Gender and Age Differences in Performance of Over 70,000 Chinese Finishers in the Half- and Full-Marathon Events

) Background: The aim of the present study was to examine the characteristics of over 70,000 long-distance nishers over the last four years in Chinese half- and full-marathon events; (2) Methods:The available data of all nishers (n= 73,485; women,n= 17,134; men,n= 56,351) who performed half- and full-marathon events in Hangzhou from 2016 to 2019 were further analyzed for the characteristics of gender, age and average running speed; (3) Results: The

examine the characteristics of over 70,000 long-distance nishers over the last four years in Chinese half- and full-marathon events; (2) Methods:The available data of all nishers (n= 73,485; women,n= 17,134; men,n= 56,351) who performed half- and full-marathon events in Hangzhou from 2016 to 2019 were further analyzed for the characteristics of gender, age and average running speed; (3) Results: The total men-to-women ratio was the lowest in the half-marathon event (1.86) and the highest in the full-marathon event (17.42). Faster running performance in males than in females and faster average running speed in short-distance runners were shown. Gender and race distance were observed to have the most signi cant effects on average running speed (p< 0.01). For both male and female nishers, the slowest running speed was shown in older age groups (p< 0.01) during the full marathon. Our results indicated that the gender difference in performance was attenuated in the longer race distances and older age groups; (4) Conclusions: Understanding the participation and performances across different running distances would provide insights into physiological and biomechanical characteristics for training protocols and sports gear development in different groups. Keywords:gender; men-to-women ratio; marathon; age; running speed 1. Introduction The health bene ts of endurance exercise might partially explain the increase in participation in marathon races during the last decades [1]. In recent years, marathon running has been considered a globally popular physical activity that can cater to the various healthy lifestyle needs of urban residents [2,3]. This running boom has gradually spread around the world. Well-known New York, London, Paris, and Berlin marathon events all had between 30,000 and 40,000 nishers [4]. While distance running used to be a male-dominated sport, today females account for 43% of marathon runners in the USA [5]. Marathon events have developed later in China than in Western countries. The number of marathon events held in China increased from 12 to 53 between 2010 and 2014 [6] to approximately 1100 in 2017 [7], which involved nearly 5 million participants and increased to over 2.2 million participants from 2016 to 2017 [7]. The number further increased to a

[5]. Marathon events have developed later in China than in Western countries. The number of marathon events held in China increased from 12 to 53 between 2010 and 2014 [6] to approximately 1100 in 2017 [7], which involved nearly 5 million participants and increased to over 2.2 million participants from 2016 to 2017 [7]. The number further increased to a total of 1900 events in 2019 [1]. The growing popularity of running has inspired a large amount of research on running biomechanics, performance and sport Int. J. Environ. Res. Public Health2022,19, 7802.

Int. J. Environ. Res. Public Health2022,19, 7802 2 of 9 gears in the past decades [8]. Studies have found that proper pace can effectively reduce the risk of musculoskeletal injury [9], and different running strategies should be used in long-distance running according to gender, age and the event the runner is training for [10]. Moreover, differences in running biomechanics between Chinese men and women were observed, with female runners showing greater range of motion in the hip and knee joints, and a smaller shoe-to-ground angle during the heel touch-down phase. This is believed to be a form of self-regulation that women use to reduce the impact of landing, and men rely more on the performance of their shoes to achieve the purpose of buffering [11]. Following the increase in female participation in distance running [12], investigations into gender differences in running mechanics were intensi ed in the Western world [12–14]. The gender difference studies indicated clear differences in females' body fat and running speed [3], resulting in distinct movement characteristics and injury etiology. Nikolaidis et al. studied the performance and age composition of different genders during marathons, where they found that women achieved their best marathon race time ~5 years earlier in life compared to men. Women's participation increased disproportionately to men's participation, leading to an increase in the ratio of men to women [15]. In addition, more and more seniors are joining in marathon races [16]. The sex gap between elite female marathon racers and elite male runners may have reached its limit [17]. The age structure of most male marathon runners is larger and older than that of females (male: 40–44 years; female: 30–34 years) [18]. Although studies have been conducted on anthropometry, physiology and training characteristics have improved our understanding of the predictors of race time [19], as well as age- and gender-related differences in pacing during endurance running [20–23]. While Western counterparts have been extensively analyzed regarding their running characteris- tics, little attention has been paid to the Chinese population. Several articles have shown differences in running between Chinese and Western populations; one study found that,

