SciELO - Scientific Electronic Library Online

vol.23 issue3The game variants in Europe. Trends and perspectives during youth competitive stagesCarbohydrate mouth rinse reduces rating of perceived exertion but does not affect performance and energy systems contribution during a high-intensity exercise author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Motriz: Revista de Educação Física

On-line version ISSN 1980-6574

Motriz: rev. educ. fis. vol.23 no.3 Rio Claro  2017  Epub Dec 21, 2017 

Original Article

Incidence of musculoskeletal injuries in professional soccer players from Aracaju/SE-Brazil.

João Bourbon de Albuquerque II1 

Walderi Monteiro da Silva Júnior2 

Matheus Melo Barreto2 

José Guilherme Vieira Bonfim2 

Marco Antônio Prado Nunes2 

1Universidade de São Paulo, USP, Ribeirão Preto, SP, Brazil

2Universidade Federal do Sergipe, UFS, Aracajú, SE, Brazil



To verify the incidence of injuries in professional soccer athletes in Aracaju-SE and to identify associated factors.


This was an observational, prospective cohort study involving 39 healthy athletes, followed for seven months and evaluated on four occasions (start on preseason and two evaluations with three months of interval between each) through data collection sheet and orthopedic physical examination.


Participants were 20 athletes from Club Sportivo Sergipe and 19 from Associação Desportiva Confiança, with mean age of 26 years (CI 95%: 25.2-28 years), and mean career time of 9.6 years (CI 95%: 7.9-11.3 years). The 13 (2.4 / 1000 hours) diagnosed injuries occurred almost exclusively in the lower limbs and thigh muscle stretch was the most common injury. Injuries were mostly mild to moderate (69%). There was no significant association with any of the variables analyzed.


Low incidence of injuries was observed in this study. Predominantly, lesions occurred on non-rainy days and were due to trauma. Lower limbs were the most affected location and thigh stretch was the most common injury.

Keywords (DeCS): athletic injuries (identificador DeCS: D001265); soccer (identificador DeCS: D012916); lower extremity (identificador DeCS: D35002); epidemiology (identificador DeCS: Q000453).


Soccer, the most practiced sport, is also responsable for attracting the largest number of spectators in the world1,2. Of a sport modality with such a capacity of public attraction and of potential income, it is expected exhibitions and disputes of high technical level, with athletes performance at its highest in the greater possible number of games2.

Nevertheless, participation in extremely competitive sports is responsible for high physical demand for athletes3. During a match, players go through an average of 10 km divided into: running (40%), walking (25%), trotting (15%), speed (10%) and running back (10%)4. In addition, the sport is characterized by short, fast and non-continuous movements, such as acceleration, deceleration and abrupt changes of direction, in addition to intense physical contact5.

As a result, injuries incidence and associated costs tend to be high. It is estimated that soccer athletes are at risk of injury a thousand times greater than workers in industries considered to be at high risk6. In general, the global economic impact of football-related injuries exceeds $ 30 billion2. Therefore, in order to create injury prevention strategies, there is great interest in determining injuries risk factors7,8.

Epidemiological studies are a useful tool in trying to define injuries risk factors and the athlete’s risk profile7. In a classic study from 1992, van Mechelen, Hlobil, Kemper.9 stated that these studies should be performed in four stages: 1) investigation of injurie's epidemiological profile, identifying risk factors; 2) attempt to define causes and mechanisms for the occurrence of injuries; 3) creation of prevention strategies and 4) assessment of the effectiveness of these strategies by repeating the first step.

More recently, experts from Union of European Football Associations (UEFA)1 and The Brazilian Football Confederation (CBF)10 have attempted to solve a problem regarding the lack of homogeneity in epidemiological studies. The authors sought to standardize injurie's definition and classifications, as well as to propose a model for the conduction of the studies. Studies11 based on these models have shown a greater number of injuries affecting soccer athletes lower limbs musculature, with greater chance of ocurrence in match rather than training days and in non-rainy days, for instance.

However, available data from scientific research in Brazil is still under the satisfactory level12. The small amount of papers available is derivated from works with elite clubs, what represents a different scenario from that experienced by most Brazilian clubs10,13. In this context, the lack of information from scientific articles represents a barrier to the creation and implementation of injury prevention strategies, the ultimate goal of epidemiological studies in sports4, 7,14.

Because it is unprecedented, the present study has the potential to generate data that can benefit more than twenty thousand athletes distributed in more than seven hundred professional soccer teams with a level of performance equivalent to that of the fourth division of the Brazilian championship or lower, which represents the majority of clubs in the country14,15. Therefore, present study's goal is to determine the incidence of injuries in male professional athletes from clubs in serie D national championship and its predisposing factors.


This was an observational prospective (cohort) study, planned according to the Declaration of Helsinki and Resolution No. 466/12 of the National Health Council and submitted to the Ethics and Research Committee of the Federal University of Sergipe (approbation number 544.973). The work was initiated after the release of this committee and data were collected after signature of informed consent from all participating athletes.

Male professional soccer players from the main clubs of the State of Sergipe, in Brazil's northern coast, were selected: Associação Desportiva Confiança and Club Sportivo Sergipe. Both are located in the city of Aracaju, the state capital and had similar technical level, being in the same category within the national scenario, at the moment in which research was conducted. The teams began the pre-season process at the same time and during the 2014 season they would participate in the same championships: Northeast Cup (beginning of the year) and Sergipano Championship (until mid-year). In addition, they would be able to participate in the serie "D" of the Brazilian Championship, depending on the performance in the regional championship.

A total of 59 athletes, who could be observed for at least 08 weeks (minimum exposure time), were selected for this study. Twenty athletes with injuries already installed at the time of the first evaluation were excluded. Despite being considered a exclusion criteria, no injuries occurring outside the professional activity, in extra-club activities, were observed.


A proportion of lesions in the lower limbs of 25% was assumed to calculate the sample volume15. Considering a significance level of 5%, a precision of 14% was determined a sample size of 37 players.


The outcome was the presence or absence of lesions (lesion vs. non-lesion). In this context, CBF10 and UEFA1 models were used for general principles, study design, and injury definitions. Thus, an injury was defined as that which prevents player from active and complete participation in the following training session or game. The player was considered injured until return to training activities occurred and was available for playing (definition of injury duration).

Lesion intensity was defined and classified using absence time as the reference: minimal injury (up to three days of absence), mild (4 to 7 days), moderate (from 7 to 28 days) and severe (more than 28 days). Regarding the mechanism, the lesions were classified as traumatic, when resulting from an identifiable acute event, and atraumatic, when they were gradually established.

Types of variables

Physical, socio-demographic characteristics and aspects regarding soccer practice were collected through a sheet for semi-coded data collection and are summarized in figure 1.

Figure 1 Variables searched and obtaining method 

Regarding self-referred categorical variables (height, weight and career duration), it should be pointed out that previous studies in several countries, including Brazil, have validated this form of collection and have demonstrated the method accuracy16-18.

For circumference measurements, 150cm flexible and inelastic tapes were used and the athletes were positioned in dorsal decubitus. Thigh circumference was obtained at a point 10 cm proximal to the proximal border of the medial condyle of the femur and the circumference of the calf at a point 10 cm distal to the tibia anterior tuberosity. Two measurements were performed for each segment and the analysis was based on the mean value obtained19,20.


This was a longitudinal, prospective and observational study that was conducted throughout 2014 season of Sergipan soccer. During the follow-up period, athletes answered the mentioned questionnaire and underwent a comprehensive physical orthopedic examination, in four occasions: at the beginning of the season (beginning of the preseason - December / 2013), every 3 months and at the end of the season. On the first occasion, the data were collected through individual physical data sheets at the beginning or at the end of each training day, so that there was no interference with teams routine.

At this stage, athletes and coaches were instructed to record any injuries that would occur throughout the year. The teams had unrestricted telephone and email access to the responsible investigator and were advised to contact him as soon as they became aware of an injury. In this way, all injuries in the period were documented at the time of their occurrence and the involved athletes reexamined by this investigator, even at dates that did not correspond to scheduled interviews.

In the other opportunities, which occurred with approximate interval of 3 months, more succinct questionnaires were applied to identify lesions. In addition, players were reexamined whenever there was recurrence of orthopedic complaints.

All researchers were trained prior to questionnaires application. The data were collected with the presence of the researcher in charge at all times. After questionnaires completion, they were discharged whenever inconsistencies were detected and new ones were applied in the next interview with the same player.

Players physical examination was carried out exclusively by the responsible researcher. This examination was performed at the end of the first interview, still during preseason, and repeated whenever there were new orthopedic complaints in the subsequent interviews.

The study lasted approximately seven months. It started in December 2013 and ended in June 2014, at the end of the Sergipe 2014 football championship, which corresponds to a full season in Sergipe football.

Data analysis

The information collected was taken to the database for further statistical analysis. The database, as well as the tables and charts, were built in Microsoft Excel 2010.

Data descriptive analysis was performed using absolute and relative frequencies for categorical variables and through central tendency and variability measures for numerical variables. The differences between proportions were tested using the Pearson Chi-square test (with a significance level of 5%). For the analysis of mean comparisons between independent samples, Student's t-statistic test was used. Program R version 3.2.3 (R Core Team, 2015) was used for calculations21.

Regarding density of incidence calculation (number of injuries per 1000 hours of training or playing) was considered, since players had different times of exposure to the risk factors. The relative risk calculation was also carried out, which is characterized by the ratio between exposed and non exposed incidence.


Fifty nine athletes were selected for the study. Of these, 20 athletes were excluded because they were already injured at the time of the initial evaluation. Thus, 39 athletes entered the study.

There were 20 players from Sergipe (51%) and 19 (49%) from Confiança. The average age of athletes was 26.6 years (95% CI: 25.2 to 28 years), with a mean career time of 9.6 years (95% CI: 7.9 to 11.3 years), mean age at which they started their careers of 17.3 years (IC95%: 16.7 to 17.9 years) and number of titles per player of 3, on average (95% CI: 2.4 to 3.9 years). The athletes' mean body mass index was 23.8 kg / m2 (95% CI: 23.3 to 24.2 kg / m2). Most players considered himself black or mulatto (16/39, 41%).

Concerning previous medical history, four players (10%) reported previous orthopedic surgeries, three (8%) reported knee infiltrations and one third of athletes (13/39) reported social alcohol consumption.

Information regarding player distribution by position, dominance leg, team situation, presence or abscence of knee complaints can be found in table 1.

Table 1 Distribution of Sergipe and Confiança 2014 football male athletes according to characteristics and conditions of performance. 

n %
Player positions
Attacker 9 23%
Goalkeeper 5 13%
Side 4 10%
Midfielder 9 23%
Box-to-box midfielder 10 26%
Defender 2 5%
Field condition*
Good 15 38%
Bad 24 62%
Dominance leg
Right 29 74%
Left 10 26%
Player stituation
First string 20 51%
Reserve bench 19 49%
Injury prevention program*
Yes 34 87%
No 5 13%
Knee complains
Yes 4 10%
No 35 90%
Total 39 100%

* According to the athlete's assessment.

Noten: sample (absolute frequency); %: Relative frequency

Thirteen injuries were diagnosed during the season, eight in Sergipe (3.5 / 1000h) and five in Confiança (2 / 1000h), with a total density of incidence of approximately 2.4 injuries per 1000 hours. Thigh muscle stretching was the main lesion (38%) observed. Seven injuries (54%) occurred during official games and the most common mechanism was trauma (11/13, 85%) (table 2).

Table 2 Characteristics of lesions diagnosed in male soccer athletes in the city of Aracaju - season 2014. 

n %
Moment of injury
Match 6 46%
Training session 7 54%
Injury etiology
Traumatic 11 85%
Non traumatic 2 15%
Trauma mechanism
Direct 1 8%
Kick 2 15%
Contact 1 8%
Stretching 2 15%
Rotational 2 15%
Jump 2 15%
Hyperextension (hand/wrist) 1 8%
Non-trau,atic injuries 2 15%
Weather condition
Rain 2 15%
Without rain 11 85%
Lesion type (diagnosis)
Thigh stretch 5 38%
Knee sprain 4 31%
ACL lesion 1 8%
Ankle sprain 1 8%
Fractures 2 15%
Total 13 100%

Note n: sample (absolute frequency); %: Relative frequency

Lesions intensity was mild to moderate in about 69% of the cases (figure 2). Of the 13 injured athletes, only one (8%) considered himselfe asymptomatic when returning to activities. Seven athletes reported a return even without syntoms complete resolution (54%) and five athletes (38%) were unable to return until the end of the season.

Figure 2 Severity of the lesions identified in male soccer athletes in the city of Aracaju - season 2014. 

Regarding the variables studied, none of them presented a significant association (p value <0.05) with the analyzed outcome, as shown in Tables 3 and 4.

Table 3 Statistical evaluation of the association between each variable and the outcome. Sergipe, Brazil, 2014. 

Lesion No lesion Total RR IC 95% X 2 p value
Sergipe 8 12 20 1.52 (0.60, 3.83) 0.821 0.365
Confiança 5 14 19
White/ hispanic 3 7 10 0.87 (0.30, 2.54) 0.067 0.795
Black 10 19 29
Field condition
Bad 10 14 24 2.08 (0.68, 6.37) 1.95 0.163
Good 3 12 15
Dominance leg
Right 10 19 29 1.15 (0.39, 3.36) 0.067 0.795
Left 3 7 10
Team situation
First-string 9 10 19 2.37 (0.87, 6.42) 3.284 0.07
Reserve bench 4 16 20
Total 13 26 39

Note RR: relative risk; 95% CI: confidence interval at 95%; X2: chi-square; p : level of statistical significance.

Table 4 Statistical evaluation of the association between each variable and the outcome. Sergipe, Brazil, 2014. 

Lesion No lesion Total RR IC 95% X 2 p value
Injury prevention program
No 2 3 5 1.24 (0.38, 4.02) 0.115 0.735
Yes 11 23 34
Knee complain
Yes 2 2 4 1.59 (0.53, 4.76) 0.557 0.455
No 11 24 35
Knee infiltration
Yes 1 2 3 1 (0.19, 5.29) 0 1
No 12 24 36
Alcohol consumption
Yes 5 8 13 1.25 (0.51, 3.07) 0.231 0.631
No 8 18 26
Total 13 26 39

NoteRR: relative risk; 95% CI: confidence interval at 95%; X2: chi-square; p : level of statistical significance.


The present prospective observational study identified a pattern of lesion occurrence similar to what has been reported in previous studies8,11,22-24, which are: lesions happening especially in non-rainy days, with greater involvement of the lower limbs, thigh stretching as the most frequent lesion, and the majority of injuries happening as a consequence of a traumatic event.

According to Longo, Loppini, Cavagnino, Maffulli, Denaro25, approximately 67% of the lesions occur in the lower limbs, followed by the upper limbs with approximately 13%25. The lower limbs are the most affected body segment since it is where there is greater physical demand and a greater propensity to fatigue with reduction of neuromuscular control12. More specifically, the thigh, knee and ankle are the most affected sites24,25.

A study by Ekstrand, Myklebust, Holme, Engebretsen, Bahr26 had similar results. A total of 2299 players from the top 51 teams in Europe were evaluated and the researchers found that muscle injury accounted for about 1/3 (31%) of the total number of injuries in the period under review. In addition, in this study, hamstrings were the main affected site (37%), followed by hip adductors (23%) and quadriceps (19%). It is worth mentioning that muscle injury is usually associated with an indirect mechanism of trauma due to sudden and vigorous contraction12.

Regarding the climatic condition, the number of injuries is lower on rainy days than on sunny or cloudy days24. The data shown in the present study are in agreement with this statement and this is probably due to the faster paced game play and with more contact between players on sunny days22.

In the present study, we also identified more injuries on training than on match days, what differs from data already published in the literature. Almeida, Scotta, Pimentel, Batista Júnior , Sampaio12, assessing players from Clube do Remo in Belém-PA-Brazil, identified a higher occurrence of injuries in matches12. Ekstrand, Hägglund, Waldén11, went further and reported a sixfold incidence of injuries in match days rather than on training days11. The difference between the present study and the data in the literature can not be explained by different loads of exposure to games and training between athletes from this and from other studies, since that calculated density of incidence for training and games was used to make comparisons.

Authors, in this study, had hypothesized that the incidence of lesions would be higher than that reported in other studies, especially when taking into account national data. The rationale was based on the fact that athletes in lower expression teams, with simpler training programs and physical structure, may be subject to a higher frequency of injuries and longer removal time10,15.

