The Effects of Playing Multiple High

School Sports on National Basketball

Association Players’ Propensity
for Injury and Athletic Performance
Caitlin Rugg,* MD, Adarsh Kadoor,y Brian T. Feeley,* MD, and Nirav K. Pandya,*z MD
Investigation performed at the University of California, San Francisco,
Benioff Children’s Hospital, Oakland, California, USA

Background: Athletes who specialize in their sport at an early age may be at risk for burnout, overuse injury, and reduced attain-
ment of elite status. Timing of sport specialization has not been studied in elite basketball athletes.
Hypothesis: National Basketball Association (NBA) players who played multiple sports during adolescence would be less likely to
experience injury and would have higher participation rates in terms of games played and career length compared with single-
sport athletes.
Study Design: Descriptive epidemiology study.
Methods: First-round draft picks from 2008 to 2015 in the NBA were included in the study. From publically available records from
the internet, the following data were collected for each athlete: participation in high school sports, major injuries sustained in the
NBA, percentage of games played in the NBA, and whether the athlete was still active in the NBA. Athletes who participated in
sports in addition to basketball during high school were defined as multisport athletes and were compared with athletes who par-
ticipated only in basketball in high school.
Results: Two hundred thirty-seven athletes were included in the study, of which 36 (15%) were multisport athletes and 201 (85%)
were single-sport athletes in high school. The multisport cohort played in a statistically significantly greater percentage of total
games (78.4% vs 72.8%; P \ .001). Participants in the multisport cohort were less likely to sustain a major injury during their
career (25% vs 43%, P = .03). Finally, a greater percentage of the multisport athletes were active in the league at time of data
acquisition, indicating increased longevity in the NBA (94% vs 81.1%; P = .03).
Conclusion: While a minority of professional basketball athletes participated in multiple sports in high school, those who were
multisport athletes participated in more games, experienced fewer major injuries, and had longer careers than those who partic-
ipated in a single sport. Further research is needed to determine the reasons behind these differences.
Keywords: professional; basketball; sport specialization; knee (general); ankle; injury prevention

Participation in organized youth sports has increased tre- 2015-2016.37 Concurrent with this rise in organized sports
mendously over the past several decades. According to the participation has been an increase in sports injuries in the
National Federation of State High School Associations, the 5- to 18-year-old group.42,43,54 Using anterior cruciate liga-
number of youth participating in organized sporting activity ment (ACL) injuries as an example, the rate of surgically
has increased from 6 million in 1995-1996 to 7.8 million in treated ACL injuries increased from 12.22 per 100,000 per-
son-years in 1994 to 17.97 per 100,000 person-years in 2006
for patients under 20 years of age.32 More recent studies
z
Address correspondence to Nirav K. Pandya, MD, Department of have also supported a rising incidence of ACL injury and
Orthopaedic Surgery, UCSF Benioff Children’s Hospital Oakland, 747 reconstruction on a population level.9,28 The reasons for ris-
52nd Street, Oakland, CA 94609, USA (email: [email protected]). ing injury rates in the youth athlete population are complex,
*Department of Orthopaedic Surgery, University of California San
yet an emphasis on skill development (ie, kicking, shooting,
Francisco, San Francisco, California, USA.
y
University of California San Diego, San Diego, California, USA. pitching) rather than generalized fitness (ie, jumping, run-
The authors declared that they have no conflicts of interest in the ning, stretching) may be largely to blame.13,14,23,48
authorship and publication of this contribution. Furthermore, a trend toward young athletes specializ-
ing in one sport at progressively younger ages has been
The American Journal of Sports Medicine
noted, even with multiple groups advocating for delayed
2018;46(2):402–408
DOI: 10.1177/0363546517738736 specialization.4,20,29,38 Single-sport specialization can be
Ó 2017 The Author(s) defined as intensive year-round training in a single sport

