American football provides a unique population within which to measure body mass and distribution of mass (3,4,9,12,14,15,21–23,25,28,31). Each position has physical demands that require different body types to be successful (10,11,16–19). Generally speaking, individuals are trying to maximize their total mass-lean mass ratio (i.e., percent body fat [%BF]), but the optimal total mass for each position is highly variable (9,12). Managing body composition to meet the demands of their position and sport presents a unique challenge for sports performance staff. To date, there is limited normative data for total and regional body composition values by position in a large sample of collegiate football players using dual x-ray absorptiometry (DXA). Identifying the positional averages for total and regional body composition, using a gold standard methodology, would help guide expectations for body composition in football players. The purpose of this study was to provide accurate and reliable measurements of total and regional body composition for Division 1 football players.
A variety of methods have been used to quantify body composition in previous studies of football players, both professional and collegiate (3,4,9,12,14,15,21–23,25,27,28,31). Although many have relied on anthropometrics, more recent reports have measured body composition using DXA. Dual x-ray absorptiometry is a 3-compartment method that quantifies fat mass, lean mass, and bone mass (and density) and is considered the “gold standard” for measuring body composition. Moreover, recent advancements in DXA software allow for regional segmentation of body composition, comparison of the right and left sides of the body, and the estimation of visceral fat mass (visceral adipose tissue [VAT]). This additional information allows for a more detailed evaluation of the distribution of mass and the estimate of VAT has important implications for the health and well-being of players (1,5,9,14,15,20,24,26). Researchers and sport performance trainers can use this information to determine mass distribution ratios that may have a stronger relationship with speed, strength, and power than traditional total measures of body composition (i.e., %BF, total lean mass, and total fat mass). Previously, using DXA, we reported positional characteristics of total and regional body composition in professional football players (12,22). Significant findings from those studies include that positions which mirror each other have similar body composition characteristics, and on average, fat mass accumulates at a greater proportion per pound increase in weight in players above 250 pounds (9).
The purpose of this study was to use DXA derived total and regional body composition measures to define position-specific characteristics in a large cohort of National Collegiate Athletic Association (NCAA) Division 1 football athletes from the Consortium of College Athlete Research (C-CAR). This information can be used to identify normative data, using DXA for each football position. Practically, this information provides coaches, performance staff, and players clear expectation of their body composition and bone density for their positions. Moreover, these data would provide distinction between the type and distribution of mass for each athlete. We hypothesized that positions which mirror each other would have similar body composition characteristics.
Experimental Approach to the Problem
Data for this study were obtained from the C-CAR research group consisting of researchers from Texas Christian University (TCU), University of Texas at Austin (UT), University of Kansas (KU), and University of Minnesota (UMN). By combining data collected from several universities, we can increase our sample size to provide accurate and reliable positional data for total and regional body composition. This study used a retrospective analysis from previously collected data to test whether total and regional body composition data differs between Division 1 football positions. All players were scanned on GE Healthcare Lunar systems (Lunar iDXA/Prodigy; GE Healthcare, Madison, WI, USA). To improve accuracy and reliability of the data, raw DXA scan files were collected from each university and analyzed at the University of Minnesota using enCore software version 16.2. The same trained individual reviewed each scan for the correct placement of the region of interest boxes to measure regional body composition. Visceral adipose tissue mass was estimated using CoreScan (GE Healthcare), as described previously (8,9,13).
A total of 467 individuals (UT n = 172, KU n = 142, TCU n = 45, and UMN n = 108) were scanned from the 4 schools from 2011 to 2016, and the number of scans for each individual ranged from 1 to 16 for a total of 1794 scans. The age of the population ranged from 17 to 24 years. This study was approved by the University of Minnesota Institutional Review Board; the data transfer from other universities was approved by each individual university. Subjects were informed of the benefits and risks of the investigation before signing an institutionally approved informed consent document to participate in the study. For individuals younger than 18 years, parental or guardian consent was also obtained.
