.39) than for women (1.50 ± 0.42) and differed between non-users and users of nutritional supplements resulting in a somewhat lower FIL in those not using supplements. Non-users of DS and SNP were mainly younger athletes in comparison to the other categories and particular users of DS + SNP were the oldest and reported the highest exercise load.
Of the total group of athletes, 61.8% reported the use of one or more nutritional supplements; 65% in men, and 56% in women. The use of DS alone was slightly lower in men (20%) than women (24%), while the use of SNP alone was more frequently reported by men (24%) than women (17%), as was the combined use of DS + SNP (men: 21%, women: 15%).
3.1. Basal Diet Intake
Food group intake is shown in
Table 2 showing that median intake of athletes in all categories shows a divers consumption pattern. The most consumed food groups contributing to micronutrient intake were: milk products, meat, vegetables, cereals and grains and potatoes, fruit and bread. The consumption of eggs, fish and vegetarian meat-replacement products was low. In most cases, vegetable and fruit consumption was below the minimum recommendations of 150 grams and 2 servings respectively.
For men, micronutrient intake from the basal diet, i.e., excluding nutritional supplements, was in most cases lower in non-users than in all subgroups of users (
Table 3,
p < 0.05). In women this difference was less pronounced except for the users of DS + SNP, who reported higher mean intakes than non-users, DS users and SNP users (
p < 0.05).
3.2. Total Micronutrient Intake
When nutritional supplements were included, users of DS and DS + SNP reported a higher total intake of all micronutrients as compared to the basal diet for both men and women (p ≤ 0.05). For SNP users (both men and women) the same trend was seen, except for vitamin A, vitamin D, iron, copper, selenium and zinc, for which values did not differ with the basal diet after including SNP in the calculation (p > 0.05).
Differences in mean intake existed between subgroups for almost all nutrients. Intake for most micronutrients was higher in users of DS and DS + SNP than in non-users and users of SNP in both men and women (p ≤ 0.05). In men, users of SNP also reported higher intake of most nutrients than non-users (p ≤ 0.05), except for vitamin A and D. Women using SNP only reported higher intakes than non-users for vitamin B2, B3, B6, folate equivalents and selenium (p ≤ 0.05).
3.3. Adequacy of Micronutrient Intake of the Basal Diet
Almost all athletes (>85%), both men and women, were not able to meet the AR of vitamin D (
Table 4). The basal diet was also low for vitamin A (20%–48%) and vitamin B1 (18%–40%) for both men and women. Also, the intake of vitamin B2 was below the AR in some of the athletes, mainly in non-users and users of SNP in men and women (with a prevalence of low intakes varying between 12% and 17%), except for men using DS + SNP, who showed only a prevalence of low intakes of 4%. In general, non-users had the highest prevalence of micronutrient intakes below the AR. They showed low intakes of folate equivalents and selenium in both men (15% and 11% respectively) and women (25% and 11% respectively) as well as low intakes of vitamin C in men (11%). In addition, the basal diet of women, but not of men, showed within all subgroups of users and non-users, a prevalence of iron intakes below the AR. However, the overall athletes’ basal diet was providing sufficient amounts of vitamin B3, vitamin B12, vitamin E, calcium, phosphorus, magnesium, copper and zinc.
3.4. Adequacy of Total Micronutrient Intake including Nutritional Supplements
If nutritional supplements were included in the analyses, vitamin D intake of DS and DS + SNP users improved. The contribution of dietary supplements reduced the prevalence of athletes below the AR of vitamin D, but still low intakes were seen in 43% and 28% of the men and 19% and 35% of the women for DS or DS + SNP, respectively. Additionally, the use of DS improved intakes of vitamin B1, B2, folate equivalents and vitamin A, and ensured that most intakes exceeded AR. When the use of DS was included in the analysis, iron intake improved in DS-users. Women using DS + SNP showed no prevalence of intakes below the AR for iron whereas 9% of women reporting only DS still showed an iron intake below AR. In general, no beneficial effect on micronutrient intake was seen because of using SNP, except in women using SNP; the prevalence of women meeting AR for vitamin B1 and vitamin A improved.
3.5. Micronutrient Intake Exceeding UL
If the defined upper level (UL) was available for a specific micronutrient, this was shown in
Table 5. The high exceedance of the UL for phosphorus in men was the result of a high intake from the basal diet. The prevalence of athletes with a micronutrient intake above the UL based on the basal diet without supplements (non-users and users of nutritional supplements excluding supplements) was in all other cases negligible.
The micronutrient intake of male users of nutritional supplements exceeded the UL for vitamin B3, in 22% and 16% using DS or DS + SNP, respectively. Of the women, 17% and 34% of users of DS or DS + SNP respectively were reporting above the UL for vitamin B3. In addition, the prevalence of high intakes of vitamin B6 (exceeding the UL with 9% and 11% for DS or DS + SNP users) and vitamin A (exceeding the UL with 8% and 17%) in women is worth mentioning. All other instances of micronutrients exceeding the UL were ≤4% in both men and women.
