Training Behaviors and Periodization Outline of Omnivorous, Vegetarian, and Vegan Recreational Runners (Part A)–Results from the NURMI Study (Step 2)

train for long-distance competitions based on underlying motivations, which may be similar to individual dietary motivations (e.g., well-being and performance). Fundamental training differences may arise in recreational runners following different diet types (omnivore, vegetarian, vegan) considering possible motive variations. Following a cross-sectional design, distance runners completed a survey (online), including a thorough assessment of training behaviors with generic training details and periodization speci cs in three phases: 1. an intermediary and rebound stage, 2. a main preparatory stage, and 3. a main event stage (tapering or interim event level/s). Kruskal–Wallis and chi-squared tests were used in the statistical analysis.

Following a cross-sectional design, distance runners completed a survey (online), including a thorough assessment of training behaviors with generic training details and periodization speci cs in three phases: 1. an intermediary and rebound stage, 2. a main preparatory stage, and 3. a main event stage (tapering or interim event level/s). Kruskal–Wallis and chi-squared tests were used in the statistical analysis. A total of 245 t recreational runners following omnivore (n= 109), vegetarian (n= 45), and vegan diets (n= 91) were included. Signi cant differences in the initial running motivation were found across dietary subgroups (p= 0.033) as well as for current motivations (p= 0.038), with vegetarians being the least health motivated (27% and 9%, respectively). No differences in each of the speci c periods were found between diet types across the outline (p> 0.05). The present evidence shows that there is a lack of fundamental training differences based on recreational runners following different generic types of diets. The results of the present investigation may be especially relevant for future studies on safety, sustainability, and performance-enhancing dietary practices among athletes. Keywords: plant-based; half marathon; marathon; recreational athlete; training; race; periodization; running 1. Introduction Running is considered a healthy physical activity for adults, with various bene ts for cognitive and physical function, that likely aids in the prevention of lifestyle-induced diseases [1]. People run long distances for various reasons (e.g., fun and tangible rewards) alongside or regardless of the health bene ts [2]. In addition, runners follow their current diet type for various purposes, including but not limited to performance, health and well- being, religion, taste, the environment, and animal welfare [3,4]. Thus, recreational runners adhering to general diet type categories may have different motivations for running, which could underlie differences in training behaviors [4]. In previous studies, diet type has been associated with a wide variety of factors such as athletic performance [5], attitudes towards food [6], physical recovery [7], and especially Nutrients2023,15, 1796.

training behaviors [4]. In previous studies, diet type has been associated with a wide variety of factors such as athletic performance [5], attitudes towards food [6], physical recovery [7], and especially Nutrients2023,15, 1796.

Nutrients2023,15, 1796 2 of 19 with outcomes of health [8,9]. An individual's dietary choices are well known to have an in uence on many different aspects of an individual's life, including but not limited to health behavior (e.g., physical activity levels) and social derogation [6,10,11]. The omnivore diet (inclusive of all food types) is by sheer prevalence the leading general type of diet in the world [12]. The Western omnivore diet is often composed of a majority of weekly calories from non-human animal-based sources (typically chicken, pig, cow, or cow by-products, including milk and dairy products) [13,14]. Likewise, with the exemption of meat, the Western vegetarian diet may also predominantly consist of animal-based products (e.g., eggs, yogurt, butter, cheese, and honey) [13,14]. However, any diet may be considered “plant-based” if fruit, vegetable, and grain consumption, for example, exceed 50% of caloric intake [15]. For distance runners switching to a vegan diet (100% plant-based), a noteworthy concern is suf cient caloric consumption and the use of dietary supplements in order to avoid speci c micronutrient de ciencies (e.g., vitamin B12, vitamin D, and zinc) [16], which could be challenging considering meal planning alongside training, working, family or social life, and adhering to regular rest and recovery times [15,17]. However, plant-based and especially vegan diets are deemed appropriate for athletes and have been associated with advanced bene ts among distance runners [16,18,19]. Training for running distance races with the basis of the periodization technique has been recommended since as early as 1999 [20]. To date, runners typically prepare for a main race with a series of three general training periods of stacking complexity (intermediary, main preparation, and race-focused) and thus follow an overall training duration (typically 3–4 months) [21]. Considerable variation in the FITT (Frequency, Intensity, Time, and Type) principle occurs between recreational runners within training periods, and the amount of focus applied to each training types is related to the individual's speci c training strat- egy [22,23]. Likewise, recreational runners typically participate in other physical activities parallel to running as the running movement occurs dominantly in the sagittal plane [24,25]. Therefore,