For this study, individuals with more than 1 scan were randomly sampled, and the randomization was stratified to provide a similar range of ages for each position. For this study, only scans conducted during the preseason were included (June 1–August 1). Participants were in their general preparation phase (June) or specific phase (July) or their training cycle. Before the scan, each participant had height and body mass measured by a standard stadiometer and electronic scale, respectively. All players scanned were instructed to maintain their hydration status before scans. As much as possible, scans were performed on off days or at least 2 hours after a practice session. Participants were scanned using standard imaging and positioning protocols. Participants were grouped into positions as follows: defensive backs (DBs), wide receivers (WRs), tight ends (TEs), linebackers (LBs), running backs (RBs), offensive linemen (OLs), defensive linemen (DLs), quarterbacks (QBs), and special teams (STs). For the purpose of this study, participants were not grouped as underclassmen or upperclassmen. This is because only age was recorded for subjects. The accuracy of classifying class level based on age would have been limited without further information.
Descriptive statistics were calculated using mean ± SD by position. An analysis of variance was used to test if positional mean values were equal to each other. The TukeyHSD (honest significant difference) method was used to compare each mean and correct for type I error from multiple comparisons with an adjusted significance level of p ≤ 0.05. In addition to standard total and regional metrics, an upper total mass to legs total mass ratio, total upper mass to lean leg mass ratio, lean upper mass to lean leg mass ratio, and gynoid (glute) lean mass to leg lean mass ratio were calculated for each position. For all these ratios, upper body equals trunk and arms and total includes fat, lean, and bone mass. These ratios were used to compare whether the type and distribution of mass were different between positions. All analyses were completed using R (R Foundation for Statistical Computing, Vienna, Austria).
Table 1 compares the physical characteristics (i.e., age, height, body mass, and body mass index [BMI]) of the cohort by position. Positions that share a letter within each row were not significantly different from one another. Age was similar across positions. LB and DL would be classified as class I obese (30.0–34.9 kg·m−2), OL as class II obese (35.0–39.9 kg·m−2), and all other positions as overweight (25.0–29.9 kg·m−2) based on standard BMI categories. The range for each of these variables is presented in Table 2. However, as shown in Table 3, only OL would be classified as obese and DL as overweight based on %BF.
Table 3 presents the total and regional body composition averages by position. OL had much higher %BF than all other positions. With the exception of OL and DL, there were no significant differences on any measures between offensive and defensive positions that mirror each other, but these values were different from other position groups. For all fat measures and fat ratios, DL was significantly lower than OL, but there were no differences between DL and OL for any lean measures and lean ratios. OL had significantly higher VAT mass (p ≤ 0.05) compared with all positions, except DL and RB. Table 4 presents the ranges of total and regional body composition variables by position. In general, each position had a wide range of values for each total and regional body composition variable.
Table 5 presents the mean total, total less head, leg and spine bone mineral densities by position, and the average z-score for total bone density (based on the enCore normative database). Table 6 presents the ranges for these bone mineral density (BMD) variables. Consistent with body composition variables, total and regional bone mineral densities were similar among positions that mirror each other; however, there were fewer positional differences. DL had the highest average density, and ST athletes had the lowest average for each measurement.
Although other studies have reported body composition in collegiate football players, this study provides the largest cohort of collegiate football players with total and regional body composition measured by DXA, the “gold standard” in body composition. This study builds on previous studies by reporting positional BMD and VAT mass differences in collegiate football players. These data provide new normative data for positional body composition characteristics of collegiate football players that can be used to guide nutrition and training plans as well as monitor player's health and wellness.
Consistent with previous work (5,13,14,23,25), most collegiate Division 1 football positions are classified as overweight or obese based on BMI. This is distinctly different from our previous report on National Football League (NFL) players where most positions were classified as class I and II obese (12). Also, consistent with other reports (9,12), positions, except OL and DL, are classified as healthy or athletic for %BF (13–20%). The %BF values reported here are slightly higher than other reports (21,23) using other methods (e.g. underwater weighing and bioelectrical impedance); however, they are consistent with previous studies on NFL players (12) and the recent report by Trexler et al. (29) that used DXA. In addition, this report builds on previous work by combining data from 4 different universities in different regions of the United States, increasing the diversity of the cohort. The values reported in this study can be used as normative body composition data for NCAA Division I football players.
With the exception of OL and DL, positions that mirror each other (DB vs. WR, LB vs. RB, and LB vs. TE) had no significant differences for any total or regional body composition measure or ratio. For OL vs. DL, there were no differences for any total or regional lean measures, but OL had higher total and regional fat measures. In general, when comparing mirrored positions against other mirrored positions, there was an increase in all total and regional body composition measures from DB and WR to LB and RB to DL and OL. Although there were no significant differences between LB and RB or LB and TE on any measure, there were differences between RB and TE. A closer look at the data indicates that, although many of the differences are not significant, RB < LB < TE for every total and regional body composition measurement. The similarities between positions that mirror each other were also observed in different body mass ratios. Although speculative, the mirroring of position for total and regional measures, as well as ratios of mass distribution, may indicate a relationship with on-field requirements because the mass distribution and ratio of distribution would influence how athletes move their body through space. Generally speaking, an OL can be successful with a higher fat mass (and total mass), because they are not required to cover much ground, whereas their counterpart, DL, is required to get up field quickly and in some systems drop back into pass coverage. Further evidence of this is observed in TE/LB/RB groups and DB/WR groups. These positions require speed and quickness and the ability to cover larger areas. In each case, the total and regional differences between offensive and defensive players were not significantly different. More research is needed to establish how regional distribution of mass effects speed, force, and on-field performance. However, the ability to accurately and reliably measure these ratios provides an opportunity for longitudinal studies, designed to answer these questions, to draw stronger conclusions.
Most positions had a wide range for each total and regional body composition variable. This may be a result of grouping similar positions together (e.g. safety and cornerback as DB) and not delineating between players who played a significant amount and those who did not. To maintain a relatively large sample size for each position, we grouped all positions together (e.g. guard, center, and tackle as OL). The total and regional body composition may differ slightly between these specific positions that results in a larger range. Future studies comparing exact positions (e.g. defensive tackle and middle line backer) would help define specific position normative data. Grouped studies such as this one examining specific positions would need to control for the scheme of the team because defensive tackles may look very different in a 3–4 scheme vs. a 4–3 scheme.
Finally, from a health perspective, there were fewer positional differences for total and regional bone density and VAT measurements (compared with fat and lean mass). Normative data on bone density is important for athletic populations. This cohort had higher bone density compared with nonathletic populations (the average total BMD for each position group was 2.1–3.4 SD above normal, measured by z-score); however, when compared within similar body types, it's possible to identify individuals who may have potential deficiencies in bone density. Individuals within a position that are near the lower limits for total or regional bone density (e.g. 10th percentile) may be at greater risk of bone injury in a collision sport. Similarly, VAT mass has been implicated and associated with significant metabolic dysfunction in children and adult populations (6–8). We have previously identified a proportional shift of accumulation of VAT, even in an athletic population, with increasing adiposity and total mass (9). Monitoring VAT both during and after a playing career, in addition to other body composition measures, will be important for athletes to maintain health and healthy metabolic function (5,13,15,24,26). This study, which shows the mean VAT values for OL and DL to be 2–4 times greater than those of the other positions, is consistent with our NFL population (9) and others that have shown linemen are at the greatest risk of cardiovascular and metabolic dysfunction after retirement (1,2,18,24,30).
In conclusion, this study has several strengths: it includes a large cohort (n = 467); gold standard body composition (DXA); and standardized analyses (every scan was analyzed using the same software and reviewed by the same technician). This study does have a few limitations. First, although several precautions were taken to control for differences between sites (each site used standard best practices), there is the potential for bias based on scan location. In addition, scanning a team in a feasible amount of time limits the control of time of day for scanning, nutrition, hydration etc. Best practices were used by each university and when possible similar position groups were scanned at the same times of day. The large sample size should limit the influence of these biases. Moreover, this report is a cross-sectional assessment of body composition and does not address the change over time in collegiate athletes.
The use of DXA to measure body composition in athletic populations has increased in recent years. Dual x-ray absorptiometry is one of the most accurate methods to assess both total and regional body composition. The data from this study provide accurate and reliable body composition and BMD data for Division 1 football positions. These data can be used by athletes and coaches as reference data when using DXA to measure total and regional body composition, as these numbers will likely differ from other body composition methods. Moreover, these data provide information regarding the type and distribution of mass for each position. More research is needed to understand how distribution of mass effects strength, force, and speed characteristics, but the similarity between mirrored positions observed in this study and our previous report in NFL athletes suggests it plays a role in on-field requirements. The data presented in this report can be used to identify ideal ranges for each position in collegiate football players. However, it is important to clarify that the data presented here is averages and ranges for each position. In practice, performance metrics must be taken into account because each individual will have their own optimal body type to maximize performance.
1. Allen TW, Vogel RA, Lincoln AE, Dunn RE, Tucker AM. Body size, body composition, and cardiovascular disease risk factors in NFL players. Phys Sportsmed 38: 21–27, 2010.
2. Baron SL, Hein MJ, Lehman E, Gersic CM. Body mass index, playing position, race, and the cardiovascular mortality of retired professional football players. Am J Cardiol 109: 889–896, 2012.
3. Binkley TL, Daughters SW, Weidauer LA, Vukovich MD. Changes in body composition in Division I football players over a competitive season and recovery in off-season. J Strength Cond Res 29: 2503–2512, 2015.
4. Bolonchuk WW, Lukaski HC. Changes in somatotype and body composition of college football players over a season. J Sports Med Phys Fitness 27: 247–252, 1987.
5. Borchers JR, Clem KL, Habash DL, Nagaraja HN, Stokley LM, Best TM. Metabolic syndrome and insulin resistance in Division 1 collegiate football players. Med Sci Sports Exerc 41: 2105–2110, 2009.
6. Bosch T, Dengel D, Kelly A, Sinaiko A, Moran A, Steinberger J. Visceral adipose tissue measured by DXA correlates with measurement by CT and is associated with cardiometabolic risk factors in children. Pediatr Obes 10: 172–179, 2015.
7. Bosch TA, Chow L, Dengel DR, Melhorn SJ, Webb M, Yancey D, Callahan H, Leon De MRB, Tyagi V, Schur EA. In adult twins, visceral fat
accumulation depends more on exceeding sex-specific adiposity thresholds than on genetics. Metab Clin Exp 64: 991–998, 2015.
8. Bosch TA, Steinberger J, Sinaiko AR, Moran A, Jacobs DR, Kelly AS, Dengel DR. Identification of sex-specific thresholds for accumulation of visceral adipose tissue in adults. Obesity (Silver Spring) 23: 375–382, 2015.
9. Bosch TA, Burruss TP, Weir NL, Fielding KA, Engel BE, Weston TD, Dengel DR. Abdominal body composition differences in NFL football players. J Strength Cond Res 28: 3313–3319, 2014.
10. Burke EJ, Winslow E, Strube WV. Measures of body composition and performance in major college football players. J Sports Med Phys Fitness 20: 173–180, 1980.
11. Crews TR, Meadors WJ. Analysis of reaction time, speed, and body composition of college football players. J Sports Med Phys Fitness 18: 169–174, 1978.
12. Dengel DR, Bosch TA, Burruss TP, Fielding KA, Engel BE, Weir NL, Weston TD. Body composition and bone mineral density of national football league players. J Strength Cond Res 28: 1–6, 2014.
13. Kaul S, Rothney MP, Peters DM, Waker WK, Davis CE, Shapiro MD, Ergun DL. Dual-energy X-ray absorptiometry for the quantification of visceral fat
. Obesity 20: 1313–1318, 2012.
14. Lambert BS, Oliver JM, Katts GR, Green JS, Martin SE, Crouse SF. DEXA or BMI: Clinical considerations for evaluating obesity in collegiate Division I-A American football athletes. Clin J Sport Med 22: 436–438, 2012.
15. Mathews EM, Wagner DR. Prevalence of overweight and obesity in collegiate American football players, by position. J Am Coll Health 57: 33–38, 2008.
16. Mayhew J, Piper F, Schwegler T, Ball T. Contributions of speed, agility and body composition to anaerobic power measurement in college football players. J Strength Cond Res 3: 101–106, 1989.
17. Melvin MN, Smith-Ryan AE, Wingfield HL, Ryan ED, Trexler ET, Roelofs EJ. Muscle characteristics and body composition of NCAA Division I football players. J Strength Cond Res 28: 3320–3329, 2014.
18. Miller MA, Croft LB, Belanger AR, Romero-Corral A, Somers VK, Roberts AJ, Goldman ME. Prevalence of metabolic syndrome in retired National Football League players. Am J Cardiol 101: 1281–1284, 2008.
19. Miller TA, White ED, Kinley KA, Congleton JJ, Clark MJ. The effects of training history, player position, and body composition on exercise performance in collegiate football players. J Strength Cond Res 16: 44–49, 2002.
20. Murphy DF, Connolly DA, Beynnon BD. Risk factors for lower extremity injury: A review of the literature. Br J Sports Med 37: 13–29, 2003.
21. Noel MB, VanHeest JL, Zaneteas P, Rodgers CD. Body composition in Division I football players. J Strength Cond Res 17: 228–237, 2003.
22. Oliver JM, Lambert BS, Martin SE, Green JS, Crouse SF. Predicting football players' dual-energy x-ray absorptiometry body composition using standard anthropometric measures. J Athl Train 47: 257–263, 2012.
23. Oppliger R, Nielsen D, Shetler A, Crowley E, Albright J. Body composition of collegiate football players: Bioelectrical impedance and skinfolds compared to hydrostatic weighing. J Orthop Sports Phys Ther 15: 187–192, 1992.
24. Potteiger JA, McGowan-Stinski M. Protecting the metabolic health of football players with high fat mass. Strength Cond J 37: 95–101, 2015.
25. Snow TK, Millard-Stafford M, Rosskopf LB. Body composition profile of NFL football players. J Strength Cond Res 12: 146–149, 1998.
26. Steffes GD, Megura AE, Adams J, Claytor RP, Ward RM, Horn TS, Potteiger JA. Prevalence of metabolic syndrome risk factors in high school and NCAA Division I football players. J Strength Cond Res 27: 1749–1757, 2013.
27. Stodden DF, Galitski HM. Longitudinal effects of a collegiate strength and conditioning program in American football. J Strength Cond Res 24: 2300–2308, 2010.
28. Thompson CW. Changes in body fat, estimated from skinfold measurements of varsity college football players during a season. Res Q 30: 87–93, 1959.
29. Trexler ET, Smith-Ryan AE, Mann JB, Ivey PA, Hirsch KR, Mock MG. Longitudinal body composition changes in NCAA Division 1 college football players. J Strength Cond Res, 31: 1–8, 2017.
30. Tucker AM, Vogel RA, Lincoln AE, Dunn RE, Ahrensfield DC, Allen TW, Castle LW, Heyer RA, Pellman EJ, Strollo PJ. Prevalence of cardiovascular disease risk factors among National Football League players. JAMA 301: 2111–2119, 2009.
31. Wickkiser JD, Kelly JM. The body composition of a college football team. Med Sci Sports 7: 199–202, 1975.