Content uploaded by Christoph Alexander Rüst
Author content
All content in this area was uploaded by Christoph Alexander Rüst on Aug 31, 2014
Content may be subject to copyright.
Corresponding author: Beat Knechtle, Ph.D., M.D., Beat Knechtle, Facharzt FMH für Allgemeinmedizin, Gesundheitszentrum St. Gallen,
Vadianstrasse 26, 9001 St. Gallen, Switzerland, Tel: +41 (0) 71 226 82 82, Fax: +41 (0) 71 226 82 72, E-mail: beat.knechtle@hispeed.ch
Received: February 23, 2011; Revised: April 3, 2011; Accepted: May 4, 2001.
2012 by The Chinese Physiological Society and Airiti Press Inc. ISSN : 0304-4920. http://www.cps.org.tw
Chinese Journal of Physiology 55(×): ×××-×××, 2012 1
DOI: 10.4077/CJP.2012.BAA014
Personal Best Times in an Olympic Distance
Triathlon and a Marathon Predict an Ironman
Race Time for Recreational Female Triathletes
Christoph Alexander Rüst1, Beat Knechtle1, 2, *, Andrea Wirth2, Patrizia Knechtle2,
Birte Ellenrieder3, Thomas Rosemann1, and Romuald Lepers4
1Institute of General Practice and Health Services Research, University of Zurich,
Zurich, Switzerland
2Gesundheitszentrum St. Gallen, St. Gallen, Switzerland
3Klinik für Orthopädische Chirurgie und Traumatologie, Kantonsspital St. Gallen, Switzerland
and
4INSERM U887, University of Burgundy, Faculty of Sport Sciences, Dijon, France
Abstract
The aim of this study was to investigate whether the characteristics of anthropometry, training or
previous performance were related to an Ironman race time in recreational female Ironman triathletes.
These characteristics were correlated to an Ironman race time for 53 recreational female triathletes in
order to determine the predictor variables, and so be able to predict an Ironman race time for future
novice triathletes. In the bi-variate analysis, no anthropometric characteristic was related to race time.
The weekly cycling kilometers (r = -0.35) and hours (r = -0.32), as well as the personal best time in an
Olympic distance triathlon (r = 0.49) and in a marathon (r = 0.74) were related to an Ironman race time
(P < 0.05). Stepwise multiple regressions showed that both the personal best time in an Olympic distance
triathlon (P = 0.0453) and in a marathon (P = 0.0030) were the best predictors for the Ironman race time
(n = 28, r2 = 0.53). The race time in an Ironman triathlon might be partially predicted by the following
equation (r2 = 0.53, n = 28): Race time (min) = 186.3 + 1.595 * (personal best time in an Olympic distance
triathlon, min) + 1.318 * (personal best time in a marathon, min) for recreational female Ironman
triathletes.
Key Words: body fat, gender, endurance, performance
Introduction
Ironman triathlons covering the distances of
3.8 km swimming, 180 km cycling and 42.2 km runn-
ing are increasing in popularity. Every year, tens of
thousands of athletes participate in these races in
order to qualify for the Ironman World Championship
in Hawaii (22, 23). Triathletes need to train in three
different disciplines in preparation for such a race
because finishing exerts such an enormous physical
effort. The question is which of these characteristics
of anthropometry, physiology, training, previous
experience or psychology, were the most important
for adequate race preparation.
In recent years, several studies have tried to
determine the predictor variables for an Ironman race
time, especially for male Ironman triathletes. Percent
body fat (14, 15), the sum of upper body skin-fold
thicknesses (13), the personal best time in both an
Olympic distance triathlon (8, 17) and a marathon
(17) were related to an Ironman race time for male
Ironman triathletes. In a recent study of male Ironman
triathletes, running speed during training, a personal
best time in a marathon and a personal best time in an
Olympic distance triathlon were related to the Ironman
race time. These three variables explained 64% of the
CJP E-prepublication Ahead of Print
2Rüst, Knechtle, Wirth, Knechtle, Ellenrieder, Rosemann and Lepers
variance in the Ironman race time (17). In another
study, the previous best performance in an Olympic
distance triathlon, coupled with the weekly cycling
distances and the longest training ride, could partially
predict an overall Ironman race performance when
both male and female Ironman triathletes were in-
cluded (8).
