PLOS ONE
RESEARCH ARTICLE
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Yu Y, Li D, Lu Y, Mi J (2023) Relationship
between methods of monitoring training load and
physiological indicators changes during 4 weeks
cross-country skiing altitude training. PLoS ONE
18(12): e0295960. https://doi.org/10.1371/journal.
pone.0295960
Editor: Rafael Franco Soares Oliveira, Instituto
Politécnico de Santarém: Instituto Politecnico de
Santarem, PORTUGAL
Received: July 27, 2023
Accepted: December 2, 2023
Published: December 15, 2023
Copyright: © 2023 Yu et al. This is an open access
article distributed under the terms of the Creative
Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in
any medium, provided the original author and
source are credited.
Data Availability Statement: Data cannot be
shared publicly because in order to protect the
privacy of national team athletes, all data in this
study are protected by the Chinese Winter Sports
Management Centre. Data are available from the
Ethics Committee of the Beijing sport University
(contact: bsukjc@bsu.edu.cn) for researchers who
meet the criteria for access to confidential data.
Funding: This research was funded by the National
key research and development program of China
PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023
Relationship between methods of monitoring
training load and physiological indicators
changes during 4 weeks cross-country skiing
altitude training
Yichao Yu1,3☯, Dongye Li2,3☯, Yifan Lu ID2,3*, Jing Mi1
1 The School of Sports Coaching, Beijing Sports University, Beijing, China, 2 The School of Sports Medicine
and Rehabilitation, Beijing Sports University, Beijing, China, 3 Laboratory of Sports Stress and Adaptation of
General Administration of Sport, Beijing Sport University, Beijing, China
☯ These authors contributed equally to this work.
* luyifan@bsu.edu.cn
Abstract
This study aimed to: (i) analyze the load characteristics of 4 weeks cross-country skiing alti-
tude training; (ii) analyze the relationships between methods of monitoring training load and
physiological indicators changes of elite male Chinese cross-country skiers during this
period. Practitioners collected load data during 4 weeks of altitude training camp. Partici-
pants performed maximal oxygen uptake, lactate threshold, body composition, and skierg
power test before and after the training camp to investigate the changes in physiological per-
formance. Edwards TRIMP, Lucia TRIMP, and session rating of perceived exertion were
collected as internal load. Training distance, time recorded by the Catapult module were col-
lected as external load. The result revealed a " pyramid " pattern in the load characteristics
during the altitude training camp. The correlation between luTRIMP and percent change in
physiological indicators was highest. Percentage changes in lactate threshold velocity (r =
.78 [95% CI -.01 to .98]), percentage changes in lactate threshold HR (r = .71 [95% CI .14-
.99]), percentage changes in maximum HR (r = .83 [95% CI .19–1.00]), percentage changes
in skierg power-to-weight ratio (r = .75 [95% CI -.28 to .98]) had very large relationships with
luTRIMP. In cross-country skiing altitude training, training loads should be reasonably con-
trolled to ensure that athletes do not become overly fatigued. Methods of training load moni-
toring that combine with athletes’ physiological characteristics and program characteristics
have the highest dose-response relationships, it is an important aspect of cross-country ski
training load monitoring. The luTRIMP could be a good monitoring tool in cross-country ski-
ing altitude training.
Introduction
Cross-country skiing is a typical endurance sports, which requires high demands on the physi-
ological performance of athletes [1–3]. The purpose of the training is to allow athletes to
1 / 15
PLOS ONE
(2018YFC2000603). The funders had no role in
study design, data collection and analysis, decision
to publish, or preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023
Relationship between methods of monitoring training load and physiological indicators changes
develop positive physiological adaptations in response to adapted training load stimuli and
improve performance levels. Physiological indicators are an objective expression of this physi-
ological adaptation [4]. The required by modern training concepts is gradual, periodic, and
specific [5], and its essence is to stimulate the athlete’s body to produce the best adaptation
through the appropriate load (including volume and intensity) [6]. Practitioners need to moni-
tor skiers’ training load and physiological stress adaptation during different training period,
which allows skiers to maintain a high level of physiological performance and avoid a severe
drop in performance due to overtraining [7–11].
With the popularity of various types of training monitoring equipment and the use of
advanced training analysis software, it is possible to systematically analyze the training load of
skiers. Training load among cross-country skiing is the cumulative stress an individual is sub-
jected to in competition or training throughout training period [12]. According to the source
of monitored indicators, there are two classification forms of load: external and internal [13].
Similar to most endurance sports, heart rate changes during cross-country skiing training are
key to load monitoring [14, 15]. In the last century, Banister pioneered the Training Impulse
(TRIMP) to help practitioners with load monitoring, which is a combination of training time,
heart rate during training period [16]. Since then, Edwards [17] and Lucia [18] proposed two
interval TRIMP calculation methods, where a linear weighting factor is used to weight the
training time for the delineated heart rate intervals. On this basis, Manzi [19] have proposed
TRIMP calculations based on heart rate and lactate changes. In addition to heart rate based
load monitoring methods, session rating of perceived exertion (sRPE) [15, 20] and external
load information collected by global positioning system (GPS) [21, 22] are also applied in ski
training practice. These internal or external load-based quantification methods can help prac-
titioners visualize the training loads to which athletes are exposed.
It is easier than ever for practitioners to access training-related load data with advances in
technology. For load monitoring to be maximally effective, it is imperative that the employed
methods are pertinent to significant physiological outcomes. But there is a great deal of uncer-
tainty about the relationship between physiological performance change and these load data.
Practitioners should select an appropriate load monitoring method based on the relationship
between physiological performance and methods of training load, thus allowing for a more
proactive approach to the development of training programmes [23]. In team sports such as
soccer, some researchers have explored this phenomenon. Oliveira et al. [24] analyzed the rela-
tionship between wellness and training and match load in 13 professional male soccer players,
they found that the intensity of training on the match day was correlated with indicators of
sleep quality and fatigue the next day. Costa et al. [25] did not find the within-subject relation-
ship between sleep indicators and HRV with training and match load in 20 elite female soccer
players, the use of statistical methods may be a possible reason for the difference. In endurance
sports, this relationship can be assessed by evaluating athletes’ physiological performance
change before and after training period. A study of 8 distance runners found that athletes
showed a significant increase in lactate threshold velocity after training. There were essentially
relationships between improvements of lactate threshold velocity and weekly individualized
TRIMP (iTRIMP). But weaker relationships between Banister TRIMP (bTRIMP) and velocity
improvements at the lactate threshold [19]. Sanders et al. [26] found that a load monitoring
approach incorporating the individual physiological characteristics of the athlete has the most
substantial dose-response relationship in 15 well-trained cyclists. The difference of load moni-
toring means and the difference of sports can be one of the reasons for the different dose-
response relationships.
Previous research has established a relationship between training load and physiological
performance that follows a dose-response relationship. However, this relationship is complex
2 / 15
PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

