Next Article in Journal
Preparation and Characterization of Soluble Dextrin Fibre from Potato Starch Obtained on a Semi-Industrial Scale
Previous Article in Journal
Small-Signal Stability Analysis and Improvement in Multi-Converter Grid-Tied System Based on Gerschgorin Disc Theorem
Previous Article in Special Issue
Editorial: Advances in Wearable Devices for Sports
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 14
Issue 4
10.3390/app14041437
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Effects of a Simulated Duathlon on Trunk Motion: A Sensor Based Approach

by
Stuart Evans
1,* and
Daniel Arthur James
2
1
School of Education, La Trobe University, Bundoora, VIC 3086, Australia
2
SABEL Labs, Brisbane, QLD 4000, Australia
*
Author to whom correspondence should be addressed.
Appl. Sci. 2024, 14(4), 1437; https://doi.org/10.3390/app14041437
Submission received: 6 January 2024 / Revised: 3 February 2024 / Accepted: 6 February 2024 / Published: 9 February 2024
(This article belongs to the Special Issue Advances in Wearable Devices for Sports)
Browse Figures
Versions Notes

Abstract

:
Duathlon consists of two durations of running separated by cycling in a format similar to triathlon. The addition of cycling and the associated loadings on the neuromuscular system can modify spatiotemporal variables in running including trunk motion, which can impact running economy. Changes to trunk motion can be inferred by measuring accelerations of the centre of mass (CoM). However, there is scarce research into trunk dynamics in duathlon. Therefore, the aim of this study was to use an inertial sensor (an accelerometer) to compare acceleration magnitudes of the trunk in the vertical, mediolateral, and anteroposterior directions during a simulated field-based duathlon. Specifically, running performance and magnitudes of trunk acceleration were compared pre and post a cycling load. Ten well-trained duathletes (seven males, three females (mean ± SD; age: 31.1 ± 3.4 years; body mass: 70.9 ± 6.9 kg; body height: 177 ± 5.82 cm; 9.45 ± 1.7 weekly training hours per week; 9.15 ± 5.2 years training experience)) completed a 5 km run performed at a self-selected pace (described as moderate intensity) prior to 20 km of continuous cycling at four varied cadence conditions. This was immediately followed by a 2.5 km run. Mean completion times for the final 2.5 km in running pre-cycling (4.03:05 ± 0.018) compared to the 2.5 km in running post-cycling (4.08:16 ± 0.024) were significantly different. Regarding trunk acceleration, the largest difference was seen in the vertical direction (y axis) as greater magnitudes of acceleration occurred during the initial 1 km of running post-cycling combined with overall significant alterations in acceleration between running pre- and post-cycling (p = 0.0093). The influence of prior cycling on trunk acceleration activity in running likely indicates that greater vertical and mediolateral trunk motion contributes to decremental running performance. In future, further advanced simulation and analysis could be performed in ecologically valid contexts whereby multiple accelerometers might be used to model a more complete set of dynamics.

1. Introduction

A duathlon is a multisport event that consists of running and cycling followed by another bout of running that is performed sequentially with no recovery between each discipline. Investigating the compounding impacts of multisport events on athletic performance involves complex experimental designs with transitions from one modality to another. In this regard, duathlon shares similarities with triathlon, but with the swimming leg swapped for an additional running section. Akin to triathlon, the amplifying effect of one activity loaded before another remains a prevalent area of exploration [1,2]. A common configuration in duathlon is the sprint distance duathlon that consists of a 5 km run, then a 20 km cycle, followed by a 2.5 km run [3]. Duathlon is internationally governed by Word Triathlon (WT), formerly the International Triathlon Union (ITU), which consents to global duathlon competitive events. Like triathlon, the most limiting (detrimental) performance effect commonly occurs once an athlete has dismounted the bicycle to commence the final running segment [4]. Nevertheless, while duathletes are known to experience similar performance detriments when running after cycling, there is an absence of research that has examined this phenomenon in the field, that is, in the athlete’s natural and familiar environment.
The trunk muscles are involved in both running and cycling. The trunk protects the spinal cord and supports the weight of the trunk, transferring it to the pelvis and lower limbs. Here, the trunk and its theoretical centre of mass (CoM) function as part of a cohesive kinetic chain as athletes grapple with the difficulties of switching between different yet equally demanding (e.g., from running to cycling, and then running). While some researchers remain active in the exploration of trunk kinematics in triathlon [5,6,7], somewhat surprisingly, given the importance of trunk stability and trunk control in both running and cycling, a scarcity of research exists when duathlon is considered. A study by Weich et al. [8] noted that cycling appears to have an adverse effect on subsequent running performance. The literature is replete with examples of how cycling modifies both kinematics and kinetics of running immediately after cycling. These modifications, specifically detrimental performance decrements, include alterations in neuromuscular [9], physiological [10], and biomechanical [11,12] parameters of running immediately followed by cycling with significant changes to athlete ground reaction forces detected when running after cycling [13]. Some researchers have found a decrease in stride length and step cadence [4]. The kinematic and kinetic modifications that have been conveyed in running post-cycling are realistic, given the onset of neuromuscular fatigue that occurs during cycling and prior to running. Because of likely alterations to neuronal and neuromuscular firing patterns in athletes running after cycling, a period of adaptation to running is needed. This adaptation of the motor program means that a delay occurs before the instantaneous switch to running [14]. In this instance, a high-intensity run may cause substantial muscle fatigue [15], which could affect subsequent cycling and running performances to a greater extent when compared with triathlon.
Established biomechanical methods, such as the evaluation of step-characteristics [16] or lower limb range of motion [17], often lack the capability to quantitatively discriminate between subtle running differences. Consequently, such approaches run the risk of discounting fundamental gait-related information [18]. To that end, various biomechanical measures have been adapted to develop and enhance methodologies. Here, running economy, which is commonly considered as a measure of optimal running performance, is subjective to different interpretations. For instance, as Barnes and Kilding [19] noted, running economy is an intricate construct that implies the sum of various metabolic, cardiorespiratory, biomechanical, and neuromuscular characteristics during submaximal running.
When the trunk is considered, the addition of cycling and the associated loadings on the neuromuscular system can modify spatiotemporal variables in running, including trunk gradient and trunk posture, which can impact running economy. Thus, from a performance optimisation perspective, a continuing area of focus for coaches and scientists is determining how to reduce these biomechanical complications. The challenge is to improve sporting performance. Yet, wearable technology can allow applied field-based data collection, which has greater ecological validity than traditional laboratory-based methods. Developments in wearable sensor technology (wearables) have allowed for the detection of triaxial movement that is capable of valid measurement in the field. Wearables play an instrumental role in sports performance and justify the adoption of more unobtrusive and portable devices that can be used in characteristic training scenarios [20,21]. Technological developments have empowered athletes and coaches to track functional movements, workload, biomechanical, and biovital markers that utilise wearable sensors to maximize performance [22]. While wearables, specifically accelerometer usage, have increased, there have been few studies (e.g., [3,9]) examining the use of these proprietary measures in duathlon as an alternative to laboratory-collected data.
In duathlon, much of the athlete’s motion in both running and cycling occurs in the sagittal plane. Therefore, if compensatory, or unwanted or excessive, movement of additional body segments occur in the frontal or transverse plane due to increased postural sway, a performance decrement is probable. Inconspicuous wearable devices such as triaxial accelerometers can be applied in the field to record variations to trunk motion. This applied method is advantageous due to data being largely more reflective of an athlete’s representative cycling and running performance. This approach may be a better reflection of an athlete’s true and self-determined kinematics [12]. In a duathlon context, knowledge, and a greater understanding of trunk accelerations, that is, using triaxial accelerometers in an applied and representative outdoor setting, is beneficial for two reasons. Firstly, the obtained data allow for quantification in a more realistic setting compared to the typical laboratory environment. Secondly, obtainment of data in a simulated duathlon may help inform training interventions in the three sequential disciplines in what remains a relatively underresearched sport compared to triathlon. Accordingly, our objective was to compare the magnitude of trunk acceleration between a single run prior to cycling and the final run of a duathlon and to compare overall performance times in running pre- and post-cycling in duathlon. We hypothesized that (1) the magnitude of trunk acceleration would increase during the initial bout of the duathlon, that is, running after cycling, and that (2) running performance time after cycling (mm:ss) would decrease compared to running prior to cycling.