improved our understanding of the predictors of race time [19], as well as age- and gender-related differences in pacing during endurance running [20–23]. While Western counterparts have been extensively analyzed regarding their running characteris- tics, little attention has been paid to the Chinese population. Several articles have shown differences in running between Chinese and Western populations; one study found that, compared with Western women, Chinese women use the medial forefoot more during the push-off phase of running [24]. Western female runners have greater ankle valgus angles than males, while there is no signi cant difference between Chinese females and males [25]. Thus, research ndings obtained from Western runners may not be directly applicable to the Chinese population because of racial differences. The present study examined trends of the men-to-women ratio, number of nishers and performances by gender and age groups across four years in half- and full-marathon events, respectively. It was expected that there would be a different men-to-women ratio and different performances across different running events. The comparison could help to better understand potential training require- ments for both elite and recreational runners of different age groups, as well as to effectively estimate the demand of running shoes for different gender and age groups. To optimize female performance and health in sport, we need to include women in our analyses in order to better understand peculiarities that may exist in physiology. Therefore, we are happy to enrich the existing pool of knowledge with more data on female participation and performance in marathon racing. Understanding the participation and performance across different running distances (half- and full-marathon events) would provide insights into the physiological and biomechanical characteristics for training protocols for different gender and age groups. 2. Materials and Methods 2.1. Participants and Data Acquisition The complete marathon event data for this study were of cially obtained from Hangzhou Marathon Organizing Committee (https://www.hzim.org) [ 26]. The records were collected from all half marathons and full marathons between 2016 and 2019, and were of cially certi ed by the World Athletics Organization. The Hangzhou marathon event included both full and half marathons

Participants and Data Acquisition The complete marathon event data for this study were of cially obtained from Hangzhou Marathon Organizing Committee (https://www.hzim.org) [ 26]. The records were collected from all half marathons and full marathons between 2016 and 2019, and were of cially certi ed by the World Athletics Organization. The Hangzhou marathon event included both full and half marathons and runs. Regretfully, the Hangzhou marathon event has been suspended due to the spread of COVID-19 in 2019. The study data included participants who completed the race in the appropriate amount of time. Age and gender

Int. J. Environ. Res. Public Health2022,19, 7802 3 of 9 information was provided for the period of 4 years. Ultimately, the study included a total of 73,485 participants (male,n= 56,351; female,n= 17,134). 2.2. Procedures Age intervals of ve years were selected to represent age groups among younger and older nishers in their categories. All runners over 71 years old were placed in one category, as there were only a few male runners in the oldest age group, while the oldest male runner was 74 years old. In total, the nishers were classi ed into 11 age groups; 21–25, 26–30, 31–35, 36–40, 41–45, 45–50, 51–55, 56–60, 61–65, 66–70 and 71+ years. Changes in gender participation are described by the men-to-women ratio (MWR, the quotient of males divided by female completers) [27]. 2.3. Statistical Analysis The of cial race time (i.e., accurate in seconds) was obtained for all nishers in both races. The average running speed in km/h was calculated using the nal race time (h) divided by race distance (km) to allow comparison of performances between two long- distance races. All descriptive statistics were reported as mean and standard deviation. Prior to statistical analyses, data distribution normality was veri ed by visual inspection of histograms and QQ plots [18]. To assess age and gender distribution among nishers in the half- and full-marathon events, a chi-square test ( 2) was performed. Statistical differences in marathon performance between 11 age groups and two events were observed. Meanwhile, their interactions were calculated using a two-way ANOVA, post hoc with Bonferroni-corrected tests, and the signi cance level was controlled at 0.05. All data were organized and summarized using Microsoft Of ce Excel 2019 (Microsoft Corporation, Redmond, WA, USA) and statistical testing was performed using SPSS 20.0 (IBM, Armonk, NY, USA). 3. Results 3.1. Participation by Gender, Race Distance, and Age Group The MWR as well as the total number of male and female nishers in each age group and race distance are presented in Table. Table 1.Distribution of male and female nishers in each age group and race distance. Half-Marathon Full-Marathon Age Groups Males Females

20.0 (IBM, Armonk, NY, USA). 3. Results 3.1. Participation by Gender, Race Distance, and Age Group The MWR as well as the total number of male and female nishers in each age group and race distance are presented in Table. Table 1.Distribution of male and female nishers in each age group and race distance. Half-Marathon Full-Marathon Age Groups Males Females Total MWR Males Females Total MWR 21–25 940 505 1445 1.86 639 158 797 4.04 26–30 3667 2166 5833 1.69 3046 714 3760 4.27 31–35 4381 2063 6444 2.12 5110 1073 6183 4.76 36–40 4360 1712 6072 2.55 6274 1249 7523 5.02 41–45 3514 1538 5052 2.28 6733 1325 8058 5.08 46–50 2820 1404 4224 2.01 6573 1386 7959 4.74 51–55 1428 633 2061 2.26 3569 678 4247 5.26 56–60 618 179 797 3.45 1540 193 1733 7.98 61–65 208 47 255 4.43 502 64 566 7.84 66–70 111 20 131 5.55 209 12 221 17.42 71+ 28 0 28 - 18 0 18 - Total 22,075 10,267 32,342 2.15 34,213 6852 41,065 4.99 MWR = men-to-women ratio. The total MWR was 2.15 and 4.99 in the half and full marathon, respectively. A gender race distance association in participation was shown ( 2= 2294.505,p< 0.01, '= 0.177). A gender age group association in participation was observed in the half marathon ( 2 = 89.091,p< 0.01,'= 0.081) and in the full marathon ( 2= 53.431,p< 0.01, '= 0.050). Furthermore, a race distance age group association in participation was

Int. J. Environ. Res. Public Health2022,19, 7802 4 of 9 shown for male nishers ( 2= 1687.054,p< 0.01,'= 0.187) and for female nishers ( 2 = 600.388,p< 0.01,'= 0.197) of different age groups. In the half marathon, the lowest MWR was observed in the age group of 26–30 years (1.69), whereas the highest MWR was observed in the age group of 36–40 years (6.12). In the full marathon, the lowest MWR was observed in the youngest age group (4.04), whereas the highest MWR was observed in the age group of 66–70 years (17.41). 3.2. Performance (Average Running Speed) 3.2.1. Overall Effects The two-way ANOVA showed significant effects of gender [F (1, 73,365)= 612.757, p< 0.001] and race distance [F (1, 73,365)= 7.914,p< 0.005], as well as age group [F (10, 73,365)= 119.550 ,p< 0.001]. Moreover, we found significant interactions for age group race distance [F (10, 73,365)= 3.763 ,p< 0.001], while no interaction was observed forgender race distance [F (1, 73,365)= 3.270,p= 0.071] and for gender age group [F (9, 73,365)= 1.329,p= 0.216]. 3.2.2. Performance by Gender and Race Distance A signi cant effect of gender on average running speed is shown (p< 0.001) in Figure, where male nishers (with the performance of 10.03 1.67 km/h) were faster than female nishers (with the performance of 9.08 1.27 km/h). In addition, a signi cant effect of race distance on average running speed was observed (p< 0.001). Figure that performance in the full-marathon event (9.95 1.71 km/h) was faster than that in the half-marathon event (9.63 1.53 km/h).Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW 4 of 9 The total MWR was 2.15 and 4.99 in the half and full marathon, respectively. A gen- der × race distance association in participation was shown (χ 2 = 2294.505, p < 0.01, φ = 0.177). A gender × age group association in participation was observed in the half mara- thon (χ 2 = 89.091, p < 0.01, φ = 0.081) and in the full marathon (χ 2 = 53.431, p < 0.01, φ = 0.050). Furthermore, a race distance ×

association in participation was shown (χ 2 = 2294.505, p < 0.01, φ = 0.177). A gender × age group association in participation was observed in the half mara- thon (χ 2 = 89.091, p < 0.01, φ = 0.081) and in the full marathon (χ 2 = 53.431, p < 0.01, φ = 0.050). Furthermore, a race distance × age group association in participation was shown for male finishers (χ 2 = 1687.054, p < 0.01, φ = 0.187) and for female finishers (χ 2 = 600.388, p < 0.01, φ = 0.197) of different age groups. In the half marathon, the lowest MWR was observed in the age group of 26–30 years (1.69), whereas the highest MWR was observed in the age group of 36–40 years (6.12). In the full marathon, the lowest MWR was observed in the youngest age group (4.04), whereas the highest MWR was observed in the age group of 66–70 years (17.41). 3.2. Performance (Average Running Speed) 3.2.1. Overall Effects The two-way ANOVA showed significant effects of gender [F (1, 73,365) = 612.757, p < 0.001] and race distance [F (1, 73,365) = 7.914, p < 0.005], as well as age group [F (10, 73,365) = 119.550, p < 0.001]. Moreover, we found significant interactions for age group × race dis- tance [F (10, 73,365) = 3.763, p < 0.001], while no interaction was observed for gender × race distance [F (1, 73,365) = 3.270, p = 0.071] and for gender × age group [F (9, 73,365) = 1.329, p = 0.216]. 3.2.2. Performance by Gender and Race Distance A significant effect of gender on average running speed is shown (p < 0.001) in Figure 1, where male finishers (with the performance of 10.03 ± 1.67 km/h) were faster than fe- male finishers (with the performance of 9.08 ± 1.27 km/h). In addition, a significant effect of race distance on average running speed was observed (p < 0.001). Figure 1 also shows that performance in the full-marathon event (9.95 ± 1.71 km/h) was faster than that in the half-marathon event (9.63 ± 1.53 km/h).

± 1.67 km/h) were faster than fe- male finishers (with the performance of 9.08 ± 1.27 km/h). In addition, a significant effect of race distance on average running speed was observed (p < 0.001). Figure 1 also shows that performance in the full-marathon event (9.95 ± 1.71 km/h) was faster than that in the half-marathon event (9.63 ± 1.53 km/h). No gender × race distance interaction on average running speed was found (p > 0.05), while the gender difference was lower in the half-marathon event (+9.98%) than in the full-marathon event (+10.25%). Half-marathon finishers were slower than full-marathon finishers among females (9.02 ± 1.22 versus 9.17 ± 1.35 km/h, respectively, p < 0.001), as well as among males (9.92 ± 1.57 versus 10.11 ± 1.73 km/h, respectively, p < 0.001). Figure 1. Race speed by race distance and gender. Error bars represent standard deviations. ˆˆp< 0.001; **p< 0.001. No gender race distance interaction on average running speed was found (p> 0.05), while the gender difference was lower in the half-marathon event (+9.98%) than in the full-marathon event (+10.25%). Half-marathon nishers were slower than full-marathon nishers among females (9.02 1.22 versus 9.17 1.35 km/h, respectively,p< 0.001), as well as among males (9.92 1.57 versus 10.11 1.73 km/h, respectively,p< 0.001).

Int. J. Environ. Res. Public Health2022,19, 7802 5 of 9 3.2.3. Performance by Age Group and Race Distance The age group race distance interaction had a signi cant effect on average running speed (p< 0.001). Under closer examination of the performance, male nishers had the fastest average running speed of 10.33 1.60 km/h, while female nishers had the slowest average running speed of 8.54 1.03 km/h, regardless of the type of event. In the half-marathon event, the fastest male age group was 61–65 years (average running speed of 10.32 1.45 km/h), while the slowest male age group was 26–30 years (average running speed of 9.61 1.57 km/h). In the full-marathon event, the male nishers had the fastest speed of 10.39 1.62 km/h in the 46–50 age interval and the slowest speed of 9.16 1.18 km/h in the 71+ age interval (Figurea).Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW 5 of 9 Figure 1. Race speed by race distance and gender. Error bars represent standard deviations. ^^ p < 0.001; ** p < 0.001. 3.2.3. Performance by Age Group and Race Distance The age group × race distance interaction had a significant effect on average running speed (p < 0.001). Under closer examination of the performance, male finishers had the fastest average running speed of 10.33 ± 1.60 km/h, while female finishers had the slowest average running speed of 8.54 ± 1.03 km/h, regardless of the type of event. In the half-marathon event, the fastest male age group was 61–65 years (average run- ning speed of 10.32 ± 1.45 km/h), while the slowest male age group was 26–30 years (av- erage running speed of 9.61 ± 1.57 km/h). In the full-marathon event, the male finishers had the fastest speed of 10.39 ± 1.62 km/h in the 46–50 age interval and the slowest speed of 9.16 ± 1.18 km/h in the 71+ age interval (Figure 2a). In the half-marathon event, female finishers had the fastest average running speed of 9.61 ± 1.21 km/h in the 66–70 age interval, but the slowest average running speed of 8.65 ± 1.13

This study analyzes gender and age differences in marathon performance among over 70,000 participants.