The paradox of sport: adolescent athletes are highly sedentary across competition, practice, and rest days

Abstract

Background

Elite athletes are highly sedentary outside of sports and must stay active after retiring to remain healthy. While sports are often recommended to improve physical activity, the effect of sport training day on daily physical activity levels in adolescent athletes is not well described. The purpose was to assess (1) physical activity (i.e., step counts and activity score [METs]) and (2) sedentary behavior (e.g., total duration and prolonged [> 1 h] bouts) in male and female adolescent athletes across different training days (i.e., competition, practice, and rest days).

Methods

Adolescent basketball and volleyball athletes wore an accelerometer continuously for 14 consecutive days while in-season. Training days (i.e., competition, practice, and rest days) were tracked for each athlete. In-person testing included anthropometrics, vertical jump height, and body composition via bioelectrical impedance.

Results

Thirty-nine athletes (15.6 ± 0.9 years; n = 24 female, n = 15 male) participated. Athletes were more physically active on competition and practice days (step count [mean ± SD]: competition: 13,021 ± 2,572, practice: 11,735 ± 1,875, rest: 6,742 ± 3,150 [p < 0.001]); activity score [METs]: competition: 35.6 ± 1.0, practice: 34.7 ± 0.9, rest: 33.0 ± 1.3 [p < 0.001]). Athletes had high total sedentary behavior across all training days (competition: 570.4 ± 89.3 min, practice: 619.5 ± 74.6 min, rest: 638.5 ± 90.8 min [p = 0.001]), but the greatest prolonged bouts of sedentary behavior on competition days (competition: 195.4 ± 100.9 min, practice: 115.9 ± 74.0 min, rest: 144.6 ± 96.2 min [p = 0.002]).

Conclusion

Adolescent athletes were highly active on competition and practice days but did not meet physical activity guidelines on rest days, achieving only ~ 6,700 steps/day. Athletes were highly sedentary across all training days, averaging ~ 10 h/day of waking sedentary behavior. Only a third met physical activity guidelines regardless of training day. Adolescent athletes are highly sedentary and many fail to meet physical activity guidelines, challenging the notion that sports alone make individuals sufficiently active.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-026-01539-6.

Background

Sports are frequently proposed to improve physical activity, as greater physical activity is strongly associated with many health benefits [2, 13, 28, 29, 32]. Former elite athletes live longer [10, 31, 47] and may have reduced incidence of cardiovascular disease and cancer compared with the general population [47], although conflicting evidence exists [21, 34]. In contrast, midlife former college varsity athletes have poorer cardiovascular health and body composition, more chronic injuries, lower self-reported physical activity, and worse function than non-athlete controls who were recreationally active during college [49]. Adequate physical activity levels must be sustained after sport retirement to maintain many of the health benefits, as a history of elite sports participation alone does not guarantee function or cardiometabolic health as individuals age [6, 55]. Some evidence suggests that former athletes are more likely to continue physical activity after retirement from sport [3, 54], although not at the same intensity as during sport competition years [51], and further research is needed [8]. Continuing physical activity after sport retirement can be difficult as some former athletes have increased physical activity limitations and disability as they age [6].

Regular, light physical activity, which breaks up prolonged bouts of sedentary behavior, is a strategy to promote optimal health given that the negative health consequences of sedentary behavior exist even in the presence of regular moderate-to-vigorous physical activity [2, 19]. Despite high levels of physical activity, athletes also engage in large amounts of sedentary behavior during leisure time [59]. A scoping review by Izzicupo and colleagues [24] concluded that relatively few studies have evaluated “off-training” physical activity behaviors and that athletes may engage in substantial prolonged sedentary behavior outside of sports training and competition. Some sedentary behavior may be a natural part of recovery for athletes, and thus influenced by practices, competitions, and rest day schedules. However, prolonged sedentary behavior may impede recovery in current athletes and have deleterious implications for their long-term health if the trend becomes a habit after sport retirement.