2. Methods

2.1. Participants

Ten well-trained amateur duathletes (seven males, three females (mean ± SD; age: 31.1 ± 3.4 years; body mass: 70.9 ± 6.9 kg; body height: 177 ± 5.82 cm; training experience: 9.4 ± 1.7 years)) participated in this study (Table 1). Participants were required to be over 18 years at the time of the study. At the time that this study was conducted, the duathletes were approximately three weeks out from their initial competitive event.
The athletes were notified that they could withdraw from the study at any time without penalty. All participants participated in recreational duathlon events that were permitted by AUS Triathlon and all were members of AUS Triathlon and a registered triathlon organisation. All participants were notified of the methodological protocol and requirements of the study, with all requested to provide informed consent before completing a brief overview of their respective medical history via a Par-Q (physical activity readiness questionnaire) and prescreening injury checklist. Participants were excluded from the study if they had any injury (acute or chronic) or illness at the time of the study that prevented them from exerting maximum effort. The participants were informed that they could withdraw from the study at any time without penalty. No participant was deemed to have an injury (acute, chronic) or disorder (illness) that would have otherwise prohibited them from participating. Ethical concerns were granted by the Institutional Review Board (HREC21114). All participants were free to withdraw from the study at any time and without penalty.

2.2. Overview

All athletes completed the simulated duathlon in one setting (Figure 1). Specifically, the athletes were required to replicate the demands of a sprint distance duathlon, that is, perform a 5 km run prior to cycling (baseline, RunPrior) that was immediately followed by a 20 km cycle, performed at varied cadence, which was then followed by a subsequent 2.5 km (RunPost). These distances are based on regulation competition as stipulated by WT. In both RunPrior and RunPost, the athletes were asked to run at a self-selected pace to ensure consistency and to limit the detrimental effect of fatigue. Along this line, the athletes were asked to abstain from intense training 24 h prior to the test and were requested to refrain from consuming caffeinated beverages six hours prior to the start of the session. Approximately 30 min prior to participants completing a self-directed warm up, the principal author obtained anthropometric details, including height and body mass. Height and body weight (mass) measurements were obtained, with participants asked to stand barefoot to ensure accuracy. Measurements were taken to the closest 0.1 cm and 0.1 kg, respectively [23].
Athlete data were collected under similar environmental conditions (approximately 13–14 °C, 48% relative humidity) as all athletes commenced the initial 5 km run (RunPrior) between 0700 and 0800. These times were intentionally selected due to the overground circuit being free from interference (i.e., vehicles, traffic, owing to the circuit being open to the public). Each athlete commenced the simulated duathlon at individual five-minute intervals (e.g., athlete one commenced at 0700, athlete two commenced at 0705) to avoid the effect of direct competition and to avoid congestion. The start and end points for RunPrior, cycling, and RunPost are shown in Figure 2. The designated start and end point also signified the location where the athletes changed cadence. The same start and end point marked the completion of one 5 km lap, which concurrently marked the start of the next lap (i.e., 5 km). The same overground running and cycle circuit was intentionally designated to avoid increased or preventable braking performance [12], as is common in the bicycle segment of a duathlon. Moreover, the overground circuit was commonly used by the athletes for training purposes and is common within the local duathlon and triathlon community owing to the low technical difficulty required to navigate the circuit at relatively high speed (i.e., lack of gradient and sharp turns). Accordingly, as the circuit was frequently used in training and performance contexts, it allowed for the appropriate evaluation of the athletes in a familiar and realistic performance setting. As the circuit was a loop, the athletes commenced and completed both cycling and running at the marked start line outlined in Figure 2. Consequently, total displacement equated to zero in accordance with Equation (1):
Δx = xf − x0
where xf is final position and x0 is initial position

2.3. Cycling

The athletes used their own time trial bicycles (i.e., a triathlete bicycle) to eliminate the effects of unfamiliarity [12]. The athletes cycled in their familiar aerodynamic position utilizing the aerodynamic bar (i.e., aerobars) that were incorporated into the frame of all the participant bicycles. As is typical in training environments, all duathletes were able to assume their familiar aerodynamic position (defined as elbows on the pads of the aero handlebars with elbow angle close to 90° and the upper part of the trunk parallel to the ground) during the 20 km cycling component. By adopting an aerodynamic position, or a streamlined position, the trunk is placed into a greater horizontal position. Furthermore, the aerodynamic position increases dependence on using the integrated gearing shifters placed at the end of the aerodynamic bars. Consequently, and akin to triathletes cycling, the duathletes are more likely to “shift up or shift down” to a lesser or greater gearing ratio to achieve the desired cadence to maintain cycling performance. Tire pressured was standardized to 110 pounds per square inch (psi). No modifications were made to the athletes’ personal bicycle settings. This is consistent with prior research (e.g., [12,24]) due to any alternations to bicycle settings leading to conceivable deviations to the length–tension relationship, muscle recruitment patterns, and general efficiency and effectiveness of the athletes’ pedal stroke. Measurements of bicycle settings, that is, bicycle seat height and seat tube angle, were made using a standard tape measure by the principal author. Further to recommendations by [25], seat (saddle) height was determined from the centre of the pedal axle to the topmost position of the bicycle saddle, with the pedal placed at the most distal end. These measurements did not restrict the participants’ movement when cycling, that is, participants were free to vary pedalling technique.
To establish the influence of prior cycling on running, Chapman et al. [26] devised a moderate-intensity protocol directed at lessening the influence of fatigue. This variable-cadence protocol was used to identify the effect of changes to neuromuscular control and economy [9], and muscle recruitment patterns [27] within running after cycling. Therefore, we believed that this protocol was satisfactory due to the duathletes’ familiarity with the cadences specified by Chapman et al. [26] and that cadence was appropriate owing to its ease of measurement. Thus, cadence, and not power, was selected due to its simplicity of measurement and, importantly, that all participants in the current study have previously used cadence as a measure of cycling speed. Cadence in cycling is defined as the number of revolutions per minute (rev/min1) that is completed at a relative speed. The power produced by the participant during cycling is the product of torque (force on the pedal) multiplied by angular velocity (pedal speed). Therefore, an increase or decrease in cycling cadence impacts the power that is produced on the bicycle. All the bicycles used by the participants included a speedometer that displayed the relative rev/min1.
To ensure consistency and to limit unintended bias, each speedometer (i.e., 10 in total) was manually calibrated by the principal author. Here, calibration was performed via the roll-out distance method, defined as the distance that the bike travels in a straight line through one revolution of the pedal cycle when in the major gear ratio. As in prior studies [12,24], changes to cadence were delivered verbally by the principal author. Preceding the start of the simulated duathlon session, the participants performed a self-selected warm-up of approximately 10 min that consisted of dynamic stretches and moderate intensity, short-duration runs. The cadence protocol (Figure 1) was the same for all athletes, with no further instructions provided. Due to the often-implemented rule of non-drafting during cycling in amateur duathlon events, pacing and cadence in cycling was monitored by the principal researcher as the effects of drafting may have caused unintentional bias.