the FITT (Frequency, Intensity, Time, and Type) principle occurs between recreational runners within training periods, and the amount of focus applied to each training types is related to the individual's speci c training strat- egy [22,23]. Likewise, recreational runners typically participate in other physical activities parallel to running as the running movement occurs dominantly in the sagittal plane [24,25]. Therefore, frequent long-distance running bouts contribute to chronic postural, exibility, and strength imbalances (e.g., hip exor muscle groups), and participation in an array of physical activities may help to balance muscular tension [24,25]. Considering the intricate background of running training (e.g., high-intensity interval training) as well as perform- ing safely in long-distance races, especially for injury avoidance and adequate nutrition, professional support is a critical recommendation for recreational athletes [3,21,24]. In a pre- vious investigation, it was concluded that differences in running and training motivations between recreational runners following different diet types may help to explain possible training behavior differences [5]. Running motivations may, however, change over time or be different for competition in athletes and may even come secondary to motivations for other sports [26]. As training for long-distance races is the foundation of the recreational runner lifestyle [27–29], analyzing possible differences between diet types could enhance the explanation for running performance or even potential health risks. Nevertheless, only a few studies have been performed on food choice and/or diet type and training behavior among recreational runners [5,30–32]. To date, however, no study has investigated a comprehensive assortment of training behaviors and motives among healthy and t recreational distance runners following plant- based (vegetarian, vegan) or animal-based diets (omnivore, vegetarian). Therefore, this investigation is the rst to assess the motivations behind training and periodization of recre- ational runners based on following an omnivorous, vegetarian, or vegan diet. Considering the results of a previous NURMI investigation with a more general participant sample [5], this study was performed to test the hypothesis that there are fundamental training differ- ences between recreational runners capable of running at least a half-marathon distance based on following an omnivorous, vegetarian, or vegan diet type. 2. Materials

following an omnivorous, vegetarian, or vegan diet. Considering the results of a previous NURMI investigation with a more general participant sample [5], this study was performed to test the hypothesis that there are fundamental training differ- ences between recreational runners capable of running at least a half-marathon distance based on following an omnivorous, vegetarian, or vegan diet type. 2. Materials and Methods The Nutrition and Running High Mileage (NURMI) Study protocol [33] was submitted to the ethical board of St. Gallen, Switzerland, and accepted in May 2015 (EKSG 14/145). The

Nutrients2023,15, 1796 3 of 19 NURMI study trial registration was completed retrospectively (ISRCTN73074080). Involving a cross-sectional design, the NURMI study was carried out in three steps. Further details regarding the NURMI study Step 2 methods have been published previously [ The main subjects targeted for enrollment in this study were recreational runners from the countries of Austria, Germany, and Switzerland. Subjects were recruited through various sources, including social media sites, marathon event websites, personal con- tacts, runner communities online, and email subscriptions to runner magazines of health, lifestyle and nutrition, sports trade fairs, and plant-based lifestyle and diet. In Table, the characteristics of the recreational runners are shown. All subjects who took part in the NURMI study received a written explanation of the study procedure and completed the informed consent before enrollment. Over a period of 11 months, subjects completed an online survey that was available in German and English from. The survey was primarily based on physical health and food frequency, which included a basic allocation to a focal running area (well-being, hobby, competition) corresponding to the subject's respective running, exercise, and racing motives, and complimentary sports activities. Ful lling the following four inclusion criteria was essential for complete study partici- pation: written informed consent, aged 18 years or older, Step 2 survey completed, and participation and completion of a half-marathon event (or longer distance) within the last two years. Control questions were implemented throughout different sections of the survey in order to verify reliable answers for running activity (including motives, performance his- tory, and training details) as well as diet type; if the datasets submitted were contradictory, lacking critical answers, or con icting, they were excluded from the investigation. A total of 91 subjects had not completed an event of at least a half-marathon distance but had completed a 10 km race, and considering the high motivation of these subjects by their accurate and appropriate answers, recreational 10 km runners were included as an additional comparator group due to the high-quality data provided. Moreover, subjects must have chosen a race (long distance preferred) for the “NURMI running event”