4. Discussion
In this survey among 553 elite and sub-elite Dutch athletes, it was shown that the basal diet alone, despite a reasonable diversity of reported food groups, often did not provide enough micronutrients to ensure adequate intake levels. However, when DS were included in the analyses, we saw significantly higher micronutrient intakes in users compared to non-users. Our study also revealed that vitamin D intake was a problem in all athletes. In addition, non-users of dietary supplements were particularly at risk for low intakes of vitamin B1, B2 and vitamin A. They also reported a low intake of vitamin B3, vitamin C and selenium, but this was less pronounced. Iron intake was below the AR in some but not all female athletes, but this appeared to be compensated for when dietary supplements were taken into account. Micronutrient intake above the upper level (UL) was only observed in those using DS, especially in case of vitamin B3.
4.1. Micronutrient Intake
The majority of the athletes (62%) involved in our study was using one or more dietary supplements, sport nutrition products or a combination of both, which is in line with a recently published meta-analysis [
34]. On the other hand, DS and (or) SNP were not consistently used ‘daily’ in the current study, and not by all athletes. Nevertheless, reported micronutrient intake was substantial higher in athletes using supplements compared to those not using these products, which is in line with previously published data for the general Dutch population [
35].
Remarkably, when nutritional supplements were not taken into account in the analysis, the intake of most micronutrients was still higher in users than in non-users, as was reported in
Table 1. This difference in intake from the basal diet between users and non-users of dietary supplements and sport nutrition products may be a result of the higher total energy intake reported by users, because micronutrient intake correlates directly to energy intake for most micronutrients [
7].
The current study did not reveal problems regarding vitamins B6 and B12, vitamins C and E, most minerals and trace elements as suggested by other studies [
1,
5,
6,
36]. This may be the result of a relatively high dairy and combined intake of bread, cereals and grains, which is common in the Netherlands [
8].
4.1.1. Vitamin D, Iron and B Vitamins
In all subcategories, the intake of vitamin D via basal diet was below the AR of 7.5 µg. Basal vitamin D intake was lower than previously reported for other groups of athletes [
3,
4,
5,
7], mainly because of the low consumption of vitamin D-containing foods such as butter and/or margarine, and of fatty fish in our population. An adequate vitamin D status in a young healthy population is associated with improved bone health and muscle strength and function [
11,
37]. Regarding bone mineral density, although low vitamin D intakes were seen, the calcium intake was above the AR and exceptionally low energy intakes were not seen in the present study [
17]. When the use of vitamin D containing dietary supplements was included in the analysis, a substantial improvement of vitamin D intake was seen. Unfortunately, we did not assess the vitamin D status itself in our population, which is mainly determined by seasonal exposure to sunlight and dietary intake. Therefore, we can only speculate on the effect of the dietary intake of vitamin D on blood status. It was previously shown in a large group of Dutch athletes that they could be at risk for developing a low 25(OH)D status, especially during winter months [
38]. Taking this into account, all athletes should be advised to use a dietary supplement containing 2.5–5 μg vitamin D in order to meet at least the suggested nutrition recommendations.
Almost one third of the female athletes experienced difficulties meeting the AR for iron through their basal diet, which confirms earlier observations in athletes [
5,
6,
7,
36]. A low iron intake increases the risk for iron deficiency or anaemia in which the oxygen-binding part of haemoglobin plays an important role [
11]. In women, iron losses are generally higher through menstrual blood loss [
39]. On the other hand, female athletes can suffer from amenorrhoea as a result of the high volume of exercise which could (temporarily) reduce the need for iron in the diet [
40]. Athletes can also suffer from exercise-induced haemolysis, due to mechanically induced damage to red blood cells, which results in a higher iron loss by the body [
41]. A low dietary iron intake in combination with exercise-induced haemolysis does not necessarily lead to iron deficiency (anaemia), as for example was seen in gymnasts [
41]. Anyway, the AR is suggested as the lower minimal level that should be met by all athletes. Athletes should be advised to select proper sources of iron, from both animal and plant-based sources, and the use of a multivitamin and mineral supplement by women could add to dietary iron intake in order to meet the recommended levels.
A substantial fraction of both non-users and users of only SNP reported an intake below the AR for vitamin B1 and B2 (thiamine and riboflavin, respectively) in both men and women, and for folate equivalents in women. Theoretically, exercise increases the need for vitamin B1 and B2, as a result of a decreased absorption and (or) an increased turnover related to tissue maintenance, repair and metabolism [
14]. It is generally assumed that athletes with a poor thiamin and riboflavin status have a reduced ability to perform physical activity, especially performing maximal work [
14]. Previously, vitamin B1 and B2 were not seen as a particular nutritional problem for Dutch athletes [
7]. A comparable prevalence for low intakes of folate were reported in 20%–25% of Dutch women of the general population [
8]. The observed energy intake in the present study was not exceptionally high in comparison to the Dutch general population. Possibly this resulted in lower B vitamin intake as confirmed by others [
1,
5,
6,
36], except for vitamin B2 [
1,
5,
6,
36]. To cover the requirements of specific B vitamins, athletes should be encouraged to frequently select vitamin B1 rich foods (for example lean meat, legumes that were almost not present in the diet of these Dutch athletes) and vitamin B2 rich foods (for example eggs, lean meat and specific dairy products).