402
AJSM Vol. 46, No. 2, 2018 NBA Sport Specialization 403

at the exclusion of other sports.35 Recently, the American size was 237 athletes. Only first-round draft picks were
Academy of Pediatrics clinical report on single-sport special- used as they represented a more ‘‘elite’’ pool of athletes
ization recommended that young athletes not begin to spe- and this subgroup of NBA players was more likely to
cialize until after puberty.5 Multiple studies have have information available on the internet with regard to
demonstrated the detrimental effects of single-sport special- past sporting participation and injury history due to
ization including social isolation, overdependence, burnout/ increased media coverage.
dropout, manipulation, and increased injury risk.13,23,31 Using publically available information on the internet,
The concept of completing 10,000 hours of deliberate we collected the following data for each player: (1) participa-
practice to achieve mastery of a skill, proposed by Ericsson tion in high school sports other than basketball (type and
et al,12 has been endorsed by many in the athletic world. number), (2) major injuries incurred during basketball-
This arguably has led to increased emphasis on high- related activities at the NBA level (type and number), (3)
intensity, high-volume skill training.7 Additionally, signif- number of games played in the NBA, and (4) whether the
icant anecdotal media attention has been paid to special- player was currently still active in the NBA. In addition,
ized child athletes who achieved professional success, age, height, weight, and body mass index (BMI) when
such as Tiger Woods and Andre Agassi. Limited studies drafted were collected as was position played in the NBA.
have demonstrated a benefit of early specialization on Positions were grouped into center, forward, and guard
achieving elite status, including a study of rhythmic gym- due to the fluidity between small/power forward and shoot-
nasts, whose athletic peak is at an earlier age than many ing guard/point guard. In addition to major sports news
other sports.21 However, despite commonly held beliefs reporting websites such as www.espn.com, www.nba.com,
by parents, coaches, and athletes that early sport speciali- www.foxsports.com, and www.si.com, the following websites
zation and high-intensity training may facilitate attain- were used to collect data for the study: www.basketball-
ment of elite status or college athletic scholarship,36,41 reference.com, www.athletic.net, www.rivals.com, www
the literature does not support this assertion.13 Gullich16,17 .MaxPreps.com, and www.draftexpress.com.
has studied multiple populations of elite European athletes To consider a player to have been a ‘‘multisport’’ athlete
and has repeatedly found that those who achieved more during his adolescence, we required evidence that the
success tended to play multiple sports until later ages player did indeed participate in a sport in addition to bas-
than their near-elite counterparts. Furthermore, in a study ketball during his time in high school. This was determined
of Danish elite and near-elite athletes, Moesch et al34 through a search of all publicly available information on the
found that elite athletes specialized later in adolescence internet including but not limited to high school sports sta-
and, rather than having high-volume training at a young tistics databases (ie, athletic.net), newspaper and magazine
age, had more training hours in their teenage years than articles, and online biographies. An athlete’s participation
near-elite athletes. Vaeyens et al49 discussed evidence in sports played recreationally or before high school (ie, dur-
that talent identification for prospective Olympic athletes ing elementary and/or middle school) did not count toward
that results in early high-volume single-sport training the data.
and competition has not necessarily resulted in interna- The tracking of a player’s injuries hinged on the type,
tional athletic success. As many parents and athletic enti- severity, circumstances, and timing of injury. With regard
ties (ie, club sports) justify single-sport specialization as to the injury type, an injury was counted toward the data
the only means to achieve ‘‘elite’’ athletic potential, further set if it affected one of the following areas of the body:
study is necessary. the neck, back, groin, torso, legs, and feet. Therefore, inju-
The purpose of our study was to (1) determine the inci- ries such as upper extremity fractures, concussions, and
dence of single-sport specialization among National Basket- jaw fractures were not counted toward this study. As
ball Association (NBA) players during adolescence, (2) well, the injury had to have occurred due to basketball-
identify differences in the rate of injury between those related activities at the professional level. For this reason,
who were multisport athletes in adolescence versus single- injuries that manifested themselves due to other activities
sport athletes, and (3) determine whether there were differ- (such as falling down the stairs or getting into a fight) were
ences in games played or career length in multisport ath- not recorded. Additionally, an injury had to have been
letes versus single-sport athletes. We hypothesized that severe enough to constitute either an acute 10-game
NBA players who played multiple sports during adolescence absence or a cumulative 10-game absence over the course
would be less likely to experience injury and would have of a season (ie, chronic recurring injury). Last, the injury
higher participation rates in terms of games played and had to have occurred while the player was still employed
career length compared with single-sport athletes. by an NBA team. All documentation about these injuries
was acquired through standard internet web searches.
For each player, we gathered both the total number of
METHODS games played (from www.Basketball-Reference.com) and
the number of possible games played. The number of possi-
The first-round draft picks for all NBA teams from 2008 to ble games played for each player was determined by look-
2015 were identified. To be included in the study, partici- ing into the player’s employment history; if a player was
pants had to have played in at least 1 regular season employed by the team then he was considered as being
NBA game. Our initial search yielded 240 players, 3 of able to participate in those team’s games. For example, if
whom did not play in any NBA game. Thus, our sample a player missed an entire 82-game season due to injury
404 Rugg et al The American Journal of Sports Medicine