event of at least a half-marathon distance but had completed a 10 km race, and considering the high motivation of these subjects by their accurate and appropriate answers, recreational 10 km runners were included as an additional comparator group due to the high-quality data provided. Moreover, subjects must have chosen a race (long distance preferred) for the “NURMI running event” (i.e., (ultra-)marathon or half-marathon distance), which they trained for and completed in Step 3 (the main NURMI study linked Steps 2 and 3) [33]. The subjects were classi ed into three dietary subgroups (with the minimum adher- ence of four weeks): omnivorous (commonly known as mixed or “Western diet”, with no dietary restrictions), vegetarian (no meat/ sh, but dairy and egg products) and vegan (no products from animal sources at all, inclusive (processed) meat, sh/seafood/shell sh, dairy products, eggs, honey) [16,38]. Analysis of control questions showed that 29 runners (11.8% of the study sample) had to be shifted to other dietary subgroups: 5 vegan runners (2 to omnivores and 3 to vegetarians) and 24 vegetarian runners (all to omnivores). Despite this, 88.2% of the total sample provided a correct self-report regarding their kind of diet. In addition, this study included a body mass index (BMI) procedure to exclude subjects with a BMI 30 kg/m 2 based on the health recommendation by the World Health Organization (WHO) [39,40]. The BMI procedure was therefore used to control for the subjects having a minimum health status associated with a minimum tness level, as safely reducing body weight requires other initial health protecting and weight management strategies besides running. In Figure, the nal subjects' ( n= 245) categorization into the dietary subgroups is shown along with the ow of study enrollment. Considering that ultra-marathon distances vary, the shortest distance that was completed by any subject in this study was 50 km, and the longest distance completed was 160 km.

Nutrients2023,15, 1796 4 of 19 Figure 1. Study enrollment owchart and dietary subgroup categorization. BMI—body mass index. HM—half-marathon; M/UM—marathon/ultra-marathon; 10 km—10 kilometers. The recreational runners' periodization outline and training details were summarized based on speci ed variables related to diet type: motivations for running (well-being, hobby, competition) seasonal (winter, spring, summer, autumn) or time of day (morning, afternoon, evening, night) running preferences (indoor and outdoor); overall training duration (from one month to over one year); source of training advice for running (none, quali ed professional, other); additional activities parallel to running (summer and winter); periodization outline, including three periods (1–3) and four training types speci ed for Period 2 (A, B, C, D) as well as weekly number of runs, weekly distance covered (km), weekly duration (hours), daily distance covered (km), and daily duration (hours) for each period and training type. Statistical analysis was performed using R software (version 4.2.2 UCRT 2022) from the R Foundation for Statistical Computing (Vienna, Austria;; accessed on 15 March 2023). Exploration analysis was conducted with descriptive analyses (including medians and interquartile ranges (IQR) as well as mean and standard deviations (SD)). Box plots were used to display the differences in the weekly number of runs and distance covered (km) as well as daily distance covered (km) by the dietary subgroups. The Kruskal–

Nutrients2023,15, 1796 5 of 19 Wallis test and the chi-square test were used to examine the differences between dietary subgroups. The level of statistical signi cance was set atp 0.05. 3. Results A total of 317 subjects submitted the questionnaire, and 72 (22.7%) were excluded due to not meeting the inclusion criteria. The nal sample was made up of 245 recreational endurance runners (42% male) of several self-reported race distances (NURMI-runners: HM, M, or UM:n= 154; 10 km runners:n= 91) with a median of 7 years (IQR 7) of experience active in running and racing. Regarding the subjects' academic quali cations (p = 0.029), 34% had completed an upper secondary education (or equivalent;n= 83) or held a university degree or higher (n= 83; 34%); however, some subjects had provided no answer for academic quali cations (n= 25; 10%), and one subject had no degree. The subjects were from various countries (Austria, Germany, or Switzerland:n= 234; Belgium, Brazil, Canada, Italy, Luxemburg, Netherlands, Poland, Spain, or United Kingdom:n= 11) and were grouped based on their self-reports of following an omnivorous diet (n= 109; 45%), a vegetarian diet (n= 45; 18%), or a vegan diet (n= 91; 37%). The vegan runners were mostly female (46% vs. 26% male;p= 0.004), whereas the omnivores were mostly male (56% vs. 36% female). The omnivores had the highest BMI with a median of 22.6 kg/m 2 (IQR 3.61), followed by the vegans with 21.3 kg/m 2 (IQR 3.21), and the vegetarians had the lowest at 20.8 kg/m 2 (IQR 3.46) (p= 0.001). Regarding the subjects' training focus, no signi cant difference was found based on the dietary subgroups (p= 0.139). A signi cant difference was found in the marital status of the runners based on diet type (p= 0.029), which showed that the omnivores were the least likely to be divorced (3%) or separated (22%), the vegetarians were most often among the single (38%), and the vegans were most common among the divorced or separated (11%). The traits, sociodemographics, and motives of the runners are provided in Table dietary subgroups; further details on these subjects are given

(p= 0.029), which showed that the omnivores were the least likely to be divorced (3%) or separated (22%), the vegetarians were most often among the single (38%), and the vegans were most common among the divorced or separated (11%). The traits, sociodemographics, and motives of the runners are provided in Table dietary subgroups; further details on these subjects are given in Part B of the sequenced investigation [41]. Table 1.Traits, Sociodemographics, and Motives of Runners Displayed by Dietary Subgroups.Total Omnivore Vegetarian Vegan Statistics 100% (245) 45% (109) 18% (45) 37% (91) Sex Female 58% (141) 36% (51) 18% (26) 46% (64) 2 (2)= 11.25 p= 0.004 Male 42% (104) 56% (58) 18% (19) 26% (27) Age (years) 39 (IQR 17) 43 (IQR 18) 39 (IQR 16) 37 (IQR 15) F (2, 242)= 2.95 p= 0.054 Body Weight (kg) 65 (IQR 14.2) 68 (IQR 16.7) 62 (IQR 11.3) 64 (IQR 10) F (2, 242)= 6.86 p= 0.001 Height (m) 1.7 (IQR 0.1) 1.7 (IQR 0.1) 1.7 (IQR 0.1) 1.7 (IQR 0.1) F (2, 242)= 2.45 p= 0.088 BMI (kg/m 2 ) 21.7 (IQR 3.5) 22.6 (IQR 3.61) 20.8 (IQR 3.46) 21.3 (IQR 3.21) F (2, 242)= 7.03 p= 0.001 Academic Quali cations No degree or certi cate <1% (1) / / 1% (1) 2 (4)= 10.78 p= 0.029 Upper secondary education/Technical quali cation/GCSE or equivalent 34% (83) 38% (41) 38% (17) 27% (25) A Levels or equivalent 22% (53) 24% (26) 16% (7) 22% (20) University degree/Higher degree 34% (83) 30% (33) 38% (17) 36% (33) No answer 10% (25) 8% (9) 9% (4) 13% (12)

Nutrients2023,15, 1796 6 of 19 Table 1.Cont. Total Omnivore Vegetarian Vegan Statistics Country of Residence Austria 18% (44) 21% (23) 18% (8) 14% (13) 2 (6)= 9.05 p= 0.171 Germany 72% (177) 70% (76) 76% (34) 74% (67) Switzerland 5% (13) 7% (8) 4% (2) 3% (3) Other 4% (11) 2% (2) 2% (1) 9% (8) Training Focus Well-being 9% (23) 9% (10) 2% (1) 13% (12) 2 (4)= 6.95 p= 0.139 Hobby 54% (133) 50% (54) 67% (30) 54% (49) Competition 36% (89) 41% (45) 31% (14) 33% (30) Preferred Racing Distance 10 km 37% (91) 34% (37) 33% (15) 43% (39) 2 (4)= 3.47 p= 0.483 HM 36% (89) 36% (39) 44% (20) 33% (30) M/UM 27% (65) 30% (33) 22% (10) 24% (22) Initial Motivation for Running Well-being 44% (108) 49% (53) 27% (12) 47% (43) 2 (2)= 6.82 p= 0.033 Hobby 56% (137) 51% (56) 73% (33) 53% (48) Current Motivation for Running Well-being 19% (47) 22% (24) 9% (4) 21% (19) 2 (4)= 10.15 p= 0.038 Hobby 46% (113) 38% (41) 64% (29) 47% (43) Competition 35% (85) 40% (44) 27% (12) 32% (29) Note. Results are presented as the percentage (%), total numbers, and median (IQR). 2statistic calculated by Pearson's chi-square test and F statistic calculated by the Kruskal–Wallis test. 10 km—10 kilometers. HM—half- marathon. M/UM—marathon/ultra-marathon. 3.1. Generic Runner Training Activity For the subjects' initial motivation for running, a signi cant difference was found across the dietary subgroups (p= 0.033) in which the vegetarians were most likely to report hobby (73%;n= 33), whereas the omnivores (49%;n= 53), followed by the vegans (47%;n= 43), were more likely to report well-being. Regarding the current motivation for running, a signi cant difference was found based on the dietary subgroups (p= 0.038), with the largest proportions of vegetarians (64%;n= 29) and vegans (47%;n= 43) running for hobby. For running preferences, no differences based on a particular season for outdoor (p= 0.351) or indoor (p= 0.669) running were identi ed for the dietary subgroups. No signi cant difference was found for the dietary subgroups based on the preferred time

This study analyzes training behaviors of omnivorous, vegetarian, and vegan recreational runners.