Training Impulse as a Tool for Linking Exercise Dose to Health Outcomes in Adolescents: Evidence from Interval-Based Interventions

Abstract

Background: The training impulse (TRIMP) method integrates exercise duration and intensity, yet its application in school-based health programs is limited. This study examined dose–response relationships between TRIMP and changes in body fat percentage (%BF), blood pressure, and cardiorespiratory fitness (CRF) in adolescents. Methods: 145 adolescents (69 males, 76 females; aged 15.01 ± 0.54 years) completed eight-week interval-based physical education programs: high-intensity interval training (HIIT) or high-intensity plyometric training (HIPT). Heart rate was continuously monitored, and TRIMP values were calculated across 16 sessions. Outcomes included %BF, systolic (SBP) and diastolic blood pressure (DBP), and maximal oxygen consumption (VO_2max), assessed pre- and post-intervention. Results: HIIT elicited higher TRIMP than HIPT in females (η²p = 0.043). Linear dose–response patterns were identified (r = 0.18–0.36) (e.g., SBP in males, r = −0.38, p = 0.001; %BF in males, r = −0.36, p = 0.002). In males, one TRIMP unit reduced SBP by 1.8 mmHg (HIIT) and DBP by 1.6 mmHg (HIPT). In females, HIPT-derived TRIMP predicted reductions of ~0.6% in %BF and~0.9 mmHg in DBP. No significant associations were found for VO_2max. Conclusions: TRIMP provides a feasible and sensitive tool for quantifying internal training load in adolescent interventions, linking exercise dose to measurable health benefits and supporting its application in preventive contexts.

1. Introduction

Promoting physical activity is a key preventive strategy, as a sedentary lifestyle contributes to increased adipose tissue mass and metabolic disorders that underlie many chronic diseases [1,2,3]. These diseases limit general physical fitness and circulatory and respiratory capacity, contribute to emotional disorders, and negatively impact mental health in young people and adults [3,4,5,6,7,8]. Since sedentary behavior and high-calorie consumption dominate child and adolescent patterns, metabolic disorders are increasingly recorded in this population. Therefore, movement and physical activity appear crucial to counteracting and treating metabolic disorders resulting from unhealthy lifestyles.

Adolescence is a time of significant physiological and structural changes that condition proper physical, mental, and emotional development [9]. Inappropriate health behaviors in this period may lead to overweight and obesity and become entrenched problems later in life. Thus, national and international health policies emphasize preventive strategies focused on promoting physical activity among youth [10].

Physical education (PE) lessons represent an important opportunity to implement health-oriented physical activity programs that are engaging for adolescents [11,12,13]. Various forms of physical activity can facilitate fat mass reduction and improve metabolic health, as well as support psychological well-being [14]. From a physiological perspective, continuous endurance aerobic exercise is effective in reducing fat tissue and improving cardiorespiratory fitness (CRF), but it is often perceived as monotonous and time-consuming [15]. Considering the time restrictions of school lessons and the potential lack of enthusiasm among students, high-intensity interval training (HIIT) has emerged as a promising alternative [16,17,18].

HIIT consists of short, high-intensity exercise bouts (>80% HRmax) interspersed with recovery periods. This structure provokes beneficial adaptive changes in a relatively short time, comparable to those induced by continuous moderate-intensity exercise [19,20]. HIIT can reduce body weight, improve CRF, and enhance strength and speed, while providing high enjoyment and adherence [21,22,23,24]. Moreover, HIIT interventions have demonstrated preventive and therapeutic effects against various metabolic diseases, including obesity and type 2 diabetes [25].

High-intensity plyometric training (HIPT) emphasizes explosive bodyweight movements such as jumps, hops, or burpees, and although it shares similarities with HIIT in terms of exercise intensity, it involves distinct neuromuscular demands. Plyometric training has been shown to enhance lower-limb power, agility, and movement dynamics, improve bone mass and stretch–shortening cycle efficiency, and reduce the risk of musculoskeletal injuries in youth populations [26,27,28,29,30,31]. Importantly, HIPT may also have cardiometabolic relevance. By promoting rapid eccentric–concentric transitions and enabling a higher number of repetitions within short work intervals, HIPT increases the total amount of work performed per bout and imposes a considerable load on cardiovascular and metabolic systems [21]. Previous studies reported that plyometric-based protocols could induce favorable changes in body composition and cardiovascular outcomes. For instance, Racil et al. [27] observed that incorporating plyometric elements into interval training improved body composition and reduced blood pressure in obese adolescent girls. Thus, HIPT represents not only a neuromuscular engaging and motivating option for adolescents but also a promising school-based strategy with the capacity to positively influence blood pressure and other health markers.