2.4. Perceived Exertion

Perceived exertion was verbally reported by the athletes at the conclusion of the 5 km (RunPrior), after each 5 km bicycle lap (i.e., when the front wheel of the bicycle crossed the white line as defined in Figure 2), and immediately after the 2.5 km run (RunPost). To evaluate participant exertion, the Borg ratings of perceived exertion scale (RPE) 6–20 [28] was used. The Borg RPE scale is a popular method for inferring a relative scale of perceived exertion and is generally classified from 6 (“no exertion at all”) to 20 (“maximal exertion”). The participants were requested to keep within a scaling range classified as moderate intensity/somewhat hard (RPE 13–14). All participants were familiar with RPE usage and the numbering conventions used. To monitor RPE levels, the participants were verbally requested to provide their perceived exertion scale at the conclusion of each 5 km cycle lap and at each 1 km of running.

2.5. Transitions (T1 and T2)

The two transition points, namely, transition one (T1), being the transition from the 5 km run (RunPrior) to cycling, and transition two (T2), being from cycling to RunPost, were timed. T1 involved the athletes removing their running shoes, and putting on their cycle shoes (i.e., clipless pedal-facilitated shoes) along with their bicycle helmet before removing their bicycle from the rack stand. T2 involved the athletes “racking” their bicycle on the rack stand before removing their helmet and cycling shoes before putting on their running shoes and commencing the 2.5 km run (RunPost) The transition time, that is, the time of entry to exit of the transition area, was recorded by the principal author.

2.6. Instrumentation and Measurement

The magnitude of trunk acceleration was collected from each participant in the 5 km RunPrior, 20 km of cycling, and 2.5 km in RunPost. Therefore, the ensuing data output was reflective of each participant’s trunk stability and control, or lack thereof, during the simulated duathlon. Accordingly, a reduced magnitude of trunk acceleration equates to a lesser magnitude of unwanted movement. Participant performances during the simulated duathlon by way of acceleration magnitudes of the trunk were obtained by an inertial measurement unit (IMU, a triaxial accelerometer). The accelerometer was placed in a waterproof, airtight, vacuum-sealed bag (Foodsaver, Brampton, ON, Canada) before being attached to each participant’s spinous process to align with the lumbar five (L5) sacrum one (S1) position [12]. From here, the principal author used double-sided adhesive tape to fix the accelerometer to each participant to diminish possible undesirable movement that may arise [24]. Alignment of the accelerometer’s axis was based on the attainment of trunk acceleration in three orthogonal directions, defined as vertical (y, upward–downward), anteroposterior (z, forward–backward), and mediolateral (x, side to side) (where 1 G is equal to 9.81 m/s2). The location of the spinous process was selected due to the point of equal distribution of the vectors (e.g., vector summation equals zero) [24]. Practically, where the vertical and mediolateral axes would be placed in different orientation for the cycling and running components, the accelerometer’s orientation remained comparable, or relative, to each participant’s gross motion.
An ActiGraph GT9X + accelerometer (ActiGraph, LLC., Pensacola, FL, USA) was used to quantity the trunk acceleration magnitude (Figure 3). The ActiGraph GT9X (3.5 × 3.5 × 1 cm, 14 g) was initialized according to manufacturer instructions to record accelerations at a sample rate of 100 Hz. This procedure permitted the control of all accelerometers during data capture. Participant information (i.e., height, weight, ethnicity, age) was entered into the ActiLife software program (version 6.13.4, ActiGraph, LLC.) with an accelerometer assigned to the relevant person.

2.7. Data Analysis

Raw data output from the accelerometers were downloaded and subsequently changed from gt3x files to a CSV format before all data were exported to Microsoft Excel (Version 2311, build 16.0.17029.20028). During the simulated duathlon, data were captured continuously from the 5 km (RunPrior), 20 km cycling, and the 2.5 km (RunPost) inclusive of the T1 and T2 transitions. However, data from individual warm-ups were excluded. Additional variables that are conventionally used in running and cycling performance (e.g., time travelled per kilometre (in km/m1)) were evaluated along with the cadence changes during the 20 km cycling condition. The vector (resultant) acceleration magnitude in RunPrior, cycling, and RunPost was calculated based on Equation (2):
a s   √ ( x 2 + y 2 + z 2 )
Here: vertical (y, upward–downward), anteroposterior (z, forward–backward), and mediolateral (x, side to side).
The accelerometer was manually synchronised by the principal author as each participant cycled past the start line to represent the conclusion of one completed 5 km lap and subsequent change of cadence (see Figure 2). Each completed 5 km lap of cycling along with each 1 km and 500 m split of running was recorded using a conventional stopwatch (Sportline 240 Econosport, New York, NY, USA). To limit possible bias in cycling, standardized cycle shoes (i.e., Shimano SPD-SL clipless shoes) along standardised yellow cleats (tolerance level of almost 6° flotation and tension) were used further to that demonstrated by Evans et al. [12]. Along this line, participants cycled in Northwave tri-sonic cycling shoes (Northwave, Via Levada, Pederobba TV, Italy). Similarly, foot placement during cycling was also standardised to place the head of the first metatarsal directly above the pedal spindle. Here, the foot was located laterally, that is, in the middle of the pedal, defined as the fore–aft position [24,29]. All participants wore a standard tightfitting synthetic triathlon racing “suit” throughout the simulated duathlon.

2.8. Statistical Analysis

All statistical analysis was completed using the software Analyze-it (version 4.92, Leeds, UK). In the preprocessing phase, any samples that had an active time of 0 m/s2 (0 G) were removed. The variables followed a normal distribution, which was analysed using a Kolmogorov–Smirnov goodness-of-fit test. Therefore, parametric methods were used for the hypothesis testing. The raw acceleration data attained by the accelerometer, that is, the magnitude of trunk acceleration in the vertical, mediolateral, and anteroposterior directions in RunPrior, the 20 km of cycling, and RunPost, were analysed using a two-way repeated-measures ANOVA with one factor being the magnitude of acceleration pertaining to each one-kilometre travelled by the participants in both running conditions. Specifically, all 1 km running split times were analysed in RunPrior along with the corresponding vertical, mediolateral, and anteroposterior magnitude of trunk acceleration (i.e., 5 km in total). The same approach was used in RunPost (i.e., 2.5 km in total). The magnitude of vertical, mediolateral, and anteroposterior trunk acceleration during each 5 km of cycling and the corresponding cycle cadence (i.e., self-selected, 55–60 rev/min1, 75–80 rev/min1, and 95–100 rev/min1) were analysed using a repeated-measures ANOVA.
To allow for a meaningful comparison in run performance, the final 2.5 km in RunPrior and the total 2.5 km in RunPost were compared. This approach was chosen given the obvious differences in running distance (i.e., 5 km and 2.5 km) in order to offer a truer reference value and that an element of fatigue would likely be present toward the conclusion of RunPrior and that the influence of cycling would likely influence RunPost. To compare 2.5 km RunPrior and 2.5 km RunPost, the Cohen effect-size benchmark (d) was calculated, giving 0.1–0.3 (small), greater than 0.3–0.5 (moderate), greater than 0.5–0.7 (large), greater than 0.7–0.9 (very large), and greater than 0.9 (extremely large) [30]. Like preceding methods (e.g., [31]), foot strike peaks were identified in the anteroposterior axis by way of manually filtering the data. Results are presented as mean ± SD. In all cases, the value p < 0.05 was used to establish statistical significance.