and influenced by multiple factors, and it remains unclear whether the same holds true for
cross-country ski training. This topic warrants further investigation, as cross-country skiing
differs significantly from summer sports. Shedding light on this dose-response relationship
can offer coaches valuable reference and data support. While a great deal of research in the
past has focused on the Nordic countries, research on Chinese cross-country skiers could also
enrich the training experience of emerging countries in cross-country ski training.
Therefore, the purpose of this study was to: (i) analyze the load characteristics of 4 weeks
cross-country skiing altitude training; (ii) analyze the relationships between methods of moni-
toring training load and physiological indicators changes of elite male Chinese cross-country
skiers during this period with an observation approach. Based on prior research in endurance
sport, it was hypothesised that the training load characteristics conform to the classical endur-
ance training model. Additionally, a load monitoring method that considers the physiological
characteristics of the individual athlete is expected to provide the best dose-effect relationship.

Materials and methods

Participants and study design
The study followed a descriptive, observational design, highlighting the relationship between
methods of monitoring training load and physiological indicators changes during 4 weeks
cross-country skiing altitude training (Fig 1). This study included 8 male athletes (age: 20.8
±1.1 years; body mass 69.7±5.1 kg and height 179.6±5.9 cm) from the Chinese national cross-
country skiing team, all of whom were elite level [27]. Although the sample size is small, it is
something that often occurs in real-world studies of elite athletes, which is supported by previ-
ous studies [11, 26].
They spent 4 weeks at the Chinese national snow sports training base (sea level: 1510–1700
m, 44.5˚ N) from 3 May 2021 to 30 May 2021. Athletes follow a regular training plan assigned
by the coaching of the national training team [28, 29], Table 1 shows the training plan example
for reference.

Fig 1. The experimental procedure of this study.

https://doi.org/10.1371/journal.pone.0295960.g001

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 3 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Table 1. Example of a weekly training plan.

Day Morning session Afternoon session
Day Free technique skiing 100 min: warm-up 45 min; Running 30 min; maximal strength training: 4
1 aerobic training (15 seconds sprint + 30 seconds exercises for upper body * 3 sets + 2 exercises for
relax) * 5 times * 3 sets, anaerobic threshold velocity lower body * 4 sets
training; cool down 20 min
Day Lactate Threshold Specialized Training Free technique roller-skiing 100 min; core training
2 30 min; 15–20 minutes stretching
Day Classical technique skiing 120 min Running 45 min; free technique ski-specific exercises
3 30 min; specialized strength training 15 min;
swimming 30 min
Day Free technique roller-skiing 110 min: warm-up 30 Classical technique roller-skiing 80 min; core
4 min; uphill 10 sets, relax for 3 minutes between sets; training 20 min
cool down 20 min
Day Classical technique roller-skiing 150 min Running 30 min; slow strength 3 exercises for upper
5 body * 5 sets, 2 exercises for lower body * 5 sets, 45s
of exercise between sets, 30s intervals
Day Free technique skiing interval Training: warm-up 25 Running 25 min; circuit strength training, 8 exercises
6 min; 5km * 2 sets, 3–4 min relaxed skating between * 2 sets, 30 seconds exercise/30 seconds rest
sets; 10 minutes skating * 2–3 sets, 3–4 minutes
relaxed skating between sets; cool down 20 min
Day Relax Relax
7
https://doi.org/10.1371/journal.pone.0295960.t001

During this period the athletes’ training and life were systematically monitored for training
load. The daily training load data was collated by the scientific staff filled in and summarized
according to the relevant templates, including the collection and calculation of internal and
external load data such as heart rate, sRPE, TRIMP, training time and distance. Before and
after the altitude training, athletes did physiological performance test at the Beijing Erqi
National Ice and Snow Research and Training Base (55m above sea level). The tests include
maximal oxygen up-take tests, lactate threshold tests, body composition tests and skierg power
tests. Athletes were examined by national team doctors to ensure that there were no injuries or
illnesses and were in good physical condition. All participants freely completed an informed
consent form before to this investigation, expressing their desire to voluntarily engage in this
study.