TABLE 1
Differences Between Single-Sport and Multisport Cohorts
in Age, Height, Weight, and Body Mass Index When Drafteda

2008-2015 Draft Classes (Aggregate) Age, y Height, in Weight, lb Body Mass Index

Multisport 21.0 6 1.4 79.4 6 3.3 215.9 6 26.0 24.2 6 1.8

Single sport 20.8 6 1.3 78.8 6 3.5 217.8 6 25.6 24.4 6 1.9
P value .35 .33 .69 .63

a
Data are reported as mean 6 SD unless otherwise noted.

but was still part of the team, then the 82 missed games TABLE 2
did indeed count toward his possible games played. This Differences Between Single-Sport and Multisport
is due to fact that the player technically could have played Cohorts in Percentage of Players Playing
in some or all of those 82 games but did not due to a health Specific Positions in the NBA
impediment. Contrarily, if a player missed an entire 82-
game season due to some other reason (ie, retirement or 2008-2015 Draft
having signed a contract to play overseas), then the 82 Classes (Aggregate) Center Forward Guard
missed games did not count toward his possible games Multisport, % (n) 25 (9) 31 (11) 44.4 (16)
played. This is due to the fact that the player made a choice Single sport, % (n) 18.4 (37) 24.9 (50) 57.0 (114)
to leave the NBA. It must be noted that only participation P value .36 .53 .2
in regular season games was counted; preseason and play-
off games were excluded from this study.
Finally, the players’ still-current status in the NBA was football (n = 15), track and field (n = 13), soccer (n = 4),
determined by their current contract status. If the player baseball (n = 3), golf (n = 2), tennis (n = 2), mixed martial
was found to still be on contract with an NBA team, he arts (n = 1), and Australian rules football (n = 1). The per-
was designated as active. If the player in question was centage of multisport athletes per draft class fluctuated
found to not be on contract with an NBA team, he was des- from year to year: 2008 (3.3%), 2009 (23.3%), 2010
ignated as inactive. (3.3%), 2011 (6.7%), 2012 (23.3%), 2013 (26.7%), 2014
(16.7%), and 2015 (13.3%). No significant differences
(Tables 1 and 2) were seen between the multisport and
Statistical Analysis
single-sport cohort in regard to age, height, weight, and
All data were pooled. The unpaired Student t test was used BMI when drafted as well as position played in the NBA.
to determine significant differences between the single- The multisport cohort played in a statistically signifi-
sport and multisport cohort in regard to age, height, weight, cantly greater percentage of total games (78.4% vs 72.8%,
and BMI. The 2-tailed Fisher exact test was used to deter- P \ .001; Table 3).
mine differences in position played between both cohorts. A total of 10 major injuries were incurred by the multisport
The 1-tailed chi-square test with Yates correction was used cohort compared with 116 major injuries for the single-sport
to determine statistically significant differences between cohort. Even with a greater percentage of games played, the
the multisport and single-sport cohort in regard to percent- multisport cohort was less likely to have sustained at least 1
age of total games played during their NBA careers. The major injury (25% vs 43%, P = .03; Table 4).
1-tailed Fisher exact test was used to determine whether sig- The most common injuries sustained in both groups
nificant differences existed between the 2 patient cohorts in included ACL, collateral ligament, meniscal, ankle, and
regard to major injuries sustained and active status in the back injuries (Table 5).
NBA. One-tailed tests were used in the statistical analysis Finally, a greater percentage of multisport athletes
due to the preexisting evidence that single-sport specializa- were still active in the league compared with the single-
tion has been associated with an increased injury risk. A sport cohort (94% vs 81.1%; P = .03).
power analysis was performed and indicated that to detect
a 16% difference in major injuries sustained between the
cohorts, with an alpha of .05 and a beta of .20, a minimum
of 36 athletes were needed in each cohort. DISCUSSION

This is one of the first studies of timing of sport specializa-

RESULTS tion and outcomes in American professional athletes.
While media attention and academic study of timing of
During the time period examined, 237 athletes were iden- sport specialization have increased recently, few studies
tified as first-round draft picks who played in the NBA. Of have compared outcomes in athletes who have specialized
these 237 athletes, 36 (15%) were multisport athletes in at an early age versus a late age. Those studies that exist
high school and 201 (85%) were involved in a single sport. have relatively small sample sizes, rely on patient report-
The multisport cohort was involved in the following sports: ing of participation in other sports, and have focused on
AJSM Vol. 46, No. 2, 2018 NBA Sport Specialization 405

TABLE 3
Differences Between Single-Sport and Multisport Cohorts in Percentage of Total Games Played

2008-2015 Draft Classes (Aggregate) Games Played, n Games Missed, n Total Possible Games, n % of Games Played

Multisport 9220 2544 11,764 78.4a

Single sport 47,901 17,903 65,804 72.8a
Total 57,121 20,447 77,568 73.6

a
P \ .001.