3. Results

3.1. Running Prior to Cycling (RunPrior)

The descriptive statistics for the mean magnitudes of trunk acceleration from RunPrior are presented in Table 2. The resultant vector represents the vector sum of the combined (i.e., x, y, z) vectors for the 5 km run. All p-values for kilometres one and two were not less than 0.05 and therefore did not indicate a statistically significant difference despite greater magnitudes of anteroposterior acceleration (z axis) seen. However, statistical significance was detected in the final three kilometres in all directions. Evaluating the results for RPE indicated that no significant differences occurred between the athletes.

3.2. Cycling

The magnitude of trunk acceleration, as determined by each completed 5 km of cycling, corresponding to the cadence condition is presented in Table 3. Four components of cadence were analysed, that is, one cadence value per 5 km completed lap of cycling (i.e., 20 km in total). Observing the cadence changes, the self-selected cadence was observed to cause the greatest magnitude of trunk acceleration in the vertical direction. In contrast, the initial 5 km of cycling resulted in the lowest magnitude in the anteroposterior path when compared to the other 5 km laps and cadences. One perceptible characteristic was that during kilometres 10–15 and 15–20, both of which were performed at the higher cadence of 75–80 rev/min1, significantly different magnitudes of trunk acceleration were observed, which ultimately resulted in larger vector magnitudes compared to the preceding cadences.

3.3. Running Prior to Cycling (RunPost)

Running after cycling was seen to trigger significant changes to the magnitude of trunk acceleration, notably in the vertical direction throughout the entire run. While anteroposterior trunk acceleration increased during the initial 1 km of running, this abated somewhat during the remaining distance, with significant differences only observed during the final 500 m. Similarly, significant differences in mediolateral acceleration were observed during the final 500 m. Notably, these increases coincided with the highest purported rate of perceived exertion (Table 4).

3.4. Comparison of Running Pre- and Post-Cycling

The athletes’ average times for the final 2.5 km in RunPrior and for the 2.5 km in RunPost were 4.03:05 (±0.018) and 4.08:16 (±0.024), respectively (in mm:ss). The 2.5 km in RunPost displayed slower performance times, which may have been influenced by prior cycling (Figure 4).
Significant differences were observed in trunk acceleration, that is, the combined mediolateral, vertical, and anteroposterior motion in the final 2.5 km during RunPrior against the 2.5 km during RunPost (p = 0.0093). Analysis revealed that the athletes’ largest variance occurred in the mediolateral direction (x axis) as the athletes displayed a greater magnitude, that is, their trunks experienced greater acceleration, during the initial 1 km of running post-cycling. From here, mediolateral acceleration was seen to increase in RunPost, although not significantly (Figure 5).

4. Discussion

This study aimed to analyse magnitudes of trunk acceleration in amateur short-distance duathletes during a simulated overground duathlon. Specifically, running performance and magnitudes of trunk acceleration were compared pre and post a cycling load.
Cycling, which was performed at a varied cadence, was also analysed. We remind the reader that this research was conducted in the field and reflected a typical training situation. Our intention here was to objectively analyse the uncoordinated feeling that duathletes commonly report during the early phase of running after a cycling load. We hypothesised that (1) the magnitude of trunk acceleration would increase during the initial bout of the duathlon, specifically running after cycling, and that (2) performance time (mm:ss) would decrease compared to running prior to cycling.
In addressing the first hypothesis, we observed a significant decrease in running performance time during RunPost compared to RunPrior. Corresponding running speeds decreased, falling from 4.03 mm: ss (14.8 km/h1) to 4:08 mm: ss (14.5 km/h1). In this amateur population, this reduction equates to a 5 s decrease per kilometre. Additional training interventions that focused on reducing this decline in running speed may provide advantages in performance.

4.1. Running Pre-Cycling (RunPrior)

At first sight, the modifications to the magnitude of trunk acceleration that were observed are difficult to place into context due to the paucity of research that has focused on the trunk as a proxy for performance in duathlon. The dynamism of the trunk requires a semblance of control and stability for athletes to maintain their accustomed running gait pattern [32,33]. Perhaps not unexpectedly, the quantity of trunk acceleration magnitude varied significantly amongst the amateur athletes in our study. The degree of between-athlete differences was not significant during kilometres 1 and 2 despite the contrasting levels of trunk movement. The possibility that elevated magnitudes of mediolateral trunk acceleration occurred during the initial 1 and 2 km suggests a gradual “settling in” to running, or a biomechanical adjustment of sorts prior to the athletes establishing their accustomed gait patterns. For the final 2.5 km kilometres, larger magnitudes of trunk acceleration were seen in all directions with slight increases to RPE. The reasons for this are unclear, although the onset of fatigue could be a mitigating factor. This highlights the need to attain a more robust dataset, possibly involving a larger sample of participants and/or a mixture of elite and amateur athletes, to allow for a wider comparison of trunk acceleration analysis.

4.2. Cycling

We analysed the effects of trunk motion by way of trunk acceleration magnitude when participants cycled in an aerodynamic cycling position. Firstly, our results add to the existing, albeit limited, body of evidence showing that trunk acceleration plays an active role in multisport (e.g., [24,29]). Our results suggest that the magnitude of mediolateral acceleration increased with a parallel rise in cadence (Table 4). The reasons for the increase in mediolateral acceleration are not fully understood, however. Factors that affect the magnitude of mediolateral acceleration may relate to a gradual “easing into” cycling following the preceding run. Alternatively, it could be that the athletes were pursuing their preferred and optimal cadence range. In this instance, when the participants endeavoured to apply a greater magnitude of force to the crank, the effort may not have been transferred effectually but instead caused undesirable trunk motion. We hypothesise that in this situation, trunk movement adjusts based on cadence. More research is needed to corroborate this, however.
Cycling performance requires considerations of internal and external criteria. For instance, Abbiss et al. [34] hypothesised that the ideal cycling cadence may differ according to the criteria implemented. Such criteria include cadence; therefore, the cadence selected by the athletes during the initial 5 km may be self-designated according to the perception of what is the most economical, produces higher power output or reduces fatigue, or what merely feels more comfortable. Though laboratory-based studies, such as that used by Chapman et al. [26], have sought to determine an optimal cadence, these studies have used elite athletes, and additionally, little consensus has emerged from these studies as to what may be considered an optimal cadence, or even if an optimal cadence exists. Regardless of this ambiguity, our results imply that larger magnitudes of total trunk acceleration occurred during cadences of 75–80 rev/min1. Specifically, these changes were observed in the final 10 km, respectively (i.e., lap 3: 10–15 km and lap 4: 15–20 km).
It would be easy to assume, given that larger accelerations of the trunk were observed during the final two laps of cycling, that an element of fatigue occurred, and this contributed to greater trunk motion. However, this could be a theory too far. We do not yet have the information to support if an unambiguous link between the level of cadence, trunk acceleration, and fatigue occurs. Notably, the RPE scale used by the athletes does not suggest that fatigue, or perceived exertion, was directly associated with greater trunk motion. Notwithstanding being somewhat of a laboured reflection, this could afford avenues of more applied research into duathlon cycle training effectiveness.