For female Ironman triathletes, however, there
is limited data regarding predictor variables for an
Ironman race time. Leake and Carter (21) reported
that training parameters were more important than
anthropometric measurements in the prediction of
performance for 16 female triathletes. In studies in-
vestigating small samples of recreational female
Ironman triathletes, weekly training hours (14, 15),
personal best times in an Ironman triathlon (14, 18),
personal best times in a marathon (18), and personal
best times in an Olympic distance triathlon (18) were
related to an Ironman race time.
The aim of this present study was to investigate,
in a larger sample of recreational female Ironman
triathletes, which of the basic variables of anthro-
pometry, training or previous performance was related
to an Ironman race time. It was an intention of this
investigation to create an equation for predicting an
Ironman race time for novice recreational female
Ironman triathletes using basic measurements any
athlete or coach could determine without the need for
highly sophisticated equipment.
Materials and Methods
Subjects
A cross-sectional, observational field study was
performed at the ‘IRONMAN SWITZERLAND’.
Since female participation in Ironman races is rather
low (14, 15, 18), data from four consecutive years
from 2007 to 2010 were collected in order to increase
the sample size to have enough statistical power. The
organizer of the ‘IRONMAN SWITZERLAND’
contacted all the female athletes via a newsletter three
months before each race and asked them to participate
in this investigation. A total of 59 non-professional
female Ironman triathletes volunteered to participate
in our investigation over this four year period. Each
subject was included only upon the first participation
in the race. The study was approved by the Institutional
Review Board for the use of Human subjects of the
Canton of St. Gallen, Switzerland. The athletes were
informed of the procedures and gave their informed
written consent. Fifty-three athletes out of our study
group finished the race successfully within the time
limit of 16 h. Six triathletes had to give up during the
run because of medical complications such as exhaus-
tion and overuse injuries of the lower limbs.
The Race
In the ‘IRONMAN SWITZERLAND’ the ath-
letes have to swim two 1.9 km laps in Lake Zurich,
cycle two 90 km laps, and then run four 10.5 km laps.
In the cycling section, the highest point of ascent
from Zurich (400 m above sea level) is the ‘Forch’
(700 m above sea level), while the run course is com-
pletely flat in the City of Zurich.
Measurements
Before the start of the race body mass, body
height, the lengths of the arm and the leg, the circum-
ferences of the limbs (upper arm, thigh, and calf), and
the thicknesses of skin-folds at eight sites (pectoralis,
axillar, triceps, subscapular, abdomen, suprailiac,
thigh, and calf) were measured on the right side of the
body. Body mass was measured using a commercial
scale (Beurer BF 15, Beurer, Ulm, Germany) to the
nearest 0.1 kg. Body height was measured using a
stadiometer to the nearest 1.0 cm. The circumferences
and lengths of limbs were measured using a nonelastic
tape measure (cm) (KaWe CE, Kirchner und Welhelm,
Germany) to the nearest 0.1 cm. The length of the
arm was measured from acromion to the end of pha-
lanx distalis of the third finger, the length of leg from
trochanter major to the middle of malleolus lateralis.
The circumference of the upper arm was measured
at mid-upper arm, the circumference of the thigh
was taken at mid-thigh and the circumference of the
calf was measured at mid-calf. The skin-fold data
were obtained using a skin-fold caliper (GPM-
Hautfaltenmessgerät, Siber & Hegner, Zurich,
Switzerland) and recorded to the nearest 0.2 mm.