The long-term effects of sport participation in adolescents have not been well studied because the limited research in this area has focused on collegiate and elite athletes. Although there are over 7.8 million adolescents participating in high school athletics each year in the United States [38], the National Health Interview Survey identified that physical activity levels begin decreasing in adolescents as early as age 15 and continue declining in adulthood [9, 15]. The benefits of physical activity, especially vigorous compared to moderate physical activity, have been described thoroughly for adolescents in general; these benefits include improved cardiorespiratory fitness, lower depression, and stronger bone health [17, 25, 50]. Increasing moderate-to-vigorous physical activity and having frequent breaks in sedentary time to limit prolonged bouts of sedentary behavior are positively associated with overall fitness in adolescents [27]. Overall adulthood physical activity levels may not differ between adolescents who frequently participated in school or club sports and individuals who did not [48].

Limited research has evaluated physical activity levels and patterns in athletes throughout the day (i.e., outside of sport) and by training day. Female collegiate basketball and lacrosse athletes have lower total daily energy expenditure, activity energy expenditure, and physical activity levels on off days compared with game, practice, or conditioning days [37]. Elite male soccer players have high amounts of sedentary behavior [59], although how training day interacts with sedentary behavior is not well studied. It is possible that prolonged sedentary behavior is highest on the days with the greatest amount of energy expenditure as athletes prepare for and recover from competition or other high training loads. Quantifying physical activity and sedentary behavior by the type of training day may be the first step in determining potential physical activity and sedentary behavior patterns that influence recovery in the short-term and, if continued after retirement from competitive sport, could have significant implications for long-term health. If deleterious sedentary behavior patterns (e.g., prolonged sedentary behavior) are present in adolescent athletes and persist in adulthood after retirement from the regularly scheduled moderate-to-vigorous physical activity of sport, identifying and ultimately designing targeted interventions could positively facilitate long-term health in millions each year.

Our purpose was to quantify physical activity and sedentary behavior in male and female adolescent athletes across different training days (i.e., competition, practice, and rest days). Specifically, we aimed to compare: (1) physical activity levels (i.e., step counts and activity score [METs]) in adolescent athletes across training days; and (2) sedentary behaviors (e.g., total duration sedentary behavior, prolonged bouts (> 1 h) of sedentary behavior) in adolescent athletes across training days. Secondarily, we explored the impact of sex on physical activity and sedentary behavior across training days in adolescent athletes. We hypothesized that (1) male and female athletes would have higher physical activity (i.e., step counts and activity score [METs]) on competition and practice days compared with rest days and (2) total and prolonged sedentary behavior would be high across training day with the greatest prolonged sedentary behavior on competition days.

Methods

Study design

The study was approved by the Marquette University Institutional Review Board (IRB #4495). The study employed an observational, quasi-experimental design where adolescent athletes were exposed to the independent variable of sport training day (i.e., competition, practice, rest) based on their teams’ schedules. Study data were collected and managed using REDCap electronic data capture tools [22, 23] hosted through the Clinical and Translational Science Institute of Southeast Wisconsin (grant #: 2UL1TR001436).

Participants

Written parental informed consent and written adolescent informed assent were acquired prior to testing. Participants were enrolled between January 2024 and May 2024 from southeastern Wisconsin. All participants were required to be current high school athletes, aged 13–18 years, who were in-season for high school varsity basketball or club volleyball. Athletes were classified as at least Tier 2 (i.e., Trained/Developmental) [35]. Athletes were excluded if any injury prevented full participation in normal sport activities. Each participant completed an in-person testing session in the community at the location of their team practices. Efforts were made to recruit both male and female athletes from different areas and backgrounds; high school varsity and club sport athletes were included.

Anthropometrics and body composition

Height was measured using a stadiometer (Tetuga). Weight, body mass index (BMI), and percent body fat were assessed via a bioelectrical impedance device (Omron Full Body Composition Monitor and Scale [HBF-514 C]; Omron, Kyoto, Japan) during in-person testing.

Vertical jump

Vertical jump is a common sports performance assessment used to capture the overall profile of an athlete. Maximal vertical jump height was assessed using a Just Jump Mat (Probotics, Huntsville, Alabama, United States of America). Participants were instructed to start on the mat with their hands on their hips and, when ready, jump as high as possible while landing on the mat. Each athlete was allowed one practice jump and then three trials were performed. The best (i.e., highest) jump height was reported.

Sport specialization

Sport specialization level was determined using a 3-question survey [26]: (1) “Do you compete and train in a single sport for more than 8 months of the year?”, (2) “Have you ever quit other sports to focus on a single sport?”, and (3) “Do you consider one sport more important to you than your other sports?”. Each yes counted as a point, with high specializers classified as 3 points, moderate specializers as 2 points, and low specializers as ≤ 1 point.

Athlete identity measurement scale

The Athlete Identity Measurement Scale (AIMS) was completed during in-person testing. AIMS is a 7-item questionnaire assessing the strength of one’s identification and perception as an athlete. AIMS is rated on Likert-type scale with responses ranging from 1 (strongly disagree) to 7 (strongly agree) [5]. The total score is the sum of all 7 items, with a minimum score of 7 and maximum score of 49. Higher AIMS scores indicate greater athlete identity.

Physical activity data using accelerometry

Physical activity was quantified using an activPAL4 activity monitor (PAL Technologies Ltd, Glasgow, Scotland, United Kingdom), a small and light triaxial accelerometer. Each participant wore the activPAL activity monitor continuously for 14 consecutive days including when they were practicing, training, and competing in sports. The activPAL activity monitor was adhered using a waterproof film on the anterior right thigh, approximately 1/3 of the distance from the anterior superior iliac spine to the patella. Participants were instructed to wear the activity monitor 24 h per day including during practices, competitions, and sleep. Participants were instructed to remove the activity monitor only if their skin became irritated or briefly when bathing if the waterproof film was no longer well-adhered. Participants were provided with additional waterproof films to re-adhere the activity monitor as needed and instructed how to do so. Per the 24-hour wear protocol for activPAL, a valid wear day was defined as a day with less than 4 h of non-wear time [40].

Physical activity (i.e., step count and activity score [METs]) and sedentary behavior (i.e., total duration, prolonged [> 1 h] bouts, secondary lying, and seated transport) were processed by the activPAL version 9 software. Step counts are the total number of reciprocal leg movements classified as steps taken per day. The activity score quantifies the total amount of METs used in a day to complete activity. Total sedentary behavior is the total amount of time, in minutes, spent in sedentary behaviors per day, not including sleep (i.e., primary lying time). School (e.g., sitting in class), leisure (e.g., watching TV, reading), and transport time are all included in this total. Prolonged bouts (> 1 h) of sedentary behavior refers to the total sedentary behavior spent in greater than 1-hour periods per day. Secondary lying is any time spent lying down for at least 60 min, not including the longest amount of time spent lying down, which is classified as primary lying (i.e., sleep) [12]. Seated transport measures the total time, in minutes, in motorized transport (e.g., car, train) per day. Each athlete completed a written activity log to track the type of training day (i.e., competition, practice, and rest days), whether the activity monitor was worn upright, when they went to bed and woke up, and if the activity monitor was removed for any reason (e.g., loose waterproof film, bathing, skin irritation). Using the activPAL software, time in bed for sleep was adjusted to match the time in and out of bed on the log if the time differed by more than 30 min.

Training day classification

Training days were classified into one of three mutually exclusive categories: competition was defined as an organized game or tournament; practice included any organized practice, conditioning, or training session with the team; rest days had no organized sport activity. Participants were instructed to pick the highest level of training if multiple types of training were performed on the same day (e.g., a day with competition and practice was recorded as a competition day). Training days were cross verified with coaches and team schedules. In the occasional instance where an activity log was different than a team schedule, the activity log classification was used.

Statistical analysis

Descriptive statistics were reported for demographic and participant characteristics. Data for each outcome measure from the 14-day wear period were averaged for each participant separately for each training day type (i.e., competition, practice, and rest). For each primary outcome measure, a repeated measure analysis of variance (ANOVA) was conducted using the average value for each participant for each training day. Mauchly’s test of Sphericity was tested for each ANOVA, and the Greenhouse-Geiser adjustment was used if the assumption of sphericity was violated (p < 0.05). Post-hoc analyses using the (LSD) method were conducted for outcomes with a statistically significant (p < 0.05) main effect of training day. The LSD method is equivalent to no adjustments, which is appropriate due to the exploratory nature of this study [46]. Descriptive statistics (mean ± standard deviation [SD]) were reported for each training day (i.e., competition, practice, and rest day) outcome; mean group differences and 95% confidence intervals (CI) were reported for post-hoc comparisons. The partial eta squared (η²) effect size and observed power were also reported for each outcome of interest for the ANOVA omnibus (i.e., main effect of training day). Partial eta squared (η²) effect sizes were interpreted as small if η² = 0.01 and above, medium if η² = 0.06 and above, and large if η² = 0.14 and above [45]. These methods were used to determine the robustness of the observed effect.