Physiological performance test

In this study the lactate threshold test was completed prior to the maximal oxygen uptake
(VO2max) test with a 5 minutes interval [30–32]. All participants performed an adequate
warm-up prior to before the test. Using the treadmill (RL2500E, Rodby, Södertalje, Sweden) to
evaluate the lactate threshold. The treadmill started at 7 km/h, the inclination angle was 10.5%
and constant throughout, the speed increased by 1km/h every 5 minutes and the athlete has a
30 second rest period before each acceleration [30–32]. The average heart rate for the last 30
seconds of each phase was the athlete’s heart rate for that level. The athlete’s lactate is collected
immediately at each step of the test and measured using a benchtop lactate meter (Boisen, EKF
Industrial Electronics, Magdeburg, Germany). Ask athletes at the end of each step about their
subjective feelings of fatigue (a 0–10 RPE scale). The lactate threshold refers to lactate level of 4
mmol L−1, treadmill speed at 4 mmol L−1 was calculated using linear interpolation [33].
VO2max test by athletes assessed using a portable gas metabolometer (MetaMax 3B, Cortex,
Leipzig, Germany) 5 minutes after the lactate threshold test [30–32]. The treadmill was
inclined at the same angle as the lactate threshold test, with the starting speed 1km/h lower

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 4 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

than the lactate threshold speed and the speed increased by 1km/h per minute until the athlete
was exhausted. During the maximal oxygen uptake test, the athlete continuously wore a venti-
lation mask to measure the athlete’s oxygen uptake. The athlete’s heart rate was record by a
heart rate belt (H10, Polar, Finland), the lactate concentration was measured 1 minute after
the test and the RPE (a 0–10 scale) was recorded. The athlete’s maximum oxygen uptake (the
average of the two highest and consecutive 30 s measurements), maximum HR (the highest 5s
heart rate measurement), final treadmill speed and respiratory exchange ratio (RER) were
recorded.
Athletes did body composition test on the morning of the test day, which used a dualenergy
X-ray bone density analyzer (Luna iDXA, General Electric Company, Schenectady, NY, USA)
to analyze the muscle mass of the athletes’ upper body.
Athletes performed a 30s power test using a ski ergometer (SKIERG, CONCEPT2 Inc.,
USA). The test was preceded by a 25-minute jogging warm-up led by a fitness coach, a 60m x 5
sets of acceleration runs, and a 1 minute x 3 sets of 75% maximum intensity warm-up using a
ski ergometer. The skier was set to gear seven and the athletes performed the 30s ski ergometer
test to the best of their ability. A researcher recorded the athlete’s average power output over
30 seconds and collated the records, the final result is a power to weight ratio.

Methods of training load monitoring

All training load data of the athletes were counted and recorded by the scientific staff accompa-
nying the team. External load such as training distance, time and speed was obtained and
recorded by the Catapult module (Catapult Sports, Melbourne, Australia). The module has a
GPS sensor, which can be connected to the GPS satellite system, with a sampling rate was
10Hz; it can be used for more than 5 hours for continuous training and competition monitor-
ing [34]. Each athlete wore a special undershirt that places the module in the middle of the
shoulder blades. The module was turned on 10 minutes before training to ensure good signal
reception. The same module was used by every athlete during the training camp for standardi-
zation. The data were collected and analyzed after each training session by using the corporate
software (Catapult openfield, Melbourne, Australia).
Athletes wore heart rate bands and watches to monitor heart rate load-related data during
daily training. In this study the athlete’s maximal heart rate and lactate threshold were
obtained through physiological testing, therefore two methods of TRIMP calculation based on
different heart rate zones standard were considered.
The first TRIMP calculation method in this study is the approach eTRIMP proposed by
Edward (Eq 1) [17].
eTRIMP ¼ 1 � T1 þ 2 � T2 þ 3 � T3 þ 4 � T4 þ 5 � T5 ð1Þ

T1: duration time when 50%HRmax <HR <60%HRmax; T2: duration time when 60%
HRmax <HR <70%HRmax; T3: duration time when 70%HRmax <HR <80%HRmax; T4:
duration time when 80%HRmax <HR <90%HRmax; T5: duration time when 90%HRmax
<HR <100%HRmax.
The second TRIMP calculation method is the way luTRIMP proposed by Lucia [18], as
shown in Eq 2.
luTRIMP ¼ 1 � T1 þ 2 � T2 þ 3 � T3 ð2Þ

T1: duration time when <HR1; T2: duration time when HR1 <HR <HR2; T3: duration
time when >HR2. The HR1 corresponding to heart rate when blood lactate is 2 mmol L−1; the
HR2 corresponding to heart rate when blood lactate is 4 mmol L−1.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 5 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

All participants were asked for RPE 30 minutes after the training, which were tallied using a
0–10 subscale and recorded, as shown in Eq 3 [7, 20].

sRPE ¼ RPE � duration time ðminÞ ð3Þ

The rate of the physiological indicators change was calculated by Eq 4.

PostTest PreTest
Physiological Indicators Changes in Percent ð%DÞ ¼ � 100% ð4Þ
PreTest

Statistical analyses
Excel 2019 and IBM SPSS 25.0 software were used to process data. All data were tested for nor-
mality using the Shapiro-Wilk test before processing, and all data were presented as
mean ± standard deviation (SD). Using the paired sample T-test to compare differences in
physiological performance change between two tests, the level of significance was set at
p < 0.05 and the level of high significance was set at p < 0.01. Choen’sd formula was used to
calculate the effect size (ES) of physiological indexes before and after the athlete’s special prep-
aration period. ES values were evaluated according to the following criteria: 0 to 0.19 trivial,
0.20 to 0.59 small, 0.6 to 1.19 moderate, 1.20 to 1.99 large, �2.00 very large. Using Pearson cor-
relation coefficients, the relationships between load quantifiers and physiological indicators
changes were identified. The correlation coefficients’ uncertainties were given as 95% confi-
dence intervals, with p < 0.05 being significant and p < 0.01 being a highly significant differ-
ence. The interpretation of the strength of the correlation coefficients is based on Hopkins: 0
to 0.09trivial, 0.1 to 0.29 small, 0.3 to 0.49 moderate, 0.50 to 0.69 large, 0.70 to 0.89 very large,
0.90 to 0.99 nearly perfect, 1.00 perfect.

Results
The statistical results of the load characteristics of 8 Chinese male cross-country skiers during
training camp are shown in and Figs 2–4. The average weekly endurance training of the 8 elite
Chinese male cross-country skiers was 206.5km, the average weekly training time was 859.7
minutes, the average weekly luTRIMP value was 941.6 AU, the average weekly eTRIMP value
was 1265.7 AU and the average weekly sRPE was 6953.3 AU.
As shown in Table 2, the lactate threshold velocity and heart rate increased after training
(p < 0.05). The RER reduced after training camp (p < 0.05). The upper body muscle mass sig-
nificant increased (p < 0.01) after the training camp. The skierg power-to-weight ratio
increased 0.28 watt�kg-1 after altitude training (p < 0.05).
Table 3 and Fig 5 show relationships between the methods of training load and physiolog-
ical indicators changes in percentage. The correlation between average weekly luTRIMP
and percent change in physiological performance was highest. Percentage changes in lactate
threshold velocity (r = .78 [95% CI -.01 to .98]), percentage changes in lactate threshold HR
(r = .71 [95% CI 0.14–0.99]), percentage changes in maximum HR (r = .83 [95% CI .19–
1.00]), percentage changes in skierg power-to-weight ratio (r = .75 [95% CI -.28 to .98]) had
very large relationships with average weekly luTRIMP. There were very large relationships
between average weekly eTRIMP and maximum HR changes in percentage. Percentage
changes in skierg power-to-weight ratio were very largely related to average weekly sRPE.
There were very large relationships between Total Distance and percentage changes in max-
imum HR. The correlation between Total Time and percent change in upperbody muscle
mass was very large.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 6 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Fig 2. Endurance distance and total training time during training camp.
https://doi.org/10.1371/journal.pone.0295960.g002

Discussion
The purpose of this study were to: (i) analyze the load characteristics of 4 weeks cross-country ski-
ing altitude training.; (ii) analyze the relationships between methods of monitoring training load
and physiological indicators changes of elite Chinese cross-country skiers. The result revealed a "
pyramid " pattern in the load characteristics of 8 Chinese male cross-country skiers during the alti-
tude training cycle. The highest relationships were found in luTRIMP, the results of this study
show that load monitoring methods that combine program characteristics and individual physio-
logical characteristics have the highest relationships with physiological performance change, as
opposed to internal and external load metrics that only combine average exercise intensity.
The essence of sports training is the precise control of the training load, and detailed statis-
tics on the training load can help to understand the athletes’ training situation in a certain
cycle and make timely adjustments accordingly [21]. The load statistics show a " pyramid "
training pattern in terms of overall training load and intensity during this period. Compared
to similar training cycles for world elite cross-country skiers (polarised training pattern),
endurance training time is relatively low at 1.4h (approximately 15.7h for the international
elite level), strength training time is 1.6h higher per week (approximately 0.7h for the interna-
tional elite level) and speed training time is similar. In terms of training load intensity, the
average weekly LIT is 1.2h lower than that of international elite athletes (about 14.4h), the MIT

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 7 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Fig 3. Training distribution during training camp.

https://doi.org/10.1371/journal.pone.0295960.g003

Fig 4. Training intensity during training camp.

https://doi.org/10.1371/journal.pone.0295960.g004

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 8 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Table 2. Physiological performance test result.

Pre-Test Post-Test Mean Difference[95%Cl] ES
Lat velocity(m�s−1) 3.02 ± 0.18 3.27 ± 0.24* 0.24 [0.05–0.44] 1.18
Lat HR (beats�min−1) 179.5 ± 9.8 188.5 ± 9.2** 9.00 [5.08–12.92] 0.95
VO2max (mL�min−1�kg−1) 73.74 ± 3.63 71.12 ± 3.14 -2.61 [-6.12 to 0.90] 0.77
−1)
VO2max (L�min 4.82 ± 0.36 4.64 ± 0.43 -0.17 [-0.33 to 0.01] 0.45
Maximum HR in VO2max Test (beats�min−1) 197.4 ± 11.6 198.9 ± 9.1 1.50 [-3.72 to 6.72] 0.14
RER 1.23 ± 0.09 1.11 ± 0.03** -0.12 [-0.23 to -0.01] 1.79
Maximum La in VO2max Test(mmol�L−1) 13.31 ± 1.42 11.33 ± 2.79 -1.99 [-4.49 to 0.52] 0.89
Muscle mass whole body(kg) 55.66 ± 4.01 56.01 ± 4.02 0.36 [-0.04 to 0.76] 0.09
Muscle mass upper body(kg) 6.54 ± 0.42 6.74 ± 0.36** 0.20 [0.09–0.31] 0.51
Muscle mass trunk(kg) 27.20 ± 2.18 27.47 ± 2.00 0.27 [-0.62 to 1.16] 0.13
Muscle mass lower body(kg) 18.60 ± 1.82 18.57 ± 1.68 -0.03 [-7.34 to 0.68] 0.02
-1
Skierg power-to-weight ratio(watt�kg ) 4.38 ± 0.18 4.66 ± 0.16* 0.28 [-0.08 to 0.47] 1.64

* indicates a significant difference compared to pre-test (

* p < 0.05
** p < 0.01)
Abbreviations: Lat = lactate threshold, VO2max = maximal oxygen uptake, HR = heart rate, RER = respiratory exchange ratio.

https://doi.org/10.1371/journal.pone.0295960.t002

is about 0.3h higher per week (about 0.5h for international elite level) and the HIT is about
0.5h lower per week (about 0.8h for international elite level). The reasons for the discrepancy
with elite international athletes may be closely related to the short years of ski-specific training.
Due to their short training period, athletes have not yet developed good glide economy and are
prone to overrange glide heart rates during the long, low-intensity aerobic glides required by
the training plan [2, 3]. The long duration of strength training is due to the fact that the
national cross-country skiing coach, in response to the lack of specific strength in our athletes,
has designed special strength training after long periods of low intensity aerobic training to
help the athletes to achieve the appropriate strength requirements as much as possible.
Scientific assessment of the relationship between external load, internal load and sports perfor-
mance can help coaches and sports researchers to better examine the dose-response relationships
resulting from sports training and to make a meaningful impact on training [10]. The strongest
dose-response relationship between luTRIMP and physiological performance was found in this
study, where it was shown that internal loads in cross-country skiers were closely related to maxi-
mal oxygen uptake and lactate threshold heart rate, and that changes in these in-ternal load moni-
toring methods were correlated to a moderate to high degree with changes in physiological
performance [35]. Consistent results were also seen in the hurling [36] and soccer players [37].
The correlation between eTRIMP and physiological indicators changes in this study was much
less than luTRIMP. The biggest reason is that eTRIMP was obtained based on data analysis of a
certain group of athletes and not based on individual physiological performance tests, and the
division of heart rate intervals deviated from the sports characteristics of cross-country skiing
[26]. Some researchers have pointed out the limitations of load monitoring tools that do not
incorporate the physiological characteristic points of the sport, as the intensity of an athlete’s train-
ing can change depending on the environment, weather and tactics. In an experiment with
cyclists, researchers found a weak dose-response relationship between bTRIMP and exercise per-
formance based solely on the generalized exercise blood lactate equation [38].
sRPE as a load monitoring indicator has been widely used in the field of sports training [20,
39]. However, the dose-response relationship between sRPE and physiological indicators
changes in this study was still lower than that between heart rate load-related metrics, which is

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 9 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Table 3. Relationship between methods of training load quantification and physiological indicators changes in percentage.
Total Distance Total Time luTRIMP eTRIMP sRPE
%Δ Lat Velocity .49[-.51 to .39] .60[-.13 to .93] .78*[-.01 to .98] .42[-.52 to .86] .62[-.31 to .94]
%Δ Lat HR .44[-.37 to .92] .49[-.28 to .97] .71*[0.14–0.99] .38[-.39 to .91] .52[-.12 to .92]
%Δ VO2maxRl .68[.31-.95] .39[-.52 to .97] .59[-.19 to .96] .61[-.06 to .99] .50[.01-.92]
%Δ VO2maxAb .69[.33-.97] .42[-.54 to .93] .63[-.26 to .91] .57[-.14 to .98] .61[.21-.94]
%Δ HRmax .81*[.36-.96] .61[-.20 to .98] .83*[.19–1.00] .79*[.29-.97] .66[-.24 to .97]
%Δ Skierg W/Kg .62[-.34 to .94] .60[-.10 to .92] .75*[-.28 to .98] .50[-.28 to .86] .72*[-.41 to .96]
%Δ Upperbody Muscle Mass .54[-.13 to .96] .78*[.44-.99] .62[-.11 to .94] .65[-.02 to .95] .53[-.34 to .97]

* Indicates a significant at the 0.05 level (2-tailed)

** indicates significant at the 0.01 level (2-tailed). Abbreviations: %Δ Lat Velocity = percentage changes in lactate threshold velocity, %Δ Lat HR = percentage changes in
lactate threshold HR, %Δ VO2maxRl = percentage changes in maximal oxygen uptake relative values, %Δ VO2maxAb = percentage changes in maximal oxygen uptake
absolute values, %Δ HRmax = percentage changes in maximum HR, %Δ Skierg W/Kg = percentage changes in skierg power-to-weight ratio, %Δ Upperbody Muscle
Mass = percentage changes in upperbody Muscle Mass, sRPE = session rating of perceived exertion, eTRIMP = Edwards training impulse, luTRIMP = Lucia training
impulse.

https://doi.org/10.1371/journal.pone.0295960.t003

consistent with the better relationship between heart rate based load monitoring methods and
oxygen consumption than RPE based methods derived by Wallace et al. [40]. It should also be
considered that the short-term training intervention in this study and the relatively high alti-
tude may have had an additional effect [41, 42]. It should also be considered that cross-country
skiing differs from other sport groups in that the training pattern of prolonged exposure to
cold temperatures may lead to a bias in the athletes’ subjective perception of fatigue, which in
turn may lead to a bias in the RPE values compared to laboratory tests, resulting in an error in
the sRPE monitoring of training load [43]. It should also be noted that load monitoring in this
study was limited to the training period, with evidence suggesting that sRPE monitoring values
during competition better reflect an athlete’s true internal load level. Furthermore, a signifi-
cant difference was found in the psychological state of athletes between training and competi-
tion, with the former being more monotonous and the latter generating stronger emotional
responses, leading to notable psychological differences. Differences in training and competi-
tion intensity can result in varying effects of load quantification [44].
The correlation between load monitoring methods and changes in physiological perfor-
mance in this study, whilst moderate, is far less than has been found in practice in other proj-
ects. Temperatures, wind speeds, snow quality and snow waxes in alpine environments differ
from those in plains, and the accumulation of fatigue due to low oxygen exposure may lead to
biased results. Differences in the duration of physiological performance tests may also contrib-
ute to this result, with some studies showing a stronger effect relationship between longer
duration physiological tests and load quantifiers [9]. Data collection from sports training prac-
tice is more difficult to control than laboratory research design, which inevitably leads to vari-
ability in the data collected [45]. Considering the above factors, the relationships in this study
should be interpreted with caution. Therefore, in future studies that wish to apply load moni-
toring methods to measure or analyze the dose effects of physiological performance in cross-
country skiers in a given cycle, more consideration should be given to the influence of complex
confounding factors on the results [46].

Limitations
As an observational study, this study still has a few flaws in it. Firstly, the sample size in this
study was relatively small due to design limitations, which is usually the norm in real-world

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 10 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Fig 5. Relationship Between Average Weekly luTRIMP and Physiological Indicators Changes in Percent-age(N = 8), (a) %Δ Lat Velocity, (b) %Δ Lat HR, (c) %
Δ HRmax, (d) %Δ Skierg W/Kg. Abbreviations: %Δ Lat Velocity = percentage changes in lactate threshold velocity, %Δ Lat HR = percentage changes in lactate
threshold HR, %Δ HRmax = percentage changes in maxi-mum HR, %Δ Skierg W/Kg = percentage changes in skierg power-to-weight ratio, luTRIMP = Lucia
training impulse.
https://doi.org/10.1371/journal.pone.0295960.g005

sports science research of elite athletes. Additionally, the training period only lasted for 4
weeks, which could limit the study’s ability to fully capture the physiological adaptations to
altitude training, as well as the acute and chronic effects of applied training. Finally, the load
monitoring methods used in this study were very limited due to practical constraints and
more different load monitoring methods should be considered for future analysis. Therefore,
future research should aim to expand the sample size and increase the number of data points
to improve statistical analysis. Linear mixed models could be considered to mitigate the impact
of individual differences on the outcomes [25]. It is important to note that competitive sports
prioritise results, hence more attention should be given to sports performance indicators.

Conclusion
The study revealed a " pyramid " pattern in the load characteristics of 8 Chinese male cross-
country skiers during the 4 weeks altitude training short cycle in preparation for the Beijing

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 11 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Winter Olympics, with a lower duration of LIT and a higher duration of MIT than at the elite
international level. In the process of training on the altitude, it is important to control LIT
while improving the other basic skills of the athletes to avoid fatigue and other phenomena.
Training load monitoring methods that combine with program characteristics and individual
physiological characteristics have the highest dose-response relationships with physiological
indicators changes. Indicating that this is an important factor in cross-country skiing training
load monitoring, luTRIMP could be used as a good internal load monitoring tool in cross-
country skiing. The use of non-invasive load monitoring methods incorporating individual
physiological characteristics of athletes should be considered in future cross-country skiing
training monitoring to reveal changes in athletes’ physiological performance after a particular
training intervention, in order to improve cross-country skiing training monitoring and load
evaluation systems.

Practical applications
Training load monitoring can better help practitioners understand the load characteristics
during a cycle, especially in specific training environments. For instance, when coaches iden-
tify that athletes are experiencing high loads based on the monitoring outcomes, they should
timely modify the training programme to avoid overreaching or overtraining. It is also valu-
able for practitioners to strengthen the dose-respond relationship between load quantification
methods and physiological indicators. This can enable the practitioners to understand the
physiological adaptations that will occur in the athlete through training loads alone. Although
the relationship can only be established for a specific group of athletes, providing an evidence-
based framework can aid in the development of training programs. Moreover, this relationship
might be impacted due to cold weather in winter sports. Practitioners ought to choose load
monitoring methods that are specific to the training environment and program characteristics.
Future research should consider better statistical methods based on more data to evaluate the
dose-response relationship between longer training and more physiological indicators in
cross-country skiing.

Acknowledgments
Thanks to the athletes who participated in the study and the Chinese National Olympic
Committee.

Author Contributions
Conceptualization: Yifan Lu, Jing Mi.
Data curation: Yichao Yu, Dongye Li.
Formal analysis: Yifan Lu, Jing Mi.
Funding acquisition: Yifan Lu.
Investigation: Yichao Yu, Dongye Li.
Methodology: Yifan Lu, Jing Mi.
Project administration: Yichao Yu, Dongye Li, Yifan Lu.
Resources: Yichao Yu, Dongye Li, Yifan Lu.
Software: Yichao Yu, Dongye Li.
Supervision: Jing Mi.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 12 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

Writing – original draft: Yichao Yu, Dongye Li.

Writing – review & editing: Yichao Yu, Dongye Li, Yifan Lu, Jing Mi.

References
1. Sandbakk O, Hegge AM, Losnegard T, Skattebo O, Tonnessen E, Holmberg HC. The Physiological
Capacity of the World’s Highest Ranked Female Cross-country Skiers. Med Sci Sports Exerc. 2016; 48
(6):1091–100. Epub 2016-6-1. https://doi.org/10.1249/MSS.0000000000000862 PMID: 26741124.
2. Sandbakk O, Holmberg HC. Physiological Capacity and Training Routines of Elite Cross-Country Ski-
ers: Approaching the Upper Limits of Human Endurance. Int J Sports Physiol Perform. 2017; 12
(8):1003–1011. Epub 2017-9-1. https://doi.org/10.1123/ijspp.2016-0749 PMID: 28095083.
3. Sandbakk O, Holmberg HC, Leirdal S, Ettema G. The physiology of world-class sprint skiers. Scand J
Med Sci Sports. 2011; 21(6):e9–16. Epub 2011-12-1. https://doi.org/10.1111/j.1600-0838.2010.01117.
x PMID: 20500558.
4. Lianshi F. The Functional Diagnosis Methods in El ite Athlete and Problems. Shanghai Sport Science
Research. 2003;(03):49–54. PubMed.
5. Gibson M, Pettitt R. Science and Practice of Strength Training. J Athl Train. 1998;33Epub 1998-1-1.
PubMed.
6. Chandler J, Brown L. Conditioning for Strength and Human Performance, 3rd Edition. 2019.
7. Foster C, Rodriguez-Marroyo JA, de Koning JJ. Monitoring Training Loads: The Past, the Present, and
the Future. Int J Sports Physiol Perform. 2017; 12(Suppl 2):S22–S28. Epub 2017-4-1. https://doi.org/
10.1123/ijspp.2016-0388 PMID: 28253038.
8. Holmberg HC. The elite cross-country skier provides unique insights into human exercise physiology.
Scand J Med Sci Sports. 2015;25 Suppl 4:100–9. Epub 2015-12-1. https://doi.org/10.1111/sms.12601
PMID: 26589123.
9. Meeusen R, Duclos M, Foster C, Fry A, Gleeson M, Nieman D, et al. Prevention, diagnosis, and treat-
ment of the overtraining syndrome: joint consensus statement of the European College of Sport Science
and the American College of Sports Medicine. Med Sci Sports Exerc. 2013; 45(1):186–205. Epub 2013-
1-1. https://doi.org/10.1249/MSS.0b013e318279a10a PMID: 23247672.
10. Mike RM, Mike M. Monitoring Training and Performance in Athletes: Human Kinetics, Inc.;Human
Kinetics. 2017.
11. Xudan C, Lijuan M, Bei Z, Yongming L, Xiaoping C. Effects of Three Weeks 1 550 m Moderate Altitude
Training on Physiological Capacity and Body Composition of High Level Young Male Cross-Country
Skiers. CHINA SPORT SCIENCE AND TECHNOLOGY. 2022; 58(01):30–37. https://doi.org/10.16470/
j.csst.2021151 PubMed.
12. Gabbett TJ, Whyte DG, Hartwig TB, Wescombe H, Naughton GA. The relationship between workloads,
physical performance, injury and illness in adolescent male football players. Sports Med. 2014; 44
(7):989–1003. Epub 2014-7-1. https://doi.org/10.1007/s40279-014-0179-5 PMID: 24715614.
13. Mujika I. The alphabet of sport science research starts with Q. Int J Sports Physiol Perform. 2013; 8
(5):465–6. Epub 2013-9-1. https://doi.org/10.1123/ijspp.8.5.465 PMID: 24058942.
14. Mujika I. Quantification of Training and Competition Loads in Endurance Sports: Methods and Applica-
tions. Int J Sports Physiol Perform. 2017; 12(Suppl 2):S29–S217. Epub 2017-4-1. https://doi.org/10.
1123/ijspp.2016-0403 PMID: 27918666.
15. Talsnes RK, Berdal T, Brattebø JM, Seeberg T, Losnegard T, Kocbach J, et al. Comprehensive analy-
sis of performance, physiological, and perceptual responses during an entire sprint cross-country skiing
competition. Eur J Appl Physiol. 2023; Epub 2023-10-7. https://doi.org/10.1007/s00421-023-05326-w
PMID: 37804364.
16. Banister EW, Carter JB, Zarkadas PC. Training theory and taper: validation in triathlon athletes. Eur J
Appl Physiol Occup Physiol. 1999; 79(2):182–91. Epub 1999-1-1. https://doi.org/10.1007/
s004210050493 PMID: 10029340.
17. Edwards S. The Heart Rate Monitor Book; 1993.
18. Lucia A, Hoyos J, Santalla A, Earnest C, Chicharro JL. Tour de France versus Vuelta a España: which
is harder? Med Sci Sports Exerc. 2003; 35(5):872–8. Epub 2003-5-1. https://doi.org/10.1249/01.MSS.
0000064999.82036.B4 PMID: 12750600.
19. Manzi V, Iellamo F, Impellizzeri F, D’Ottavio S, Castagna C. Relation between individualized training
impulses and performance in distance runners. Med Sci Sports Exerc. 2009; 41(11):2090–6. Epub
2009-11-1. https://doi.org/10.1249/MSS.0b013e3181a6a959 PMID: 19812506.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 13 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

20. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, et al. A new approach to monitor-
ing exercise training. J Strength Cond Res. 2001; 15(1):109–15. Epub 2001-2-1. PMID: 11708692.
21. Halson SL. Monitoring training load to understand fatigue in athletes. Sports Med. 2014; 44 Suppl 2
(Suppl 2): S139–47. Epub 2014-11-1. https://doi.org/10.1007/s40279-014-0253-z PMID: 25200666.
22. Talsnes RK, Solli GS, Kocbach J, Torvik PØ, Sandbakk Ø. Laboratory- and field-based performance-
predictions in cross-country skiing and roller-skiing. PLoS One. 2021; 16(8):e0256662. Epub 2021-1-
20. https://doi.org/10.1371/journal.pone.0256662 PMID: 34428258.
23. Akubat I. TRAINING LOAD MONITORING IN SOCCER. 2014.
24. Oliveira R, Canário-Lemos R, Peixoto R, Vilaça-Alves J, Morgans R, Brito JP. The relationship between
wellness and training and match load in professional male soccer players. PLoS One. 2023; 18(7):
e0289374. Epub 2023-1-20. https://doi.org/10.1371/journal.pone.0289374 PMID: 37523395.
25. Costa J, Figueiredo P, Nakamura F, Rago V, Rebelo A, Brito J. Intra-individual variability of sleep and
nocturnal cardiac autonomic activity in elite female soccer players during an international tournament.
PLoS One. 2019; 14(9):e0218635. Epub 2019-1-20. https://doi.org/10.1371/journal.pone.0218635
PMID: 31527865.
26. Sanders D, Abt G, Hesselink M, Myers T, Akubat I. Methods of Monitoring Training Load and Their
Relationships to Changes in Fitness and Performance in Competitive Road Cyclists. Int J Sports Phy-
siol Perform. 2017; 12(5):668–675. Epub 2017-5-1. https://doi.org/10.1123/ijspp.2016-0454 PMID:
28095061.
27. McKay A, Stellingwerff T, Smith ES, Martin DT, Mujika I, Goosey-Tolfrey VL, et al. Defining Training
and Performance Caliber: A Participant Classification Framework. Int J Sports Physiol Perform. 2022;
17(2):317–331. Epub 2022-2-1. https://doi.org/10.1123/ijspp.2021-0451 PMID: 34965513.
28. Grushin AA, Nageykina SV. Elite female cross-Country skiers1 training for major international sport
competitions in mid-altitude areas. 2016:66–69. Epub 2016-1-1. PubMed.
29. Tonnessen E, Sylta O, Haugen TA, Hem E, Svendsen IS, Seiler S. The road to gold: training and peak-
ing characteristics in the year prior to a gold medal endurance performance. PLoS One. 2014; 9(7):
e101796. Epub 2014-1-20. https://doi.org/10.1371/journal.pone.0101796 PMID: 25019608.
30. Talsnes RK, Hetland TA, Cai X, Sandbakk O. Development of Performance, Physiological and Techni-
cal Capacities During a Six-Month Cross-Country Skiing Talent Transfer Program in Endurance Ath-
letes. Front Sports Act Living. 2020; 2:103. Epub 2020-1-20. https://doi.org/10.3389/fspor.2020.00103
PMID: 33345092.
31. Xudan C, Lijuan M, Bei Z, Kjøsen TR, SANDBAKK Ø, Tor-Arne H, et al. The Development of Physical
Capacity of Talent-Transferring Athletes in Long Term Cross-Country Skiing Training—Based on
Sports Physiological Evaluations. CHINA SPORT SCIENCE. 2021; 41(08):3–13. Epub 2021-8-15.
PubMed.
32. Yu Y, Wang R, Li D, Lu Y. Monitoring Physiological Performance over 4 Weeks Moderate Altitude Train-
ing in Elite Chinese Cross-Country Skiers: An Observational Study. Int J Environ Res Public Health.
2022;20(1)Epub 2022-12-24. https://doi.org/10.3390/ijerph20010266 PMID: 36612586.
33. Sjodin B, Jacobs I, Svedenhag J. Changes in onset of blood lactate accumulation (OBLA) and muscle
enzymes after training at OBLA. Eur J Appl Physiol Occup Physiol. 1982; 49(1):45–57. Epub 1982-1-
19. https://doi.org/10.1007/BF00428962 PMID: 6213407
34. Parmley J, Jones B, Whitehead S, Rennie G, Hendricks S, Johnston R, et al. The speed and accelera-
tion of the ball carrier and tackler into contact during front-on tackles in rugby league. J Sports Sci.
2023:1–9. Epub 2023-11-5. https://doi.org/10.1080/02640414.2023.2273657 PMID: 37925647.
35. Mourot L, Fabre N, Andersson E, Willis S, Buchheit M, Holmberg HC. Cross-country skiing and postex-
ercise heart-rate recovery. Int J Sports Physiol Perform. 2015; 10(1):11–6. Epub 2015-1-1. https://doi.
org/10.1123/ijspp.2013-0445 PMID: 24806737.
36. Malone S, Hughes B, Collins K, Akubat I. Methods of Monitoring Training Load and Their Association
With Changes Across Fitness Measures in Hurling Players. J Strength Cond Res. 2020; 34(1):225–
234. Epub 2020-1-1. https://doi.org/10.1519/JSC.0000000000002655 PMID: 29985218.
37. Akubat I, Patel E, Barrett S, Abt G. Methods of monitoring the training and match load and their relation-
ship to changes in fitness in professional youth soccer players. J Sports Sci. 2012; 30(14):1473–80.
Epub 2012-1-20. https://doi.org/10.1080/02640414.2012.712711 PMID: 22857397.
38. Borresen J, Lambert MI. The quantification of training load, the training response and the effect on per-
formance. Sports Med. 2009; 39(9):779–95. Epub 2009-1-20. https://doi.org/10.2165/11317780-
000000000-00000 PMID: 19691366.
39. Haddad M, Stylianides G, Djaoui L, Dellal A, Chamari K. Session-RPE Method for Training Load Moni-
toring: Validity, Ecological Usefulness, and Influencing Factors. Front Neurosci. 2017; 11:612. Epub
2017-1-20. https://doi.org/10.3389/fnins.2017.00612 PMID: 29163016.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 14 / 15

PLOS ONE Relationship between methods of monitoring training load and physiological indicators changes

40. Wallace LK, Slattery KM, Impellizzeri FM, Coutts AJ. Establishing the criterion validity and reliability of
common methods for quantifying training load. J Strength Cond Res. 2014; 28(8):2330–7. Epub 2014-
8-1. https://doi.org/10.1519/JSC.0000000000000416 PMID: 24662229.
41. Garvican LA, Pottgiesser T, Martin DT, Schumacher YO, Barras M, Gore CJ. The contribution of hae-
moglobin mass to increases in cycling performance induced by simulated LHTL. Eur J Appl Physiol.
2011; 111(6):1089–101. Epub 2011-6-1. https://doi.org/10.1007/s00421-010-1732-z PMID: 21113616.
42. Pinot J, Grappe F. A six-year monitoring case study of a top-10 cycling Grand Tour finisher. J Sports
Sci. 2015; 33(9):907–14. Epub 2015-1-20. https://doi.org/10.1080/02640414.2014.969296 PMID:
25357188.
43. Morton RH, Fitz-Clarke JR, Banister EW. Modeling human performance in running. J Appl Physiol
(1985). 1990; 69(3):1171–7. Epub 1990-9-1. https://doi.org/10.1152/jappl.1990.69.3.1171 PMID:
2246166
44. van Erp T, Foster C, de Koning JJ. Relationship Between Various Training-Load Measures in Elite
Cyclists During Training, Road Races, and Time Trials. Int J Sports Physiol Perform. 2019; 14(4):493–
500. Epub 2019-4-1. https://doi.org/10.1123/ijspp.2017-0722 PMID: 30300025.
45. Hoogeveen AR. The effect of endurance training on the ventilatory response to exercise in elite cyclists.
Eur J Appl Physiol. 2000; 82(1–2):45–51. Epub 2000-5-1. https://doi.org/10.1007/s004210050650
PMID: 10879442.
46. Vermeire KM, Van de Casteele F, Gosseries M, Bourgois JG, Ghijs M, Boone J. The Influence of Differ-
ent Training Load Quantification Methods on the Fitness-Fatigue Model. Int J Sports Physiol Perform.
2021; 16(9):1261–1269. Epub 2021-9-1. https://doi.org/10.1123/ijspp.2020-0662 PMID: 33691278.

PLOS ONE | https://doi.org/10.1371/journal.pone.0295960 December 15, 2023 15 / 15