4.3. Run Post-Cycling (RunPost)

Our results share similarities with others (e.g., [9,26,27]) in that a varied cadence protocol modifies the magnitude of trunk acceleration in running after cycling compared to a standalone run. The foremost effect we observed was a significant increase in vertical and anteroposterior trunk acceleration combined with a modest increase in mediolateral acceleration (Figure 5). Here, it appears that the athletes applied a different running strategy during the initial 1 km, in contrast to the same temporal timestamp during 1 km in RunPrior. These findings closely agree with those reported by De Vito et al. [35] in that running after cycling causes a negative effect on running performance. Our results suggest that a trade-off between vertical and anteroposterior trunk motion could be needed to ensure that a more balanced “bandwidth” of trunk acceleration is achieved. This is because the trunk transfers and controls force and motion in a cohesive and dynamic chain [36]. This is central to athletic function and performance. Regrettably, research into the synchronic motions of trunk acceleration in duathlon is infrequent [37]. More research is warranted to better comprehend the impact of cycling on running in duathlon.
Running and cycling mainly require sagittal movement and very little transverse plane movement. However, changes to movement away from the sagittal plane are possible because of fatigue on spatiotemporal parameters (e.g., step length, step time, and ground contact time) and the CoM trajectory [38]. Here, a modicum of movement variability will be participant-dependent, which will also be based on the constraint to self-pace effectively to maintain effectiveness. Prior interventions have shown that increasing vertical oscillation via greater whole-body trunk acceleration may compromise running economy [12]. This means that the magnitude of trunk acceleration can alter as running velocity changes, which can subsequently influence running economy and efficiency. Indeed, past results have suggested that when running is intersected with cycling, significant differences occur to the vertical acceleration of the body’s CoM in both sinusoidal curves and foot strike peaks [39]. Under such circumstances as running “off the bike”, a more conservative or balanced running strategy that allows reduced trunk motion may be beneficial for overall performance. Nonetheless, this suggestion is speculative, and its relevance requires additional research.
Maintaining an aerodynamic trunk position on a bicycle during a duathlon places athletes in a protracted shoulder girdle and trunk flexion position. This, then, may affect spatiotemporal parameters and the magnitude of trunk acceleration and therefore influence performance. The increased trunk acceleration that was observed during RunPost could partially explain the increase in running time (i.e., in mm:ss) observed, since duathletes would have to overcome increases in vertical loading rates and vertical ground reaction forces. The influence of prior cycling may alter the neuromuscular system and/or perception of vertical loading vertical ground reaction forces, which may help explain the increased vertical acceleration seen in our study (Figure 5).
Our results indicate that significantly different mean run times of 04.03 (mm:ss) and 4.08 (mm:ss) occurred when RunPrior and RunPost were contrasted, respectively. These differences in run performance may be influenced by trunk kinematics, given the changes in movement patterns between running and cycling. Here, cycling adversely affected the 2.5 km run performance. Coordination patterns between the timing and magnitude of relative body motions can be influenced by trunk motion. An example of this is changes in running ground reaction force and braking impulse [40]. Although trunk accelerations can and do occur in the mediolateral and anteroposterior directions combined with deviations from the sagittal plane, to maintain trunk control and restrict motion to the sagittal plane, a greater extent of trunk, or core, strength is needed.
Running and cycling are characterised by distinct types of upper and lower limb motion and physiological correlates [14,41] along with both kinetic and kinematical variations [24,29,42]. However, caution is required if directing an athlete to assume new running mechanics based only on reducing trunk acceleration, as running is a multifaceted motion that requires metabolic, cardiorespiratory, and neuromuscular considerations. Despite the intricacies involved in running, if the disadvantageous impact of unnecessary trunk acceleration could be reduced, there exists likely opportunities to develop training-based interventions for trunk control when running after cycling.
This study does have limitations inherent to this type of field-based study design. Firstly, given the relatively small sample size used for this study (i.e., 10 participants were recruited), this limits the statistical power; therefore, a larger sample would be necessary for a greater interpretation of the results. Secondly, though the participants were asked to retain their customary running and cycling mechanics, when external variables such as headwind (drag) during cycling are considered, it may be that the participants had to modify their trunk, albeit marginally. Here, a small adjustment to the trunk may have influenced participant performance. It is uncertain how the varying components of drag affected the participants’ trunk motion and acceleration; this warrants further investigation. This study included both male and female participants. The differences between the sexes in body composition are well proven; therefore, studying running performance and magnitudes of trunk acceleration pre and post a cycling load between male and female athletes in duathlon is warranted.
Running and cycling economy, that is, efficiency and effectiveness, is a variable that will likely alter depending on individual factors. Here, participant variability relative to biomechanical factors such as stride rate, stride strength, and cycling pedalling technique are multifactorial measures. Such measures, while out of scope in our study, are determined by training, metabolic, cardiorespiratory, biomechanical, and neuromuscular factors [43]. Finally, we allocated five minutes between each participant commencing the initial 5 km run. A duration of five minutes was selected to allow sufficient time for each participant to self-select their running speed. Despite this, we acknowledge that some participants may have felt a level of competitiveness or the feeling of chasing or being chased by another participants. This was mitigated as best as possible by use of the RPE scale at each 1 km in running to monitor exertion levels. Yet, the contribution of each factor to duathlon performance will likely depend on how technical (i.e., skilled) the movement is. This acknowledges the need for additional research to encourage further insight into the intricacies of the influence of cycling in running performance in duathlon.
Our study reported magnitudes of trunk acceleration in running and cycling from an amateur athlete standpoint. This limits the application of our results to the semiprofessional and elite category of duathletes. Future studies should apply a similar protocol using elite athletes instead of amateur level athletes. Arguably, however, both modes of research design (e.g., elite vs. amateur) are necessary to allow further scientific evidence for coaches, athletes, and sports scientists to analyse respective datasets with the aim of improving performance in all disciplines of duathlon. This will permit athletes to prepare more judiciously for multisport events.
Although there is significant inconsistency in the literature regarding preferred and optimal cadence, there is general agreement that cyclists use a comparatively high cadence as they are more efficient [44]. Coast and Welch [45] reported that optimal cadence (minimal oxygen uptake) changes linearly, increasing from just over 40 rpm at 100 W to nearly 80 rpm at 300 W. It is recognized that a given power output can be accomplished at a variety of cadences, so, in effect, there would be a number of cadence-resistance combinations at which an athlete could achieve the target power output [46].
While cadence, and not power, was used in our paper, the identification of an efficient cadence–power relationship during duathlon cycling could have important implications for both cycle performance and the proceeding run.
Finally, our novel approach involved taking a readable available technological device to infer performance outcomes via a relatively simple data metric. That our study was conducted in the field and represented a typical duathlon highlights the simplicity in research design and methodology that can be applied. Additionally, we used existing research, for instance, the cadence protocol devised by Chapman et al. [26] combined with a prior validation for accelerometer usage in cycling [47], as a basis to progress the sport of duathlon. This raises an important point. The purpose and function of applied sport sciences are to investigate questions about performance, which must be solved on a scientific basis. Analysing more specifically the different fields of study combined with relatively unexplored sports, it is possible to affirm that some fields have more robust growth, while in others, their growth is more moderate [48]. This could be solved by additional collaboration across the different fields of studies and sporting codes. Specifically, prior areas of research such as the bicycle-to-run transition in triathlon could be applied to duathlon to share knowledge and create possible training interventions.