However, results obtained here show a density of incidence much lower than 10 to 35 injuries per 1000 hours of practice, which is the estimated value in literature2. In fact, it has been suggested by other authors24 that players in football teams with higher performance levels tend to have more injuries and this is probably due to the differences in training load and the demands placed on clubs that participate in more expressive and competitive championships.

It is worth mentioning that a small number of injuries were identified, despite the fact that the present study took the model proposed by UEFA1 and adopted by CBF10 into consideration. The recommendation from this models is that soccer epidemiological studies should be prospective and athletes should be followed for one season, at least. The authors in the present study believe that for small football teams that participate in short championships/ short seasons and whose athletes remain for limited periods (because of temporary contracts), this recommendation should be extended for a minimum of two seasons.

In addition, it was not possible to establish a significant association between any of the variables and the outcome analyzed, although other authors have already demonstrated a greater number of injuries for older athletes2,27 and for those who plays as midfielders or attackers26, for example. Even so, data resulting from this study were presented to the board and the medical department of the clubs in order that targeted prevention strategies could be adopted.


A low incidence of lesions was observed in the present study, with predominant involvement of the lower limbs of the athletes (thigh stretching being the most common lesion). There was no significant association between the variables studied and the outcome.



1. Hägglund M, Waldén M, Ekstrand J. Methods for epidemiological study of injuries to professional football players: developing the UEFA model. Br J Sports Med. 2005; 39:340-346. doi: 10.1136/bjsm.2005.018267 [ Links ]

2. Vanlommel L, Vanlommel J, Bollars P, Quisquater L, Crombrugge KV, Corten K, et al. Incidence and risk factors of lower leg fractures in Belgian soccer players. Int. J. Care Injured. 2013;(44):1847-1850. doi: 10.1016/j.injury.2013.07.002 [ Links ]

3. Arliani GG, Astur DC, Yamada RKF, Yamada AF, Fernandes ARC, Ejnisman B, et al. Professional football can be considered a healthy sport? Knee Surg Sports Traumatol Arthrosc. 2015. doi: 10.1007/s00167-015-3636-2 [ Links ]

4. Bjordal JM, Arnly F, Hannestad B, Strand T. Epidemiology of anterior cruciate ligament injuries in soccer. Am J Sports Med. 1997;25(3). doi: 10.1177/036354659702500312 [ Links ]

5. Palacio EP, Candeloro BM, Lopes AA. Lesões nos Jogadores de Futebol Profissional do Marília Atlético Clube: Estudo de Coorte Histórico do Campeonato Brasileiro de 2003 a 2005. Rev Bras Med Esporte. 2009;15(1). Doi: 10.1590/S1517-86922009000100007 [ Links ]

6. Hawkins RD, Hulse MA, Wilkinson C, Hodson A, Gibson M. The association football medical research programme: an audit of injuries in professional football. Br J Sports Med. 2001;35:43-47. doi: 10.1136/bjsm.35.1.43 [ Links ]

7. Hägglund M, Waldén M, Til L, Pruna R. Editorial. The importance of epidemiological research in sports medicine. Apunts Med Esport. 2010;45(166):57-59. doi: 10.1016/j.apunts.2010.02.006 [ Links ]

8. van Beijsterveldt AMC, van de Port IGL, Krist MR, Schmikli SL, Stubbe JH, Frederiks JEB et al. Effectiveness of an injury prevention programme for adult male amateur soccer players: a cluster-randomised controlled trial. Br J Sports Med. 2012;46:1114-1118. doi: 10.1136/bjsports-2012-091277 [ Links ]

9. Van Mechelen W, Hlobil H, Kemper HC. Incidence, severity, aetiology and prevention of sports injuries. A review of concepts. Sports Med. 1992; 14(2):82-99. [ Links ]

10. Arliani GG, Belangero PS, Runco JL, Cohen M. The Brazilian Football Association (CBF) model for epidemiological studies on professional soccer player injuries. Clinics. 2011;66(10):1707-1712. doi: 10.1590/S1807-59322011001000007 [ Links ]

11. Ekstrand J, Hägglund M, Waldén M. Injury incidence and injury patterns in professional football: the UEFA injury study. Br J Sports Med. 2011;45:553-558. doi: 10.1136/bjsm.2009.060582 [ Links ]

12. Almeida PSM, Scotta AP, Pimentel BM, Batista Júnior S, Sampaio YR. Incidência de lesão musculoesquelética em jogadores de futebol. Rev Bras Med Esporte. 2013;19(2). [ Links ]

13. Freitas RJ, Fachina G, Andrade MS, Silva FR, Waszczuk-Junior S, Montagner PC, et al. Descriptive epidemiology of injuries in a Brazilian premier league soccer team. Journal of Sports Medicine. 2013;4:171-174. doi: 10.2147/OAJSM.S44384 [ Links ]

14. Schiff MA, Caine DJ, O’Halloran R. Injury prevention in sports. Am J Lifestyle Med. 2010;4:42-64.doi: 10.1177/1559827609348446 [ Links ]

15. Bastos FN, Vanderlei FM, Vanderlei LCM, Júnior JN, Pastre CM. Investigation of characteristics and risk factors of sports injuries in young soccer players: a retrospective study. Int Arch Med. 2013;6:14. doi: 10.1186/1755-7682-6-14. [ Links ]

16. Bolton-Smith C, Woodward M, Tunstall-Pedoe H, Morrison C. Accuracy of the estimated prevalence of obesity from self reported height and weight in an adult Scottish population. J Epidemiol Community Health. 2000;54:143-148. [ Links ]

17. Osuna-Ramírez I, Hernández-Prado B, Campuzano JC, Salmeron J. Indice de masa corporal y percepción de la imagen corporal en una población adulta mexicana: la precisión del autorreporte. Salud Pública de México. 2006;48(2): 94-103. [ Links ]

18. Peixoto MRG, Benicio MHD, Jardim PCBV. Validade do peso e da altura auto-referidos: o estudo de Goiânia. Rev Saúde Pública. 2006;40(6):1065-72. [ Links ]

19. Snijder MB, Dekker JM, Visser M, Bouter LM, Stehouwer CDA, Kostense PJ, et al. Associations of hip and thigh circumferences independent of waist circumference with the incidence of type 2 diabetes: the Hoorn Study. Am J Clin Nutr. 2003;77:1192-7. [ Links ]

20. Singh R, Gupta S. Relationship of calf circumference with bone mineral density and hip geometry: a hospital-based cross-sectional study. Arch Osteoporos. 2015;10:17. doi: 10.1007/s11657-015-0221-8 [ Links ]

21. R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL: [ Links ]

22. Azubuike SO, Okojie OH. An epidemiological study of football (soccer) injuries in Benin City, Nigeria. Br J Sports Med. 2009;43:382-386. doi: 10.1136/bjsm.2008.051565 [ Links ]

23. Oztekin HH, Boya H, Ozcan O, Zeren B, Pinar P. Foot and ankle injuries and time lost from play in professional soccer players. The Foot. 2009;19:22-28. doi: 10.1016/j.foot.2008.07.003 [ Links ]

24. Aoki H, O’Hata N, Kohno T, Morikawa T, Seki J. A 15-Year Prospective Epidemiological Account of Acute Traumatic Injuries During Official Professional Soccer League Matches in Japan. Am J Sports Med. 2012;40 (5):1006-1014. doi: 10.1177/0363546512438695 [ Links ]

25. Longo UG, Loppini M, Cavagnino R, Maffulli N, Denaro V. Muscoskeletal problems in soccer players: current concepts. Clin Cases Miner Bone Metab. 2012;9(2):107-111. [ Links ]

26. Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Intrinsic risk factors for acute knee injuries among male football players: a prospective cohort study. Scand J Med Sci Sports. 2011;21:645-652. doi: 10.1111/j.1600-0838.2010.01095.x [ Links ]

27. Kristenson K, Waldén M, Ekstrand J, Hägglund M. Lower injury rates for newcomers to professional soccer: a prospective control study for 9 consecutive seasons. Am J Sports Med. 2013;41(6):1419-1425. doi: 10.1177/0363546513485358 [ Links ]

Received: May 06, 2017; Accepted: July 06, 2017

Corresponding author João Bourbon de Albuquerque II. Av. Gonçalo Prado Rollemberg, 460. Centro. Aracaju-SE-Brasil. 49.010-410.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License