For our secondary analyses exploring the impact of sex on physical activity and sedentary behavior outcomes and how sex interacts with training day in adolescent athletes, a multi-step process was followed to address sex as a biological variable according to Diester et al. [14] Specifically, we first ran repeated measures ANOVAs for each outcome of interest separately for males and females to determine whether there was an effect of training day within males and females, respectively. Next, we conducted a repeated measures ANOVA with sex as a fixed factor to determine whether there was an interaction effect (sex*training day) or main effect of sex. Linear mixed method analyses were also performed using individual participant data with training day and sex as fixed factors, sex*training day as an interaction effect, and individual participants as a random effect. The statistical analyses were conducted using SPSS Version 29 (IBM, Armonk, New York, United States of America).

Results

Thirty-nine adolescent athletes participated in the study (Table 1). One participant did not have activity data due to the device not recording data (Fig. 1). Of the remaining 38 participants, the number of valid wear days averaged 12.9 ± 2.2 (median: 14, range: 5–14 days). The average number of days competing was 3.0 ± 2.0 (median: 2, range: 0–8 days), practicing was 5.3 ± 1.7 (median: 5, range: 2–10 days), and resting was 5.7 ± 2.5 (median: 6, range: 0–11 days) during the 14-day wear time. This equates to approximately 1.5 competition, 2.7 practice, and 2.9 rest days per week. Thirty-one participants had valid activity data for at least one day for each training day type (i.e., competition, practice, and rest day) (Supplemental Table 1).

Table 1.

Participant descriptive characteristics; values are presented as mean ± standard deviation or number (%)

Participant Characteristic	n	Descriptive Statistics
Age (years)	39	15.6 ± 0.9
Sex	39	Female n = 24 (61.5%) Male n = 15 (38.5%)
Race	39	White/Caucasian n = 27 (69.2%) Black/African American n = 5 (12.8%) More than one race n = 7 (17.9%)
Sport	39	Club Volleyball n = 27 (69.2%, n = 12 female, 15 male) High School Basketball n = 12 (30.8%, n = 12 female)
Sport Specialization Level	39	High n = 23 (59.0%) Moderate n = 10 (25.6%) Low n = 6 (15.4%)
Participant Characteristic	n	Total	Female	Male
Height (cm)	39	171.6 ± 10.1	167.0 ± 7.6*	178.8 ± 9.5*
Weight (kg)	39	65.2 ± 10.1	64.3 ± 9.7	66.6 ± 11.0
Body Mass Index (BMI) (kg/m2)	39	22.1 ± 2.9	23.0 ± 3.1*	20.7 ± 2.0*
Percent Body Fat (%)	38	24.7 ± 10.9	31.9 ± 5.4*	12.5 ± 5.3*
Maximum Vertical Jump Height (cm)	39	44.8 ± 8.6	39.3 ± 5.3*	53.5 ± 4.5*
Athlete Identity Measurement Scale (AIMS, range: 7–49)	39	39.6 ± 6.0	39.9 ± 6.5	39.1 ± 5.3

*Represents a p-value < 0.01 between female and male

Fig. 1.

CONSORT flow diagram

Physical activity

There was a significant effect of training day for step counts (Table 2). Athletes took 1,286 (95% CI: 462, 2,110; p = 0.003) steps more on competition days than practice days and 6,278 (95% CI: 4,765, 7,792; p < 0.001) steps more on competition days compared with rest days. They also took 4,992 (95% CI: 3,758, 6,227; p < 0.001) steps more on practice days than rest days.

Table 2.

Average daily physical activity and sedentary behavior values (n = 31); values are presented as mean ± standard deviation

Outcome	Competition	Practice	Rest	Main Effect of Training Day
				P-Value	Effect Size (η2)	Observed Power
Step Count	13,021 ± 2,572	11,735 ± 1,875	6,742 ± 3,150	p  < 0.001	0.671	1.0
Activity Score (METs)	35.6 ± 1.0	34.7 ± 0.9	33.0 ± 1.3	p  < 0.001	0.614	1.0
Total SB (min)	570.4 ± 89.3	619.5 ± 74.6	638.5 ± 90.8	p  = 0.001	0.205	0.939
Prolonged (> 1 h) Bouts of SB (min)	195.4 ± 100.9	115.9 ± 74.0	144.6 ± 96.2	p  = 0.002	0.184	0.904
Seated Transport (min)	176.6 ± 80.5	136.3 ± 94.2	117.8 ± 71.1	p  < 0.001	0.268	0.970
Secondary Lying (min)	45.9 ± 64.2	37.4 ± 39.0	70.5 ± 67.5	p = 0.065	0.097	0.500