TABLE 4
Differences Between Single-Sport and Multisport Cohorts in Percentage of Players With At Least One Major Injury

2008-2015 Draft Sustained At Least Sustained No Major

Classes (Aggregate) 1 Major Injury, n Injuries, n % Major Injury % No Major Injury

Multisport 9 27 25.0 75.0a

Single sport 86 115 42.8 57.2a
Total 95 142 40.1 59.9

a
P = .03.

collegiate, national team, and Olympic athletes rather TABLE 5

than professional athletes.16,34,49 Injuries Sustained During Participation in the NBA
In this study, early specialization was defined as lack of for Multisport and Single-Sport Athletes
participation in multiple sports in high school. The Amer-
ican Academy of Pediatrics position statement has defined Injury Type Multisport, n Single Sport, n
early specialization in sport as limiting participation to Anterior cruciate ligament 3 8
a single sport at the exclusion of other sports for a majority Lateral collateral ligament 0 1
of the year before the onset of puberty.5 According to the Medial collateral ligament 1 6
National Health and Nutrition Examination Survey, the Posterior cruciate ligament 0 2
average age of initiation of pubarche for boys is approxi- Meniscus 1 14
mately 12 years, and sexual maturity is reached on aver- Fracture 1 25
age at 15.5 years.46 The average age of starting high Hamstring 0 2
school in the United States is 13 to 14 years, which is Tendon 0 9
Muscle 0 2
around the time of puberty for the majority of boys. Hence,
Tissue scarring 0 1
by using multiple sport participation in high school as an Sprain 1 6
indication of late specialization (ie, after onset of puberty) Ankle or knee injury (unspecified, 0 14
and single-sport participation in high school as early spe- but required surgery)
cialization, we anticipated that the groups in this study Back 1 bulging/herniated disc 1 7
roughly represented pre- and postpubertal specialization. Hernia 0 4
This method may be a helpful designation in future Soreness 0 1
research examining timing of sport specialization, but it Hip 1 4
may be more applicable to male athletes rather than Dislocation 0 2
female athletes given the earlier age of puberty in girls. Bone bruise 0 1
Shoulder (unspecified) 0 7
The population of interest for this study was first-round
Nerve 1 0
draft picks in the NBA from 2008 to 2015. Since 2006, ath- Total 10 116
letes must be at least 19 years of age and 1 year beyond
high school graduation. Approximately 30 athletes are
selected in the first-round each year. By limiting the study articles. While theoretically an athlete could have been
to first-round draft picks, the study population was more misclassified as a single-sport athlete when in fact he
homogeneous in terms of age, projected career length, was a multisport athlete, the inclusion of athlete data after
and skill relative to the entire NBA population. Addition- the advent of the internet makes this less likely. Further-
ally, first-round draft picks were more likely to have robust more, the use of publically available records to collect data
documentation of their athletic career before their profes- on injuries, games played, and career length in profes-
sional career given their elite status within the NBA. We sional athletes has been successfully used by prior
used multiple internet-based sources to determine whether studies.6,40,47,50,51,53
an athlete was multisport, including player biographies, We found that the majority (85%) of first-round draft
high school athletic sites, and newspaper and magazine picks in the NBA had specialized in basketball by the
406 Rugg et al The American Journal of Sports Medicine