Despite the broad use of HIIT and HIPT, there is a lack of studies investigating their effects during PE lessons in adolescents. Previous studies suggest that both modalities are safe and effective in promoting beneficial physiological changes in this population [32,33,34]. Their variability ensures a positive reception by young people, potentially increasing participation and enhancing long-term health outcomes [35,36].

The concept of training impulse (TRIMP) is commonly used to quantify internal training load, integrating both intensity (average HR) and duration (minutes). While TRIMP has been applied primarily in adult or laboratory settings, its application in school-based adolescent PE contexts remains underexplored. Furthermore, there is a lack of studies analyzing the dose–response relationship between TRIMP and health outcomes such as body fat percentage (%BF), blood pressure, and CRF in real-world school settings [37,38,39,40]. A recent study comparing original and alternative TRIMP methods provided important insights into quantifying internal training load [41], supporting its use in intervention studies with adolescents.

Therefore, the present study aimed to investigate the relationship between the training impulse (TRIMP) induced by HIIT and HIPT interventions and the magnitude of changes in body composition and cardiovascular health markers among adolescents during PE lessons. To the best of the authors’ knowledge, this is the first study presenting the dose–response relationship for health outcomes using TRIMP in a school-based intervention.

In this study, we deliberately applied two distinct intervention formats—HIIT and HIPT—to examine the usefulness of TRIMP for quantifying internal training load. While both protocols share high-intensity characteristics, they differ in neuromuscular and metabolic demands, with HIIT emphasizing structured intervals of resistance-type movements and HIPT relying on explosive plyometric actions. By comparing these two modalities, we aimed to determine whether TRIMP could reliably capture dose–response relationships independent of the training format. This design ensured that the validity of TRIMP as a universal indicator of internal training load could be evaluated beyond a single type of intervention, thereby increasing the ecological value and applicability of our findings in school-based health programs.

2. Materials and Methods

2.1. Study Design

The trial was conducted over an 8-week period. For the present analysis, only participants from the two experimental groups—high-intensity interval training (HIIT) and high-intensity plyometric training (HIPT)—were included, as the main aim was to examine dose–response effects of the delivered training load (TRIMP). Participants were recruited from two secondary schools participating in the PEER-HEART project. A total of 429 students were initially enrolled and randomly allocated by class to either high-intensity interval training (HIIT), high-intensity plyometric training (HIPT), or corresponding control groups. From this sample, 145 adolescents who completed either HIIT or HIPT interventions were analyzed in the current study. Both interventions were conducted during regular physical education lessons twice per week for eight weeks and included a progressive Tabata-style structure (4–8 bouts of 20 s work, 10 s rest). HIIT sessions emphasized bodyweight resistance exercises, whereas HIPT incorporated explosive plyometric movements. Heart rate was continuously monitored using Polar Verity Sense sensors (Polar Electro, Kempele, Finland) to ensure high-intensity effort. Further methodological details regarding recruitment and intervention protocols are provided in our previous publication [42]. For the purpose of the present analysis, only data from participants allocated to the experimental groups were included, as the primary focus was on quantifying delivered training load (TRIMP) and its relationship with health outcomes.

2.2. Clinical Trial Registration

This investigation was carried out within the framework of the national project “Science for Society II,” supported by the Polish Ministry of Science and Higher Education (project no. NdS-II/SP/0521/2023/01). The study was prospectively registered at ClinicalTrials.gov (identifier: NCT06431230) under the acronym PEER-HEART (Physical Education dosE Response Health markErs Adolescents inteRval Training).