Effect Size is partial eta squared (η²)

Abbreviations: METs Metabolic equivalents, SB Sedentary behavior

P-Values are bolded when statistically significant (p < 0.05)

There was also a significant effect of training day for the activity score (Table 2). Adolescent athletes had 0.9 (95% CI: 0.5, 1.3; p < 0.001) METs more on competition days compared with practice days, 2.5 (95% CI: 1.9, 3.2; p < 0.001) METs more on competition days than rest days, and 1.6 (95% CI: 1.1, 2.2; p < 0.001) METs more on practice days than rest days.

Sedentary behaviors

There was a significant effect of training day for total sedentary behavior (Table 2). Athletes were least sedentary on competition days, with 49.1 (95% CI: -82.8, -15.5; p = 0.006) minutes less total sedentary behavior on competition days compared with practice days and 68.1 (95% CI: -110.5, -25.7; p = 0.003) minutes less total sedentary behavior on competition days compared with rest days. There were no statistically significant differences for total sedentary behavior between practice and rest days, with athletes having 19.0 (95% CI: -51.8, 13.8; p = 0.245) minutes less on practice days.

Prolonged (> 1 h) bouts of sedentary behavior also differed between training day (Table 2). Athletes had 79.5 (95% CI: 31.6, 127.5; p = 0.002) minutes more of prolonged (< 1 h) sedentary behavior on competition days compared with practice days and 50.8 (95% CI: 1.1, 100.5; p = 0.046) more minutes of prolonged sedentary behavior on competition days compared with rest days. There was no significant difference between practice and rest days, with athletes spending 28.7 (95% CI: -64.0, 6.6, p = 0.107) minutes less in prolonged bouts of sedentary behavior on practice days.

There was a significant effect of training day for seated transport (Table 2). Athletes spent 40.3 (95% CI: 8.6, 72.0; p = 0.015) minutes more in seated transport on competition days compared with practice days and 58.7 (95% CI: 33.2, 84.3; p < 0.001) minutes more on competition days than rest days. Athletes tended to engage in 18.5 (95% CI: -1.4, 38.4; p = 0.068) minutes more seated transport time on practice days compared with rest days.

For secondary lying time, there was not a significant effect of training day (Table 2).

Sex factors (Exploratory Analysis)

Univariate ANOVAs for only male athletes had similar main effects as above except for seated transport (Supplemental Table 2). Univariate ANOVAs for only female athletes had similar main effects of training day for step counts, activity score, and seated transport. However, female athletes did not have differences between training day for total sedentary behavior and prolonged bouts of sedentary behavior, and had a significant difference for secondary lying (Supplemental Table 2).

Total sedentary behavior and secondary lying were the only variables to have a significant interaction effect (training day * sex) (Fig. 3A). Among female athletes, there was no significant difference in total sedentary behavior between training days (p > 0.200). Male athletes, in contrast, had significantly less total sedentary behavior on competition days: 87.1 (95% CI: -136.5, -37.7; p = 0.001) minutes fewer compared with practice days and 143.4 (95% CI: -198.6, -88.3; p < 0.001) minutes fewer compared with rest days. Males also had 56.3 (95% CI: -104.4, -8.3; p = 0.023) minutes fewer total sedentary behavior on practice days compared with rest days.

Fig. 3.

Sedentary behavior duration and patterns by training day and sex among adolescent athletes (n = 31): 3A) total sedentary behavior, 3B) prolonged (> 1 h) bouts of sedentary behavior, 3C) seated transport time, and 3D) secondary lying time. Values are presented as average daily values across female and male athletes with error bars representing ± 1 standard deviation (SD). *Represents a p-value < 0.05 for post-hoc comparison between groups. Abbreviations: SB – sedentary behavior

Regarding secondary lying, females had the most time spent on rest days with 56.8 (95% CI: 13.6, 99.9; p = 0.012) minutes more compared with competition days and 44.2 (95% CI: 14.9, 73.5; p = 0.004) minutes more compared with practice days. There was not a significant difference between females on competition and practice days for secondary lying (p = 0.388). Males spent 37.6 (95% CI: 3.1, 72.1; p = 0.034) minutes more in secondary lying on competition days compared with practice days, and there were no differences between competition and rest days (p = 0.431) or between practice and rest days (p = 0.301).