time they reached high school. Specialization rates vary game schedule, age, and prior injury.35,44,47,52 An injury
widely between different sports and tend to be poorly was included in the analysis if it involved the neck, back,
described in team sports such as football, basketball, and torso, or lower extremities and resulted in a 10-game
baseball. Rather, individual sports and those requiring absence either acutely or over the course of the season.
highly technical skills, such as tennis, dance, gymnastics, Upper extremity injuries and concussions were not
and swimming and diving, have been more comprehen- included in the analysis because in basketball, these inju-
sively studied in regard to specialization.13,22 Post et al38 ries are less likely to represent overuse or muscle imbal-
studied the specialization rates in 343 Division I athletes ance–related injuries. Yet, this is an important area to
and found that by the senior year of high school, only examine in the future as concussions are potentially
41.1% of athletes were highly specialized and not partici- more likely to occur in overuse situations and/or fatigue
pating in another sport. However, only 13 men’s basketball situations in which the patient is not alert or able to avoid
players were included in that investigation and their spe- injury.
cialization patterns were not individually analyzed. The use of a 10-game cutoff meant that chronic overuse
DiFiori et al8 studied collegiate athletes in a Division I pro- injuries that affected a significant number of games were
gram and found that the vast majority had participated in counted. Studies of young athletes who specialize early in
multiple sports as children and that 70% did not specialize their career have demonstrated increased overuse inju-
until after age 12. No studies of sport specialization pat- ries,18,24,39 but it is unknown whether the specialized group
terns in any form of elite basketball have been published. in our study had a more extensive injury history than the mul-
While the early specialization rate was high in the current tisport group. Sport diversification may be injury protective to
study, it is unknown whether this rate is similar to that of athletes from a neuromuscular standpoint. For example,
all NBA players or of collegiate-level basketball players, female youth athletes who play multiple sports were found
and this is a potential area of future study. to be less likely to experience overuse injuries or muscle imbal-
The NBA season consists of 82 games per team over a 6- ance injuries of the knee.18,27 Our findings suggest that the
to 9-month period depending on playoff berth. In studying protective effect of being a multisport athlete in high school
an athlete’s performance and contribution to his team, may persist for years into an athlete’s professional career.
‘‘games played’’ is a commonly used proxy in the National Of note, injury types in our study were similar to those
Football League, National Basketball Association, and described in large epidemiologic studies of NBA athletes and
Major League Baseball.11,30,33,45 In this study, this variable most frequently included knee, ankle, and back injuries.10,44
was calculated as a percentage of total possible games In this study, a greater percentage of multisport athletes
played, to account for differences in career length. Multi- were still active in the NBA at the 2015 time point, indicat-
sport athletes played in more games than single-sport ath- ing longer careers than their single-sport counterparts. Few
letes by approximately 5.4%, or 4.4 games per season. data are available regarding causes for decreased career
Several factors may have likely contributed to the increased length among professional basketball players. The most
games played by multisport athletes in this study; most common causes for reduced longevity in the literature are
notably, multisport athletes had fewer injuries that affected lower extremity and back injuries, including ACL injury,26
participation during their career, allowing them to partici- Achilles tendon rupture,2 and lumbar disc herniation.3
pate more while healthy. Numerous studies have demon- Our results mirror the findings in these studies in that
strated that lower extremity and back injuries may result the single-sport group with increased injury rates addition-
in fewer games played in the NBA population.3,19,26 Future ally had decreased longevity compared with multisport
research could be performed to determine how the impact of athletes.
games played translates to actual statistical performance, This study had a number of methodological limitations.
although multiple other confounding factors can determine Use of internet-based records to determine whether an ath-
statistical performance including teammate skill-set, coach- lete was multisport could have resulted in misclassification;
ing philosophy, and position. however, inclusion of athletes after 2008 (when the internet
Furthermore, Fransen et al15 studied male athletes was in widespread use for athletic reporting) makes this
between ages 6 and 12 and found that participation in mul- less likely. We also could not gauge the level of participation
tiple sports had a positive influence on gross motor coordi- of the multisport athletes due to the web-based search we
nation, speed, endurance, and strength. Abernethy et al1 performed. In addition, ethical issues should be considered
found that formation of elite decision making was more fac- when using publically available information,25 although
ile in a group of athletes who had diverse sporting experi- we did not report any protected health information. Internet
ence in their youth. It is possible that the diverse exposure reporting of injuries has been described as an acceptable
to sports experienced by the multisport athlete resulted in method in other studies of professional athletes.5,49,50 More-
achieving a more elite status within the first-round draft over, major injuries were highly likely to be discovered
pick group, contributing to the increased games played using internet search given that they affected game partic-
observed in this group. ipation and were potentially season-ending. The NBA injury
Despite participating in more competition, basketball registry, to which we do not have access, would potentially
players who played multiple sports in high school had provide the most accurate information in regard to injury
fewer major injuries during their careers in the NBA. Pre- risk. Our sample size of multisport athletes was relatively
viously described risk factors for injury in the NBA have small, although our analyses did have statistical and clini-
included participation in competition rather than practice, cal significance.
AJSM Vol. 46, No. 2, 2018 NBA Sport Specialization 407