4.1.2. Anti-Oxidants
Exercise is associated with increased oxidative stress [
42,
43,
44]. Notwithstanding the continuing debate on the relation between their oral intake and the anti-oxidant status of the body, athletes are currently advised not to use anti-oxidant supplements as they could negatively influence protein signaling and adaptation in relation to oxidative stress [
42,
43,
44]. Although the bioactive anti-oxidative substances with the diet transcends the micronutrients included in this study, the basal dietary intake of vitamin C, vitamin E, selenium and zinc was in most cases above the AR in the present study. Only the basal intake of vitamin A, reported in this study as retinol activity equivalents (i.e., including b-carotene), was low in a substantial number of all athletes. Therefore, consuming a well-chosen diet, rich in a large variety of fruits and vegetables containing b-carotene and retinol containing products, should be considered to optimize the antioxidant capacity of the diet [
17,
45].
4.2. Micronutrients Exceeding UL
The prevalence of micronutrient intake exceeding the UL based on dietary supplement use was very low, i.e., a prevalence of 1%–4%, with the exception of users of dietary supplements containing nicotinic acid (vitamin B3), vitamin B6 and (or) vitamin A. High vitamin B intakes, as a result of the use of dietary supplements, could lead to side effects like flushing, and to more severe health problems (vitamin B3) [
11] and neuropathy (vitamin B6) [
46].
In the current study, eighteen athletes reported an intake of vitamin B6 above the generally accepted UL of 25 mg [
47]. Therefore, athletes should be advised to check their dietary supplements and avoid a combination of (highly dosed) nutritional supplements.
The UL is defined to prevent toxicity due to excessive micronutrient consumption over a long term. However, for example, high intake levels of vitamin C and E below the UL have been associated with attenuated skeletal muscle adaptations and protein signaling [
43,
48]. In the current study, no high doses of vitamin E were reported, but a substantial number of athletes reported the use of high-dosed vitamin C supplements (i.e., 1000 mg or higher). At present no consensus exists on the effect of high doses of vitamin C on redox signaling [
23], thus athletes should preferably maintain an appropriate vitamin C status by selecting a large variety of fruits and vegetables.
4.3. Strengths and Limitations
The large elite athletic population and the unannounced 24-h recalls and questionnaires were a strong aspect of our study. Also, dietary reports included two week days and one weekend day. By using the method as described by Nusser et al. (1996), we removed the within-person variation and were able to compare the distribution of usual intake with the average requirement [
32]. With the inclusion of three days for almost all participants, the method was suitable to sufficiently correct for within-person variation [
49]. Based on low D
t, we conclude that the effect of random errors for most nutrients was low, except for vitamin B1 in women not using nutritional supplements which makes it more difficult to draw conclusions for this value than for the others [
30].
A limitation of this study is that we did not have the possibility to measure micronutrient concentrations in blood. On the other hand, good biochemical indicators of dietary intake or status are not available for all micronutrients, or they only reflect short-term micronutrient intake. There is only a limited number of biomarkers available to assess usual micronutrient intake (for example folic acid concentrations in erythrocytes reflecting intake of the past six weeks), especially when only single measurements are taken [
29]. Also, concentration biomarkers in blood cannot be used to estimate absolute levels of intake. Thus, although self-reports of micronutrients have errors and inter-individual differences exist regarding individual needs, it can be assumed that the percentage of micronutrient intakes below the AR reflect the best estimate of the percentage of the study population with intakes that are below their individual requirements [
32]. Nevertheless, we recommend that future research also takes into account micronutrient status because it provides objective information in addition to the self-reports.
All self-reporting methods are prone to misreporting and may not completely reflect true dietary intake [
12,
49]. Based on our previous validation study we know that our method, similarly to other self-reports, underestimates intake [
13]. Therefore, we calculated the FIL as an estimator for possible energy under-reporting. The FIL is especially lower for the non-users of supplements in both men and women. As the non-users were reporting the lowest workload (exercise minutes per day) it is expected that they also report lower energy intakes. Although FIL values for non-users are substantially lower than those of users of nutritional supplements, this falls within an acceptable range [
13]. If we assume that micronutrient intake was not biased by selective under-reporting, correcting for under-reporting would overall result in a higher intake per person and thus actual micronutrient intake would be somewhat higher and inadequacy lower than presented.
Our study shows that there is still room for improvement of the athletes’ diet. With regards to the use of dietary supplements, several suggestions can be made for well-informed use. These suggestions are provided by
Table 6.