There was no main effect of sex for any of the physical activity or sedentary behavior outcomes (i.e., step count, activity score, total sedentary behavior, prolonged [> 1 h] bouts of sedentary behavior, seated transport, and secondary lying time) (Figs. 2 and 3).

Fig. 2.

Physical activity by training day and sex among adolescent athletes (n = 31): 2A) step counts and 2B) activity score (METs). Values are presented as average daily values across female and male athletes with error bars representing ± 1 standard deviation (SD)

Linear mixed methods

The results of the linear mixed methods analyses were consistent with the ANOVA analyses (Supplemental Table 3). In addition to the statistically significant findings above, the linear mixed method results showed that secondary lying had a main effect of training day (p = 0.021). Total sedentary behavior and seated transport had main effects of sex (p = 0.011 and p < 0.001, respectively). There was also an interaction effect of training day*sex for step count (p = 0.011), activity score (p = 0.009), and prolonged bouts (> 1 h) of sedentary behavior (p = 0.001).

Discussion

Competitive adolescent athletes (high school varsity and club) demonstrated different levels in both physical activity and sedentary behavior depending on their type of sport training day. Our first hypothesis, that adolescent athletes would have higher physical activity levels on competition and practice days compared with rest days, was supported. On rest days, athletes were alarmingly physically inactive and failed to meet recommended minimum physical activity guidelines. Our second hypothesis was also supported, as adolescent athletes averaged approximately 10 h sedentary behavior per day (excluding sleep) and had the greatest prolonged sedentary behavior on competition days. While our study partly supports the common notion that sports enhance moderate-to-vigorous physical activity, our findings also suggest that competitive sports alone are not sufficient at keeping athletes active when they are not playing sports and may even promote high levels of total and prolonged sedentary behaviors.

As expected, adolescent athletes were most physically active on days with structured athletic activities (i.e., competition and practice days). Adolescent athletes averaged 1.5 days of competition, 2.7 days of practice, and 2.9 days of rest per week. Moderate-to-vigorous physical activity has been associated with overall fitness in youth, independent of sedentary behavior [27], suggesting that the adolescent athletes in the present study who, on average, competed or practiced over 4 days/week would be at a health advantage due to the regular moderate-to-vigorous physical activity. However, adolescent athletes averaged only 6,742 steps on rest days, which is considerably less than the typical daily step guidelines of 11,000–12,000 steps per day for adolescents [11, 33]. One explanation for the low step count on rest days is relating to the recovery process from sport. The National Athletic Trainers’ Association recommends that adolescent athletes should have a minimum of two days of rest per week from sport competition and training [1]. Adolescent athletes in our study averaged 2.9 rest days per week suggesting that adequate time was available for recovery and thus should not alone explain the low step count on rest days. Furthermore, active recovery on rest days is more productive than passive or sedentary recovery in trained individuals due to improving blood flow, alleviating muscle soreness, and decreasing inflammation [20, 52].

Children aged 6–11 years walk 10,000–16,000 steps per day [57], but average step counts decrease in adolescence, and by age 18 individuals average ~ 8,000–9,000 steps per day [57]. While different step count minimums have been suggested, 11,000–12,000 steps per day is often recommended for adolescents [11, 33]. Averaging daily step counts across training day in the present study showed that only 37% (n = 14/38) achieved 10,000 steps/day, 34% (n = 13/38) achieved 11,000 steps/day, and 18% (n = 7/38) achieved 12,000 steps/day (Fig. 4). Only a third (34%, n = 13/38) of the adolescent athletes would have met the threshold of 11,111 steps/day proposed by Mayorga-Vega and colleagues, which may distinguish between adolescents who do and do not meet overall physical activity recommendations [33]. Adolescent athletes’ lack of meeting daily step physical activity guidelines on days without structured athletic activities may be concerning and imply underlying physical inactivity outside of sports. Additionally, all the athletes in the present study were cleared for full sports participation, but prior research indicates that injured athletes are less active than healthy athletes and non-athletes [4, 30, 49, 56], suggesting these concerns may be even greater in athletes with a history of injury.

Fig. 4.

Percentage of individuals (n = 38) in the present study who met different recommended daily step count thresholds from 8,000 to 12,000 steps per day

Several studies show that the long-term health benefits associated with participating in competitive sports are only maintained if physical activity is continued after sport retirement [55], suggesting that current physical activity may be more important for disease prevention than prior participation in athletics [6]. Scicluna et al. found that adolescent sport participation did not result in different levels of physical activity in adulthood compared with non-athletes [48], whereas Ravi et al. found that adolescent sport participation is associated with higher leisure-time physical activity in midlife [44]. However, former adolescent athletes only had higher leisure-time physical activity in self-reported physical activity and not in accelerometer-measured physical activity in midlife [44], potentially indicating a discrepancy in perception versus reality of physical activity. One potential explanation is that once individuals are no longer participating in sport, they lose the structured physical activity associated with competitions and practices and thus do not independently maintain higher levels of physical activity. Former athletes may maintain their strong sense of athletic identity after sport retirement [7], which could be a reason for the discrepancy in actual physical activity (accelerometry) versus perceived physical activity (self-reported). Further research using accelerometers in former athletes as well as longitudinal tracking of (former) athletes throughout their sport retirement is warranted.

Adolescent athletes were highly sedentary, averaging approximately 10 h per day of total sedentary behavior across training day. High amounts of total sedentary behavior during leisure time in athletes is similar to what was previously found in professional male soccer players [59]. The professional soccer players averaged just over 8 h/day of sedentary behavior, but their average wear time of the accelerometer was only about 10.5 h/day, potentially underestimating the true amount of sedentary behavior compared with our study. A threshold of 6–8 h/day of total sitting or sedentary time has been suggested for risk of all-cause and cardiovascular disease mortality [41]. Regardless of training day, male and female adolescent athletes were well above this threshold for total sedentary behavior [41], indicating that competitive sports do not ameliorate and may even exacerbate risk for high levels of sedentary behavior. Although adolescent athletes had more total sedentary behavior on rest days, they had more prolonged bouts of sedentary behavior and more time spent in seated transport on competition days compared with both practice and rest days. A possible explanation for the increase in prolonged bouts of sedentary behavior includes athletes wanting to rest before competition or recover after competition. Another potential explanation for the prolonged sedentary behavior on competition days includes the increased time in seated transport on competition days, perhaps due to the increased travel needed to arrive at games or tournaments. Regardless of training day, athletes were attending school on most weekdays at the time of data collection. A possible explanation for the overall high sedentary behaviors could be explained in part by large amounts of time sitting in class, highlighting a potential benefit of creating more active classrooms. However, athletes were also highly sedentary on weekend days when they were not in school (averaging over 9.5 h total sedentary behavior on weekends) and had greater prolonged bouts of sedentary behavior on weekends than weekdays (Supplemental Table 4). More breaks in sedentary behavior, or reducing prolonged bouts of sedentary behavior, is positively associated with overall fitness [27] and may be an important consideration for educating athletes for long-term health particularly after they retire from competitive sports.

We also explored sex differences in physical activity and sedentary behavior in adolescent athletes in the present study. Female adolescent athletes had consistent total sedentary behavior across all training days whereas males had more total sedentary behavior on practice and rest days, reaching nearly 11 h per day. Palomaki et al. found that people who were active in youth sports had more healthy habits in adulthood than those that were non-active in their youth; however, only females had significantly greater odds of having many healthy habits in adulthood whereas males did not [39]. Traditionally, research has focused more on male athletes; [42] further research on sex differences in athletes is warranted.

Several anthropometrics and physical performance measures differed between male and female athletes. As expected, males were taller, had lower BMI, were leaner, and jumped higher than females, which is consistent with prior literature [53]. Athletes did not differ in weight or AIMS score by sex. In the general population, percent body fat decreases from 21% to 13% for boys from 12 to 18 years, while girls remain consistent at 24–27% [58]. Male adolescent athletes in our sample were slightly below these norms and female adolescent athletes were above these norms. Adolescent athletes score similarly on the AIMS as previously reported averages ranging from 37 to 40 [18, 36]. Males are occasionally reported to have higher AIMS scores [5], but this was not seen in our sample of adolescent athletes.

Clinical and behavioral implications

Our results highlight a potential need to increase physical activity and limit sedentary behaviors in adolescent athletes. One possible solution is to implement active breaks during school hours, such as longer walking breaks between classes, short movement breaks during classes, recesses, or a daily gym class. Regular movement outside of school and on weekends is important too, including social interactions like going for a bike ride with friends. Movement reminders on wearable devices (e.g., smartwatches) may facilitate movement breaks. Athletes may also benefit from using active recovery strategies such as walking, dynamic stretching, yoga, Pilates, hiking, or other recreational activities to break up and reduce sedentary behavior, particularly on rest days. Educating coaches, teachers, parents, and adolescents to promote physical activity and reduce prolonged sedentary behavior – even among adolescent athletes who are seen as highly active – may be beneficial.

Limitations

The sample size of adolescent athletes was small, recruited from a single geographic region, and limited in sport variety, which was done to ensure that participants were in-season and enhance internal consistency. However, sport type and seasonality may influence physical activity and sedentary behavior; future research should include a greater variety of adolescent sports and report the presence and impact of simultaneous multiple sport participation. Only athletes who were cleared for full sports activity were eligible for this study as musculoskeletal injuries, which are sustained at much higher levels in athletes than non-athletes, are known to decrease physical activity levels. Incorporating injury history into future research is essential for understanding long-term health of adolescent athletes. Another limitation is that the activPAL software algorithm may underestimate the step counts at higher running speeds [43], although this limitation should not affect sedentary behaviors, and it is likely that the discrepancy between step counts on rest days versus competition and practice days could be even greater than reported.

A strength of our study was that both males and females were included, although sex comparisons were exploratory in nature given the small sample size of male athletes. Further research in adolescent athletes is needed to fully elucidate sex differences in larger sample sizes of athletes from diverse sports. Longitudinal follow-up would be beneficial for future research to determine how physical activity and sedentary behavior change after retirement from sport. The use of objective, continuous accelerometry for 14 consecutive days was another major strength, as it measures physical activity and sedentary behavior more accurately than self-report. Self-reported physical activity often over-reports vigorous physical activity and under-reports sedentary behavior [16], thus accelerometry should continue to be the chosen method in future research on adolescent athletes. We relied on self-reported classification of training day, and although the activity logs were cross-referenced with team schedules, there is a possibility some days were misclassified. Future research should also consider physical activity and sedentary behavior in-season compared with off-season.

Conclusion

Sports training day impacts adolescent athletes’ physical activity and sedentary behaviors, including step count, METs, total sedentary behavior, prolonged (> 1 h) bouts of sedentary behavior, and seated transport time. Adolescent athletes were highly physically active on competition and practice days but are also highly sedentary across all training days, averaging approximately 10 h sedentary behavior per day. Moreover, athletes did not meet physical activity guidelines on non-sport (i.e., rest) days, averaging only 6,742 steps per day on rest days, and only a third met daily step count guidelines regardless of training day.

Abbreviations

AIMS

Athlete Identity Measurement Scale

METs

Metabolic Equivalents

SB

Sedentary Behavior

Authors’ contributions

JHS contributed to conception and design, funding acquisition, data acquisition, analysis and interpretation, and manuscript drafting and incorporating revisions. RDG and AML contributed to data acquisition, analysis and interpretation, and critical review. NB contributed to data analysis and interpretation, and critical review. SH contributed to conception and design and critical review. JJC contributed to conception and design, funding acquisition, data analysis and interpretation, manuscript drafting and incorporating revisions and critical review. All authors guarantee the accuracy of the data and approve the final version of the submitted manuscript.

Funding

This study was funded by a Marquette University Athletic and Human Performance Research Center Pilot Grant (JHS, JJC). The study was also supported in part by the National Institutes of Health (NIH), Office of the Director (OD) and the National Institute of Dental and Craniofacial Research (NIDCR), through an NIH Director’s Early Independence Award to Dr. Jacob J. Capin (NIH DP5-OD031833). Data management was supported by the NIH grant 2UL1TR001436. The views expressed herein are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. The funding sources did not have any role in designing the study or interpreting the results.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Marquette University Institutional Review Board (IRB #4495). Written parental informed consent and written adolescent informed assent were acquired prior to testing. This study adhered to the ethical principles outlined in the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.