Another limitation of our study was exclusion of postsea- athletes who participate in multiple sports into high school
son games. The exclusion of data collected from the postsea- may have more productive, healthy careers in regard to
son can be primarily justified by the fact that not all teams games played, avoidance of injury, and career length com-
in the NBA play the same number of games in the postsea- pared with those limited to a single sport in high school.
son or may not even reach the postseason altogether. Due to While these findings are mainly applicable to elite, profes-
the variability in the access to postseason from team to team sional basketball players, their implications should pro-
and season to season, there is also an inequity in the num- mote further research in timing of sport specialization
ber of games played by the players, which could affect injury and the effects on attainment of elite status and healthy
susceptibility during the playoffs. The logic follows that if participation at the elite level.
a player logs more games during the postseason, the likeli-
hood of getting injured is greater for that player compared
with a player whose team is not playing in the postseason.
REFERENCES
Thus, to standardize the number of games played, the scope
of our study was narrowed to only regular season games, to 1. Abernethy B, Baker J, Cote J. Transfer of pattern recall skills may
which all players hypothetically have equal access. Further- contribute to the development of sport expertise. Appl Cognitive Psy-
more, the adding of playoff games to certain players’ careers chol. 2005;19(6):705-718.
could artificially inflate the perception of how durable or 2. Amin NH, Old AB, Tabb LP, Garg R, Toossi N, Cerynik DL. Perfor-
healthy the player is. Similarly, durable players on teams mance outcomes after repair of complete Achilles tendon ruptures
in National Basketball Association players. Am J Sports Med.
that do not make the postseason would have their durability
2013;41(8):1864-1868.
deflated due to not having the opportunity to play more 3. Anakwenze OA, Namdari S, Auerbach JD, et al. Athletic performance
games. Both of these assumptions can be drawn from the outcomes following lumbar discectomy in professional basketball
premise that the opportunity to play in and advance within players. Spine (Phila Pa 1976). 2010;35(7):825-828.
the postseason is a result of collective team success, which 4. Bell DR, Post EG, Trigsted SM, Hetzel S, McGuine TA, Brooks MA.
cannot necessarily be positively correlated with any given Prevalence of sport specialization in high school athletics: a 1-year
observational study. Am J Sports Med. 2016;44(6):1469-1474.
specific player’s individual durability or lack thereof. Fur-
5. Brenner JS. Sports specialization and intensive training in young ath-
ther study that controls for postseason participation can letes. Pediatrics. 2016;138(3): e20162148.
be performed in the future. Additionally, if a player sat 6. Casals M, Martinez JA, Cayla JA, Martin V. Do basketball players
out a game or was benched without an injury, these games have a high risk of pulmonary embolism? A scoping review. Med
were still counted within the ‘‘possible games played’’ variable. Sci Sports Exerc. 2016;48(3):466-471.
The decision to not play an athlete can be multifactorial— 7. Cheatham SA, Little BA. Early sports specialization: helpful or harm-
ful? Orthopedics. 2015;38(12):724-725.
minor injury not reported, performance related, and/or
8. DiFiori J, Gray A, Kimlin E. Sports participation history and early
fatigue—all of which may be affected by multisport participa- sports specialization in National Collegiate Athletic Association Divi-
tion in high school. The goal of our study was to look at overall sion I athletes. Clin J Sport Med. 2011;21(2):168.
participation in games once drafted, with all players being 9. Dodwell ER, Lamont LE, Green DW, Pan TJ, Marx RG, Lyman S. 20
subject to the same probability of being benched or not played. years of pediatric anterior cruciate ligament reconstruction in New
We believe that the data were not affected by the small, if any, York State. Am J Sports Med. 2014;42(3):675-680.
percentage of games affected. 10. Drakos MC, Domb B, Starkey C, Callahan L, Allen AA. Injury in the
National Basketball Association: a 17-year overview. Sports Health.
Finally, these findings can be applied to only elite male
2010;2(4):284-290.
basketball players. First-round draft picks do not repre- 11. Erickson BJ, Gupta AK, Harris JD, et al. Rate of return to pitching and
sent the NBA population as a whole. For the purposes of performance after Tommy John surgery in Major League Baseball
our study, they represent an ‘‘elite’’ level of athlete enter- pitchers. Am J Sports Med. 2014;42(3):536-543.
ing the NBA, and they had the most readily available infor- 12. Ericsson KA, Krampe RT, Heizmann S. Can we create gifted people?
mation about their sports participation on the internet due Ciba Found Symp. 1993;178:222-231; discussion 232-249.
13. Feeley BT, Agel J, LaPrade RF. When is it too early for single sport
to their high level of athletic achievement at a young age.
specialization? Am J Sports Med. 2016;44(1):234-241.
It is unknown whether these findings would also apply to 14. Finley B. A single goal in common. New York Times. December 17,
female athletes or those who play other sports. 2006: L13. http://www.nytimes.com/2006/12/17/nyregion/nyregions-
Although no differences were seen in regard to age, pecial2/17Rsports.html?scp=1&sq=Bill%20Finley%20a%20single
height, weight, BMI, and position played in our study, fur- %20goal%20in%20common%202006&st=cse&_r=0. Accessed
ther prospective study is necessary to look at the interplay December 16, 2006.
15. Fransen J, Pion J, Vandendriessche J, et al. Differences in physical
of body type with injury risk. Certain body types and bas-
fitness and gross motor coordination in boys aged 6-12 years spe-
ketball positions may be less compatible with other sports cializing in one versus sampling more than one sport. J Sports Sci.
and affect injury risk. For example, a very tall basketball 2012;30(4):379-386.
center may be unable to play other sports (precluding mul- 16. Gullich A. International medallists’ and non-medallists’ developmen-
tisport specialization) but may have an increased risk of tal sport activities—a matched-pairs analysis. J Sports Sci.
foot or ankle injuries due to his height (not lack of multi- 2017;35(23):2281-2288.
sport participation). 17. Gullich A, Emrich E. Considering long-term sustainability in the devel-
opment of world class success. Eur J Sport Sci. 2014;14(suppl
Despite these limitations, this study is the first of its
1):S383-397.
kind to evaluate clinical outcomes between early and late 18. Hall R, Barber Foss K, Hewett TE, Myer GD. Sport specialization’s
specialization athletes in an elite professional athletic association with an increased risk of developing anterior knee pain
population. Our findings suggest that elite professional in adolescent female athletes. J Sport Rehabil. 2015;24(1):31-35.
408 Rugg et al The American Journal of Sports Medicine