2.3. Ethics Committee

Ethical approval was obtained from the Ethics Committee of the Wroclaw University of Health and Sport Sciences (decision no. ECUPE 33/2018, dated 31 October 2018). All procedures adhered to the ethical principles for research involving human participants outlined in the Declaration of Helsinki. Written informed consent was obtained from both the students and their legal guardians prior to enrollment.

2.4. Participants

The analysis focused on adolescents who participated in the experimental arms of a larger randomized controlled trial (RCT). The final analytic sample consisted of 145 adolescents (HIIT: 45 males, 30 females; HIPT: 24 males, 46 females; mean age 15.01 ± 0.54 years). Participants were included based on compliance and attendance during the intervention. A post hoc power analysis (GPower v.3.1) confirmed that the available sample (n = 145) provided sufficient power (0.99) to detect small-to-moderate effects (f = 0.20) at α = 0.05. Details of recruitment, eligibility, and participant flow are provided in the previous publication [42].

2.5. Data Collection

2.5.1. Procedures

All measurements were performed at baseline, after eight weeks of intervention, and at an eight-week follow-up. Data collection occurred in school sports halls between 8:00 a.m. and 1:00 p.m. under standardized conditions.

All outcome measurements were performed by members of the research team with prior training and standardized procedures. The HIIT and HIPT sessions were delivered by physical education teachers who received specific training and detailed written instructions to ensure full adherence to the intervention protocols, under regular supervision from the researchers.

2.5.2. Body Morphology

Height was measured to the nearest 0.1 cm with a GPM anthropometer (DKSH Ltd., Zurich, Switzerland) following ISAK standards [43]. Body weight and %BF were assessed with a Tanita BC-601 bioelectrical impedance device (Tanita Co., Tokyo, Japan). Standardized pre-measurement routines were ensured (hydration, meals, bathroom use). BMI (kg/m²) was calculated using the formula:

$$\mathrm{BMI}={\displaystyle \frac{\mathrm{body}\mathrm{mass}\left[\mathrm{kg}\right]}{{\mathrm{body}\mathrm{height}[\mathrm{m}}^2]}}$$

2.5.3. Blood Pressure

SBP and DBP were measured using an Omron BP710 automatic monitor (Omron Healthcare, Inc., Hoffman Estates, IL, USA) [44]. Cuff size was adjusted individually, and participants rested for 10 min before the first of three measurements taken at 10 min intervals. Mean values were used in the analyses.

2.5.4. Multistage Fitness Test

VO_2max was estimated via the Multi-Stage Fitness Test (MSFT) with continuous HR monitoring (Polar Verity Sense Sensors, Polar Electro, Finland) [45]. VO_2max was calculated according to Ramsbottom et al. [22]. The following formula was used:

$${VO}_{2max}=3.46\times \left(\mathrm{L}+\mathrm{SN}/\left(\mathrm{L}\times 0.4325+7.0048\right)\right)+12.2,$$

where L is the level, and SN is the number of shuttles.

2.6. Intervention

Both interventions were implemented during PE lessons, twice weekly, over eight weeks. Two protocols have been conducted: (1) HIIT protocol: Tabata-style sessions with progressive volume (4 to 8 × 20 s effort, 10 s rest) using bodyweight exercises (e.g., squats, lunges, push-ups), and (2) HIPT protocol: Plyometric movements emphasizing explosive strength and power (e.g., squat jumps, ankle hops, burpees, mountain climbers).

Each session began with a 10 min standardized warm-up. Supervisors ensured compliance, providing feedback and encouraging maximal effort during work intervals. The remainder of PE classes followed the standard school curriculum. Control groups in the broader RCT continued standard PE lessons but were not included in this analysis.

Heart rate intensity was tracked weekly. Across eight weeks, mean HR in HIPT ranged from 75.9% to 80.7% HRmax, and in HIIT from 78.5% to 85.7% HRmax, confirming high-intensity load.

Training intensity monitoring: Mean weekly heart rate values, expressed as %HRmax, confirmed that both interventions consistently reached high-intensity levels across eight weeks (

Table 1. Mean weekly heart rate values expressed as percentage of maximum heart rate (%HRmax) during high-intensity interval training (HIIT) and high-intensity plyometric training (HIPT) sessions across eight weeks. Data are presented as mean ± standard deviation (SD), 95% confidence intervals (CI), and observed ranges.