The skin-fold measurements were taken once for all
eight skin-folds and then the procedure was repeated
twice more and the mean of the three measurements
was used for the analyses. The timing of the taking
of the skin-fold measurements was standardized to
ensure reliability. According to Becque et al. (2)
the readings were performed 4 s after applying the
caliper. One trained investigator took all the skin-
fold measurements as inter-tester variability is a major
source of error in skin-fold measurements. An intra-
tester reliability check was conducted on 11 female
runners prior to testing. Intra-class correlation (ICC)
was excellent for each of the two judges for all the
anatomical measurement sites (ICC > 0.9). Agree-
ment tended to be better individually than between
the measurers, but still reached excellent reliability
(ICC > 0.9) for the summary measurements of the
skin-fold thicknesses (11). Percent body fat was
calculated using the following anthropometric formula
for women: Percent body fat = –6.40665 + 0.41946 *
(Σ3SF) – 0.00126 * (Σ3SF)2 + 0.12515 * (hip) +
Predictor Variables in Triathlon 3
0.06473 * (age), according to Ball et al. (1) where
Σ3SF means the sum of three skin-folds (triceps,
suprailiacal and thigh) and hip means the circum-
ference of the hip in cm. The circumference of the
hip was determined at the level of the trochanter
major to the nearest 0.1 cm.
Upon inscription to the investigation, each athlete
was asked to maintain a comprehensive training diary,
recording each endurance training session and showing
both the distance and duration in each discipline, since
training volume is important for endurance athletes
(27). The athletes received an EXCEL-sheet to record
each training session. Speed per unit was calculated
using kilometres and time per training unit. The
athletes also recorded their personal best time in both
an Olympic distance triathlon and in a marathon. The
personal best times were defined as the best time ever
achieved over these distances, independent of the course
or any environmental factors.
Statistical Analysis
Normally distributed data are presented as
means ± standard deviations (SD). A potential asso-
ciation between the characteristics of anthropometry,
training and previous performance was investigated
using Pearson correlation analysis. Stepwise multiple
regression analysis was then used to determine the
best variables correlated with the Ironman race time.
The significant variables were used to create an
equation for predicting an Ironman race time. A
power calculation was performed according to
Gatsonis and Sampson (7). To achieve a power of
80% (two-sided Type I error of 5%) to detect a
minimal association between race time and anthro-
pometric characteristics of 20% (i.e. coefficient of
determination r2 = 0.2) a sample of 40 participants
was required. Bland-Altman analysis was used to
determine absolute limits of agreement between
Table 1. Age and anthropometric characteristics and bivariate association with the Ironman race time (n = 53)
Pearson r
Age (years) 37.0 ± 6.7 0.23
Body height (m) 1.67 ± 0.06 -0.17
Body mass (kg) 59.9 ± 5.9 0.05
Body mass index (kg/m2)21.3 ± 1.6 0.25
Length of leg (cm) 82.8 ± 5.3 -0.19
Length of arm (cm) 74.4 ± 3.6 -0.15
Circumference of upper arm (cm) 26.4 ± 1.5 0.16
Circumference of thigh (cm) 53.1 ± 2.9 0.15
Circumference of calf (cm) 36.0 ± 2.1 0.12
Percent body fat (%) 23.8 ± 5.7 0.10
Table 2. Training characteristics and pre race experience of the subjects and bivariate association with the
Ironman race time (n = 53)
Pearson r
Weekly training hours (h) 14.1 ± 3.5 -0.20
Weekly kilometers swimming (km) 6.2 ± 2.7 -0.14
Weekly hours swimming (h) 2.8 ± 1.1 0.12
Speed during swimming (km/h) 2.8 ± 0.6 -0.25
Weekly kilometers cycling (km) 196.6 ± 83.5 -0.35, P = 0.0150
Weekly hours cycling (h) 7.4 ± 2.5 -0.32, P = 0.0201
Speed during cycling (km/h) 26.0 ± 3.6 -0.10
Weekly kilometers running (km) 41.0 ± 10.7 -0.06
Weekly hours running (h) 4.1 ± 1.0 0.03
Speed during running (km/h) 10.7 ± 1.4 -0.24
Personal best time Olympic distance triathlon (min) (n = 41) 152.5 ± 15.3 0.49, P = 0.0013
Personal best time in a marathon (min) (n = 36) 230.7 ± 26.1 0.74, P < 0.0001
P-value is inserted in case of a significant association. Significance level was set at P < 0.05.