19. Harris JD, Walton DM, Erickson BJ, et al. Return to sport and perfor- 37. The National Federation of State High School Associations. 2015-16
mance after microfracture in the knees of National Basketball Asso- High school athletics participation survey. http://www.nfhs.org/Part
ciation players. Orthop J Sports Med. 2013;1(6):2325967113512759. icipationStatistics/PDF/2015-16_Sports_Participation_Survey.pdf.
20. Hill GM, Simons J. A study of the sport specialization on high school Accessed February 1, 2017.
athletics. J Sport Social Issues. 1989;13(1):1-13. 38. Post EG, Thein-Nissenbaum JM, Stiffler MR, et al. High school sport
21. Hume PA, Hopkins WG, Robinson DM, Robinson SM, Hollings SC. specialization patterns of current division I athletes. Sports Health.
Predictors of attainment in rhythmic sportive gymnastics. J Sports 2017;9(2):148-153.
Med Phys Fitness. 1993;33(4):367-377. 39. Post EG, Trigsted SM, Riekena JW, et al. The association of sport
22. Jayanthi N, Dechert A, Durazo R, Dugas L, Luke A. Training and specialization and training volume with injury history in youth ath-
sports specialization risks in junior elite tennis players. J Med Sci letes. Am J Sports Med. 2017;45(6):1405-1412.
Tennis. 2011;16:14-20. 40. Ruestow PS, Duke TJ, Finley BL, Pierce JS. Effects of the NFL’s
23. Jayanthi N, Pinkham C, Dugas L, Patrick B, Labella C. Sports spe- amendments to the free kick rule on injuries during the 2010 and
cialization in young athletes: evidence-based recommendations. 2011 seasons. J Occup Environ Hyg. 2015;12(12):875-882.
Sports Health. 2013;5(3):251-257. 41. Schwebel DC, Yang J. Elite US college admissions: could the quest
24. Jayanthi NA, LaBella CR, Fischer D, Pasulka J, Dugas LR. Sports- for admission increase overuse injury risk? J Inj Violence Res.
specialized intensive training and the risk of injury in young athletes: 2016;8(2):109-110.
a clinical case-control study. Am J Sports Med. 2015;43(4):794-801. 42. Sheu Y, Chen LH, Hedegaard H. Sports- and recreation-related injury
25. Keller HE, Lee S. Ethical issues surrounding human participation episodes in the United States, 2011-2014. Natl Health Stat Report.
research using the internet. Ethics Behav. 2003;13(3):211-219. 2016(99):1-12.
26. Kester BS, Behery OA, Minhas SV, Hsu WK. Athletic performance 43. Smith NA, Chounthirath T, Xiang H. Soccer-related injuries treated in
and career longevity following anterior cruciate ligament reconstruc- emergency departments: 1990-2014. Pediatrics. 2016;138(4):e20160346.
tion in the National Basketball Association [published online March 44. Starkey C. Injuries and illnesses in the national basketball associa-
12, 2016]. Knee Surg Sports Traumatol Arthrosc. doi:10.1007/ tion: a 10-year perspective. J Athl Train. 2000;35(2):161-167.
s00167-016-4060-y. 45. Steingrover C, Wattie N, Baker J, Schorer J. Does relative age affect
27. Ladenhauf HN, Graziano J, Marx RG. Anterior cruciate ligament pre- career length in North American professional sports? Sports Med
vention strategies: are they effective in young athletes—current con- Open. 2016;2(1):18.
cepts and review of literature. Curr Opin Pediatr. 2013;25(1):64-71. 46. Sun SS, Schubert CM, Chumlea WC, et al. National estimates of the
