Performance Trends in Age Group Triathletes in the Olympic Distance Triathlon at the World Championships

Chinese Journal of Physiology 60(3): xxx-xxx, 2017 1 DOI: 10.4077/CJP.2017.BAF448 Performance Trends in Age Group Triathletes in the Olympic Distance Triathlon at the World Championships 2009-2014 Martin Wonerow 1, Christoph Alexander Rüst 2, Pantelis Theodoros Nikolaidis 3, Thomas Rosemann 2, and Beat Knechtle 2, 4 1 Department of Orthopedic Surgery, Hospital Grabs, Grabs, Switzerland 2 Institute of Primary Care, University of Zurich, Zurich, Switzerland 3 Exercise Physiology Laboratory, Nikaia, Greece and 4 Gesundheitszentrum St. Gallen, St. Gallen, Switzerland Abstract Performance trends in elite triathletes competing in age group classes have been investigated for Ironman triathlon in Ironman Hawaii but not for short distance triathletes competing at international level in event such as the Olympic distance triathlon at the World Championships. The aim of the present study was to evaluate participation and performance trends of age group triathletes competing in the Olympic distance triathlon at the ITU (International Triathlon Union) World Championships 2009-2014. During this period, the number of participants remained constant. Swimming performance improved in athletes in age groups 25-29 years to 55-59 years, but not in younger (18-24 years) and older (> 60-64 years) age groups. Cycling performance improved in athletes in age groups 18-24 years to 70-74 years, but not in age group 75-79 years. In running, athletes in age groups 18-24 years, 30-34 years, 35-39 years and 65-69 years improved, but not in the other age groups. Overall race time was improved by athletes in age groups 18-24 to 65-69 years, but not in age groups 70-74 years and 75-79 years. Transition times were improved by all age group athletes. Women were slower than men in swimming, cycling, running and overall race time in age groups 18-24 years to 70-74 years, but not in the age group 75-79 years. For transition times, women were slower than men in age groups 18-24 years to 65-69 years, but not in age groups 70-74 years to 75-79 years. In summary, women and men improved performance in most age groups across all years, men were faster than women except in the age group 75-79 years and the sex difference between women and men remained constant. This knowledge should be considered when future age group triathletes train for and compete in Olympic distance triathlons at international level. Key Words: age group, female athletes, participation, performance trend, sex difference Introduction Triathlon is characterized by the successive completion of three sport disciplines (i.e. swimming, cycling and running). The subsequent stress put on varying muscle groups, the quick changing of sport clothes and material, as well as the athlete s fast adaptation to the requirements of the next discipline, are all demands of this sport (7, 26). Triathlon can be defined as one sport, three disciplines and two transitions with specific biomechanical, physiological, and sensorial adaptations required for the second transition from cycling to running (25, 47). Most of the existing literature on triathlon has focused on profiling Corresponding author: Prof. Dr. med. Beat Knechtle, Facharzt FMH für Allgemeinmedizin, Gesundheitszentrum St. Gallen, Vadianstrasse 26, 9001 St. Gallen, Switzerland, Tel: +41 (0) 71 226 93 00, Fax: +41 (0) 71 226 93 01, E-mail: beat.knechtle@hispeed.ch Received: June 8, 2016; Revised: October 23, 2016; Accepted: November 3, 2016. 2017 by The Chinese Physiological Society and Airiti Press Inc. ISSN : 0304-4920. http://www.cps.org.tw

2 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle the anthropometric (3, 10, 30, 32, 35, 52), physiological (9) and psychological characteristics (11, 27, 49) of athletes. The age-related performance decline was also investigated for different sports and distances (6, 18, 34, 41, 47). In triathlon, participation and performance trends have been thoroughly investigated for age group athletes in the Ironman distance and by sex in ultra distance triathlons (25, 33). However, as the influence of the competition distance could be demonstrated for marathon and half-marathon events, it seems reasonable to also consider the Olympic distance triathlon (2, 60). Until now there has been no literature focusing on the participation and performance trends in elderly athletes (> 75 years) in the Olympic distance triathlon. Studies investigating marathon runners indicated an increasing number of participants, as well as an improved performance for athletes older than 75 years (1). Existing studies about participation, age and sex interactions in performance of age group athletes in triathlon used data from national competitions (18), but international events were previously not included. Following the definition of ITU (International Triathlon Union), age group triathletes are recreational athletes (and not professional like elite athletes). 1 As only 0.1% of the 2.3 million registered triathletes are professionals, the majority take part in competitions using the age group system. 2 The ITU Triathlon World Championship over the Olympic distance for age group athletes is held annually. The standard distance for an ITU race is 1.5 km swimming, 40 km cycling, and 10 km running. 3 In 1994, this format was adopted by the Olympic Games and in 2000 the first Olympic Triathlon took place in Sydney. As a result, a race consisting of these distances is also classified as the Olympic distance. 4 The Olympic distance format of triathlon has unique characteristics compared to the Ironman triathlon. It would therefore be of great practical value for both athletes and coaches to be aware of specific participation and performance trends of age group starters in the Olympic Triathlon. Less information is available with regards to participation and performance trends across years. Previous studies that examined these kind of parameters were carried out mostly on the Ironman triathlon format (21, 59). Moreover, these trends might vary by age and sex, as revealed by analysis of the Ironman triathlon (59), so these parameters should also be considered when assessing the Olympic distance triathlon. The aims of our study were therefore to investigate (i) the participation and performance trends of age group athletes at the ITU World Championships over the Olympic distance triathlon from 2009 to 2014, and (ii) the changes in terms of sex difference over the same period. Ethics Approval Materials and Methods All procedures used in the study were approved by the Institutional 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 only involved the analysis of publicly available data. Data Sampling The data for this study was taken from the World Championships result lists over the Olympic distance from 1989 to 2014 that are digitally available. 