4Rüst, Knechtle, Wirth, Knechtle, Ellenrieder, Rosemann and Lepers
Table 3. Multiple linear regression analysis with race time as the dependent variable (n = 28)
ßSEP
Mean weekly cycling kilometers 0.21 0.23 0.35
Mean weekly cycling hours -7.09 7.58 0.35
Personal best time in an Olympic distance triathlon 2.99 0.71 0.0453
Personal best time in a marathon 1.07 0.35 0.0030
The coefficient of determination (r2) of the model was 53%. Significance level was set at P < 0.05.
Fig. 1. The personal best time in an Olympic distance triathlon
was significantly and positively related to the Ironman
race time (n = 41) (r = 0.49, P = 0.0013).
Fig. 2. The personal best time in a marathon correlated signifi-
cantly and positively to the Ironman race time (n = 36)
(r = 0.74, P < 0.0001).
predicted and effective race time. A level of 0.05 was
used to indicate significance.
Results
The 53 athletes completed the race within 751 ±
89 min. In the bi-variate analysis, none of the anthro-
pometric characteristics were related to the Ironman
race time (see Table 1). Considering the charac-
teristics of training and previous performance, the
mean weekly cycling kilometers, the mean weekly
cycling hours and the personal best time in both an
Olympic distance triathlon and a marathon were related
to the Ironman race time after bivariate analysis (see
Table 2). Stepwise multiple regressions showed that
the personal best time in both the Olympic distance
triathlon (see Fig. 1) and in the marathon (see Fig. 2)
were the best predictor variables for an Ironman race
time when corrected with all significant variables
after bivariate analysis (see Table 3). The personal
best marathon time was significantly and positively
related to the run split time in the Ironman race (r =
0.57, P < 0.0001). The race time in an Ironman
triathlon might be partially predicted by the following
equation for recreational female Ironman (r2 = 0.52,
n = 28): Race time (min) = 186.3 + 1.595 * (personal
best time in an Olympic distance triathlon, min) +
1.318 * (personal best time in a marathon, min). The
predicted Ironman race time was 725 ± 51 min and
correlated highly significantly to the achieved Ironman
race time (see Fig. 3). Fig. 4 shows the level of agre-
ement using Bland-Altman method (Bias = -93.5 ±
93.5 min) between the effective and the predicted
race time. Intra class correlation (ICC) between ef-
fective and predicted Ironman race time was 0.70.
Discussion
The aim of this study was to investigate which
of the basic variables of physical characteristics,
training or previous performance were related to an
Ironman race time for recreational female triathletes
in order to create an equation for predicting an Ironman
race time for future novice female Ironman triathletes.
1000
950
900
850
800
750
700
600
550
500
650
Ironman Race Time (min)
120 140 160
Personal Best Time in an Olympic Distance Triathlon (min)
180 200
1000
950
900
850
800
750
650
600
550
500
700
Ironman Race Time (min)
180 200 220
Personal Best Time in a Marathon (min)
240 280260
Predictor Variables in Triathlon 5
After multi-variate analysis, both the personal best
times in an Olympic distance triathlon and in a
marathon were related to the Ironman race time;
anthropometric and training characteristics, however,
were not.
The association of a personal best time in a
marathon and in an Olympic distance triathlon with
the Ironman race time showed highly significant
correlation coefficients after bi-variate analysis and
remained the single predictor variables after multi-
variate analysis. This specific finding, that a personal
best time in a race shorter than the actual race is a
predictor variable for race time has been reported for
male Ironman triathletes (8, 17), female Ironman
triathletes (18), and male ultra-endurance runners
(12, 16). Gulbin and Gaffney (8) described that pre-
vious best performances in Olympic distance triathlons
coupled with weekly cycling distances and longest
training rides could partially predict an Ironman race
time when investigating both male and female Ironman
triathletes in the same sample. For male Ironman
triathletes (17), speed in running during training,
together with a personal best time in a marathon and
an Olympic distance triathlon, were related to the
Ironman race time. These three variables explained
64% of the variance in the Ironman race time (17). In
male ultra-runners during a 24-hour run, only the
personal best marathon time was related to race
performance, not anthropometry or training (16). In
multi-stage mountain ultra-marathoners, again the
personal best marathon time was related to total race
time (12).
