Abstract
investigated age and performance in distance-limited ultra-marathons held from 50 km to 1,000 km. Age of peak running speed and running speed of the fastest competitors from 1969 to 2012 in 50 km, 100 km, 200 km and 1,000 km ultra-marathons were analyzed using analysis of variance and multi-level regression analyses. The ages of the ten fastest women ever were 40 ± 4 yrs (50 km), 34 ± 7 yrs (100 km), 42 ± 6 yrs (200 km), and 41 ± 5 yrs (1,000 km). The ages were significantly different between 100 km and 200 km and between 100 km and 1,000 km. For men, the ages of the ten fastest ever were 34 ± 6 yrs (50 km), 32 ± 4 yrs (100 km), 44 ± 4 yrs (200 km), and 47 ± 9 yrs (1,000 km). The ages were significantly younger in 50 km compared to 100 km and 200 km and also significantly younger in 100 km compared to 200 km and 1,000 km. The age of the annual ten fastest women decreased in 50 km from 39 ± 8 yrs (1988) to 32 ± 4 yrs (2012) and in men from 35 ± 5 yrs (1977) to 33 ± 5 yrs (2012). In 100 km events, the age of peak running speed of the annual ten fastest women and men remained stable at 34.9 ± 3.2 and 34.5 ± 2.5 yrs, respectively. Peak running speed of top ten runners increased in 50 km and 100 km in women (10.6 ± 1.0 to 15.3 ± 0.7 km/h and 7.3 ± 1.5 to 13.0 ± 0.2 km/h, respectively) and men (14.3 ± 1.2 to 17.5 ± 0.6 km/h and 10.2 ± 1.2 to 15.1 ± 0.2 km/h, respectively). In 200 km and 1,000 km, running speed remained unchanged. In summary, the best male 1,000 km ultra-marathoners were ~15 yrs older
(10.6 ± 1.0 to 15.3 ± 0.7 km/h and 7.3 ± 1.5 to 13.0 ± 0.2 km/h, respectively) and men (14.3 ± 1.2 to 17.5 ± 0.6 km/h and 10.2 ± 1.2 to 15.1 ± 0.2 km/h, respectively). In 200 km and 1,000 km, running speed remained unchanged. In summary, the best male 1,000 km ultra-marathoners were ~15 yrs older than the best male 100 km ultra-marathoners and the best female 1,000 km ultra-marathoners were ~7 yrs older than the best female 100 km ultra-marathoners. The age of the fastest 50 km ultra-marathoners decreased across years whereas it remained unchanged in 100 km ultra-marathoners. These findings may help athletes and coaches to plan an ultra-marathoner’s career. Future studies are needed on the mechanisms by which the fastest runners in the long ultra-marathons tend to be older than those in shorter ultra-marathons. Keywords:Ultra-marathon; Age of peak running speed; Running speed Background Ultra-marathon running is defined as any running race longer than the classical marathon distance of 42.195 km (ULTRAmarathon Running, 2013). Currently, the most common race distances held for distance-limited ultra- marathons are 50 km and 100 km as well as 50 miles and 100 miles (Deutsche Ultramarathon Vereiningung DUV 2013). During the last 20 years, ultra-marathon running be- came more and more popular (Da Fonseca et al. 2013; Hoffman 2010; Knoth et al. 2012). In particular, the par- ticipation of women and master athletes has increased over time in ultra-marathons (Hoffman and Weglin 2009; Eichenberger et al. 2012) where Reaburn and Dascombe (2008) have described master athletes as competitors older than 35 yrs and training regularly for organized sport events designed for older adults. Indeed, recent studies have demonstrated that the age of peak running speed for ultra-marathoners seemed to be above 35 yrs (Knechtle et al. 2012; Zingg et al. 2014a, b). The age of peak running speed in endurance sports depends highly on the kind of the athletic challenge (Young and Starkes 2005). The mean age of top finishers performing tasks requiring endurance, experience and knowledge, such as ultra-marathon running is in the early 30s (Knechtle et al.
be above 35 yrs (Knechtle et al. 2012; Zingg et al. 2014a, b). The age of peak running speed in endurance sports depends highly on the kind of the athletic challenge (Young and Starkes 2005). The mean age of top finishers performing tasks requiring endurance, experience and knowledge, such as ultra-marathon running is in the early 30s (Knechtle et al. 2012), whereas athletic chal- lenges such as short distance running were performed at a younger age in the early 20s (Schulz and Curnow 1998). However, it has been demonstrated that peak * Correspondence:beat.knechtle@hispeed.ch 2 Gesundheitszentrum St. Gallen, Vadianstrasse 26, 9001 St. Gallen, Switzerland Full list of author information is available at the end of the article a SpringerOpen Journal © 2014 Romer et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Romeret al. SpringerPlus2014,3:693 http://www.springerplus.com/content/3/1/693
running speed in endurance sports is not reached until the age of ~30 to ~35 yrs (Baker et al. 2003; Donato et al. 2003). Moreover, a decline in endurance running speed occurs after the age of ~50 yrs (Leyk et al. 2007; Leyk et al. 2009). A recent study performed by Zingg et al. (2013) showed that the age of peak running speed in ultra- marathons longer than 200 km was significantly higher than in ultra-marathons shorter than 200 km. In addition, an increase in running speed over time was reported (Zingg et al. 2013). Similar findings regarding the age of peak running speed have been reported for long-distance triathletes competing in the‘Ironman Hawaii’where the age of the annual ten fastest triathletes increased over time (Gallmann et al. 2014). Moreover, the best finishers im- proved performance over consecutive years (Gallmann et al. 2014). A long period of consistent training is required for a successful finish in an ultra-marathon (Knechtle et al. 2010a). For example, an analysis of the‘Western States 100 Mile Endurance Run’showed that the age of the fastest runners increased gradually over the years (Hoffman and Wegelin 2009). In addition, Hoffman and Krishnan (2013) reported in their‘UltraStudy’that the average ex- perience in endurance running was ~7 yrs before running the first ultra-marathon and that ~25% of the finishers had ~3 yrs or more of experience in ultra-marathon run- ning. Indeed, Hoffman and Parise (2014) recently demon- strated that an elite level of ultra-marathon performance could be maintained into the fourth decade of life (Hoffman and Parise 2014). These recent findings lead to the suggestion that the age of peak running speed may be higher in longer ultra-marathon distances compared to shorter running races. However, previous studies analyzed only individual events such as the‘Spartathlon’(Da Fonesca et al. 2013), the‘Badwater’(Da Fonesca et al. 2013) and the‘Western States 100 Mile Endurance Run’(Hoffman and Wegelin 2009). An analysis of the different lengths of ultra- marathon events over decades of time is lacking. There- fore, the assumption that the age of peak ultra-marathon performance increases with increasing race distance needs verification. The aim of
events such as the‘Spartathlon’(Da Fonesca et al. 2013), the‘Badwater’(Da Fonesca et al. 2013) and the‘Western States 100 Mile Endurance Run’(Hoffman and Wegelin 2009). An analysis of the different lengths of ultra- marathon events over decades of time is lacking. There- fore, the assumption that the age of peak ultra-marathon performance increases with increasing race distance needs verification. The aim of the present study was to deter- mine the age of peak ultra-marathon performance in distance-limited ultra-marathons covering 50 km, 100 km, 200 km and 1,000 km held worldwide over the period from 1969 to 2012. It was hypothesized that the age of peak ultra-marathon performance would be higher in the longer race distances. Methods Ethics All procedures used in the study were approved by the In- stitutional Review Board of Kanton St. Gallen, Switzerland with a waiver of the requirement for informed consent of the participants given the fact that the study involved the analysis of publicly available data. Data sampling and data analysis The data set for this study was obtained from the race website of the‘Deutsche Ultramarathon Vereinigung’ (Deutsche Ultramarathon Vereinigung (DUV, 2013) including all data of ultra-marathon finishes with name, gender, age, ranking and race time since 1969. The rela- tionship between age, race length and mean running vel- ocity were analyzed for all worldwide ultra-marathoners who raced from 1969 to 2012. Prior to 1969, no reliable data for a successful ultra-marathon finish were available. First, the performances of the ten fastest women and men ever for each race distance were determined. Sec- ond, to analyse the change in both the age of peak run- ning speed and running speed across years, race times of the annual fastest and the annual ten fastest for each race distance were determined. All race times were con- verted to running speed (km/h). The density in running speed between the winner and the 10th place was deter- mined (i.e.at least ten finishers in the respective year and race distance). Differences in running speed of the winner and 10th place finisher were analyzed using the equation [running speed-density between 1st and 10th
were determined. All race times were con- verted to running speed (km/h). The density in running speed between the winner and the 10th place was deter- mined (i.e.at least ten finishers in the respective year and race distance). Differences in running speed of the winner and 10th place finisher were analyzed using the equation [running speed-density between 1st and 10th placed athlete] = [running speed of 10th placed–running speed of 1st placed athlete]/[running speed of 1st placed athlete] × 100, and expressed as a percentage of the winner’s performance. Statistical analysis Each set of data was tested for normal distribution and for homogeneity of variances prior to statistical analyses. Normal distribution was tested using a D’Agostino and Pearson omnibus normality test and homogeneity of var- iances was tested using a Levene’s test. Trends in partici- pation were analyzed using regression analysis with ‘straight line’and‘exponential growth equation’model, whereas for each set of data (e.g.each gender) both models where compared using Akaike’s Information Cri- teria (AICc) to decide which model showed the highest probability of correctness. Differences in performance and athlete’s age of peak performance between race distances were investigated using one-way analysis of vari- ance (ANOVA) with subsequent Tukey-Kramer post-hoc analysis. Multi-level regression analyses were used to investigate changes in age and running speed of the fastest across calendar years. A hierarchical regression model was used to avoid the impact of a cluster-effect on results in case one athlete finished more than once in the top per- formers. Regression analyses of performance were cor- rected for age of athletes to prevent a misinterpretation of an‘age-effect’as a‘time-effect’. Statistical analyses were Romeret al. SpringerPlus2014,3:693 Page 2 of 13 http://www.springerplus.com/content/3/1/693
performed using IBM SPSS Statistics (Version 21, IBM SPSS, Chicago, IL, USA) and GraphPad Prism (Version 6.01, GraphPad Software, La Jolla, CA, USA). Significance was accepted atp<0.05(two-sidedfort-tests). Data in the text are given as mean ± standard deviation (SD). Results Between 1969 and 2012, a total of 47,393 women (17.5%) and 223,402 men (82.5%) finished a 50 km, a 100 km, a 200 km or a 1,000 km ultra-marathon. A total of 5,139 women and 22,079 men had to be excluded from data analysis due to missing information about age in the race results. Finally, full data from 205,577 finishers (4,254 women and 201,323 men) were available for data analysis. Participation trends The 100 km was the most popular distance with a total of 148,017 finishes (54.7% of all finishes) (18,998 women and 129,019 men) (Table 1, Figure 1). The 50 km was also a popular race distance with a total of 122,372 fin- ishes (45.2% of all finishes) (28,364 women and 94,008 men). Despite the fact that the popularity of the 200 km (0.11% of all finishes) (9 women and 285 men) and 1,000 km (0.04% of all finishes) (22 women and 90 men) ultra-marathons has also increased during the last years, the number of participants was low in these races. In 50 km, 100 km and 1,000 km, the number of finishers increased exponentially, whereas in 200 km the increase was linear (Figure 2). The age of the ultra-marathoners The most frequent participation of male and female age group athletes in 50 km was observed in the age group 40–44 yrs (Figure 3). In 100 km and 200 km, most of the runners were recorded in the age group 45–49 yrs for both genders. In 1,000 km, most female runners were in age group 35–39 yrs, whereas most male runners were in age group 55–59 yrs. Age and running speed of the ten fastest ever The athlete’s ages of the ten fastest women ever were 40 ± 4 yrs (50 km), 34 ± 7 yrs (100 km), 42 ± 6 yrs (200 km), and 41 ± 5 yrs
female runners were in age group 35–39 yrs, whereas most male runners were in age group 55–59 yrs. Age and running speed of the ten fastest ever The athlete’s ages of the ten fastest women ever were 40 ± 4 yrs (50 km), 34 ± 7 yrs (100 km), 42 ± 6 yrs (200 km), and 41 ± 5 yrs (1,000 km) (Figure 4). The athlete’sageof the ten fastest women was significantly different between 100 km and 200 km and between 100 km and 1,000 km (Table 2). No differences were found between the other race distances. For men, the athlete’s ages of the ten fastest ever were 34 ± 6 yrs (50 km), 32 ± 4 yrs (100 km), 44 ± 4 yrs (200 km), and 47 ± 9 yrs (1,000 km). The athlete’s ages were significantly younger in 50 km compared to 100 km and 200 km and also younger in 100 km com- pared to 200 km and 1,000 km (Table 2). No differences were found between 50 km and 100 km and between 200 km and 1,000 km. Running speed was fastest in 50 km and lowest in 1,000 km (Figure 4) in both women and men (Table 2). Changes in the age of peak running speed of the annual fastest runners across years In the 50 km events (Figure 5A), the age of the annual fastest women decreased significantly (Table 3) from 41 yrs (1977) to 26 yrs (2012). However, the age of peak running speed of the annual fastest men remained un- changed at 35.5 ± 7.3 yrs (Figure 5A). In contrast, in the 100 km events (Figure 5B), the age of the annual fastest men increased significantly (Table 3) from 27 yrs (1969) to 40 yrs (2012). The age of peak running speed of the annual fastest women remained unchanged (Table 2) at 34.6 ± 9.7 yrs (Figure 5B). In both 200 km (Figure 5C) and 1,000 km events (Figure 5D), the age of peak run- ning speed remained unchanged (Table 3). In 200 km ultra-marathons the age of peak running speed of
to 40 yrs (2012). The age of peak running speed of the annual fastest women remained unchanged (Table 2) at 34.6 ± 9.7 yrs (Figure 5B). In both 200 km (Figure 5C) and 1,000 km events (Figure 5D), the age of peak run- ning speed remained unchanged (Table 3). In 200 km ultra-marathons the age of peak running speed of the annual fastest women remained unchanged at 42.5 ± 5.1 yrs and of the annual fastest men at 42.8 ± 5.8 yrs. Similar results were found in 1,000 km where the age of peak running speed remained stable in women at 42.1 ± 5.5 yrs and in men at 45.3 ± 10.6 yrs. Changes in the age of peak running speed of the annual ten fastest In the 50 km events, the age of peak running speed de- creased significantly (Table 4) across years in both gen- ders (Figure 6A and B). The age of female finishers in 50 km decreased significantly (Table 4) from 39 ± 8 yrs (1988) to 32 ± 4 yrs (2012). The mean age of male 50 km ultra-marathoners decreased significantly (Table 4) from 35 ± 5 yrs (1977) to 33 ± 5 yrs (2011). In the 100 km events, the age of peak running speed of the annual ten fastest women and men remained stable at 34.9 ± 3.2 and 34.5 ± 2.5 yrs, respectively (Table 4). There were not enough data available for an analysis of age of peak run- ning speed in the 200 km and the 1,000 km events. Change in running speed of the annual fastest runners In the 50 km (Figure 7A) and the 100 km events (Figure 7B), running speed increased significantly in the annual fastest women and men. Peak running speed of all Table 1 Overall finishers by gender and race distance Distance Women Men Overall 50 km 28,364 (23.2%) 94,008 (76.8%) 122,372 (45.2%) 100 km 18,998 (12.8%) 129,019 (87.2%) 148,017 (54.7%) 200 km 9 (3.1%) 285 (96.9%) 294 (0.1%) 1,000 km 22 (19.6%) 90 (80.4%) 112 (0.1%) Overall 47,393 (17.5%) 223,402 (82.5%) 270,795 Finishers per distance
men. Peak running speed of all Table 1 Overall finishers by gender and race distance Distance Women Men Overall 50 km 28,364 (23.2%) 94,008 (76.8%) 122,372 (45.2%) 100 km 18,998 (12.8%) 129,019 (87.2%) 148,017 (54.7%) 200 km 9 (3.1%) 285 (96.9%) 294 (0.1%) 1,000 km 22 (19.6%) 90 (80.4%) 112 (0.1%) Overall 47,393 (17.5%) 223,402 (82.5%) 270,795 Finishers per distance with the number of both male and female finishers, the percentage of finishers of events in brackets and the overall finishers from 50 km to 1,000 km. Romeret al. SpringerPlus2014,3:693 Page 3 of 13 http://www.springerplus.com/content/3/1/693
Figure 1Number of all finishers sorted by gender and race distance. Figure 2Number of all 50 km (Panel A), 100 km (Panel B), 200 km (Panel C) and 1,000 km (Panel D) ultra-marathon finishes sorted by sex. Romeret al. SpringerPlus2014,3:693 Page 4 of 13 http://www.springerplus.com/content/3/1/693
female 50 km finishers increased linearly (Figure 7A) from 9.7 km/h (1977) to 15.3 km/h (2012) (Table 5). Men’speak running speed increased linearly from 16.9 km/h (1977) to 18.1 km/h (2012). Peak running speed in 100 km in- creased linearly (Figure 7B) from 8.1 km/h (1969) to 13.2 km/h (2012) in women and from 12.6 km/h (1969) to 15.7 (2012) km/h in men. In the 200 km events, running speed decreased linearly from 9.6 km/h (1988) to 7.1 km/h (2011) in women and from 11.1 km/h (1988) to 8.9 km/h (2011) in men (Figure 7C). There were no significant changes in running speed in 1,000 km events (Figure 7D) where peak running speed remained unchanged at 4.6 ± 0.6 km/h for women and at 5.5 ± 0.6 km/h for men. Changes in running speed of the annual ten fastest runners In the 50 km (Figure 8A) and 100 km events (Figure 8B), the annual ten fastest female and male runners became significantly faster across years (Table 6). In the 50 km events, peak running speed increased linearly from 10.6 ± 1.0 km/h (1988) to 15.3 ± 0.7 km/h (2012) in women and from 14.3 ± 1.2 km/h (1977) to 17.5 ± 0.3 km/h (2012) in men (Figure 8A). In the 100 km events, peak running speed increased linearly from 7.2 ± 1.5 km/h (1975) to 13.0 ± 0.2 km/h (2012) in women and from 10.2 ± 1.2 km/h (1969) to 15.1 ± 0.3 km/h (2012) in men (Figure 8B). Not enough data available for the analysis of the running speed of the top ten runners in 200 km and 1000 km. Density in running speed The density in running speed in the 50 km events (Figure 9A) decreased significantly in men from 125.0% (1977) to 104.7% (2012). Although the density in the 50 km events also decreased in women, it was not statisti- cally significant (p= 0.63). In 100 km (Figure 9B), the density decreased in both genders significantly (p< 0.0001). In women, it decreased from 145.1% (1975) to 103.9% (2012) and in men from 128.6% (1969) to 105.4% (2012).
Description
The study analyzes age and performance trends in ultra-marathons from 1969 to 2012.