5 In this time period, a total of 20,100 athletes participated in different competitions. Due to incorrect or incomplete data sets the races from 1989 to 2000 were not considered reliable for data analysis. For 1989, the data for age group participants was collected without transition times; we subsequently excluded this data from our analysis. From 1990 to 2003, no result lists for the various age groups were provided. The majority of data sets were either incorrect or incomplete for the years from 2004 to 2008; these could also not be included in our study. What remained were the data sets for triathletes who competed in the years 2009 to 2014 as age group participants in the Olympic distance at the World Championships. Therefore, this study includes primarily the data of 8,605 participants. As a result of poor weather the swimming distance was halved to 750 m at the London World Championships in 2013. The results of the 2,108 participants were excluded from this study because of the non-comparable swimming times. Due to incomplete or incorrect data sets an additional 13 participants in the period from 2009 to 2014 could not be considered. The reasons for data exclusion are listed in Table 1. In the end, the data sets 1 ITU. Age Group. 2012. http://www.triathlon.org/agegroup 2 Lacke, S. Rookie pros: Tips & experiences. 2015. http://triathlete-europe.competitor.com/2015/04/07/rookie-pros-year-one-experiences 3 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf 4 Furlong, C. TRIATHLON: History of Triathlon at the Olympic games. 1 4, 2015.http://www.olympic.org/Assets/OSCSection/pdf/QR_ sports_summer/sports_olympiques_triathlon_eng.pdf 5 ITU. Result Listings.http://www.triathlon.org/results/search?event_name=&event_category=348&event_sport=357&event_ distance=&event_year=all&event_region=&event_country=&button=search

Performance Trends in Olympic Distance Triathlon 3 Table 1. Excluded data, sorted by race. T1 stands for transition time 1 and T2 stands for transition time 2 Venue Year Total Female Male Reason Gold Coast 2009 1 1 only overall race time Beijing 2011 3 1 2 only overall race time 1 missing T2 Auckland 2012 1 1 only overall race time 1 1 only overall race time and swim time 2 2 only swim-, run -, and overall race time 2 1 1 missing bike time and T2 1 1 missing swim time and T1 Edmonton 2014 1 1 missing T2 and run time Number of Female Athelets 100 90 80 70 60 50 40 30 20 10 0 Female Participation 2009 2010 2011 2012 2014 Year Fig. 1. Calendar year participation for all age groups (upper figure for female athletes, lower figure for male athletes). Number of Female Athelets 160 140 120 100 80 60 40 20 0 Male Participation 2009 2010 2011 2012 2014 Year 18-24 25-29 30-34 35-39 40-44 45-49 50-54 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 55-59 60-64 65-69 70-74 75-79 80-84 from 6,484 triathletes were considered for statistical analysis. To allow for comparisons with other studies, the data sets for the age groups 18-19 and 20-24 years were integrated into the more commonly accepted age group of 18-24 years. Statistical Analysis To analyze changes in performance of finishers, a mixed-effects regression model with each participant as a random variable was used to consider participants who completed several races. We included sex and calendar year as fixed variables in our analysis. Models were calculated for each age group and the final model was selected by means of Akaike Information Criterion (AIC). Sex difference was calculated using the equation 100 ([time in women] - [time in men]/[time in men]). All sex differences were transformed to absolute values before analyzing and subsequently investigated for changes using linear regression models. Statistical analyses were performed using IBM SPSS Statistics (Version 22, IBM SPSS, Chicago, IL, USA) and GraphPad Prism (Version 6.01, GraphPad Software, La Jolla, CA, USA). Significance was accepted at P < 0.05 (two-sided for t-tests). Data in the tables are given as mean ± standard deviation (SD). Participation Trends Results No significant change in the number of female or male participants for any of the age groups was demonstrated from 2009 to 2014 (Fig. 1). The limits set by the ITU - every National Federation can register up to

4 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle Table 2. Participation per sex, year and age group of the eight largest triathlon nations and host nations divided Women 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 2009 Gold Australia 36 28 29 30 29 28 29 26 11 2 Coast Canada 8 6 5 4 7 7 6 2 3 1 France 1 3 1 1 Australia Great Britain 10 18 14 9 10 5 9 2 4 1 Germany 2 1 1 1 1 1 1 Japan 1 3 1 New Zealand 17 15 9 16 9 14 8 1 2 1 2010 Budapest Hungary 2011 Beijing China 2012 Auckland New Zealand USA 8 3 9 7 12 9 10 10 11 7 5 2 Australia 5 9 5 5 6 6 4 2 3 Canada 1 6 7 9 7 2 10 3 2 France 1 1 Great Britain 12 18 19 18 15 12 15 9 3 2 1 Germany 2 2 3 3 2 1 1 Japan 1 2 New Zealand 4 6 1 1 3 4 2 3 1 USA 11 4 5 8 8 4 13 10 7 6 4 Hungary 6 13 9 8 2 6 1 1 Australia 2 4 5 6 6 5 1 1 Canada 2 4 1 3 7 3 3 1 1 France Great Britain 2 5 8 8 4 5 4 2 2 Germany 1 1 1 1 Japan 1 3 1 1 New Zealand 3 8 5 3 4 1 3 1 USA 2 2 4 5 3 7 6 9 4 2 4 China 2 3 2 3 2 6 2 Australia 18 19 18 17 18 19 17 10 3 1 Canada 8 8 7 11 10 10 11 7 1 France 1 1 1 Great Britain 2 16 19 18 12 5 14 4 1 1 Germany 1 1 Japan 1 2 3 1 1 1 New Zealand 19 20 20 22 22 21 20 13 6 4 USA 10 7 12 12 7 9 8 9 9 7 5 1 2014 Australia 10 10 10 5 10 8 3 1 1 Edmonton Canada 12 14 11 20 16 15 21 13 6 4 France 1 1 Canada Great Britain 6 9 12 10 8 9 2 7 2 1 1 Germany 1 1 1 Japan 2 2 1 1 1 2 1 New Zealand 4 6 6 2 3 1 2 2 1 USA 24 9 9 7 9 8 11 15 16 14 7 2 1 Men 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 2009 Gold Coast Australia 2010 Budapest Hungary 2011 Beijing China 2012 Auckland New Zealand 2014 Edmonton Canada Australia 45 29 27 29 30 30 28 30 29 15 4 1 Canada 7 11 7 4 7 4 9 7 2 1 2 France 1 2 2 7 5 9 2 1 2 Great Britain 16 23 22 18 25 18 14 10 8 7 1 1 Germany 4 2 2 2 2 1 1 Japan 1 4 2 1 1 1 New Zealand 19 19 18 14 20 19 10 10 11 4 2 1 USA 8 6 11 10 10 7 13 10 8 8 7 3 Australia 7 13 7 7 12 10 8 1 7 3 Canada 4 5 3 6 4 9 5 8 3 4 1 France 1 1 2 1 1 Great Britain 17 17 20 20 19 18 20 16 14 8 2 1 Germany 2 4 3 2 5 5 2 1 4 1 1 Japan 1 1 1 2 New Zealand 5 6 7 3 3 5 5 4 2 1 1 USA 6 5 8 6 8 8 12 10 9 14 6 1 1 Hungary 18 24 37 35 32 27 24 10 6 5 Australia 3 10 17 11 6 5 1 1 Canada 4 3 5 2 7 3 3 1 1 1 France Great Britain 2 8 11 11 4 5 4 2 2 Germany 1 1 1 Japan 2 3 1 2 1 1 New Zealand 6 3 1 4 4 3 1 USA 6 3 7 11 3 7 6 9 4 2 4 2 1 China 8 24 21 27 32 26 20 16 12 1 1 Australia 26 19 18 18 20 19 19 18 8 6 2 Canada 6 8 8 13 6 12 13 12 8 2 1 France 1 1 1 2 5 5 3 Great Britain 9 17 18 15 17 12 17 10 7 4 2 Germany 1 Japan 1 2 1 1 2 1 1 2 New Zealand 15 19 19 23 23 22 22 20 16 17 2 2 USA 5 3 4 7 4 6 7 6 9 9 10 2 Australia 9 11 7 8 12 9 9 12 1 1 3 1 Canada 17 16 16 23 23 28 27 28 17 9 6 3 France 1 1 2 Great Britain 8 12 9 12 7 10 15 4 6 8 2 Germany 2 2 2 1 Japan 1 1 1 1 4 1 1 New Zealand 6 3 1 1 7 6 7 3 1 2 2 USA 25 6 11 5 10 9 14 7 10 12 8 4 1

Performance Trends in Olympic Distance Triathlon 5 Table 3. Results of the mixed-effects regression analysis for times Age group Estimate SE df t P Swimming 18-24 constant term 117.91 148.89 575.88 0.79 0.429 [sex=female] 2.28 0.31 585.57 7.42 <0.0001 year -0.05 0.07 575.88-0.64 0.524 25-29 constant term -451.10 187.00 648.87-2.41 0.016 [sex=female] 1.66 0.36 692.91 4.62 <0.0001 year 0.24 0.09 648.86 2.54 0.011 30-34 constant term -631.56 216.05 695.87-2.92 0.004 [sex=female] 3.00 0.40 715.75 7.50 <0.0001 year 0.33 0.11 695.86 3.04 0.002 35-39 constant term -1250.40 228.91 808.21-5.46 <0.0001 [sex=female] 2.50 0.40 715.25 6.22 <0.0001 year 0.64 0.11 808.21 5.58 <0.0001 40-44 constant term -937.59 204.27 711.37-4.59 <0.0001 [sex=female] 2.79 0.40 700.13 6.97 <0.0001 year 0.48 0.10 711.36 4.72 <0.0001 45-49 constant term -784.30 235.00 732.78-3.34 0.001 [sex=female] 4.02 0.44 634.34 9.23 <0.0001 year 0.40 0.12 732.77 3.45 0.001 50-54 constant term -959.46 241.15 701.20-3.98 <0.0001 [sex=female] 4.15 0.45 610.33 9.13 <0.0001 year 0.49 0.12 701.20 4.09 <0.0001 55-59 constant term -804.45 277.93 497.68-2.89 0.004 [sex=female] 2.98 0.56 413.89 5.34 <0.0001 year 0.41 0.14 497.67 3.00 0.003 60-64 constant term -601.05 305.17 369.66-1.97 0.050 [sex=female] 4.11 0.64 299.97 6.42 <0.0001 year 0.31 0.15 369.66 2.07 0.039 65-69 constant term -711.66 415.46 232.15-1.71 0.088 [sex=female] 4.24 0.93 170.19 4.56 <0.0001 year 0.37 0.21 232.16 1.79 0.075 70-74 constant term 284.45 747.55 101.69 0.38 0.704 [sex=female] 5.01 1.83 66.95 2.74 0.008 year -0.12 0.37 101.69-0.33 0.739 75-79 constant term -1648.27 1.80 29.14-913.61 <0.0001 [sex=female] 4.71 4.85 29.06 0.97 0.340 year 0.84 0.00 141 26321.78 <0.0001 Cycling 18-24 constant term -2124.42 279.34 646.00-7.61 <0.0001 [sex=female] 8.07 0.53 646.00 15.37 <0.0001 year 1.09 0.14 646.00 7.83 <0.0001 25-29 constant term -3263.66 317.82 762.00-10.27 <0.0001 [sex=female] 5.87 0.55 762.00 10.68 <0.0001 year 1.65 0.16 762.00 10.47 <0.0001 30-34 constant term -2594.74 336.17 759.18-7.72 <0.0001 [sex=female] 6.80 0.58 574.70 11.76 <0.0001 year 1.32 0.17 759.22 7.91 <0.0001 35-39 constant term -2909.82 320.22 816.63-9.09 <0.0001 [sex=female] 6.71 0.54 665.05 12.52 <0.0001 year 1.48 0.16 816.63 9.29 <0.0001 40-44 constant term -3174.33 334.19 792.34-9.50 <0.0001 [sex=female] 7.22 0.60 594.27 12.06 <0.0001 year 1.61 0.17 792.35 9.69 <0.0001 45-49 constant term -2493.09 331.81 732.28-7.51 <0.0001 [sex=female] 8.62 0.60 579.38 14.46 <0.0001 year 1.27 0.17 732.29 7.71 <0.0001 50-54 constant term -2741.18 369.06 715.99-7.43 <0.0001 [sex=female] 8.41 0.67 608.67 12.64 <0.0001 year 1.40 0.18 715.99 7.61 <0.0001 55-59 constant term -2128.57 454.10 498.59-4.69 <0.0001 [sex=female] 6.65 0.88 409.92 7.54 <0.0001 year 1.09 0.23 498.60 4.84 <0.0001 60-64 constant term -2254.64 596.36 366.45-3.78 <0.0001 [sex=female] 8.20 1.16 270.97 7.07 <0.0001 year 1.16 0.30 366.46 3.90 <0.0001 65-69 constant term -2710.77 718.04 233.00-3.78 <0.0001 [sex=female] 8.60 1.50 233.00 5.74 <0.0001 year 1.39 0.36 233.00 3.88 <0.0001 70-74 constant term -270.25 1350.36 100.73-0.20 0.842 [sex=female] 7.92 2.97 49.00 2.66 0.010 year 0.18 0.67 100.74 0.26 0.794 75-79 constant term -1739.35 2331.60 29.76-0.75 0.462 [sex=female] 4.18 6.81 29.64 0.61 0.544 year 0.91 1.16 29.76 0.79 0.438 Table 3. (connected) Age group Estimate SE df t P Running 18-24 constant term 455.86 224.76 609.35 2.03 0.043 [sex=female] 5.72 0.45 576.03 12.60 <0.0001 year -0.21 0.11 609.35-1.85 0.065 25-29 constant term 233.45 250.92 661.99 0.93 0.353 [sex=female] 4.49 0.48 692.02 9.34 <0.0001 year -0.10 0.12 661.98-0.77 0.444 30-34 constant term 476.02 235.34 364.94 2.02 0.044 [sex=female] 4.76 0.49 719.15 9.63 <0.0001 year -0.22 0.12 364.93-1.84 0.066 35-39 constant term 505.82 241.66 682.87 2.09 0.037 [sex=female] 3.85 0.45 733.75 8.57 <0.0001 year -0.23 0.12 682.87-1.91 0.056 40-44 constant term -24.78 234.06 563.74-0.11 0.916 [sex=female] 4.64 0.49 706.36 9.52 <0.0001 year 0.03 0.12 563.73 0.29 0.770 45-49 constant term 339.78 237.48 571.43 1.43 0.153 [sex=female] 5.76 0.49 654.66 11.79 <0.0001 year -0.15 0.12 571.43-1.24 0.214 50-54 constant term -14.02 245.44 457.75-0.06 0.954 [sex=female] 5.86 0.54 630.10 10.81 <0.0001 year 0.03 0.12 457.74 0.25 0.806 55-59 constant term 59.06 325.02 451.57 0.18 0.856 [sex=female] 5.10 0.70 414.57 7.26 <0.0001 year 0.00 0.16 451.57-0.03 0.976 60-64 constant term 797.10 440.54 257.50 1.81 0.072 [sex=female] 7.94 1.09 298.35 7.30 <0.0001 year -0.37 0.22 257.49-1.69 0.092 65-69 constant term 1659.14 567.32 228.12 2.93 0.004 [sex=female] 9.89 1.35 170.59 7.32 <0.0001 year -0.80 0.28 228.11-2.83 0.005 70-74 constant term -233.07 900.84 53.71-0.26 0.797 [sex=female] 11.32 3.07 65.56 3.69 <0.0001 year 0.15 0.45 53.71 0.33 0.742 75-79 constant term -2142.91 2406.90 27.32-0.89 0.381 [sex=female] 0.72 7.28 28.91 0.10 0.922 year 1.10 1.20 27.32 0.92 0.364 Overall race time 18-24 constant term -2641.41 570.16 645.65-4.63 <0.0001 [sex=female] 16.73 1.10 518.03 15.16 <0.0001 year 1.38 0.28 645.64 4.86 <0.0001 25-29 constant term -4679.33 676.42 758.72-6.92 <0.0001 [sex=female] 12.27 1.23 675.15 10.00 <0.0001 year 2.39 0.34 758.71 7.12 <0.0001 30-34 constant term -3915.63 716.24 752.68-5.47 <0.0001 [sex=female] 14.95 1.29 708.66 11.56 <0.0001 year 2.01 0.36 752.67 5.66 <0.0001 35-39 constant term -5052.87 695.97 811.95-7.26 <0.0001 [sex=female] 13.40 1.21 693.26 11.04 <0.0001 year 2.58 0.35 811.95 7.46 <0.0001 40-44 constant term -5377.92 689.36 735.32-7.80 <0.0001 [sex=female] 15.08 1.33 697.84 11.32 <0.0001 year 2.74 0.34 735.31 8.00 <0.0001 45-49 constant term -4301.03 704.74 696.65-6.10 <0.0001 [sex=female] 19.03 1.36 646.37 13.99 <0.0001 year 2.21 0.35 696.64 6.30 <0.0001 50-54 constant term -4937.85 761.47 694.19-6.49 <0.0001 [sex=female] 19.40 1.46 637.90 13.32 <0.0001 year 2.53 0.38 694.18 6.68 <0.0001 55-59 constant term -4310.00 930.39 493.29-4.63 <0.0001 [sex=female] 15.33 1.90 413.81 8.07 <0.0001 year 2.22 0.46 493.28 4.80 <0.0001 60-64 constant term -2814.39 1225.13 369.98-2.30 0.022 [sex=female] 21.42 2.53 289.61 8.46 <0.0001 year 1.48 0.61 369.98 2.43 0.016 65-69 constant term -2848.29 1381.61 230.00-2.06 0.040 [sex=female] 24.00 3.03 152.12 7.92 <0.0001 year 1.50 0.69 230.01 2.19 0.030 70-74 constant term -136.20 2737.28 99.63-0.05 0.960 [sex=female] 26.03 6.95 62.50 3.75 <0.0001 year 0.16 1.36 99.62 0.12 0.905 75-79 constant term -8396.80 5397.49 29.86-1.56 0.130 [sex=female] 9.29 16.00 29.14 0.58 0.566 year 4.29 2.68 29.86 1.60 0.121

6 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle Table 3. (connected) Age group Estimate SE df t P Transition Times 18-24 constant term -1054.11 53.25 646.00-19.80 <0.0001 [sex=female] 0.54 0.10 646.00 5.37 <0.0001 year 0.53 0.03 646.00 19.88 <0.0001 25-29 constant term -1189.13 63.56 803.91-18.71 <0.0001 [sex=female] 0.38 0.11 823.23 3.45 0.001 year 0.59 0.03 803.90 18.79 <0.0001 30-34 constant term -1159.74 69.00 775.00-16.81 <0.0001 [sex=female] 0.51 0.12 775.00 4.33 <0.0001 year 0.58 0.03 775.00 16.88 <0.0001 35-39 constant term -1245.74 73.98 819.00-16.84 <0.0001 [sex=female] 0.35 0.12 819.00 2.80 0.005 year 0.62 0.04 819.00 16.91 <0.0001 40-44 constant term -1198.22 68.82 794.03-17.41 <0.0001 [sex=female] 0.38 0.12 794.03 3.13 0.002 year 0.60 0.03 794.03 17.49 <0.0001 45-49 constant term -1158.75 68.83 732.29-16.84 <0.0001 [sex=female] 0.66 0.12 557.55 5.35 <0.0001 year 0.58 0.03 732.30 16.92 <0.0001 50-54 constant term -1293.33 84.67 715.65-15.28 <0.0001 [sex=female] 0.89 0.15 595.13 5.86 <0.0001 year 0.65 0.04 715.65 15.35 <0.0001 55-59 constant term -1318.33 104.04 497.17-12.67 <0.0001 [sex=female] 0.45 0.20 390.07 2.28 0.023 year 0.66 0.05 497.18 12.74 <0.0001 60-64 constant term -1426.49 126.40 367.69-11.29 <0.0001 [sex=female] 1.08 0.25 288.12 4.31 <0.0001 year 0.71 0.06 367.70 11.34 <0.0001 65-69 constant term -1457.87 160.79 233.00-9.07 <0.0001 [sex=female] 1.11 0.34 233.00 3.31 0.001 year 0.73 0.08 233.00 9.12 <0.0001 70-74 constant term -1184.20 434.59 102.00-2.73 0.008 [sex=female] 1.56 0.90 102.00 1.73 0.086 year 0.59 0.22 102.00 2.75 0.007 75-79 constant term -2908.91 530.72 27.52-5.48 <0.0001 [sex=female] -0.16 1.56 29.78-0.10 0.919 year 1.45 0.26 27.53 5.50 <0.0001 20 athletes and the hosting National Federation can register 25 athletes - were reached only by a limited number of national federations (mostly the hosting nations) (Table 2). 6 During the examined period from 2009 to 2014, an overall total of 13,683 participants were registered for all starter groups at the ITU World Championships. The year with the most registered participants was 2012 (Auckland) with 3,098 athletes; the smallest number of registered participants was in 2011 (Beijing) with 1,840 athletes. The largest age groups were 40-44 years for male athletes with 500 and 35-39 years for female athletes with 337, all of whom finished the race (Fig. 1). The number of participants per nation varied depending on the country hosting the event. For example, 264 Hungarian age group athletes participated in Budapest 2010, but only two in Beijing 2011 (Table 2). Performance Trends across Years For the performance trends, the results of each age group in 2009 were taken as the baseline and then tracked through until 2014. Table 3 presents the results of the mixed-effects regression analyses and Table 4 shows the split times, overall race times and transition Sex Difference in % 30 25 20 15 10 5 0 Overall Race Time 2009 2010 2011 2012 2014 Year 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 Fig. 2. Sex difference (%) in overall race time. times for women and men acrossall age groups. In swimming, athletes improved in age groups 25-29 years to 55-59 years, but not in the younger (18-24 years) and older (> 60-64 years) age groups. In cycling, athletes in age groups 18-24 years to 70-74 years improved, but not in the age group 75-79 years. In running, athletes in age groups 18-24 years, 30-34 years, 35-39 years and 65-69 years improved, but not in the other age groups. Overall race time was improved in age groups 18-24 to 65-69 years, but not in age groups 70-74 years and 75-79 years. Transition times were improved by all age group athletes. For the analysis of performance trends by sex, the performance of men was taken as the baseline. Specific analysis of the women compared the difference in performance with this baseline. Men were faster than women in most age groups. Women were slower than men in swimming, cycling, running and overall race time in age groups 18-24 years to 70-74 years, but not in the age group 75-79 years. Regarding transition times, women were slower than men in age groups 18-24 years to 65-69 years, but not in age groups 70-74 years to 75-79 years. Sex Differences in Performance Figures 2-6 show the trends in sex difference in performance for overall race time, transition times, swimming, cycling, and running. The difference in overall race time between male and female athletes was 2% in 2011 for athletes in age group 25-29 years and 24% for athletes in age group 80-84 years in 2014. The female athletes achieved an average rate of 112% of the male race time over all age groups. Except for an improvement in running performance for women in the age group 75-79 years compared to the same age group of men, along with an increase in transition times for women as opposed to men in the age group 50-54 years, no variation was shown. 6 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf

Performance Trends in Olympic Distance Triathlon 7 Table 4. Race times for all age group athletes. Times are expressed as h:min with SD Sex Age group 2009 2010 2011 2012 2014 Swimming women 18-24 26:13 ± 2:39 23:56 ± 3:25 26:14 ± 3:11 25:49 ± 3:41 23:53 ± 2:44 25-29 26:25 ± 2:59 25:07 ± 4:22 25:42 ± 2:14 28:19 ± 4:33 24:47 ± 3:05 30-34 28:05 ± 3:23 25:47 ± 3:07 29:06 ± 4:20 32:18 ± 5:40 26:15 ± 3:14 35-39 26:46 ± 3.11 27:02 ± 4:30 29:41 ± 4:42 34:19 ± 5:59 26:49 ± 2:49 40-44 27:28 ± 3:02 26:35 ± 3:58 30:22 ± 5:16 32:29 ± 6:17 27:16 ± 3:15 45-49 28:44 ± 3:04 28:12 ± 5:29 32:34 ± 5:39 35:05 ± 7:09 27:06 ± 3:31 50-54 29:16 ± 3:21 27:58 ± 3:57 33:09 ± 4:40 37:20 ± 6:28 28:43 ± 4:12 55-59 31:37 ± 5:11 29:11 ± 3:11 36:28 ± 5:43 36:02 ± 4:26 29:26 ± 3:29 60-64 32:13 ± 4:23 34:04 ± 6:17 36:14 ± 5:08 38:19 ± 6:01 31:16 ± 4:14 65-69 37:05 ± 4:28 34:58 ± 3:32 39:08 ± 4:22 38:24 ± 8:49 34:39 ± 3:21 70-74 40:34 ± 11:17 39:44 ± 5:52 41:32 ± 5:29 43:02 ± 5:16 35:50 ± 4:45 75-79 42:33 ± 8:49 49:44 ± 22:05 80-84 45:41 men 18-24 21:49 ± 2:38 21:45 ± 3:58 26:42 ± 7:22 23:40 ± 3:26 22:40 ± 2:37 25-29 21:49 ± 2:15 23:07 ± 4:05 30:08 ± 8:13 24:59 ± 3:51 24:13 ± 3:44 30-34 22:28 ± 2:20 24:33 ± 6:29 29:31 ± 6:53 27:09 ± 4:32 23:59 ± 3:58 35-39 23:35 ± 2:39 25:29 ± 4:43 29:58 ± 6:52 30:04 ± 5:49 24:55 ± 2:11 40-44 23:32 ± 2:48 24:36 ± 4:16 31:23 ± 6:39 28:14 ± 4:45 24:40 ± 3:13 45-49 23:51 ± 2:44 25:00 ± 4:05 30:51 ± 6:22 28:29 ± 6:01 25:14 ± 3:16 50-54 24:22 ± 2:19 26:35 ± 5:17 31:56 ± 5:20 30:29 ± 5:17 25:05 ± 3:11 55-59 25:42 ± 2:49 27:48 ± 3:59 34:10 ± 5:34 33:56 ± 6:37 27:18 ± 3:47 60-64 27:11 ± 3:25 28:19 ± 4:27 35:35 ± 5:56 31:13 ± 3:59 28:04 ± 2:56 65-69 27:22 ± 2:43 30:52 ± 4:14 38:04 ± 7:13 33:19 ± 5:43 30:44 ± 5:52 70-74 31:59 ± 5:58 32:08 ± 4:12 39:35 ± 10:25 40:20 ± 7:33 31:41 ± 3:46 75-79 34:19 ± 5:31 38:20 ± 4:19 50:15 ± 3:58 44:52 ± 8:42 40:38 ± 9:49 80-84 45:21 40:29 37:17 ± 0:29 Cycling women 18-24 1:10:00 ± 4:19 1:01:50 ± 5:06 1:17:13 ± 5:57 1:16:50 ± 5:31 1:11:55 ± 4:48 25-29 1:08:05 ± 3:40 1:00:25 ± 5:14 1:15:07 ± 4:07 1:16:22 ± 5:21 1:01:05 ± 5:28 30-34 1:09:30 ± 4.42 1:00:25 ± 4:22 1:16:50 ± 7:21 1:16:57 ± 5:23 1:09:51 ± 3:59 35-39 1:09:55 ± 3:34 1:00:29 ± 4:36 1:17:16 ± 5:31 1:17:38 ± 5:16 1:10:56 ± 4:43 40-44 1:09:31 ± 3:28 1:01:11 ± 4:46 1:18:16 ± 9:17 1:17:47 ± 6:24 1:11:53 ± 5:02 45-49 1:12:38 ± 5:05 1:03:23 ± 7:19 1:18:38 ± 6:53 1:19:43 ± 7:07 1:12:43 ± 5:46 50-54 1:13:35 ± 4:39 1:03:21 ± 6:43 1:20:12 ± 7:08 1:22:58 ± 6:13 1:14:46 ± 7:56 55-59 1:16:50 ± 5:11 1:04:26 ± 6:01 1:23:53 ± 8:11 1:24:50 ± 7:00 1:16:08 ± 5:46 60-64 1:19:28 ± 7:30 1:17:54 ± 4:45 1:29:34 ± 14:11 1:32:46 ± 9:10 1:19:17 ± 6:15 65-69 1:23:53 ± 8:46 1:11:02 ± 3:28 1:31:30 ± 7:32 1:33:28 ± 8:40 1:25:07 ± 6:11 70-74 1:32:08 ± 15:20 1:16:00 ± 6:41 1:35:42 ± 6:51 1:41:44 ± 12:16 1:27:42 ± 8:34 75-79 1:36:36 ± 14:53 1:40:05 ± 8:45 80-84 1:40:15 men 18-24 1:01:26 ± 3:36 55:46 ± 5:27 1:10:10 ± 8:41 1:17:07 ± 4:34 1:03:20 ± 4:44 25-29 1:01:19 ± 3.66 54:38 ± 4:07 1:12:53 ± 9:46 1:08:19 ± 5:04 1:04:05 ± 5:57 30-34 1:01:13 ± 3:05 55:50 ± 4:56 1:11:32 ± 12:32 1:17:35 ± 4:38 1:03:41 ± 5:25 35-39 1:01:31 ± 3:24 57:05 ± 5:43 1:13:02 ± 8:31 1:08:11 ± 4:32 1:04:06 ± 5:16 40-44 1:01:49 ± 4:13 56:05 ± 5:28 1:13:26 ± 8:20 1:09:38 ± 6:19 1:04:58 ± 5:03 45-49 1:03:35 ± 3:51 56:43 ± 4:26 1:13:16 ± 7:45 1:09:37 ± 4:37 1:05:13 ± 5:01 50-54 1:06:05 ± 4:56 57:47 ± 5:03 1:14:19 ± 6:35 1:12:13 ± 6:44 1:06:11 ± 5:37 55-59 1:08:41 ± 3:53 1:01:29 ± 7:56 1:19:50 ± 7:14 1:17:26 ± 7:08 1:09:07 ± 7:03 60-64 1:12:30 ± 5:46 1:01:08 ± 4:38 1:22:11 ± 6:56 1:19:34 ± 8:10 1:11:00 ± 5:31 65-69 1:16:05 ± 7:25 1:04:29 ± 5:35 1:23:22 ± 7:13 1:22:56 ± 4:35 1:15:35 ± 11:23 70-74 1:25:44 ± 7:08 1:05:32 ± 4:14 1:30:32 ± 9:19 1:33:25 ± 12:04 1:18:24 ± 7:21 75-79 1:29:37 ± 15:34 1:12:58 ± 6.60 106.07 ± 13:36 1:41:08 ± 7:48 1:33:14 ± 8:10 80-84 1:25:33 1:50:20 1:25:37 ± 1:10 Running women 18-24 46:48 ± 4:31 46:31 ± 5:41 47:31 ± 6:43 45:11 ± 4:41 46:53 ± 6:16 25-29 45:01 ± 5:16 47:04 ± 8:52 44:41 ± 3:36 45:07 ± 5.17 46:23 ± 4:33 30-34 46:34 ± 4:31 47:31 ± 7:37 47:57 ± 6:10 46:05 ± 5:01 46:51 ± 4:44 35-39 47:34 ± 4:44 47:14 ± 6:19 48:39 ± 5:13 46:14 ± 4:02 47:00 ± 3:54 40-44 47:37 ± 4:41 48:24 ± 7:01 49:37 ± 9:00 46:58 ± 5:06 48:31 ± 3:36 45-49 50:58 ± 6:31 49:23 ± 6:06 50:24 ± 6:04 48:49 ± 5:41 50:21 ± 8:47 50-54 53:28 ± 7:36 50:33 ± 6:49 53:09 ± 6:13 50:11 ± 4:35 52:20 ± 6:20 55-59 56:05 ± 6:59 53:11 ± 5:03 56:53 ± 6:41 53:31 ± 5:35 53:40 ± 6:28 60-64 1:03:40 ± 11:46 57:14 ± 7:28 1:03:48 ± 15:38 57:38 ± 6:14 57:39 ± 6:19 65-69 1:11:06 ± 9:52 1:05:26 ± 2:50 1:02:46 ± 5:08 1:01:53 ± 6:56 1:05:10 ± 7:23 70-74 1:19:49 ± 15:25 1:15:28 ± 5:26 1:16:04 ± 5:48 1:14:43 ± 9:20 1:12:41 ± 8:59 75-79 1:16:19 ± 16:52 1:18:48 ± 6:41 80-84 1:28:36

8 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle Table 4. (connected) Sex Age group 2009 2010 2011 2012 2014 Running men 18-24 40:58 ± 4:26 41:02 ± 6:48 41:33 ± 7:44 40:24 ± 5:48 39:50 ± 4:08 25-29 40:29 ± 3:51 41:18 ± 7:59 45:13 ± 7:46 39:28 ± 3:47 40:47 ± 5:54 30-34 40:54 ± 3:55 43:04 ± 8:33 44:41 ± 10:34 39:45 ± 3:47 41:01 ± 4:43 35-39 42:05 ± 4:43 44:29 ± 7:59 45:47 ± 8:47 41:02 ± 4:22 42:55 ± 4:48 40-44 42:20 ± 5:02 43:49 ± 7:30 46:02 ± 6:50 42:04 ± 5:10 43:15 ± 6:11 45-49 43:45 ± 4:44 44:41 ± 6:36 47:09 ± 6:59 41:53 ± 3:53 43:44 ± 4:13 50-54 45:56 ± 4:42 45:54 ± 6:20 48:30 ± 8:32 44:23 ± 7:40 46:22 ± 6:58 55-59 49:06 ± 5:58 48:23 ± 5:29 52:23 ± 6:49 49:02 ± 7:50 49:11 ± 9:05 60-64 53:35 ± 7:16 51:08 ± 10:02 54:38 ± 7:28 50:39 ± 7:05 50:05 ± 6:25 65-69 1:00:42 ± 8:53 55:11 ± 9:35 56:51 ± 6:38 54:08 ± 5:44 54:00 ± 7:04 70-74 1:10:31 ± 12:58 58:49 ± 9:20 1:04:51 ± 9:01 1:01:50 ± 9:28 1:01:52 ± 12:11 75-79 1:14:56 ± 16:53 1:02:08 ± 11:13 1:22:20 ± 8:32 1:08:09 ± 7:41 1:24:02 ± 16:59 80-84 1:47:48 1:32:25 1:13:59 ± 4:53 Overall race time women 18-24 2:25:47 ± 9:13 2:17:39 ± 12:43 2:35:03 ± 14:20 2:34:53 ± 11:56 2:28:28 ± 12:20 25-29 2:22:24 ± 10:19 2:18:20 ± 17:32 2:29:20 ± 8:46 2:37:14 ± 12:38 2:28:22 ± 11:16 30-34 2:27:14 ± 10:22 2:19:32 ± 12:53 2:38:28 ± 16:22 2:42:46 ± 14:19 2:29:07 ± 10:28 35-39 2:27:34 ± 9:33 2:20:44 ± 14:44 2:40:16 ± 13:26 2:46:07 ± 12:50 2:31:12 ± 10:20 40-44 2:28:00 ± 9:22 2:22:27 ± 15:04 2:42:45 ± 22:37 2:45:20 ± 16:11 2:34:15 ± 10:40 45-49 2:36:14 ± 13:43 2:27:32 ± 18:19 2:46:23 ± 18:05 2:52:05 ± 18:54 2:37:14 ± 16:49 50-54 2:40:21 ± 14:11 2:28:41 ± 15:32 2:51:21 ± 15:51 2:59:40 ± 15:46 2:43:52 ± 16:32 55-59 2:49:07 ± 14:56 2:34:03 ± 10:47 3:02:46 ± 17:12 3:04:10 ± 16:01 2:46:56 ± 13:53 60-64 3:00:39 ± 22:42 2:47:41 ± 18:20 3:15:10 ± 34:47 3:20:18 ± 20:04 2:57:01 ± 15:42 65-69 3:17:59 ± 18:34 3:01:04 ± 5:57 3:19:37 ± 16:25 3:26:01 ± 21:50 3:15:37 ± 15:05 70-74 3:39:49 ± 39:40 3:24:41 ± 15:55 3:39:59 ± 15:04 3:53:08 ± 20:56 3:28:22 ± 22:48 75-79 3:43:14 ± 38:32 4:03:47 ± 38:28 80-84 4:19:13 men 18-24 2:06:58 ± 9:25 2:03:28 ± 14:37 2:22:02 ± 23:02 2:17:37 ± 12:35 2:10:50 ± 10:23 25-29 2:06:26 ± 8:02 2:04:10 ± 16:04 2:32:45 ± 25:34 2:19:29 ± 11:24 2:14:34 ± 14:38 30-34 2:07:29 ± 8:14 2:09:12 ± 19:11 2:29:52 ± 27:59 2:21:14 ± 11:28 2:14:07 ± 13:13 35-39 2:10:24 ± 9:28 2:13:08 ± 17:22 2:33:22 ± 23:29 2:26:23 ± 13:10 2:18:07 ± 11:56 40-44 2:11:06 ± 10:57 2:10:15 ± 15:57 2:35:32 ± 20:53 2:27:13 ± 15:22 2:19:01 ± 13:56 45-49 2:14:48 ± 10:14 2:12:26 ± 14:31 2:35:52 ± 19:48 2:27:40 ± 12:50 2:20:31 ± 11:22 50-54 2:20:10 ± 10:10 2:16:43 ± 16:08 2:39:18 ± 18:45 2:34:59 ± 19:16 2:24:09 ± 15:14 55-59 2:27:37 ± 10:05 2:24:27 ± 15:25 2:51:44 ± 17:37 2:50:02 ± 20:36 2:32:49 ± 18:45 60-64 2:37:48 ± 14:10 2:27:30 ± 18:09 2:57:59 ± 17:57 2:50:58 ± 18:33 2:36:47 ± 12:47 65-69 2:49:02 ± 16:17 2:38:41 ± 18:03 3:04:31 ± 20:29 3:00:47 ± 12:55 2:48:46 ± 19:54 70-74 3:14:35 ± 22:25 2:45:55 ± 16:59 3:21:42 ± 27:38 3:30:08 ± 27:51 3:01:05 ± 20:08 75-79 3:25:37 ± 35:04 3:03:53 ± 22:43 4:08:32 ± 21:01 3:50:01 ± 22:50 3:51:59 ± 33:10 80-84 4:13:11 4:10:04 3:29:10 ± 6:42 Transition Times women 18-24 2:48 ± 0:28 5:22 ± 1:00 4:05 ± 0:37 7:00 ± 0:37 5:45 ± 0:50 25-29 2:53 ± 0:31 5:44 ± 1:26 3:50 ± 0:33 7:23 ± 0:57 6:04 ± 0:53 30-34 3:07 ± 0:26 5:51 ± 0:56 4:36 ± 1:19 7:46 ± 1:05 6:08 ± 0:55 35-39 3:20 ± 0:35 5:59 ± 1:14 4:40 ± 1:05 7:54 ± 1:02 6:25 ± 0:58 40-44 3:24 ± 0:33 6:17 ± 1:11 4:31 ± 0:57 8:04 ± 1:26 6:34 ± 0:46 45-49 3:55 ± 0:55 6:35 ± 1:52 4:47 ± 0:43 8:25 ± 1:25 7:03 ± 1:14 50-54 4:02 ± 0:45 6:49 ± 1:10 4:51 ± 0:47 9:07 ± 1:36 8:01 ± 1:38 55-59 4:37 ± 0:56 7:14 ± 1:11 5:32 ± 0:48 9:46 ± 1:40 7:41 ± 1:04 60-64 5:19 ± 1:10 8:30 ± 1:50 5:34 ± 1:20 11:32 ± 2:31 8:47 ± 1:27 65-69 5:55 ± 1:24 9:40 ± 1:01 6:14 ± 1:18 12:13 ± 1:59 10:40 ± 1:44 70-74 7:19 ± 1:46 12:47 ± 2:30 6:43 ± 1:22 13:37 ± 1:29 12:07 ± 3:47 75-79 7:48 ± 2:01 15:07 ± 0:56 80-84 24:39 men 18-24 2:46 ± 0:24 4:55 ± 0:53 3:37 ± 0:57 6:23 ± 0:41 4:58 ± 0:41 25-29 2:50 ± 0:20 5:07 ± 1:26 4:31 ± 1:40 6:40 ± 0:46 5:26 ± 1:11 30-34 2:56 ± 0:24 5:46 ± 1:52 4:08 ± 1:16 6:43 ± 0:56 5:24 ± 0:51 35-39 3:14 ± 0:36 6:04 ± 1:41 4:34 ± 1:28 7.:04 ± 0:56 6:09 ± 2:37 40-44 3:25 ± 0:38 5:46 ± 1:46 4:41 ± 1:14 7:14 ± 1:10 6:07 ± 1:23 45-49 3:37 ± 0:42 6:03 ± 1:25 4:37 ± 0:58 7:13 ± 1:02 6:17 ± 0:56 50-54 3:47 ± 0:38 6:28 ± 2:17 4:35 ± 1:02 7:52 ± 1:40 6:29 ± 1:05 55-59 4:09 ± 0:39 6:47 ± 1:34 5:23 ± 2:04 9:13 ± 2:06 7:11 ± 1:42 60-64 4:34 ± 0:50 6:56 ± 1:27 5:35 ± 1:00 9:29 ± 2:07 7:36 ± 1:31 65-69 4:54 ± 0:50 8:08 ± 1:37 6:14 ± 1:50 10:22 ± 1:42 8:25 ± 1:53 70-74 6:22 ± 2:09 9:27 ± 4:13 6:44 ± 1:29 14:31 ± 5:40 9:06 ± 1:22 75-79 6:46 ± 2:20 10:27 ± 1:07 9:54 ± 1:43 15:49 ± 3:01 14:02 ± 3:31 80-84 14:30 6:49 12:16 ± 1:08

Performance Trends in Olympic Distance Triathlon 9 120 Transition Time 18 Cycling Sex Difference in % 100 80 60 40 20 Sex Difference in % 16 14 12 10 8 6 4 0 2009 2010 2011 2012 2014 Year 18-24 25-29 30-34 35-39 40-44 45-49 50-54 2 0 2009 2010 2011 2012 2014 Year 55-59 60-64 65-69 70-74 75-79 80-84 18-24 25-29 30-34 35-39 40-44 45-49 50-54 Fig. 3. Sex difference (%) in transition time. 55-59 60-64 65-69 70-74 75-79 80-84 Fig. 5. Sex difference (%) in cycling. 40 Swimming 30 Running Sex Difference in % 35 30 25 20 15 10 5 Sex Difference in % 25 20 15 10 5 0 2009 2010 2011 2012 2014 Year 0 2009 2010 2011 2012 2014 Year 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 Fig. 4. Sex difference (%) in swimming. Fig.6. Sex difference (%) in running. Discussion The aim of this study was to examine the performance trends of age group triathletes competing in the Olympic distance triathlons at international level between 2009 and 2014 by age group and sex. The main findings of the present study were that (i) women improved performance in most age groups across years and (ii) the sex difference in performance between women and men remained constant. Participation Trends Participation rates were very constant, as no change in the number of participants in any of the age groups was shown. These findings did not initially seem surprising, because the participant quotes for the World Triathlon Series were limited by the ITU. 7 However, we could demonstrate that these quotes were only utilized by a few of the federations (i.e. Australia, Great Britain, New Zeeland, Hungary, China and Canada) between 2009 and 2014 (Table 2). This means that the number of participants could have increased, but the composition of the starter field varied depending on the venue (Table 2). Reasons for this could have been travel costs and the time for the journey and the competition (14). The constant numbers of participants at international level might correlate with the ongoing enthusiasm for the triathlon sport. In 2013, a total of 2.1 million active triathletes were registered worldwide. 7 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf

10 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle The years from 2007-2013 revealed a yearly increase averaging 15% and the worldwide number of triathlon clubs was reported to be about 9,000. 8 The results in the present study were in line with those of Etter et al. (17), who reported constant participation numbers in all male age groups, as well as in younger and older female age groups for the Olympic distances at the Zurich Triathlon held from 2000 to 2010. Increasing numbers of participants were only shown in the three female age groups from 40-44 to 50-54 years (18). In previous studies, there seemed to be a trend of increasing participation in longer distances and in more extreme competitions, such as long-distance swimming (33), long-distance running (28, 64, 65) and long-distance cycling (24, 53, 55, 63). However, the trend of an increasing number of participants, as shown for longer distances in triathlon and in the Ironman distance or ultra-triathlon distances (31, 46), was not confirmed by the present study. For the Ironman triathlon, Lepers et al. showed in their study that there was an age-related effect on the number of participants (i.e. a lower number of under 40-yearold athletes, while the number of older athletes increased) (40). Perhaps this shows a trend towards longer distances. Performance Trends in Swimming In swimming, athletes improved in age groups 25-29 years to 55-59 years, but not in younger (18-24 years) and older (> 60-64 years) age groups. The positive performance trend in the 25-59-year-old athletes may be the result of increased experience in swimming and competing (36). Bongard et al. (8) showed that the 1-year age-related declines in the performance of athletes were more than four times as great at 80 years than at 20 years of age, so perhaps the effect of more experience in swimming and competition is less pronounced than the effect of an agerelated decline. An improvement in swimming times for these older athletes would therefore be implausible. Performance Trends in Cycling In cycling, athletes in age groups 18-24 years to 70-74 years improved, but not in the age group 75-79 years. The positive performance trend in cycling can certainly be partly explained by the ongoing advancements in racing bike technology. Bicycle frames, especially for triathlon and time trail races, have become more aerodynamic and also more affordable (29). In addition, electronic gear changing, which was introduced in 2009 by Shimano for example, allows faster and more precise gear changes with less energy expenditure and consequently has a positive effect on performance. 9 Optimized nutrition and liquid intake can result in a better general feeling and consequently better performance levels in cycling, as found by Maunder et al. (45). Peiffer et al. (49) showed a lower efficiency for cycling amongst master triathletes compared to younger athletes. This could explain why athletes in the age group 75-79 failed to improve their cycling performance. Performance Trends in Running In running, athletes in age groups 18-24 years, 30-34 years, 35-39 years and 65-69 years improved, unlike those in the other age groups. An explanation for the improvement of running times could be found in better performance in transition zones caused by training and experience, resulting in better preconditions for running and reduced neuromuscular fatigue (47). A study by Rendos et al. (51) showed an increased risk of running injuries during the first 14 min of running after a minimum of 30 min cycling. Beside a stronger focus on pure running training, an additional targeted transition zone training can also lead to a reduction in transition times and an improvement in running performance (26). Due to the conservation of energy reserves, ascertaining optimal seat position while cycling has a direct effect on running performance and this is independent of an athlete s training schedule for running (23, 56). As with swimming, drafting could also help explain the positive performance trend (12), as according to ITU competition rules, drafting is not forbidden 10. The significant effect of drafting on elite runners running performance over 3000 m was highlighted by Zouhal et al. (68). The lack of an improvement in the running performance of athletes in age groups >70 years can be explained by the higher energy cost of running in master athletes compared to younger participants, as described by Peiffer et al. (20). Performance Trends in Overall Race Time Overall race time was improved in athletes in age groups 18-24 to 65-69 years, but not in age groups 70-74 years and 75-79 years. In these oldest age groups 8 Roethenbaugh, G. International Triathlon Union (ITU) Global triathlon participation ITU.2014. http://triathlonquebec.objectif226.ca/wp-content/uploads/2012/03/itu-global-triathon-participation Survey-NF.pdf 9 Schütz, A. Wenn jede Zehntelsekunde zählt. 2012. http://www.maxonmotor.de/maxon/view/application/electronic-shift-system-ab. 10 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf

Performance Trends in Olympic Distance Triathlon 11 only 142 (101 men and 31 women) athletes participated at the World Championships in the Olympic distance triathlon from 2009 to 2014. This small number of athletes may influence the results. The analyzed time period of five years is too short to detect an improvement in the performance of the oldest athletes because, as shown by Forster et al. (19), the development of performance occurs slower than in younger age groups. Some of the older athletes may also have reached their upper limit of performance. Performance Trends in Transition Times Transition times were improved by all age group athletes. This may be the result of frequent training of the changes between the single disciplines and increasing experience. Both Heiden & Burnett (26) and Millet & Vleck (47) recommend specific training to enable efficient progression through the transition zones. They reported: this is particularly so at the top end of the field. For example, at the 1997 World Championships, Chris McCormack took the lead over his opponents with a gap of eight seconds earned in the transition area, and was never caught (47). The proportionate share of transitions represents only 1% of the total time consumed (20). Despite the fact that this represents only a small part of the total time, with an increasingly high level of competition, this aspect can be decisive. Men Were Not Faster in Older Age Groups Compared to Women Women were slower than men in swimming, cycling, running and overall race time in age groups 18-24 years to 70-74 years, but not in the age group 75-79 years. In the age group 70-75 years in 2009 only 2 women (USA) and in 2014 only 3 women (2x USA, 1x GBR) participated and in 2014 one athlete did not finish the race. Compared to this small group, 26 male athletes took part. With a look at the expectancy of life in these countries (USA m/f 75.6/80.8 y, GBR m/f 77.2/81.6 y) one would expect more female athletes than males. 11 This leads to the assumption that men and women participate for different reasons (38). It is possible that women only take part in competitions when they feel sufficiently prepared and equipped, so that they are really able to compete with male athletes of the same age. Sex Difference in Performance The performance difference between male and female athletes remained constant from 2009 to 2014. In agreement with Etter et al. (17), who reported in their study of Olympic distance triathlons held from 2000 to 2010 at the Zurich Triathlon, no changes in sex difference in total times or in the three individual disciplines for the five top elite triathlon athletes were shown. In running, numerous studies have reported a rapprochement of female performance levels to those of men at increasing distances (4, 5, 13, 28). Similar results were reported by Lepers et al. (40) for the Ironman distance triathlon over 25 years from 1986 to 2010, although the sex difference for the age group 18-39 years remained constant. When comparing the Olympic distance to longer distances, the energy-generation advantages of the female organism reported for ultralong distances (4, 5, 13) cannot be a significant factor. Alongside the distance of the competition, Salihu et al. (54) found that drafting (which is not allowed in cycling for age group athletes in Olympic distance triathlon) generally has an influence on the sex difference. 12 Strengths and Weaknesses of the Study The strength of this study is the sizeable amount of data provided by all World Championship age groups athletes in Olympic distance triathlons from 2009 to 2014. A total of 6,484 data sets were analyzed. The results from London 2013 had to be excluded due to the different swimming distance. The varying effect of external influences such as terrain, race route, altitude, air and water temperature, transition zone set-ups, wind strength and wind direction were not assessed (15, 16, 20, 57, 62). The fact that the ITU will record water and air temperature, as well as details such as whether or not wetsuits were worn, could help clarify some of the parameters mentioned above. 13 Future studies should examine performance trends in the Olympic triathlon over a longer period (21) and at the same competition site, to show the direct influence of these specific parameters on outcomes, as Lepers et al. (40) have demonstrated with their study of performance trends at the Ironman Hawaii. The range of technical equipment used and worn by the athletes could also not be taken into in this study. A comparative study between athletes competing with, for example, mechanical as opposed to electronic gear shifting could 11 https://de.wikipedia.org/wiki/liste_der_durchschnittlichen_lebenserwartung_in_den_staaten_der_erde) 12 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf 13 ITU Executive Board. ITU Competition Rules (2014). 2014. http://www.triathlon.org/uploads/docs/itusport_competition-rules_19022014v2- highlighted.pdf

12 Wonerow, Rüst, Nikolaidis, Rosemann and Knechtle help triathletes with their choice of equipment. Similar study models would likewise be meaningful for setting up training and nutrition programs. Even though the age-related performance peaks are in the younger age groups, as already investigated for various distances (6, 17, 18, 39, 58), we showed in our study that athletes can still improve their personal performance in higher ages. The positive performance trend for the female sex indicated an improvement in training, optimization of competition tactics and better equipment, meaning that women can improve their performance over the entire age range and in all disciplines. Further studies should be performed to find out the specific influence of equipment, training, tactics and techniques on competition performance. Today s society demands body fitness and performance up to a relative old age (38). Regular training cannot stop people getting old, but it can decelerate the aging process (42-44, 50, 61). In addition to the increasing number of events and participants in a wide range of endurance competitions (66 67), in triathlons with longer distances, the top athletes are getting older (37). Moreover, the trend shows that winners are increasingly coming from higher age groups and that the winning times are improving. Gallman et al. (21) were able to provide evidence for this from the Ironman Hawaii. As part of optimized training, target split times can predict the final position in the race, which helps to define race tactics (48). This knowledge should be considered for training, competition and career planning (22, 37). Practical Applications For athletes and coaches, age group athletes competing in Olympic distance triathlons in recent years showed an improvement in split times, overall race times and transition times. As already mentioned, the focus should be placed on target split times to predict the final position in the race (48). Transition zone training is also important, as shown by Heiden & Burnett (26) and Millet & Vleck (47), because as levels of competition, this can also be decisive. Based upon these findings for a relatively short period of time, we expect a further improvement in performance. As women older than 75 years were able to achieve a similar performance tomen, coaches need to encourage elderly women to continue training and competing in order to beat men in the higher age groups. More substantial physiology-based research is recommended, especially in the area of ageing and athlete performance. References 1. Ahmadyar, B., Rosemann, T., Rüst, C.A. and Knechtle, B. Improved race times in marathoners older than 75 years in the last 25 years in the World's largest marathons. Chinese J. Physiol. 59: 139-147, 2016. 2. Anthony, D., Rüst, C.A., Cribari, M., Rosemann, T., Lepers, R. and Knechtle, B. Differences in participation and performance trends in age group half and full marathoners. Chinese J. Physiol. 57: 209-219, 2014. 3. Arslan, E. and Aras, D. Comparison of body composition, heart rate variability, aerobic and anaerobic performance between competitive cyclists and triathletes. J. Phys. Ther. Sci. 28: 1325-1329, 2016. 4. Bam, J., Noakes, T.D., Juritz, J. and Dennis, S.C. Could women outrun men in ultramarathon races? Med. Sci. Sports Exerc. 29: 244-247, 1997. 5. Beneke, R., Leithäuser, R.M. and Doppelmayr, M. Women will do it in the long run. Brit. J. Sports Med. 39: 410, 2005. 6. Bernard, T., Sultana, F., Lepers, R., Hausswirth, C. and Brisswalter, J. Age-related decline in olympic triathlon performance: effect of locomotion mode. Exp. Aging Res. 36: 64-78, 2010. 7. Bijker, K.E., de Groot, G. and Hollander, A.P. Differences in leg muscle activity during running and cycling in humans. Eur. J. Appl. Physiol. 87: 556-561, 2002. 8. Bongard, V., McDermott, A.Y., Dallal, G.E. and Schaefer, E.J. Effects of age and gender on physical performance. Age (Dordr.) 29: 77-85, 2007. 9. Brisswalter, J., Wu, S.S.X., Sultana, F., Bernard, T. and Abbiss, C.R. Age difference in efficiency of locomotion and maximal power output in well-trained triathletes. Eur. J. Appl. Physiol. 114: 2579-2586, 2014. 10. Canda, A.S., Castiblanco, L.A., Toro, A.N., Amestoy, J.A. and Higueras, S. Morphological characteristics of the triathlete according to sex, category and competitive level. Apunts Med. Esport. 49: 75-84, 2014. 11. Cazorla, R.L., Mendo, A.H., Garrido, R.E.R. and Sánchez, V.M. Relationship between self-concept and psychological sport profile in triathletes. C. P. D. 15: 95-102, 2015. 12. Chatard, J.C. and Wilson, B. Drafting distance in swimming. Med. Sci. Sports Exerc. 35: 1176-1181, 2003. 13. Coast, J.R., Blevins, J.S. and Wilson, B.A. Do gender differences in running performance disappear with distance? Can. J. Appl. Physiol. 29: 139-145, 2004. 14. Dähler, P., Rüst, C.A., Rosemann, T., Lepers, R. and Knechtle, B. Nation related participation and performance trends in 'Ironman Hawaii' from 1985 to 2012. BMC Sports Sci. Med. Rehabil. 6: 16, 2014. 15. El Helou, N., Tafflet, M., Berthelot, G., Tolaini, J., Marc, A., Guillaume, M., Hausswirth, C. and Toussaint, J.F. Impact of environmental parameters on Marathon running performance. PLoS One 7: 37407, 2012. 16. Ely, M.R., Cheuvront, S.N., Roberts, W.O. and Montain, S.J. Impact of weather on marathon-running performance. Med. Sci. Sports Exerc. 39: 487-493, 2007. 17. Etter, F., Knechtle, B., Bukowski, A., Rüst, C.A., Rosemann, T. and Lepers, R. Age and gender interactions in short distance triathlon performance. J. Sports Sci. 31: 996-1006, 2013. 18. Etter, F., Knechtle, B., Rüst, C.A., Rosemann, T. and Lepers, R. The age-related decline in Olympic distance triathlon performance differs between males and females. J. Sports Med. Phys. Fitness 53: 261-267, 2013. 19. Foster, C., Wright, G., Battista, R.A. and Porcari, J.P. Training in the aging athlete. Curr. Sports Med. Rep. 6: 200-206, 2007. 20. Fröhlich, M., Balter, J., Pieter, A., Schwarz, M. and Emrich, E. Model-theoretic optimization approach to triathlon performance under comparative static conditions results based on the Olympic Games 2012. Int. J. Kinesiol. Sports Sci. 1: 9-14, 2012. 21. Gallmann, D., Knechtle, B., Rüst, C.A., Rosemann, T. and Lepers, R. Elite triathletes in 'Ironman Hawaii' get older but faster.