28. Leathers MP, Merz A, Wong J, Scott T, Wang JC, Hame SL. Trends timing of sexual maturation and racial differences among US chil-
and demographics in anterior cruciate ligament reconstruction in the dren. Pediatrics. 2002;110(5):911-919.
United States. J Knee Surg. 2015;28(5):390-394. 47. Teramoto M, Cross CL, Cushman DM, Maak TG, Petron DJ, Willick
29. Lord M. Too much, too soon? Doctors group warns against early SE. Game injuries in relation to game schedules in the National Bas-
specialization. US News World Rep. 2000;129(3):46. ketball Association. J Sci Med Sport. 2017;20(3):230-235.
30. Mai HT, Alvarez AP, Freshman RD, et al. The NFL Orthopaedic Sur- 48. Tritto C. Paying for performance: Velocity, HammerBodies cash in as
gery Outcomes Database (NO-SOD): the effect of common ortho- parents invest thousands in child athletes. St. Louis Business Jour-
paedic procedures on football careers. Am J Sports Med. nal. July 9, 2006. http://www.bizjournals.com/stlouis/stories/2006/
2016;44(9):2255-2262. 07/10/story2.html?page=1. Accessed November 5, 2008.
31. Malina RM. Early sport specialization: roots, effectiveness, risks. Curr 49. Vaeyens R, Gullich A, Warr CR, Philippaerts R. Talent identification
Sports Med Rep. 2010;9(6):364-371. and promotion programmes of Olympic athletes. J Sports Sci.
32. Mall NA, Chalmers PN, Moric M, et al. Incidence and trends of ante- 2009;27(13):1367-1380.
rior cruciate ligament reconstruction in the United States. Am J 50. Watkins RG, Hanna R, Chang D, Watkins RG III. Return-to-play out-
Sports Med. 2014;42(10):2363-2370. comes after microscopic lumbar diskectomy in professional athletes.
33. Minhas SV, Kester BS, Larkin KE, Hsu WK. The effect of an orthopae- Am J Sports Med. 2012;40(11):2530-2535.
dic surgical procedure in the National Basketball Association. Am J 51. Webner D, Iverson GL. Suicide in professional American football
Sports Med. 2016;44(4):1056-1061. players in the past 95 years. Brain Inj. 2016;30(13-14):1718-1721.
34. Moesch K, Elbe AM, Hauge ML, Wikman JM. Late specialization: the 52. Yeh PC, Starkey C, Lombardo S, Vitti G, Kharrazi FD. Epidemiology
key to success in centimeters, grams, or seconds (cgs) sports. of isolated meniscal injury and its effect on performance in athletes
Scand J Med Sci Sports. 2011;21(6):e282-e290. from the National Basketball Association. Am J Sports Med.
35. Myer GD, Jayanthi N, Difiori JP, et al. Sport specialization, part I: 2012;40(3):589-594.
does early sports specialization increase negative outcomes and 53. Yengo-Kahn AM, Zuckerman SL, Stotts J, et al. Performance follow-
reduce the opportunity for success in young athletes? Sports Health. ing a first professional concussion among National Basketball Asso-
2015;7(5):437-442. ciation players. Phys Sportsmed. 2016;44(3):297-303.
36. Myer GD, Jayanthi N, DiFiori JP, et al. Sports Specialization, Part II: 54. Zhang AL, Sing DC, Rugg CM, Feeley BT, Senter C. The rise of con-
Alternative solutions to early sport specialization in youth athletes. cussions in the adolescent population. Orthop J Sports Med.
Sports Health. 2016;8(1):65-73. 2016;4(8):2325967116662458.

For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav.