Baseline Aerobic Fitness in High School and College Football Players: Critical for Prescribing Safe Exercise Regimens

Abstract

Background:

Nontraumatic fatalities occur on a regular basis in high school (HS) and college football athletes, primarily in obese linemen performing high-intensity exercise. One contributing factor to these deaths may be a mismatch between baseline aerobic (cardiorespiratory) fitness and exercise regimens.

Hypothesis:

There is a wide range of aerobic fitness in HS and college football players. Body mass index (BMI) is a safe and simple method for estimating baseline aerobic fitness.

Study Design:

Retrospective cohort study.

Level of Evidence:

Level 3.

Methods:

A retrospective review was performed on 79 HS football athletes who had VO_2Peak (mL·kg⁻¹·min⁻¹) measured during the offseason. Multivariate regression analysis was used to determine if BMI (obese, overweight, and normal; kg/m²), position played (linemen vs other), year in school (freshmen vs other), and/or race (African American vs White) were risk factors for poor aerobic fitness. A separate cohort of 135 (48 HS; 87 college) football athletes performed a 6-minute run test to determine speed (miles/min), extrapolate VO_2Max, and calculate reference values for suggested upper threshold safe starting speeds (85% of maximum) for aerobic training based on BMI. The relationship between BMI and VO_2Peak was assessed. The exercise regimens (speeds) of 2 collegiate football fatalities from the public domain were used to predict their VO_2Max values.

Results:

Mean VO_2Peak (mL·kg⁻¹·min⁻¹) was 38.5 ± 8.6 (range 19.1-60.6); when grouped by BMI, low scores (<40) were found in 87.5% of obese (32.4 ± 7.7), 47.8% of overweight (40.8 ± 7.6), and 45.2% of normal (41.4 ± 7.8) athletes. VO_2Peak was significantly lower in linemen (32.8 ± 6.4; P = 0.007) compared with nonlineman (41.8 ± 7.9), and in obese players (by BMI; 32.4; P = 0.019) compared with nonobese players (41.4 ± 7.6), but did not differ by age, year in school, or race. Means for speed (min/mile) and extrapolated VO_2Max (mL·kg⁻¹·min⁻¹) for the 6-minute run test by BMI groups were both significantly different (P = 0.001) for normal (7.0 ± 0.6; 51.1 ± 2.6), overweight (7.6 ± 0.8; 46.5 ± 3.2), and obese (8.9 ± 1.5; 36.8 ± 5.9) athletes. There was a significant negative correlation (r = −0.551; P = 0.001; R² = 0.304) between VO_2Peak and BMI. Safe starting speed recommendations for running 1 mile range from 7.3 to 12.1 min/mile for BMIs 20 to 40 kg/m² for HS and college athletes. For the 2 fatalities (mean, BMI of 36.5 kg/m²) repetitive sprint speeds were 49 and 89% higher than our safe starting speeds for their BMI.

Conclusion:

A large spectrum of baseline aerobic fitness was noted in HS and college football players. Obese players and linemen had statistically lower baseline aerobic fitness, a major risk factor for possible heat illness. BMI is an acceptable surrogate for VO_2Peak and can be employed to develop safe training regimens without the need for a maximum fitness test, which can place the athlete at risk for a medical event.

Clinical Relevance:

Knowledge of BMI provides an estimate of baseline aerobic fitness and a foundation for prescribing safe, individualized exercise regimens.

The literature on prescribing appropriate exercise regimens based on aerobic (cardiorespiratory) fitness level in football players is limited. As a result, overexertion of athletes during conditioning sessions can lead to overuse injuries, rhabdomyolyis,^9,15 heat stroke, and even fatalities.^7,8 One study of high school (HS) and college football players between 1998 and 2008 reported an average of 9.4 nontraumatic fatalities annually. ⁸ The majority (86.6%) of fatalities occurred when high intensity training, such as repetitive sprints or fitness tests with limited rest periods, was being performed during practice, often in conjunction with punishment drills. ⁸ Linemen represented two-thirds of all fatalities and 97% of the exertional heat stroke fatalities. ⁸ The average body mass index (BMI) was 32.6 kg/m² for all fatalities and 36.4 kg/m² for those with exertional heat stroke fatalities. ⁸ The report revealed a lack of individualized training programs such that athletes, regardless of aerobic fitness, were often required to perform the same drill. ⁸ In contrast to traumatic fatalities in HS and college football, which have declined 4- to 5-fold since the 1960s, the annual incidence of nontraumatic fatalities has remained constant over the past 60 years. ⁸ In addition to fatalities, numerous cases of nonfatal exertional rhabdomyolysis (ER) occur during training. ¹⁵ The incidence of team ER outbreaks due to excessive conditioning drills and a requirement for all athletes to exert at the same pace and intensity, are equally concerning. ¹⁵ The incidence of ER cases presenting to emergency departments in the United States has increased over the past decade, with football representing about 50% of all sports-induced ER cases. ⁹ The recurring pattern is football players being exposed to drills that they are unaccustomed to, which results in overexertion.

Knowledge of aerobic fitness is essential for developing appropriate football exercise regimens that maximize aerobic training and minimize risk of injuries and fatalities. Maximal oxygen uptake exercise testing (VO_2Max), the gold standard for measuring aerobic fitness,^14,17 is confirmed when oxygen uptake plateaus during maximal physiologic exertion. Since the plateau portion of the VO_2Max test can be difficult to achieve, VO_2Peak—the highest VO₂ observed during a maximal effort exercise test—is a valid substitute for VO_2Max in assessing aerobic fitness.^14,17,37,38 For the purposes of this study, the terms VO_2Max and VO_2Peak will be used throughout the article to describe aerobic fitness. Peak heart rate (HR_Peak) was also measured, as it is critical for monitoring exercise programs for cardiac drift and exercise above VO_2Max. ¹²

Although VO_2Max testing is supported by strong scientific evidence, data on professional and university football players are limited.^33,47 Importantly, no known studies provide data on football players at the HS level despite its being the most popular HS sport in the United States. ³¹ Instead, analysis of the physical characteristics of football players has focused on body composition, strength, jumping, and sprinting capabilities.^18,19

The 6-minute run test is an effective tool for predicting aerobic fitness with limited data on football players. ²⁷ The disadvantages of both VO_2Max and the 6-minute run tests is that both require a maximum effort, which has a small risk of a cardiac event, ²² heat illness, or untoward event in an athlete with sickle cell trait. ⁸ To avoid these risks, we also assessed whether BMI could be an accurate method for predicting aerobic fitness.

The purpose of this study was to determine the spectrum of baseline aerobic fitness in HS football players by measuring VO_2Peak and determining if VO_2Peak could be predicted from BMI, position played, year in school, and/or race. A secondary purpose was to compare 6-minute run test times and VO_2Peak in both HS and college athletes for use as a potential substitute for VO_2Max. Based on speeds from the 6-minute run, an upper limit for initial, safe training speeds was determined for individual and group BMI values. A third goal was to assess whether BMI can accurately predict baseline fitness in order to avoid a maximum aerobic test. Last, speeds employed during exercise that led to fatalities in 2 obese collegiate football players were analyzed to predict VO_2Max and assess the relative safety of these conditioning regimens.

Methods

The current study was deemed exempt by the institutional review board since retrospective, de-identified data collected were used. Demographic data included age, height, weight, position played (linemen or other), year in school (freshmen vs upperclassmen defined as sophomores, juniors, and seniors), and race (White vs African American) for each athlete. BMI (kg/m²) was calculated from height and weight and grouped as normal (18.5-24.9), overweight (25-29.9), or obese (≥30). ¹⁶ All testing was performed at 2 HSs and 1 college during the offseason month of May. Initially, coaches agreed to participate as a potential method to monitor baseline fitness, develop safe aerobic fitness programs to improve performance, and reduce the risk of injury, particularly heat illness. Athlete participation was encouraged, but voluntary.

Athletes (n = 79) from 2 HS teams were connected to a Vista-MX2 Metabolic Measurement System^21,40 for incremental treadmill exercise testing. VO₂ (mL·kg⁻¹·min⁻¹), HR, and respiratory exchange ratio (RER) were recorded each minute throughout the testing protocol. ²¹ Athletes walked at a 3.0 mph pace with a 3% incline. After 3 minutes, the athlete was asked to jog on a 5% incline at 4.5 to 5 mph with incline increasing 2% every minute with an optional speed increase until the athlete reached a RER of 1.15—a level above the criterion for maximal effort attainment. ²⁹ At this point, VO_2Peak and HR_Peak were recorded. VO_2Peak was usually achieved within 8 to 10 minutes of initiation of exercise. When a treadmill was not available, a vertical cardio climber (VersaClimber) ⁴¹ was employed as an alternative exercise machine with a similar protocol. This procedure elicits a comparable VO_2Peak with treadmill running. ¹¹ The athlete started climbing at a rate of 20 to 40 ft/min, with speed increasing by 20 ft/min every minute until VO_2Peak was achieved.

A 6-minute run test ²⁷ was performed on a separate cohort of 87 collegiate (1 team) and 48 HS (1 team, different from team tested for VO_2Peak) football players during the offseason month of May. Athletes ran as far as possible in 6 minutes. The total distance was recorded as miles/min and yards/s and then converted into time to run 1 mile (inverse of miles/min). VO_2Peak was estimated from the mile speed (in minutes), age, gender, and BMI by using a validated multiple regression equation: 8.41 × (time) + 0.34 × (time) ² + 0.21 × (age × gender) − 0.84 × (BMI) + 108.94. ¹³ Women are classified as 0 and men as 1.

Using linear regression, maximum speeds were calculated for individual BMI values of 20 to 40 kg/m². Maximum speeds for each BMI were multiplied by 85% to calculate the upper threshold for safe starting speeds to promote aerobic training. ²⁸

Statistical Analysis

Analysis performed using STATA 15.1 (StataCorp LLC) resulted in sufficient power for the VO_2Peak and 6-minute run tests. Data were analyzed by using SAS 9.4 (SAS Corp) and IBM SPSS 25 (IBM Corp). Descriptive statistics for age, BMI groups, VO_2Peak, and HR_Peak included calculating means, standard deviations, and ranges. The frequency of VO_2Peak values was assessed and further subdivided as percentage less than 40 mL·kg⁻¹·min⁻¹, which has previously been classified as poor heat tolerance ⁴ and aerobic fitness in the lowest 25th percentile for the general population of 20- to 29-year-old men. ³⁶ Plots for the distribution of VO_2Peak values and VO_2Peak versus BMI were prepared.

Pearson correlation coefficients were performed to assess correlation of VO_2Peak and BMI groups. Simple linear models were used to assess relationship between BMI values and VO_2Peak and time to run 1 mile. Multivariate linear models, based on unstandardized coefficients, were evaluated to determine predictors of miles per minute and VO_2Peak. The following potential predictors were assessed in these models: age, HR_Peak, BMI groups, position played, year in school, and race.

Analysis of variance was used to assess differences in mean (±SD) maximum speeds (yards/s and time to run 1 mile) and VO_2Max (mL·kg⁻¹·min⁻¹) extrapolated between BMI groups in HS and college athlete groups. Bonferonni correction was used for multiple comparison of BMI categories. Independent t test was used to detect differences in mean (±SD) maximum speeds (yards/s and time to run 1 mile) and VO_2Peak (mL·kg⁻¹·min⁻¹) extrapolated between HS and college athletes in normal, overweight, and obese groups. A BMI optimal cutoff point was calculated using the Youden Index to distinguish between low and high VO_2Peak.

Results

VO_2Peak Parameters

The mean VO2_Peak in HS athletes was 38.5 ± 8.6 (range, 19.1-60.6) mL·kg⁻¹·min⁻¹ (Figure 1). More than half (n = 46, 58%) had a VO_2Peak less than 40 mL·kg⁻¹·min⁻¹, which included 87.5% of obese (32.4 ± 7.7), 47.8% of overweight (40.8 ± 7.6), and 45.2% of normal (41.4 ± 7.8) weight athletes (Figure 2). Mean age and BMI were 16.9 ± 1.1 years (range, 15-18 years) and 27.8 ± 6.5 kg/m² (range, 18.8-50.2 kg/m²), respectively. Participants consisted of 49 (62.0%) African Americans and 29 (36.7%) White athletes; ethnicity/race was unknown in 1 (1.3%) athlete. Thirty (37.9%) athletes were linemen and 47 (59%) were other than linemen; the positions of 2 (2.5%) were unknown. Athletes were tested either on a treadmill (n = 42; 53.2%) or the VersaClimber (n = 37; 46.8%).

Figure 1.

Frequency of VO_2Peak.

Figure 2.

Percentage of athletes with VO_2Peak values less than or greater than 40 mL·kg⁻¹·min⁻¹ in each body mass index category.

Figure 3 presents a scatter plot of the relation between VO_2Peak and BMI: A significant negative correlation (r = −0.551; P = 0.001, R² = 0.304) was noted. Every 1-unit increase in BMI resulted in a VO_2Peak decrease of 0.73 mL·kg⁻¹·min⁻¹. Mean VO_2Peak values of BMI classifications were not equal (F = 10.63; P = 0.001). According to Bonferonni correction used for multiple comparison of BMI categories, the mean VO_2Peak values were significantly lower in obese (32.4 ± 7.7 mL·kg⁻¹·min⁻¹, P = 0.019) compared with overweight (40.8 ± 7.6 mL·kg⁻¹·min⁻¹; P = 0.001) and normal weight (41.4 ± 7.8 mL·kg⁻¹·min⁻¹; P = 0.001) football players. Also, mean VO_2Peak values were significantly lower for lineman (32.8 ± 6.4 mL·kg⁻¹·min⁻¹, P = 0.007) compared with other positions (41.8 ± 7.9 mL·kg⁻¹·min⁻¹) (P = 0.001). No significant differences were noted in VO_2Peak between normal and overweight players (P = 0.999), and VO_2Peak did not differ significantly by years in school (P = 0.853), race (P = 0.865), or age (P = 0.368). Mean HR_Peak values were 181 ± 12.0 for normal, 185.5 ± 11.9 for overweight, and 180.3 ± 9.2 for obese BMI groups. Overall, BMI was a good predictor for classifying players with low VO_2Peak (area under the curve = 0.7). A significant positive correlation was observed between VO_2Peak and HR_Peak (r = 0.359; P = 0.001).

Figure 3.

Scatter plot of VO_2Peak and body mass index (BMI).

Six-Minute Run Test

The mean age was 15.3 ± 1.1 (range, 14-18) and 21.7 ± 1.4 (range, 19-25) years, and the mean BMI was 26.3 ± 5.1 (range, 17.8-40.4) and 29.2 ± 5.0 (range, 21.2-44.3) kg/m² for HS and college participants, respectively. Mean yards per second, time to run 1 mile, and predicted VO_2Max values demonstrated significant differences by BMI groups (P = 0.001) (Table 1). All HS athletes with normal BMI and all college athletes with normal or overweight BMI had predicted VO_2Max values more than 40 mL·kg⁻¹·min⁻¹. In contrast, 42.9% of overweight HS athletes and 100% of HS and 66.7% of college obese athletes had predicted VO_2Max values less than 40 mL·kg⁻¹·min⁻¹. A significant difference in mean speeds were noted between HS and college overweight and obese BMI athletes (Table 2).

Table 1.

Maximum speeds and extrapolated VO_2Peak between body mass index groups

		College					High School
		N	Mean	SD	F	P	N	Mean	SD	F	P
Yards/s	Normal	17	4.2	0.4	16.5	0.001a	23	3.8	0.3	9.5	0.001a
	Overweight	43	3.9	0.5			16	3.7	0.5
	Obese	27	3.4	0.6			9	3.1	0.5
	Total	87	3.8	0.6			48	3.7	0.5		0.109
Time to run 1 mile	Normal	17	7.0	0.6	18.1	0.001a	23	7.7	0.6	8.5	0.001a
	Overweight	43	7.6	0.8			16	8.0	1.5
	Obese	27	8.8	1.5			9	9.8	2.0
	Total	87	7.9	1.3			48	8.2	1.5		0.157
VO2Max, mL·kg−1·min−1	Normal	17	51.1	2.6	73.3	0.001a	23	48.7	2.2	67.2	0.001a
	Overweight	43	46.5	3.2			16	44.6	3.9
	Obese	27	36.8	5.9			9	34.1	4.0
	Total	87	44.4	6.8			48	44.6	6.3		0.853

^aSignificant at α = 0.05.

Table 2.

Maximum speeds and extrapolated VO_2Max between high school and college athletes

Body Mass Index	Measure	Group	n	Mean	SD	t-value	P
Normal	Yards/s	High school athletes	23	3.8	0.3a	−3.45	0.001*
		College athletes	17	4.2	0.4
	Time to run 1 mile	High school athletes	23	7.7	0.6a	3.48	0.001*
		College athletes	17	7.0	0.6
	VO2Peak	High school athletes	23	48.7	2.2a	−3.13	0.003*
		College athletes	17	51.1	2.6
Overweight	Yards/s	High school athletes	16	3.7	0.5	−1.24	0.221
		College athletes	43	3.9	0.5
	Time to run 1 mile	High school athletes	16	8.0	1.5	1.45	0.153
		College athletes	43	7.6	0.8
	VO2Peak	High school athletes	16	44.6	3.9	−1.95	0.055
		College athletes	43	46.5	3.2
Obese	Yards/s	High school athletes	9	3.1	0.5	−1.46	0.153
		College athletes	27	3.4	0.6
	Time to run 1 mile	High school athletes	9	9.8	2.0	1.43	0.162
		College athletes	27	8.8	1.5
	VO2Max	High school athletes	9	34.1	4.0	−1.23	0.229
		College athletes	27	36.8	5.9

^aSignificant at α = 0.05.

Figure 4 presents a scatterplot depicting the relation between time to complete 1 mile and BMI for HS (r = 0.574, P = 0.001; R² = 0.329) and college (r = 0.686; P = 0.001; R² = 0.47) athletes. Every 1-unit increase in BMI resulted in a 9.6- and 10.2-second increase in time to complete 1 mile for HS and college students, respectively. Every 1-unit increase in BMI resulted in a decrease in speed of 0.07 yards per second for both HS and college students.

Figure 4.

Scatter plot of body mass index (BMI) and speed for high school and college athletes.

Based on the data collected from the collegiate athletes and the moderate to strong relationships between speed, VO_2Peak, and BMI, ¹ we developed safe starting speeds for initial football training for HS and college athletes (Table 3).

Table 3.

Extrapolated VO_2Max and suggested maximum safe “starting run times” ^a

		Extrapolated VO2Max, mL·kg−1·min−1		Yards/s		50 Yards, s		110 Yards, s		Mile, s
	BMI, kg/m2	HS	College	HS	College	HS	College	HS	College	HS	College
Normal	20	51.7	55.9	4.1	5.5	14.3	12.5	31.4	27.5	501.9 (8 min 22 s)	440.5 (7 min 21 s)
	21	50.6	54.6	4.0	5.4	14.6	12.9	32.1	28.3	513.2 (8 min 33 s)	452.5 (7 min 33 s)
	22	49.5	53.4	3.9	5.2	14.9	13.2	32.8	29.0	524.5 (8 min 45 s)	464.5 (7 min 45 s)
	23	48.4	52.1	3.9	5.1	15.2	13.5	33.5	29.8	535.8 (8 min 56 s)	476.5 (7 min 57 s)
	24	47.2	50.9	3.8	5.0	15.5	13.9	34.2	30.5	547.1 (9 min 7 s)	488.5 (8 min 9 s)
Overweight	25	46.1	49.6	3.7	4.9	15.9	14.2	34.9	31.3	558.4 (9 min 18 s)	500.5 (8 min 21 s)
	26	45.0	48.4	3.6	4.8	16.2	14.6	35.6	32.0	569.7 (9 min 30 s)	512.5 (8 min 33 s)
	27	43.9	47.1	3.5	4.6	16.5	14.9	36.3	32.8	581.0 (9 min 41 s)	524.5 (8 min 45 s)
	28	42.8	45.9	3.5	4.5	16.8	15.2	37.0	33.5	592.3 (9 min 52 s)	536.5 (8 min 57 s)
	29	41.6	44.6	3.4	4.4	17.1	15.6	37.7	34.3	603 (10 min 4 s)	548.5 (9 min 9 s)
Obese	30	40.5	43.4	3.4	4.3	17.5	15.9	38.4	35.0	614.9 (10 min 15 s)	560.5 (9 min 21 s)
	31	39.4	42.1	3.3	4.3	17.8	16.3	39.1	35.8	626.2 (10 min 26 s)	572.5 (9 min 33 s)
	32	38.3	40.9	3.2	4.2	18.1	16.6	39.8	36.5	637.5 (10 min 37 s)	584.5 (9 min 45 s)
	33	37.2	39.6	3.2	4.1	18.4	16.9	40.5	37.3	648.8 (10 min 49 s)	596.5 (9 min 57 s)
	34	36.0	38.4	3.1	4.0	18.8	17.3	41.3	38.0	660.1 (11 min 0 s)	608.5 (10 min 9 s)
	35	34.9	37.1	3.1	3.9	19.1	17.6	42.0	38.8	671.4 (11 min 11 s)	620.5 (10 min 21 s)
	36	33.8	35.9	3.0	3.9	19.4	18.0	42.7	39.5	682.7 (11 min 23 s)	632.5 (10 min 30 s)
	37	32.7	34.6	3.0	3.8	19.7	18.3	43.3	40.3	694.0 (11 min 34 s)	644.5 (10 min 45 s)
	38	31.5	33.4	2.9	3.7	20.0	18.7	44.1	41.0	705.3 (11min 45 s)	656.5 (10 min 57 s)
	39	30.4	32.1	2.9	3.6	20.4	19.0	44.8	41.8	716.6 (11 min 57 s)	668.5 (11 min 9 s)
	40	29.3	30.9	2.8	3.6	20.7	19.3	45.5	42.5	727.8 (12 min 8 s)	680.5 (11 min 21 s)

BMI, body mass index; HS, high school.

^aTimes were derived based on estimated 85% of maximum speed for BMI values (20-40 kg/m²) and are expressed in seconds (s) or seconds and minutes (min) and seconds. The estimated confidence interval for time to run 1 mile is approximately 20 seconds. This cannot be translated for distances other than 1 mile. Thus, the mile time for a BMI of 20 kg/m² would range from 420.5 to 500.5 seconds.

Discussion

VO_2Peak

One primary finding in this study is the wide spectrum of baseline aerobic fitness in HS and college football players. The range of VO_2Peak values from 19.1 to 60.6 mL·kg⁻¹·min⁻¹ in HS and 25.9 to 55.8 mL·kg⁻¹·min⁻¹ in college football athletes indicates that tailored aerobic exercise training is a necessity. The American College of Sports Medicine position stand recognizes that individuals with VO_2Peak values under 40 mL·kg⁻¹·min⁻¹ are less heat tolerant than those above this threshold. ⁴ Moreover, the mean VO_2Peak of 38.5 mL·kg⁻¹·min⁻¹ in our HS athletes is classified as poor aerobic fitness and a concern for heat tolerance. ⁴ For our obese HS football players, the mean VO_2Peak of 32.4 mL·kg⁻¹·min⁻¹ is also poor and equivalent to an average VO_2Max for 50- to 60-year-old men.^4,23 Prior studies on aerobic fitness in football players are limited, but defensive linemen have VO_2Max values 10% to 20% lower than those in other positions at the professional and university levels.^33,47 Football is a unique sport as the relative body fat may range from 5.7% to 21.8%.^33,47 Reports on football fatalities caused by medical conditions have documented that the majority of deaths occur during conditioning sessions. ⁸ Risk factors for these fatalities were obesity, position as lineman, teams exercising as a unit, repetitive sprints, fitness tests, and punishment drills.^{2,4,5,8,24,25,46,49} The results of our study further demonstrate the need to develop exercise regimens according to aerobic fitness levels.

Measuring or even estimating aerobic fitness in football players is not common. For example, in the National Football League Combine, where multiple tests—power, sprint, jumping, and agility—are performed, there is no aerobic fitness assessment. Football training has traditionally been based on a warrior attitude and toughness with the false belief that athletes can be pushed mentality and have no physical limits.^3,8 Training often focuses on power and anaerobic capacity with the misperception that aerobic training will compromise power and the anaerobic system. Running sprints above the maximal aerobic capacity elicits anaerobic metabolism whereby lactic acid is produced, the fatigue threshold is exceeded, and the aerobic system is not utilized. ³⁴ In many cases, athletes are encouraged to continue exercising despite signs and symptoms of physiologic failure. ⁸ Developing rational conditioning programs according to established exercise principles and baseline aerobic fitness is critical. Ideally, an estimate of baseline aerobic fitness is essential before initiating an exercise program. Coaches and sports medicine personnel should develop individualized training programs based on the estimated aerobic fitness to maximize effectiveness as well as lower the risk of injury, especially overexertion injuries such as rhabdomyolysis and fatalities.

The optimal method for establishing and monitoring appropriate aerobic training programs is by measuring VO_2Max or VO_2Peak and HR_Peak. However, because of risks of a medical event during maximum testing, especially in those who are either overweight or obese, we recommend using starting speeds based on individual BMI values. The results of this study showed a statistically significant negative correlation between VO_2Peak and BMI, which demonstrates BMI is an acceptable alternative to a maximum exercise test. Maximum fitness tests should be avoided in obese athletes and athletes with medical conditions, such as sickle cell trait. If a maximum fitness test is performed, we recommend medical personnel be readily available along with a defibrillator. ²² Knowledge of maximal aerobic fitness allows for development of safe, submaximal exercise where the intensity and duration can be slowly increased over time to improve aerobic fitness.^28,45 Optimal regimens for improving aerobic fitness are performed at intensities ranging from 50% to 85% of VO_2Peak, ²⁸ as intensities above 85% can lead to fatigue and are associated with increased injury risk. ²⁸ Monitoring HR during exercise is important because it will show if there is cardiac drift, ¹² a condition associated with progressive, small reductions in stroke volume mirrored by increases in HR and core temperature. Cardiac drift may occur in nonendurance athletes when prolonged (>10 minutes) moderate or intense exercise is performed in the presence of heat and/or humidity, dehydration, illness, and other factors.^12,48

Since VO₂ testing may not be practical in football players, and carries a small risk of a medical event, we assessed other potential predictors of VO_2Peak. Both obesity (BMI >30 kg/m²) and being a lineman were associated with VO_2Peak values consistent with low aerobic fitness (<40 mL·kg⁻¹·min⁻¹). Therefore, even without VO₂ testing, it should be assumed that obese football athletes and linemen will likely have low aerobic capacities and receive individualized training programs during group workouts. In contrast, age, race, and grade in school were not predictors of aerobic capacity in HS athletes.

Interestingly, HS football players with both normal and overweight BMI demonstrated a wide range of VO_2Peak (24-60.6 mL·kg⁻¹·min⁻¹) with 45% and 48%, respectively, demonstrating poor aerobic capacity. Therefore, one cannot assume that normal and overweight HS athletes are aerobically fit; they also require exercise regimens dependent on their aerobic fitness. A normal BMI athlete may have tremendous power, sprinting, and jumping ability but have low aerobic capacity and difficulty competing at a high tempo pace for an entire game without fatigue and a decrease in performance. At the college level, all normal and overweight BMI athletes had extrapolated VO_2Peak values greater than 40 mL·kg⁻¹·min⁻¹ indicating that college athletes are likely more fit than comparable HS athletes.

Six-Minute Run Test

In the absence of VO₂ testing, the 6-minute run test offers the advantages of easy administration and reliability in predicting VO_2Max without the time, equipment, or expense of VO₂ testing.^20,26,27 The 12-minute Cooper test is another acceptable test for estimating VO_2Max. Although such running tests are not as reliable as VO_2Max testing, the literature shows a strong correlation with VO_2Max.^{20,23,26,27,33,43} The correlation between predicted VO_2Max for the 12-minute test (0.90) is higher than for the 6-minute test (0.85), but the 12-minute test requires 6 extra minutes of running at a high intensity, which could predispose athletes, especially obese athletes, to a medical event. ¹⁰ The disadvantages of both run tests is that they do not confirm whether an athlete is giving a maximal effort unless HR is measured, and they both could place the athlete at risk of a medical event. A HR monitor is preferred when determining HR_Peak, which is essential for monitoring exercise programs for cardiac drift and exercise above VO_2Max.

The results of the 6-minute run test can be used to approximate aerobic fitness and indicate safe exercise levels as a function of aerobic fitness when starting a rigorous athletic program, but also place the athlete at risk for a medical event. ¹⁰ The distance run in a set time at maximum effort can also be employed to estimate aerobic capacity and risk stratify safe sprinting and training speeds. Running sprints at a fast pace relies on anaerobic metabolism with resultant accumulation of lactic acid after 1 or 2 sprints. In the absence of sufficient recovery time, the result would be subsequent fatigue. The run test performed in our study revealed significantly lower maximum speeds for the obese (HS, 22.6%; college, 23.5%) and overweight (HS, 2.7%; college, 7.7%) groups, respectively, than for the normal weight groups. There was a significant difference in mean speeds and predicted VO_2Max between the HS and college normal BMI groups but no difference in overweight and obese categories (Table 2). This indicates better fitness levels in normal BMI college athletes but no differences in HS and college athletes once BMI is in the overweight or obese range.

Although VO_2Max and the 6-minute run test are the optimal techniques for measuring aerobic fitness, both require a maximal effort, which can predispose high-risk athletes to a medical event. Therefore, we recommend training programs be based on individual BMI levels instead of a maximal fitness test. The run test was used to help provide guidelines for safe training regimens for athletes by BMI. These maximum speeds were divided by 85% to obtain an upper threshold for performing submaximal running intervals for aerobic training. ²⁸ Obese HS and collegiate BMI groups, respectively, should begin training at speeds that are at least 26.5% (22.5%/0.85) and 27.6% (23.5%/0.85) slower than normal weight athletes, until their aerobic fitness improves. Overweight HS and collegiate BMI groups should be training at speeds that are at least 3.2% (2.7%/0.85) and 9.1% (7.7%/0.85) slower than normal weight athletes. These suggestions conform to safe training guidelines, and although individualizing training by fitness is not easy, the US Army has shown that when low-fit recruits are screened and provided appropriate exercise progression fitness programming, there are fewer injuries and less attrition. ³⁰ Moreover, the literature shows that fitness level when beginning any type of training is a critically important factor to consider when designing a physical activity curriculum. ³² Individual BMI analysis revealed that the time to run 1 mile increased 9.6 and 10.2 seconds for HS and college athletes, respectively, and the speed in yards per second decreased 0.07 seconds (7 s/100 y) in HS and college athletes, respectively, for every 1-unit increase in BMI.

Suggested Initial Submaximal Training Speeds Based Solely on BMI

Table 3 provides suggested initial submaximal training speeds based solely on individual BMI values for college athletes. Although these BMI-derived reference speeds are less accurate than actual VO_2Max values or results from a run test, they can be used as a guide for developing aerobic training programs without the need for VO_2Max testing or a 6- or 12-minute run test. Although the initial training speeds may seem slow, they are a safe starting point for HS and collegiate athletes from which incremental increases in speeds and decreases in rest periods can be employed as both aerobic and anaerobic capacity improve. For aerobic training submaximal (upper threshold of 85% of maximum) intensities are recommended with work intervals of at least 1 to 2 minutes to sufficiently stress the oxygen delivery and utilization systems and thereby facilitate physiologic improvements/adaptations.^28,39 For example, an average college athlete with a BMI of 35 kg/m² should not start the training season by performing repetitive runs faster than 38.8 seconds for 110-yard gassers and 10.3 minutes for a mile. After the initial run, the coach can slowly increase the speed and decrease the rest periods. This is particularly important when the weather is hot. Importantly, the suggested starting speeds do not account for medical conditions, illnesses, or unusually hot temperatures, which may require further adjustments in the training protocol.

One important factor in aerobic training that has received limited research is the rest or recovery period between work bouts. The work to rest ratio choice depends on the energy system (phosphagen system/phosphocreatine, anaerobic glycolysis/lactic acid system, and aerobic) being trained—and reflects the distance to be run. Overall rest periods allow for adequate recovery and are applied to enhance performance outcomes. ³⁹ When training to improve sprint speeds with maximal or supramaximal effort of repetitive sprints less than 15 seconds (phosphagen system), work to rest ratios of 1:5 or longer rest periods are recommended. ³⁹ Sprints lasting between 15 and 30 seconds (anaerobic glycolysis) require a work to rest ratios of 1:4 or longer rest periods. ³⁹ For aerobic conditioning involving submaximal training at distances over 1 to 2 minutes, a recovery period of 3 minutes is usually sufficient to prevent premature fatigue. ³⁹ As with all sports, football players should train each energy system so rest periods should vary. However, in principle, as aerobic training (>1- to 2-minute runs) progresses, speeds run by the athletes can be slowly increased and rest periods decreased to achieve a 1:1 work:rest ratio.

Case Examples

It is important to show how failing to apply proper exercise science concepts when developing training regimens can result in fatalities. To that end, we identified 2 fatalities from the public domain and examined their training regimens. ⁴² Both athletes from the public domain were obese linemen performing repetitive sprints on the first day of offseason or preseason conditioning and they died from exertional heat stroke.^8,42,44 The low 80°F temperatures during the conditioning session for both cases indicate that exertion was likely more important than temperature. ⁸ One collegiate exertional heat stroke fatality, ⁴⁴ a lineman with a BMI of 35 kg/m² was required to run 10 × 110-yard gassers during the first day (80°F) of off-season conditioning. The athlete collapsed after seven 110-yard sprints. The athletes were required to run each gasser in 19 seconds, a mean speed of 5.8 yards/s (5.1 min/mile) or 49% higher than our recommended starting speed (3.9 yards/s) in Table 3. The extrapolated VO_2Max for that running speed is 49.5 mL·kg⁻¹·min⁻¹ or 33% higher than the estimated value (37.1 mL·kg⁻¹·min⁻¹; Table 3) based on his BMI of 35 kg/m². Importantly, the 45-second rest periods between the 110-yard sprints resulted in a work:rest ratio of 1:2.4, ⁴⁴ which is much shorter than the recommended rest period (1:4) for sprints in the 15- to 30-second range. ³⁹

The second collegiate exertional heat stroke fatality was a lineman with a BMI of 38 kg/m². He was required to run 36 × 50-yard sprints on the first day (84°F) of preseason conditioning, each in 8 seconds for a mean pace of 6.3 yards/s ⁴² (4.7 min/mile) or 88% higher than the recommended starting speed of 3.2 yards/s (Table 3) based on his BMI. The extrapolated VO_2Max (49.5 mL·kg⁻¹·min⁻¹) at the required speed was 48% higher than the extrapolated VO_2Max (33.4 mL·kg⁻¹·min⁻¹; Table 3) predicted based on a BMI of 38 kg/m². Again, the work:rest ratio (1:3.8) had a shorter rest period than recommended (1:5) for sprints less than 15 seconds. ³⁹ Although running only 1 sprint at an individual’s VO_2Max or speed may be appropriate, multiple sprints at maximum effort clearly require adequate rest periods.^4,8 These real-life examples illustrate flaws in current training techniques employed during football conditioning sessions, which are similar to fatalities reported in the US military ⁶ and the Israeli Defense Forces. ³⁵ The fatalities stress the importance of progressive, individualized training regimens based on fitness level. Having an index reflective of aerobic fitness is key to overall health and performance.

Limitations

VO_2Peak was utilized as the primary measure of maximum aerobic capacity instead of VO_2Max. Although VO_2Max is considered the gold standard for quantifying aerobic fitness, 30% to 50% of patients tested, ¹⁴ especially children and adolescents, ³⁸ are unable to achieve the VO₂ plateau required, so VO_2Peak is more practical. Multiple studies have demonstrated VO_2Peak is a valid surrogate for VO_2Max when assessing aerobic fitness.^14,17,37,38 A second limitation is that both a treadmill and a climber were used for measuring VO_2Peak. Ideally, 1 test could have been employed; however, the literature suggests that the differences will not be statistically different. ¹¹ A third limitation is that our suggested training speeds were derived from HS and collegiate football athletes. We cannot confirm whether the speeds predicted can be employed for other sports and age categories.

Conclusion

This study demonstrates a large spectrum of baseline aerobic fitness in HS and collegiate football players. Obese players and linemen had poor aerobic fitness, a risk factor for exertional heat stroke. Most obese and approximately 50% of normal and overweight HS football athletes, based on VO_2Peak, had poor aerobic fitness, which suggests a need for individualized exercise regimens. BMI is an appropriate measure for estimating aerobic fitness and developing aerobic training regimens that optimize aerobic conditioning, while at the same time reducing the risks of injury including the risk of a medical event during a maximum exercise test. The speeds provided in this study can be employed for safely starting exercise regimens based on individual BMI levels. Knowledge of baseline aerobic fitness level, based on BMI, is critically important for designing and monitoring appropriate, safe exercise programs.