5. Summary and Conclusions

This novel approach used a field-based methodology and applied an unobtrusive wearable device to analyse the motion pattern of trunk acceleration to infer variations in running pre and post a cycling load. The data that we obtained demonstrated that the first kilometre of a duathlon run, that is, running post-cycling, is prone to increased magnitudes of trunk acceleration compared to a standalone run, which adds to the prevailing literature that running after cycling causes uncoordinated running that has been commonly reported by athletes. Additionally, while the magnitude of trunk acceleration may be expected to dissipate throughout the run, this was not the case when running after a cycling load. The results indicate that some of this effect can be linked to trunk acceleration variability. In duathlon, explicitly the running after cycling component where athletes strive for marginal improvements, our findings could offer additional perspective when considering trunk kinematics under ecologically valid conditions.

Author Contributions

Conceptualization, S.E.; methodology, S.E.; formal analysis, S.E. and D.A.J.; investigation, S.E.; writing—original draft preparation, S.E.; writing—review and editing, D.A.J. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (HREC 21114, December 2022).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are possibly available on request. This will be subject to institutional ethical approval for release by the researchers.

Acknowledgments

The authors wish to thank the athletes who kindly volunteered to be part of this study.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Landers, G.J.; Blanksby, B.A.; Ackland, T.R.; Smith, D. Morphology and performance of world championship triathletes. Annals Hum. Biology 2000, 27, 387–400. [Google Scholar] [CrossRef] [PubMed]
  2. Vleck, V.E.; Bürgi, A.; Bentley, D.J. The consequences of swim, cycle, and run performance on overall result in elite Olympic distance triathlon. Int. J. Sports Med. 2006, 27, 43–48. [Google Scholar] [CrossRef] [PubMed]
  3. Sparks, S.A.; Cable, N.T.; Doran, A.; Maclaren, P.M. The influence of environmental temperature on duathlon performance. Ergonomics 2007, 48, 1558–1567. [Google Scholar] [CrossRef] [PubMed]
  4. Millet, G.P.; Vleck, V. Physiological and biomechanical adaptations to the cycle to run transition in Olympic triathlon: Review and practical recommendations for training. Br. J. Sports Med. 2000, 34, 384–390. [Google Scholar] [CrossRef]
  5. Suriano, R.; Vercruyssen, F.; Bishop, D.; Brisswalter, J. Variable power output during cycling improves subsequent treadmill run time to exhaustion. J. Sci. Med. Sport. 2007, 10, 244–251. [Google Scholar] [CrossRef]
  6. Bernard, T.; Vercruyssen, F.; Grego, F.; Housswirth, C.; Lepers, R.; Vallier, J.-M.; Brisswalter, J. Effect of cycling cadence on subsequent 3 km running performance in well trained triathletes. Br. J. Sports Med. 2003, 37, 154–158. [Google Scholar] [CrossRef]
  7. Chapman, A.R.; Vicenzino, B.; Blanch, P.; Dowlan, S.; Hodges, P.W. Does cycling effect motor coordination of the leg during running in elite triathletes? J. Sci. Med. Sport 2008, 11, 371–380. [Google Scholar] [CrossRef]
  8. Weich, C.; Jensen, R.L.; Vieten, M. Triathlon transition study: Quantifying differences in running movement pattern and precision after bike-run transition. Sports Biomech. 2019, 18, 215–228. [Google Scholar] [CrossRef]
  9. Bonacci, J.; Saunders, P.U.; Alexander, M.; Blanch, P.; Vicenzino, B. Neuromuscular control and running economy is preserved in elite international triathletes after cycling. Sports Biomech. 2011, 10, 59–71. [Google Scholar] [CrossRef]
  10. Hue, O.; Le Gallais, D.; Boussana, A.; Chollet, D.; Prefaut, C. Ventilatory responses during experimental cycle-run transition in triathletes. Med. Sci. Sports Exer. 1999, 31, 1422. [Google Scholar] [CrossRef]
  11. Millet, G.P.; Millet, G.Y.; Hofmann, M.; Candau, R. Alterations in running economy and mechanics after maximal cycling in tri-athletes: Influence of performance level. Int. J. Sports Med. 2000, 21, 127–132. [Google Scholar] [CrossRef] [PubMed]
  12. Evans, S.A.; James, D.A.; Rowlands, D.; Lee, J.B. Assessment of accelerometer-derived data in the context of seat height changes in cycling and the impact on running in triathlon. Proc. Inst. Mech. Eng. Part. P J. Sports Eng. Tech. 2023, 237, 252–262. [Google Scholar] [CrossRef]
  13. Quigley, E.J.; Richards, J.G. The effects of cycling on running mechanics. J. Appl. Biomech. 1996, 12, 470–479. [Google Scholar] [CrossRef]
  14. Millet, G.P.; Vleck, V.E.; Bentley, D.J. Physiological differences between cycling and running: Lessons from triathletes. Sports Med. 2009, 3, 179–206. [Google Scholar] [CrossRef] [PubMed]
  15. Gohlitz, D.; Große, S.; Witt, M. Darstellungen von Veränderungen der Schrittlänge und Schrittfrequenz beim Übergang vom Radfahren zum Laufen zur Kennzeichnung der Dauer von Übergangsphasen im Duathlon (Pilotuntersuchung). In Triathlon—Medizinische und Methodische Probleme des Trainings; Engelhardt, M., Franz, B., Neumann, G., Pfützner, A., Eds.; Triathlon und Sportwissenschaft; Czwalina: Hamburg, Germany, 1994; ISBN 978-3-88020-249-8. [Google Scholar]
  16. Connick, M.J.; Li, F.-X. Prolonged cycling alters stride time variability and kinematics of a post-cycle transition run in triathletes. J. Electromyogr. Kinesiol. 2015, 25, 34–39. [Google Scholar] [CrossRef] [PubMed]
  17. Chapman, A.; Vicenzino, B.; Blanch, P.; Hodges, P. Is running less skilled in triathletes than runners matched for running training history? Med. Sci. Sports Exerc. 2008, 40, 557–565. [Google Scholar] [CrossRef]
  18. Vieten, M.M.; Sehle, A.; Jensen, R.L. A novel approach to quantify time series differences of gait data using attractor attributes. PLoS ONE 2013, 8, e71824. [Google Scholar] [CrossRef]
  19. Barnes, K.R.; Kilding, A.E. Running economy: Measurement, norms, and determining factors. Sports Med. Open 2015, 1, 8. [Google Scholar] [CrossRef]
  20. Lai, A.; James, D.; Hayes, J. Semi-automatic calibration technique using six inertial frames of reference. SPIE 2004, 5274, 531–542. [Google Scholar]
  21. Kobsar, D.; Osis, S.; Hettinga, B.; Ferber, R. Classification accuracy of a single tri-axial accelerometer for training background and experience level in runners. J. Biomech. 2014, 47, 508–2511. [Google Scholar] [CrossRef]
  22. Seshadri, D.R.; Li, R.T.; Voos, J.E.; Rowbottom, J.R.; Alfes, C.M.; Zorman, C.A.; Drummond, C.K. Wearable sensors for monitoring the internal and external workload of the athlete. NPJ Digit. Med. 2019, 29, 71. [Google Scholar] [CrossRef]
  23. Jackson, A.S.; Pollock, M.L. Practical assessment of body composition. Physic. Sport. Med. 1985, 13, 76–90. [Google Scholar] [CrossRef]
  24. Evans, S.A.; James, D.A.; Rowlands, D.; Lee, J.B. The effect of cleat position on running using acceleration-derived data in the context of triathlons. Sensors 2021, 21, 5899. [Google Scholar] [CrossRef]
  25. Gregor, R.J.; Broker, J.P.; Ryan, M.M. The biomechanics of cycling. Exerc. Sport Sci. 1991, 19, 127–169. [Google Scholar] [CrossRef]
  26. Chapman, A.R.; Vicenzino, B.; Hodges, P.W.; Blanch, P.; Hahn, A.G.; Milner, T.E. A protocol for measuring the direct effect of cycling on neuromuscular control of running in triathletes. J. Sports Sci. 2009, 27, 767–782. [Google Scholar] [CrossRef] [PubMed]
  27. Bonacci, J.; Chapman, A.; Blanch, P.; Vicenzino, B. Neuromuscular adaptations to training, injury and passive interventions: Implications for running economy. Sports Med. 2009, 39, 903–921. [Google Scholar] [CrossRef] [PubMed]
  28. Borg, G. Borg’s Perceived Exertion and Pain Scales; Human Kinetic: Champaign, France, 1998. [Google Scholar]
  29. Bini, R.R.; Hume, P.A.; Croft, J. Cyclists and triathletes have different body positions on the bicycle. Eur. J. Sports Sci. 2014, 14, S109–S115. [Google Scholar] [CrossRef]
  30. Cohen, J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.; Lawrence Erlbaum Associates: Hillside, NJ, USA, 1988. [Google Scholar]
  31. Lee, J.B.; Sutter, K.J.; Askew, C.D.; Burkett, B.J. Identifying symmetry in running gait using a single inertial sensor. J. Sci. Med. Sport. 2010, 13, 559–563. [Google Scholar] [CrossRef] [PubMed]
  32. Turnwald, J.; Sousa, C.V.; Andrade, M.S.; Thuany, M.; Cuk, I.; Nikolaidis, P.T.; Weiss, K.; Knechtle, B. Participation and performance trends in short-, medium, and long-distance duathlon. Sci. Rep. 2023, 13, 9303. [Google Scholar] [CrossRef] [PubMed]
  33. Nikolaidis, P.T.; Villiger, E.; Victor Sousa, C.; Rosemann, T.; Knechtle, B. The effect of aging on pacing strategies in short and long-distance duathlon. Exp. Aging Res. 2019, 45, 223–233. [Google Scholar] [CrossRef] [PubMed]
  34. Abbiss, C.R.; Peiffer, J.J.; Laursen, P.B. Optimal cadence selection during cycling. Int. Sports Med. 2009, 10, 1–15. [Google Scholar]
  35. De Vito, G.; Bernardi, M.; Sproviero, E.; Figura, F. Decrease of endurance performance during Olympic triathlon. Int. J. Sports Med. 1995, 16, 24–28. [Google Scholar] [CrossRef] [PubMed]
  36. Kibler, W.B.; Press, J.; Sciascia, A. The role of core stability in athletic function. Sports Med. 2006, 36, 189–198. [Google Scholar] [CrossRef] [PubMed]
  37. Saeterbakken, A.H.; Andersen, V.; Behm, D.G.; Cumming, K.T.; Prieske, O.; Solstad, T.E.J.; Shaw, M.; Stien, N. The role of trunk training for physical fitness and sport-specific performance. Protocol for a meta-analysis. Front. Sports Act. Living 2021, 3, 625098. [Google Scholar] [CrossRef]
  38. Tesio, L.; Rota, V. The motion of body center of mass during walking: A review oriented to clinical applications. Front. Neurol. 2019, 10, 999. [Google Scholar] [CrossRef]
  39. Evans, S.A.; James, D.A.; Rowlands, D.R.; James, J.B.; Gleadhill, S. Cycling effects vertical acceleration of the center of mass in running in triathletes. J. Phys. Fit. Treat. Sports 2020, 8, 555737. [Google Scholar]
  40. Wille, C.M.; Lenhart, R.L.; Wang, S.; Thelen, D.G.; Heiderscheit, B.C. Ability of sagittal kinematic variables to estimate ground reaction forces and joint kinetics in running. J. Orthop. Sports Phys. Ther. 2014, 44, 825–830. [Google Scholar] [CrossRef]
  41. Bentley, D.; Millet, G.P.; Vleck, V.E.; McNaughton, L.R. Specific aspects of contemporary triathlon: Implications for physiological analysis and performance. Sports Med. 2002, 32, 345–359. [Google Scholar] [CrossRef]
  42. Vallier, J.M.; Mazure, C.; Hausswirth, C.; Bernard, T.; Brisswalter, J. Energy cost of running during a specific transition in duathlon. Can. J. Appl. Physiol. 2003, 28, 673–684. [Google Scholar] [CrossRef] [PubMed]
  43. Barnes, K.R.; Kilding, A.E. Strategies to improve running economy. Sports Med. 2015, 45, 37–56. [Google Scholar] [CrossRef] [PubMed]
  44. Marsh, A.P.; Martin, P.E. Effect of cycling experience, aerobic power, and power output on preferred and most economical cycling cadences. Med. Sci. Sports Exerc. 1997, 29, 1225–1232. [Google Scholar] [CrossRef] [PubMed]
  45. Coast, J.R.; Welch, H.G. Linear increase in optimal pedal rate with increased power output in cycle ergometry. Eur. J. Appl. Physiol. 1985, 53, 339–342. [Google Scholar] [CrossRef] [PubMed]
  46. Macintosh, B.R.; Neptune, R.R.; Richard, R.; Horton, J.F. Cadence, power, and muscle activation in cycle ergometry. Med. Sci. Sports Exerc. 2000, 32, 1281–1287. [Google Scholar] [CrossRef] [PubMed]
  47. Evans, S.A.; Ballhause, K.; James, D.A.; Rowlands, D.; Lee, J.B. The development and validation of an inertial sensor for measuring cycling kinematics: A preliminary study. J. Sci. Cycl. 2021, 10, 34–44. [Google Scholar] [CrossRef]
  48. Maneiro, R.; Losada, J.L.; Casal, C.A.; Papadopoulou, S.; Sarmento, H.; Ardá, A.; Iglesias, X.; Amatria, M. Editorial: Advances in Sport Science: Latest Findings and New Scientific Proposals. Front. Psychol. 2022, 13, 891906. [Google Scholar] [CrossRef]
Applsci 14 01437 g001
Figure 1. Overview of study design protocol, where rev/min1 is revolutions per minute.
Figure 1. Overview of study design protocol, where rev/min1 is revolutions per minute.
Applsci 14 01437 g001
Applsci 14 01437 g002
Figure 2. Map view and location of experiment setting. The start line signified a change of cadence and represented one completed lap of the circuit (i.e., 5 km). The rightmost panel represents the index of elevation that contains the course variables experienced by the duathletes during the simulated duathlon. The mean gradient across the circuit was 0%. Evans et al. [24].