Week	HIPT			HIIT
	Mean ± SD (%HRmax)	95% CI	Range	Mean ± SD (%HRmax)	95% CI	Range
1	80.7 ± 6.8	79.1–82.3	65–97	85.7 ± 6.1	84.3–87.1	66.0–97.0
2	79.7 ± 6.8	78.1–81.3	61.9–94.8	82.4 ± 5.8	81.0–83.7	66.1–96.2
3	78.4 ± 8.2	76.4–80.3	52.6–90.1	81.3 ± 6.2	79.9–82.7	65.3–92.5
4	78.7 ± 7.7	76.9–80.6	53.1–92	81.2 ± 6.3	79.8–82.7	62.7–93.8
5	76.4 ± 8.6	74.3–78.4	48.7–90.4	79.2 ± 8.0	77.4–81.1	52.9–93.0
6	76.0 ± 9.5	73.7–78.3	49.1–89.9	78.5 ± 8.0	76.6–80.3	41.3–91.7
7	77.4 ± 6.6	75.8–79.0	62.4–91.8	79.5 ± 6.3	78.1–81.0	57.9–90.9
8	75.9 ± 8.6	73.8–77.9	56.5–90.6	80.9 ± 8.5	79.0–82.9	50.7–96.5

).

2.7. Training Impulse (TRIMP) Calculation

Training load was assessed using Banister’s TRIMP [46], which combines exercise duration and intensity based on HR data, calculated as:

$$\mathrm{TRIMP}=\mathrm{t}\times \Delta \mathrm{HR}\times \mathrm{y}$$

where t is duration (min),

ΔHR is the HR reserve:

$$\Delta \mathrm{HR}={\displaystyle \frac{\mathrm{HRavg}-\mathrm{Hrrest}}{\mathrm{HRmax}-\mathrm{Hrrest}}}$$

and y is a weighting factor of $0.64\times \mathrm{e}\left(1.92\times \Delta \mathrm{HR}\right)$ for males or $0.86\times \mathrm{e}\left(1.67\times \Delta \mathrm{HR}\right)$ for females.

Footnote: HRavg is the average HR, HRrest is the resting HR, and HRmax is the maximum HR, estimated as 220 minus age.

HR data were recorded with an HR monitor. This method quantifies training stress in arbitrary units, as described by Morton et al. [47].

2.8. Statistics

Normality was assessed with the Shapiro–Wilk test. Descriptive statistics included means, SDs, medians, and 95% confidence intervals. Two-way ANOVA tested for differences in TRIMP by intervention (HIIT vs. HIPT) and sex (male vs. female), both treated as between-subjects factors, with Bonferroni post hoc corrections. Partial eta squared (η²p) values were calculated as measures of effect size and interpreted according to conventional thresholds: 0.01 (small), 0.06 (medium), and 0.14 (large). Dose–response modeling was performed using relationships between TRIMP and outcomes (%BF, SBP, DBP, VO_2max), explored via Pearson’s r and Kendall’s τ. Non-linearity was checked with Xi correlation [48]. Depending on results, simple, polynomial, or restricted cubic spline regressions (RCS) were applied. Depending on results, simple, polynomial, or restricted cubic spline regressions (RCS) were applied. Restricted cubic spline regression was used to flexibly model potential non-linear dose–response relationships by fitting smooth polynomial segments between predefined knots, while constraining the tails to linearity. This approach allows curvilinear patterns to be captured without overfitting. Model fits were compared using R² and AIC. Further linear models were selected based on parsimony. Effects of intervention type were assessed via regression slopes (b₁) and intercepts (b₀). Differences in slopes were tested following Andrade & Estévez-Pérez [49]. Analyses were performed in Statistica v13.0 (Statsoft Polska, Kraków, Poland). RCS analyses used the rms package in R v.4.4.3 (RStudio 2024.12.0 + 467). A significance level of α = 0.05 was adopted for all tests.

3. Results

In this study, no significant baseline differences were observed between intervention types across sex for any of the analyzed variables (all p > 0.10). Expected sex-related differences were present, particularly in anthropometric characteristics and %BF. Furthermore, a main effect of intervention was identified independent of sex (F = 3.91, p = 0.023, η²p = 0.01), and temporal changes were significantly influenced by HIIT (F = 9.90, p < 0.001, η²p = 0.01).

Overall, the interventions significantly reduced body fat independently of sex, with greater improvements observed in the HIIT group compared with HIPT. Both interventions also lowered SBP in males and females and increased VO_2max, particularly among males participating in HIIT (p < 0.001), confirming a beneficial effect of HIIT on CRF. In contrast, effects on DBP were minimal in both groups. Despite these significant findings, the effect sizes were small, suggesting a modest overall impact of each protocol. Moreover, substantial inter-individual variability in responses was noted.

The present analysis focused on the training load applied each week, quantified using TRIMP, with pre–post intervention differences expressed as deltas (Δ). Descriptive statistics for Δ%BF, ΔSBP, ΔDBP, ΔVO_2max, and mean TRIMP values by sex and intervention type are provided in

Table 2. Descriptive statistics for body fat percentage, systolic and diastolic blood pressure, maximal oxygen consumption, and mean training impulse values.

Variable	HIPT				HIIT
	Mean	±SD	Lower 95%CI	Upper 95% CI	Mean	±SD	Lower 95%CI	Upper 95% CI
Male
Δ%BF [%]	−0.67	1.55	−1.32	−0.01	−1.31	1.66	−1.81	−0.81
ΔSBP [mm/Hg]	−5.38	4.32	−7.20	−3.55	−2.73	5.28	−4.32	−1.15
ΔDBP [mm/Hg]	−2.63	6.25	−5.27	0.02	−0.22	5.00	−1.72	1.28
ΔVO2max [ml/kg/min]	2.14	3.91	0.49	3.79	3.81	4.67	2.41	5.21
TRIMP [score]	12.46	1.74	11.72	13.19	12.90	1.21	12.54	13.27
Female
Δ%BF [%]	−1.54	2.66	−2.33	−0.75	−0.63	2.37	−1.52	0.25
ΔSBP [mm/Hg]	−3.07	4.70	−4.46	−1.67	−4.00	6.89	−6.57	−1.43
ΔDBP [mm/Hg]	−2.00	5.09	−3.51	−0.49	−1.37	5.63	−3.47	0.74
ΔVO2max [ml/kg/min]	0.67	2.84	−0.17	1.52	0.41	3.25	−0.80	1.63
TRIMP [score]	11.59	1.70	11.07	12.08	13.33	1.50	12.78	13.91

Footnote: %BF—body fat percentage; SBP—systolic blood pressure; DBP—diastolic blood pressure; VO_2max—maximal oxygen consumption; TRIMP—training impulse; SD—standard deviation; HIPT—high-intensity plyometric training; HIIT—high-intensity interval training.

. Comparisons of TRIMP are illustrated in Figure 1, where a significant sex × intervention interaction was detected (F = 6.27, p = 0.0133, η²p = 0.043). Post hoc tests with Holm–Bonferroni correction confirmed that females performing HIIT accumulated significantly higher TRIMP compared with females in HIPT (p < 0.001) and with males in HIIT (p < 0.001), while no differences were observed between the male groups (p = 1.000).

Table 3. Correlation coefficients between training impulse and body fat, systolic blood pressure, diastolic blood pressure, and maximal oxygen consumption after applying Pearson’s r, Spearman’s ρ, Kendal’s τ, and Xi ξ. p-values in brackets.

Method	Male				Female
	%BF	SBP	DBP	VO2max	%BF	SBP	DBP	VO2max
Pearson’s r	−0.36 (0.002)	−0.38 (0.001)	−0.34 (0.005)	0.18 (0.131)	−0.15 (0.168)	−0.26 (0.039)	−0.18 (0.161)	0.04 (0.952)
Kendal’s τ	−0.23 (0.002)	−0.28 (<0.001)	−0.22 (0.004)	0.10 (0.257)	−0.09 (0.217)	−0.16 (0.056)	−0.15 (0.088)	0.01 (0.682)
Xi ξ	−0.21 (0.007)	−0.15 (0.005)	−0.10 (0.593)	0.06 (0.912)	−0.14 (0.120)	−0.19 (0.082)	−0.08 (0.222)	0.04 (0.698)

Footnote: %BF—body fat percentage; SBP—systolic blood pressure; DBP—diastolic blood pressure; VO_2max—maximal oxygen consumption.

presents correlation coefficients (Pearson’s r, Spearman’s ρ, Kendall’s τ, and Xi ξ) used to examine the nature of the associations between TRIMP and health-related outcomes. Pearson’s r consistently showed the highest values, indicating predominantly linear relationships. Therefore, linear regression models were selected for the subsequent analyses.

Simple and quadratic regression models were computed and compared across groups. Goodness-of-fit indices for both approaches are reported in

Table 4. Comparisons of linear and quadratic regression models for predicting outcomes.

Group	DV	%BF		SBP		DBP		VO2max
	Model	x	x2	x	x2	x	x2	x	x2
M HIIT	R2	0.06	0.08	0.20	0.20	0.11	0.11	0.07	0.11
	p	0.096	0.169	0.002	0.008	0.035	0.081	0.083	0.083
	AIC	175.5	176.6	272.2	274.1	272.2	274.1	268.2	268.0
M HIPT	R2	0.24	0.25	0.21	0.22	0.20	0.21	0.00	0.02
	p	0.014	0.050	0.024	0.069	0.030	0.080	0.999	0.806
	AIC	87.5	89.3	137.7	139.2	155.8	157.3	138.6	140.1
F HIIT	R2	0.01	0.04	0.03	0.04	0.03	0.10	0.03	0.06
	p	0.601	0.609	0.379	0.602	0.336	0.240	0.366	0.449
	AIC	141.6	142.8	205.1	206.8	192.8	192.6	160.0	161.1
F HIPT	R2	0.14	0.14	0.08	0.10	0.06	0.06	0.02	0.03
	p	0.010	0.036	0.051	0.099	0.108	0.267	0.340	0.575
	AIC	218.4	220.3	273.9	275.0	282.5	284.4	230.7	232.5

Footnote: DV—dependent variable; %BF—body fat percentage; SBP—systolic blood pressure; DBP—diastolic blood pressure; VO_2max—maximal oxygen consumption; HIIT—high-intensity interval training; HIPT—high-intensity plyometric training; M—males; F—females. Statistical significance (p < 0.05) in bold. Gray fields are p-values.

, showing highly comparable results. Pairwise comparisons using F-tests for R² (all p > 0.05) and differences in AIC values (ΔAIC < 2) indicated no meaningful advantage of quadratic models over simple linear models. Accordingly, and consistent with the principle of parsimony, subsequent analyses were conducted using simple linear regressions.

Table 4 summarizes the regression parameters, including intercepts (b₀), raw slopes (b₁), and standardized coefficients (β). For VO_2max in females undertaking HIIT, one outlier was excluded from the analysis. The corresponding dose–response associations between TRIMP and outcome changes (Δ%BF, ΔSBP, ΔDBP, ΔVO_2max) are illustrated in Figure 2.

Comparisons of TRIMP effects on health outcomes were based on slopes (b₁ and β), whereas differences between intervention types were assessed through intercepts. To evaluate whether the dose–response relationship varied by intervention type, interactions between slopes were tested (Table 4). Results indicated outcome-specific and group-specific effects.

Males: Significant associations were observed for both SBP and DBP, independent of intervention type. In HIIT, each TRIMP unit reduced SBP by 1.8 mmHg, compared with 1.13 mmHg in HIPT. Conversely, HIPT exerted stronger effects on DBP (−1.6 mmHg per TRIMP unit) than HIIT (−1.34 mmHg per unit). Intercept comparisons confirmed a distinct upward shift in the regression line for HIIT: SBP (b₀ = 20.48 vs. 8.75; p = 0.006) and DBP (p = 0.022). A significant relationship with %BF was found only in HIPT (−0.4% per TRIMP unit), though the difference in deltas was nonsignificant (p = 0.2267). TRIMP effects on VO_2max were weak and nonsignificant in both groups, although HIIT showed a slightly stronger trend (~+1.0 mL/kg/min per unit). No slope interactions were detected (all p > 0.05). Overall, TRIMP most strongly affected blood pressure, with modality-specific differences: HIIT on SBP and HIPT on DBP.

Females: Associations were weaker overall than in males. TRIMP derived from HIPT demonstrated stronger effects than HIIT, particularly for %BF, where one TRIMP unit corresponded to ~0.6% reduction (p = 0.016). The steeper HIPT slope reflected this effect, as shown by the significant downward shift of the regression line. For DBP, HIPT also induced reductions (~0.9 mmHg per TRIMP unit), though intercepts between interventions were not significantly different. No significant effects were found for VO_2max, and no slope interactions were confirmed (

Table 5. Statistics (b-coefficients and corresponding p-values) derived from multivariate regression analysis to test the differences between intercepts and slopes calculated in simple regressions for relationships between training impulse and body fat percentage, systolic blood pressure, diastolic blood pressure, and maximal oxygen consumption in males and females undergoing high-intensity interval training and high-intensity plyometric training.

DV	M HIPT			M HIIT			F HIPT			F HIIT
	Coeff			Coeff			Coeff			Coeff
	b0	b1	β	b0	b1	β	b0	b1	β	b0	b1	β
%BF	4.62	−0.42	−0.48	2.32	−0.28	−0.20	4.97	−0.56	−0.36	0.35	−0.07	−0.04
SBP	8.75	−1.13	−0.46	20.48	−1.80	−0.41	5.84	−0.77	−0.28	11.35	−1.15	−0.25
DBP	17.25	−1.60	−0.44	17.06	−1.34	−0.32	8.18	−0.88	−0.29	9.19	−0.82	−0.22
VO2max	1.80	0.03	0.01	−8.67	0.97	0.25	−1.36	0.18	0.11	2.88	−0.15	−0.08
test b0	M	%BF	SBP	DBP	VO2max		F	%BF	SBP	DBP	VO2max
Intercepts	p	0.2267	0.0063	0.0217	0.1995		p	0.0165	0.6381	0.2017	0.8701
test b1	M	%BF	SBP	DBP	VO2max		F	%BF	SBP	DBP	VO2max
Slopes	p	0.6008	0.4036	0.7740	0.2203		p	0.2044	0.6602	0.9397	0.4605

Footnote: %BF—body fat percentage; SBP—systolic blood pressure; DBP—diastolic blood pressure; VO_2max—maximal oxygen consumption; coeff—coefficient; HIIT—high-intensity interval training; HIPT—high-intensity plyometric training; M—males; F—females. Statistical significance (p < 0.05) in bold.

).

In summary, TRIMP demonstrated clear dose–response associations with blood pressure and body fat, with distinct patterns by sex and intervention type, whereas effects on VO_2max were negligible.

4. Discussion

This study demonstrated that incorporating HIIT and HIPT protocols into PE lessons over an eight-week period promoted beneficial changes in adolescents’ health indicators. Importantly, by applying TRIMP as a quantitative measure of internal training load, we were able to demonstrate dose–response associations with body fat percentage (%BF), systolic blood pressure (SBP), and diastolic blood pressure (DBP). These associations were primarily linear, confirming that TRIMP provides a sensitive and reproducible method for linking delivered exercise dose to health-related responses. Moreover, intervention effects were sex-specific: in males, HIIT and HIPT produced comparable TRIMP values, while in females, HIIT elicited higher TRIMP than HIPT. Regarding outcomes, HIIT more strongly influenced SBP in males, whereas HIPT was more effective in reducing DBP and %BF in females.

HIIT is extensively studied for its effectiveness in improving physical and cardiorespiratory fitness across different age groups, including adolescents [16]. Evidence consistently shows that very short, intensive sessions can yield substantial physiological benefits within limited time frames. Tabata-style HIIT, as applied in this study [42], has been widely documented to reduce body weight, body fat, and blood pressure, while also improving CRF [50], muscular strength, and metabolic markers [51,52,53,54,55].

Because of these rapid health benefits, HIIT is increasingly implemented in preventive and therapeutic strategies for metabolic diseases such as obesity and type 2 diabetes [56,57,58]. Meta-analyses confirm its dose-dependent effects; e.g., Wang et al. [39] reported significant CRF improvements in obese adults, while Martin-Smith et al. [33] demonstrated similar benefits in adolescents across weight categories. These findings underline the role of short-term, low-volume HIIT interventions within school PE programs. Deng and Wang [39] further emphasized that variations in HIIT protocols can yield comparable improvements in overweight and obese youth.

The present study applied a 2:1 work-to-rest ratio following the original Tabata framework. Our results support prior evidence that both HIIT and plyometric-based HIPT can reduce blood pressure [59,60,61,62,63,64,65], including in hypertensive populations [64]. The reductions observed here align with studies such as Molmen-Hansen et al. [63] and Popowczak et al. [66], who also reported decreases in DBP following structured training.

However, consistent with previous reviews, the short duration and relatively low frequency of our protocol likely limited CRF improvements. Deng and Wang [39] highlighted that HIIT programs longer than 10 weeks and delivered three times per week are most effective in enhancing VO_2max in adolescents. Our intervention, lasting eight weeks with two weekly sessions, may therefore have been insufficient to elicit significant aerobic adaptations. Additionally, we did not control for maturation stage, which could have introduced variability in VO_2max responses [67].

A key contribution of this study is the demonstration of TRIMP’s methodological utility. While HIIT and HIPT both reduced blood pressure and body fat, it was the TRIMP index that allowed us to quantify these effects in a dose–response framework. Notably, HIPT-derived TRIMP was more strongly related to %BF reduction in females, echoing findings by Racil et al. [27], who reported enhanced body composition changes when plyometric elements were added to interval training in adolescent girls. This reinforces the importance of applying TRIMP not only in sports science but also in health-oriented interventions, where understanding the delivered dose is crucial.

Although VO_2max did not improve significantly in our cohort, the observed reductions in %BF, SBP, and DBP represent meaningful health gains. TRIMP proved sufficient to capture these changes, suggesting its potential as a methodological bridge between performance-focused training load research and public health applications.

4.1. Limitations

The relatively small sample size within sex- and intervention-specific subgroups limits the generalizability of these results. Biological maturation was not directly assessed, which may have influenced variability in responses. TRIMP, while a robust proxy for training load, does not capture qualitative aspects of exercise (e.g., movement mechanics or muscle recruitment), which may differ between HIIT and HIPT. Moreover, high inter-individual variability was observed, possibly masking some group-level effects. Nonlinear dose–response patterns were not supported by our analyses, but alternative models (e.g., threshold or plateau effects) may warrant future exploration. Another limitation of this study is the limited availability of comparable research on high-intensity plyometric training (HIPT) in the school setting, particularly in relation to cardiovascular and blood pressure outcomes. While HIIT has been widely investigated, evidence on HIPT remains scarce, which restricts the ability to comprehensively contextualize our findings. Consequently, the interpretation of HIPT-related results should be considered exploratory, and further studies are required to confirm its potential cardiometabolic benefits in adolescents. Finally, as only pre–post changes were analyzed, long-term adaptations to repeated TRIMP exposure remain unknown.

4.2. Future Directions

Future research should: (1) Investigate session-by-session TRIMP values to provide a more granular view of dose–response dynamics; (2) Differentiate responders and non-responders to better understand individual variability in TRIMP-related outcomes; (3) Explore TRIMP interactions with intrinsic and extrinsic factors such as age, maturation status, dietary habits, and habitual physical activity; (4) Extend analyses to longer interventions, incorporating follow-up periods, to assess sustained health adaptations.

5. Conclusions

TRIMP proved to be a practical and sensitive tool for quantifying internal training load in adolescent school-based interventions. In this study, TRIMP was significantly associated with reductions in body fat percentage, systolic blood pressure, and diastolic blood pressure, while no meaningful associations were observed for VO_2max. Importantly, the effects differed by intervention and sex: HIIT produced stronger reductions in SBP, particularly among males, whereas HIPT was more effective in lowering DBP and %BF, especially in females.

These findings demonstrate that TRIMP can capture intervention-specific and sex-specific adaptations, supporting its value as a methodological framework for linking exercise dose with health outcomes in youth. By integrating intensity and duration into a single index, TRIMP extends beyond performance monitoring and offers a reproducible approach for tailoring interval-based training to public health objectives. Future research should verify these associations in longer interventions and explore inter-individual variability, including responder and non-responder patterns.