Since Leake and Carter (21) reported that train-
ing parameters were more important than anthropo-
metric measurements in the prediction of performance
for 16 female triathletes, we also investigated a po-
tential association between training characteristics
and the Ironman race time. O’Toole (24) and Gulbin
and Gaffney (8) concluded that distances in training
were more important than intensity, whereas Hendy
and Boyer (9) described the opposite. In the bi-
variate analysis, the weekly cycling kilometers (r =
-0.35) and the weekly cycling hours (r = -0.32) were
related to an Ironman race time in these female
triathletes. However, neither the volume in kilometres
nor the speed in training was found to be associated
with the Ironman race time when corrected for the
variables of anthropometry and previous performance.
These disparate findings might be explained by the
different distances and genders. Gulbin and Gaffney
(8) found, in their 242 lower level Ironman triathletes
of 230 male and 12 female athletes, that weekly cy-
cling distances were related to an Ironman race time.
However, they did not include anthropometric
measurements in their considerations as was con-
sidered in this investigation. Furthermore, they had
more male than female athletes in their sample.
For male triathletes, body fat was an important
predictor variable for race time over both the Olympic
(19, 28) and Ironman (14, 15) distances. In contrast
to these studies, no association between anthro-
pometric characteristics and the Ironman race was
found for these recreational female Ironman triathletes,
as has already been described in studies using smaller
samples of recreational female Ironman triathletes
(14, 15, 18). This might be due to gender or the longer
Fig. 4. Bland-Altman plots comparing predicted with effective
race time.
Fig. 3. The predicted Ironman race time correlated significantly
to the achieved Ironman race time (n = 28) (r = 0.73, P <
0.0001).
850
800
750
650
600
700
550 600 650 700
Effective Race Time (min)
900750 800 850
Predicted Race Time (min)
100
80
60
40
20
0
-20
-60
-80
-100
-40
Difference (effective-predicted)
550 600 650 700
Mean of All
750 800 850 900
6Rüst, Knechtle, Wirth, Knechtle, Ellenrieder, Rosemann and Lepers
distance involved compared with the Olympic
distance. Laurenson et al. (20) concluded that no
ideal or unique anthropometric profile could be
established for female triathletes competing over the
Olympic distance. Also for female marathoners,
body fat percentage did not correlate with the finish
time (3). The present findings now confirm that
anthropometric characteristics were not able to predict
an Ironman race time for female Ironman triathletes.
The personal best time in an Olympic triathlon
will be influenced by a wide range of physiological,
psychological and behavioural factors. The inclusion
of a personal best time in the Olympic distance as
a predictive variable in the multi-variate analyses
can identify the influence of additional factors which
may include the difficulty in defining the concept of
‘pre- race experience’. In future studies, recreational
female Ironman triathletes should be compared with
recreational female marathoners in order to find
similarities or differences between these two groups
of athletes.
This study is limited because of the rather small
number of subjects. However, Gulbin and Gaffney
(8) had in their study of 242 Ironman triathletes a total
of 230 male and only 12 female athletes. This study
is also limited as nutritional intake was not assessed.
It is very likely that race nutrition will influence
overall race time in Ironman events (10). The problem
of fluid intake and, especially, exercise-associated
hyponatremia might have an influence on race time
(29, 30). In further studies, nutrition should also be
considered. The determination of physiological
characteristics such as maximum oxygen uptake or
lactate threshold would be useful (4, 25). In addition,
the determination of muscle fiber composition would
give more physiological insights (6). Apart from the
variables of physiology, anthropometry, training and
previous experience, the aspect of motivation might
also considerably influence an Ironman race outcome
(26, 31). This data analysis is also limited since en-
vironmental conditions were not included. Environ-
mental factors might influence race performance (5,
32). It has been shown that marathon performance
progressively slows when the temperature increases
from 5 °Celsius to 25 °Celsius (5). Furthermore, the
time between the personal best times and the Ironman
race might be very different for each athlete and
therefore influence this association.
To summarize, the present findings suggest that
a previous performance in marathon running and
Olympic distance triathlons is of greater importance
than anthropometric or training characteristics in
recreational female Ironman triathletes. Race time
in an Ironman triathlon might be predicted by the
following equation (r2 = 0.52, n = 28): Race time
(min) = 186.3 + 1.595 * (personal best time in an
Olympic distance triathlon, min) + 1.318 * (personal
best time in a marathon, min) for recreational female
Ironman triathletes. Further studies examining the
physiological and psychological characteristics of
Ironman triathletes are required to better understand
the determinants of an Ironman triathlon performance.
The inclusion of physiological variables might in-
crease the coefficient of correlation for the equation
to predict an Ironman race time.
Acknowledgments
We thank the crew of ‘IRONMAN
SWITZERLAND’ for their generous support and the
athletes for their promptness in the collection of data
during the race. For her help in translation, we thank
Mary Miller from Stockton-on-Tees, Cleveland,
England.
Author disclosures
The authors have no conflict of interest and
received no external funding.
References
1. Ball, S.D., Swan, P. and Desimone, R. Accuracy of anthropometry
compared to dual energy x-ray absorptiometry. A new generaliz-
able equation for women. Res. Q. Exerc. Sport 75: 248-258, 2004.
2. Becque, M.D., Katch, V.L. and Moffatt, R.J. Time course of skin-
plus-fat compression in males and females. Hum. Biol. 58: 33-42,
1986.
3. Christensen, C.L. and Ruhling, R.O. Physical characteristics of
novice and experienced women marathon runners. Brit. J. Sports
Med. 17: 166-171, 1983.
4. Dengel, D.R., Flynn, M.G., Costill, D.L. and Kirwan, J.P. Deter-
minants of success during triathlon competition. Res. Q. Exerc.
Sport 60: 234-238, 1989.
5. 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.
6. Flynn, M.G., Costill, D.L., Kirwan, J.P., Fink, W.J. and Dengel,
D.R. Muscle fiber composition and respiratory capacity in triath-
letes. Int. J. Sports Med. 8: 383-386, 1987.
7. Gatsonis, C. and Sampson, A.R. Multiple correlation: exact power
and sample size calculations. Psychol. Bull. 106: 516-524, 1989.
8. Gulbin, J.P. and Gaffney, P.T. Ultraendurance triathlon participa-
tion: typical race preparation of lower level triathletes. J. Sports
Med. Phys. Fitness 39: 12-15, 1999.
9. Hendy, H.M. and Boyer, B.J. Specifity in the relationship between
training and performance in triathlons. Percept. Mot. Skills 81:
1231-1240, 1995.
10. Kimber, N.E., Ross, J.J., Mason, S.L. and Speedy, D.B. Energy
balance during an Ironman triathlon in male and female triathletes.
Int. J. Sport Nutr. Exerc. Metab. 12: 47-62, 2002.
11. Knechtle, B., Joleska, I., Wirth, A., Knechtle, P., Rosemann, T.
and Senn, O. Intra- and inter-judge reliabilities in measuring
the skin-fold thicknesses of ultra-runners under field conditions.
Percept. Mot. Skills 111: 1-2, 2010.
12. Knechtle, B., Knechtle, P. and Rosemann, T. Race performance in
male mountain ultra-marathoners: anthropometry or training?
Percept. Mot. Skills 110: 721-735, 2010.
Predictor Variables in Triathlon 7
13. Knechtle, B., Knechtle, P. and Rosemann, T. Upper body skin-fold
thickness is related to race performance in male Ironman triathletes.
Int. J. Sports Med. 32: 20-27, 2011.
14. Knechtle, B., Wirth, A., Baumann, B., Knechtle, P. and Rosemann,
T. Personal best time, percent body fat and training volume are
differently associated with race time in male and female Ironman
triathletes. Res. Q. Exerc. Sport 81: 62-68, 2010.
15. Knechtle, B., Wirth, A., Knechtle, P., Baumann, B., Rosemann, T.
and Senn, O. Differential correlations between anthropometry,
training volume and performance in male and female Ironman
triathletes. J. Strength Cond. Res. 24: 2785-2793, 2010.
16. Knechtle, B., Wirth, A., Knechtle, P., Zimmermann, K. and Kohler,
G. Personal best marathon performance is associated with per-
formance in a 24-h run and not anthropometry or training volume.
Brit. J. Sports Med. 43: 836-839, 2009.
17. Knechtle, B., Wirth, A. and Rosemann, T. Predictors of race time
in male Ironman triathletes: physical characteristics, training, or
prerace experience? Percept. Mot. Skills 111: 437-446, 2010.
18. Knechtle, B., Wirth, A. and Rosemann, T. Is body fat a predictor
variable for race performance in recreational female Ironman
triathletes? Med. Sport. 15: 6-12, 2011.
19. Landers, G.J., Blanksby, B.A., Ackland, T.R. and Smith, D. Mor-
phology and performance of World championship triathletes.
Ann. Hum. Biol. 27: 387-400, 2000.
20. Laurenson, N.M., Fulcher, K.Y. and Korkia, P. Physiological
characteristics of elite and club level female triathletes during
running. Int. J. Sports Med. 14: 455-459, 1993.
21. Leake, C.N. and Carter, J.E. Comparison of body composition
and somatotype of trained female triathletes. J. Sports Sci. 9: 125-
135, 1991.
22. Lepers, R. Analysis of Hawaii Ironman performances in elite
triathletes from 1981-2007. Med. Sci. Sports Exerc. 40: 1828-1834,
2008.
23. Lepers, R. and Maffiuletti, N.A. Age and gender interactions in
ultraendurance performance: insight from the triathlon. Med. Sci.
Sports Exerc. 43: 134-139, 2011.
24. O’Toole, M.L. Training for ultraendurance triathlons. Med. Sci.
Sports Exerc. 21: S209-S213, 1989.
25. O’Toole, M.L., Douglas, P.S. and Hiller, W.D. Lactate, oxygen
uptake and cycling performance in triathletes. Int. J. Sports
Med. 10: 413-418, 1989.
26. Parry, D., Chinnasamy, C., Papadopoulou, E., Noakes, T. and
Micklewright, D. Cognition and performance: anxiety, mood and
perceived exertion among Ironman triathletes. Brit. J. Sports
Med., epub ahead of print, 2011.
27. Schumacher, Y.O., Mroz, R., Mueller, P., Schmid, A. and Ruecker,
G. Success in elite cycling: A prospective and retrospective
analysis of race results. J. Sports Sci. 24: 1149-1156, 2006.
28. Sleivert, G.G. and Rowlands, D.S. Physical and physiological
factors associated with success in the triathlon. Sports Med. 22:
8-18, 1996.
29. Speedy, D.B., Noakes, T.D., Rogers, I.R., Thompson, J.M.,
Campbell, R.G., Kuttner, J.A., Boswell, D.R., Wright, S. and
Hamlin, M. Hyponatremia in ultradistance triathletes. Med. Sci.
Sports Exerc. 31: 809-815, 1999.
30. Speedy, D.B., Rogers, I.R., Noakes, T.D., Wright, S., Thompson,
J.M., Campbell, R., Hellemans, I., Kimber, N.E., Boswell, D.R.,
Kuttner, J.A. and Safih, S. Exercise-induced hyponatremia in
ultradistance triathletes is caused by inappropriate fluid retention.
Clin. J. Sport Med. 10: 272-278, 2000.
31. Stoeber, J., Uphill, M.A. and Hotham, S. Predicting race perfor-
mance in triathlon: the role of perfectionism, achievement goals and
personal goal setting. J. Sport Exerc. Psychol. 31: 211-245, 2009.
32. Wegelin, J.A. and Hoffman, M.D. Variables associated with odds
of finishing and finish time in a 161-km ultramarathon. Eur. J. Appl.
Physiol., epub ahead of print, 2011.