Figure 2. Map view and location of experiment setting. The start line signified a change of cadence and represented one completed lap of the circuit (i.e., 5 km). The rightmost panel represents the index of elevation that contains the course variables experienced by the duathletes during the simulated duathlon. The mean gradient across the circuit was 0%. Evans et al. [24].
Applsci 14 01437 g002
Applsci 14 01437 g003
Figure 3. Leftmost: Location of sensor on lumbar 5 sacrum 1 area. Rightmost: Reference system used by manufacturer of ActiGraph GT9X + accelerometer used on all participants (vertical (y, upward–downward), anteroposterior (z, forward–backward), and mediolateral (x, side to side)). Note that graphical effects were added for emphasis.
Figure 3. Leftmost: Location of sensor on lumbar 5 sacrum 1 area. Rightmost: Reference system used by manufacturer of ActiGraph GT9X + accelerometer used on all participants (vertical (y, upward–downward), anteroposterior (z, forward–backward), and mediolateral (x, side to side)). Note that graphical effects were added for emphasis.
Applsci 14 01437 g003
Applsci 14 01437 g004
Figure 4. Comparison of the final 2.5 km performance in running prior to cycling (RunPrior) and 2.5 km in running after cycling (RunPost) in a simulated duathlon (n = 10). Total effect between conditions d ≥1.
Figure 4. Comparison of the final 2.5 km performance in running prior to cycling (RunPrior) and 2.5 km in running after cycling (RunPost) in a simulated duathlon (n = 10). Total effect between conditions d ≥1.
Applsci 14 01437 g004
Applsci 14 01437 g005
Figure 5. Magnitudes of trunk acceleration during a 2.5 km run in RunPrior and RunPost in a simulated duathlon in ten duathletes (n = 10). Mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. Applsci 14 01437 i001 Significant at p < 0.05; d > 1.9 (extremely large).
Figure 5. Magnitudes of trunk acceleration during a 2.5 km run in RunPrior and RunPost in a simulated duathlon in ten duathletes (n = 10). Mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. Applsci 14 01437 i001 Significant at p < 0.05; d > 1.9 (extremely large).
Applsci 14 01437 g005
Table 1. Descriptive statistics of 10 amateur duathletes who participated in the simulated duathlon.
Table 1. Descriptive statistics of 10 amateur duathletes who participated in the simulated duathlon.
Participants (n = 10)Mean ± SDMinMax
Ages (years)30.25 ± 2.32637
Mass (kg)68.24 ± 1.96179
Body height (cm) 175.7 ± 2.3169186
Weekly training (hours per week)9.45 ± 1.7712
Training experience (years)9.15 ± 5.2710
Table 2. Ten duathletes (n = 10) running prior to cycling (RunPrior) in a simulated duathlon with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. RPE = ratings of perceived exertion. ** p < 0.01. ≠ Represents 2.5 km used for analysis. In gravitational acceleration (G).
Table 2. Ten duathletes (n = 10) running prior to cycling (RunPrior) in a simulated duathlon with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. RPE = ratings of perceived exertion. ** p < 0.01. ≠ Represents 2.5 km used for analysis. In gravitational acceleration (G).
Running
Distance		Mean
Running Pace (km/h1)		x
(G)		py
(G)		pz
(G)		pResultant VectorRPE
Kilometre 14.06 ± 0.30.56 ± 0.2 0.053.23 ± 0.30.051.69 ± 0.10.051.93 ± 0.386 ± 0.1
Kilometre 24.05 ± 0.10.35 ± 0.10.053.25 ± 0.20.051.79 ± 0.10.055.11 ± 1.19 6 ± 0.1
Kilometre 34.08 ± 3.00.58 ± 0.2<0.01 **3.28 ± 0.5<0.01 **1.81 ± 0.2<0.01 **6.44 ± 0.77 ± 0.2
500 m ≠4.09 ± 2.80.61 ± 0.2<0.01 **3.35 ± 0.5<0.01 **1.80 ± 0.2<0.01 **7.20 ± 0.87 ± 0.1
Kilometre 4 ≠4.09 ± 2.90.62 ± 0.1<0.01 **3.32 ± 0.8<0.01 **1.76 ± 0.3<0.01 **7.19 ± 0.97 ± 0.3
Kilometre 5 ≠4.09 ± 2.80.71 ± 0.8<0.01 **3.33 ± 0.8<0.01 **1.78 ± 0.2<0.01 **7.72 ± 0.98 ± 0.4
Table 3. Ten duathletes (n = 10) cycling distance and cadence per 5 km lap with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. ** p < 0.01. T1 = the first transition marked the transition from RunPrior to the cycle segment. In gravitational acceleration (G).
Table 3. Ten duathletes (n = 10) cycling distance and cadence per 5 km lap with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. ** p < 0.01. T1 = the first transition marked the transition from RunPrior to the cycle segment. In gravitational acceleration (G).
DistanceCadence (rev/min1)x (G)py (G)pz (G)pRPEResultant Vector
Transition (T1)
1.23 (mm:ss) ± 0.4		 0.056 ± 0.2-
Kilometre 0–5Self-selected0.95 ± 0.10.050.99 ± 0.20.050.59 ± 0.20.056 ± 0.11.93 ± 0.3
Kilometre 5–1055–60 rev/min10.99 ± 0.20.050.77 ± 0.20.050.65 ± 0.40.057 ± 0.25.11 ± 0.4
Kilometre 10–1575–80 rev/min11.21 ± 0.20.050.88 ± 0.30.050.89 ± 0.10.059 ± 0.15.79 ± 0.4
Kilometre 15–2075–80 rev/min11.64 ± 0.8<0.01 **0.81 ± 0.6<0.01 **0.88 ± 0.3<0.01 **10 ± 0.26.51 ± 0.4
Table 4. Ten duathletes (n = 10) running post-cycling (RunPost) in a simulated duathlon with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. RPE = ratings of perceived exertion. ** p < 0.01. T2 = the second transition marked the transition from cycling to RunPost. In gravitational acceleration (G).
Table 4. Ten duathletes (n = 10) running post-cycling (RunPost) in a simulated duathlon with mediolateral (x), vertical (y), and anteroposterior (z) acceleration magnitude. RPE = ratings of perceived exertion. ** p < 0.01. T2 = the second transition marked the transition from cycling to RunPost. In gravitational acceleration (G).
Running
Distance		Mean
Running Pace km/h1		x
(G)		py
(G)		pz
(G)		pRPEResultant Vectorp
Transition 2 (T2)
2.03 (mm: ss) ± 0.5		  0.0510 ± 0.2 0.05
Kilometre 14.09 ± 0.80.66 ± 0.30.054.25 ± 0.4<0.01 **2.010.055 ± 0.16.40 ± 0.6
Kilometre 24.08 ± 0.80.71 ± 0.60.053.95 ± 0.2<0.01 **1.990.056 ± 0.16.29 ± 0.6
500 m4.07 ± 0.90.69 ± 0.7<0.01 **3.92 ± 0.9<0.01 **1.92<0.01 **8 ± 0.26.21 ± 0.5
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Evans, S.; James, D.A. The Effects of a Simulated Duathlon on Trunk Motion: A Sensor Based Approach. Appl. Sci. 2024, 14, 1437. https://doi.org/10.3390/app14041437

AMA Style

Evans S, James DA. The Effects of a Simulated Duathlon on Trunk Motion: A Sensor Based Approach. Applied Sciences. 2024; 14(4):1437. https://doi.org/10.3390/app14041437

Chicago/Turabian Style

Evans, Stuart, and Daniel Arthur James. 2024. "The Effects of a Simulated Duathlon on Trunk Motion: A Sensor Based Approach" Applied Sciences 14, no. 4: 1437. https://doi.org/10.3390/app14041437

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop