The ‘Sequence of Prevention’ for musculoskeletal injuries among recreational basketballers: a systematic review of the scientific literature

ABSTRACT

Objective: Currently, there is no overview of the incidence and (basketball-specific) risk factors of musculoskeletal injuries among recreational basketball players, nor any insight into the effect of preventive measures on the incidence of basketball injuries. This study aimed to gather systematically the scientific evidence on the incidence, prevalence, aetiology and preventive measures for musculoskeletal injuries among recreational basketball players.

Methods: Highly sensitive search strategies were built based on three groups of keywords (and related search terms). Two electronic databases were searched, namely Medline (biomedical literature) via Pubmed, and SPORTDiscus (sports and sports medicine literature) via EBSCOhost.

Results: The incidence of musculoskeletal injuries among recreational basketball players ranged from 0.0047 injuries per 1,000 athlete-exposures (AE) for dental injuries to 10.1 injuries per 1000 AE for overall injuries during match play. Significant risk factors for injuries were defending, postural sway, high vertical ground reaction force during jumping and weight >75 kg. All prevention studies have shown to have a significant effect on reducing the risk of injury ranging from an odds ratio (95% confidence interval (CI)) of 0.175 (0.049–0.626) for training injuries and a relative risk (95% CI) of 0.83 (0.57–1.19) achieved with FIFA 11+ prevention exercises and sport-specific balance training, relatively.

Conclusion: In order to gain insight in the aetiology of basketball-specific injuries and consequently facilitate the development of preventive strategies, more high quality basketball-specific and injury-specific studies among recreational basketball players are needed.

KEYWORDS:

• Basketball
• athletic injuries
• sport medicine
• preventive medicine
• epidemiology

Introduction

Van Mechelen’s ‘sequence of prevention’ has been recognized as the most influential model of the past 25 years when it comes to sport injury prevention [1,2]. This model relies on four subsequent steps, from establishing the incidence and etiology of musculoskeletal injuries (steps 1 and 2) to the development and evaluation of preventive interventions (steps 3 and 4) [1]. These four steps of van Mechelen’s ‘sequence of prevention’ are the prerequisite to any successful implementation of interventions aiming to reduce or prevent the occurrence of musculoskeletal injuries in sports.

Eleven percent of the world population plays recreational (nonelite) basketball, which makes basketball one of the most popular sports worldwide [3]. While being undoubtedly beneficial to physical, mental, and social well-being, recreational basketball is like other sports also associated with a risk for musculoskeletal injuries. Basketball not only appears to have the highest frequency of injuries among noncontact sports, but is also referred as being more dangerous, with a higher injury risk, than contact sports [4,5]. These injuries are likely to lead to substantial direct and indirect healthcare costs, as well as to induce impairments in daily life, sport and/or work [6]. Although prevention studies have shown to be effective, basketball injuries remain common [7,8]. In order to gain a greater focus on injury prevention among recreational basketball players, a systematic overview presenting the scientific evidence related to all steps of van Mechelen’s ‘sequence of prevention’ for musculoskeletal injuries among recreational basketball players is needed but remains at present time lacking.

Consequently, the objective of this study was to gather epidemiological information in order to answer following four research questions: 1) What are the most common musculoskeletal injuries occurring among recreational basketball players? 2) What are the risk factors and mechanisms of these most common musculoskeletal injuries among recreational basketball players? 3) What are the available primary preventive interventions in order to prevent these most common musculoskeletal injuries among recreational basketball players? 4) What is the effectiveness of the available primary preventive interventions on the reduction or prevention of musculoskeletal injuries among recreational basketball players?

Methods

A systematic review of the scientific literature was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [9].

Search strategies and database

Highly sensitive search strategies (Appendix 1) were built based on three groups of keywords (and related search terms), namely ‘injury’, ‘basketball’, and ‘cohort study’. Two electronic databases were searched up to May 2017, namely Medline (biomedical literature) via Pubmed (from 1966), and SPORTDiscus (sports and sports medicine literature) via EBSCOhost (from 1985). The following filters were applied: Humans, English, Randomized Controlled Trial, and/or Academic Journal. Within each keyword, all search terms were combined by the Boolean command OR, and the keywords (and respective search terms) were linked by the Boolean command AND. In Medline, existing medical subject headings [MeSH] were used if possible, while search terms were truncated with *.

Eligibility criteria

To retrieve articles relevant to all research questions, criteria for inclusion were:

• (1) The population of interest consists exclusively of recreational (nonelite) basketball players.

• (2) The article presents an available original study written in English.

• (3a) If related to descriptive epidemiology, prospective cohort design is used.

• (4a) If related to descriptive epidemiology, incidence (relative to exposure) or prevalence rate (overuse injuries) are reported.

• (3b) If related to etiology, prospective cohort design is used.

• (4b) If related to etiology, a risk estimate is reported.

• (3c) If related to primary prevention, randomized controlled trial is conducted.

• (4c) If related to primary prevention, incidence rates and/or effect sizes are reported.

Study selection

All studies identified through our search strategies were separately imported in a citation database (EndNote). To identify potentially relevant articles, titles and abstracts were screened independently by three researchers (KE, vOV, GV) by using the relevant eligibility criteria. If the title and abstract did not provide sufficient information to determine whether the eligibility criteria were met, it was included for the full text selection. Then, full text articles were assessed independently for eligibility by three researchers (KE, vOV, GV). Any disagreements regarding the inclusion or exclusion of articles were resolved by consulting a fourth researcher (KO). To avoid missing any relevant publications, the references of included studies were screened.

Data extraction

Data from the included articles were extracted and cross-checked by three authors (KO, vOV, GV). Therefore, two standardized extraction forms were used (one for our first and second research question, and one for our third and fourth research question) in order to report: study information (author, year), study population (sample size, age, gender, level of sport), injury definition and registration, injury incidence (inclusive pathology), risk factors and mechanism (if applicable), preventive measure (if applicable), effect (if applicable).

Risk of bias appraisal

The risk of bias of all included articles was assessed and cross-checked by two authors (KO, GV). If there was a difference in scoring an item, a consensus was reached by the authors. Any disagreements regarding the methodological appraisal of articles were resolved by consulting a third researcher (BM). For the articles related to descriptive epidemiology and etiology (first and second research question), the Quality in Prognosis Studies (QUIPS) tool was used (Appendix 2), exploring the following six bias domains: study population, study attribution, prognostic factor information, measurement of and controlling of confounding variables, measurement of outcomes, and analysis approaches [10]. Each of the six potential bias domains was rated as having a high, moderate, or low risk of bias [9,11]. We considered a study to have an overall low risk of bias when the methodological risk of bias was rated as low or moderate in all six domains, with at least four domains being rated as low [9,11]. A study was rated as having an overall high risk of bias if two or more of the domains were scored as high [9,11]. In-between quality was scored as moderate [9,11]. For the articles related to prevention (third and fourth research question), the Cochrane Collaboration’s tool was used (Appendix 2), exploring the following six bias domains: sequence generation, allocation concealment, blinding of participant and personnel, blinding of outcome, incomplete data, and selective reporting [12]. Studies were classified as having a low risk of bias when all items were rated as low [10,11]. A high risk of bias was assigned when at least one item was rated as high. A moderate risk of bias was assigned when at least one item was classified as moderate [10,11]. The results of the risk of bias appraisal for all included articles can be found in Appendix 3.

Synthesis of evidence

Because van Mechelen’s ‘sequence of prevention’ remains essential when it comes to retrieve and present epidemiological information related to injury and injury prevention in sports, its four steps were applied in order to visually synthetize and present the gathered scientific information.

Results

Search strategies

A total of 785 potentially relevant citations were retrieved from the literature search in Medline (265 citations) and SPORTDiscus (520 citations). After applying the inclusion criteria to the titles and abstracts, only 93 potentially relevant studies were included for the full text review. From those potentially relevant studies, 55 studies were excluded for various reasons (not an original study: N = 12; inappropriate study design: N = 15; data not solely about injuries among recreational basketball players: N = 28); incidence rate not presented: N = 16. Since the reference check of the included studies resulted in a single additional relevant study, 38 relevant original studies were included in our systematic review: 34 studies describing the occurrence and etiology (first and second research question) of musculoskeletal injuries [13–46], and four studies describing the effectiveness of preventive interventions [11,47–49]. The PRISMA flow chart of the search procedure can be found in Appendix 4.

Musculoskeletal injuries in basketball: occurrence and etiology

The included studies showed that the incidence of musculoskeletal injuries among recreational basketball players ranged from 0.47 injuries per 100,000 AE (0.0047 injuries per 1000 AE) for dental injuries in male basketball players to 10.10 injuries per 1000 AE for overall injuries in match play during regular season [22,24]. Especially ankle and knee injuries are often reported as there are studies only focusing on these injuries [14–16,20,21,25,29,30,34,36,44,45]. There are two studies focusing only on shoulder injuries [17,42]. For detailed information on the incidence of injuries see Table 1 and Figure 1, the van Mechelen’s ‘sequence of prevention’ [2]. For the incidence in the figure we only presented studies describing their injury rates as injury per 1000 AE as this description of injury rates was used in the majority of the studies. With respect to Figure 1 we divided the injury rates per 10,000 and 100,000 AE by respectively 10 and 100 to obtain injury per 1000 AE in order to maintain consistency in the figure.

Table 1. Musculoskeletal injuries among recreational basketball players: occurrence and etiology.

Reference	Participation and design	Injury definition	Incidence	Risk factors
Agel [13]	N: 10.345–14.596 G: female A:? C:USA D: Cohort F: 16 years	Injury: defined as one that (1) occurred as a result of participation in an organized intercollegiate practice or competition and (2) required medical attention by a team certified athletic trainer or physician and (3) resulted in restriction of the student-athlete’s participation or performance for 1 or more calendar days beyond the day of injury.	-Practice situation: 3.99 injuries per 1000 AE (65.6% lower extremity; 10.4% upper extremity; 10.4% trunk/back; 8.9% head/neck) - Game situation: 7.68 injuries per 1000 AE (60.8% lower extremity; 14.7% head/neck; 14.1% upper extremity; 7.4% trunk/back) - Preseason: 6.75 injuries per 1000 AE - Regular season: 2.84 injuries per 1000 AE - Post season: 5.52 injuries per 1000 AE - Anterior cruciate ligament (ACL) injuries (games): 0.66 injuries per 1000 AE - Stress fractures: 0.10–0.34 injuries per 1000 AE - Ankle injuries (games): 1.89 injuries per 1000 AE - Concussion (games): 0.50 injuries per 1000 AE
Arendt [14]	N: ? G: male/female A: ? C: USA D: Cohort F: 5 years	Knee injury: defined as one that occurs as a result of participation in an organized intercollegiate practice or game, requires medical attention by a team athletic trainer or physician, and results in restriction of the student athlete’s participation for 1 or more days beyond the day of injury.	- Male 0.70 injuries per 1000 AE - Female: 1.0 injuries per 1000 AE - Collateral ligament: Male: 0.21 injuries per 1000 AE Female: 0.28 injuries per 1000 AE - Torn cartilage (meniscus): Male: 0.13 injuries per 1000 AE Female: 0.29 injuries per 1000 AE - Patella or patellar tendon: Male: 0.26 injuries per 1000 AE Female: 0.24 injuries per 1000 AE - ACL: Male: 0.07 injuries per 1000 AE Female: 0.29 injuries per 1000 AE - Posterior cruciate ligament: Male: 0.01 injuries per 1000 AE Female: 0.01 injuries per 1000 AE
Beynnon [15]	N: 901 G: 357 male; 544 female A: 13–22 C: USA D: cohort F: 4 years	Ankle injury: inversion trauma that forced the athlete to miss at least 1 practice or game (registered by athletic trainer).	- Male: 0.42 injuries per 1000 participation-days - Female: 1.90 injuries per 1000 participation-days
Beynnon [16]	N: 901 G: male; female A: 13–22 C: USA D: cohort F: 4 years	ACL injury: complete grade 3 disruption of the ligament in a person with no previous ACL injury to either leg, occurring as a result of participation in an organized practice, or game and not involving any direct contact to the knee from external forces such as those produced by equipment, other athletes, or the ground (registered by study coordinator).	- College: Male: 0.051 injuries per 1000 participation-days Female: 0.143 injuries per 1000 participation-days - High school: Male: 0.037 injuries per 1000 participation-days Female: 0.061 injuries per 1000 participation-days
Bonza [17]	N: ? G: male; female A: ? C: USA D: cohort F: 2 years	Shoulder injury: occurred as a result of participation in an organized high school practice or competition, and required medical attention by an athletic trainer or a physician, and resulted in restriction of the athlete’s participation for at least 1 day beyond the day of the injury (registered by athletic trainer).	- Male: Overall: 0.47 injuries per 10,000 AE Competition: 0.90 injuries per 10,000 AE Training: 0.30 injuries per 10,000 AE - Female: Overall: 0.45 injuries per 10,000 AE Competition: 0.76 injuries per 10,000 AE Training: 0.32 injuries per 10,000 AE
Borowski [18]	N: ? G: male; female A: ? C: USA D: Cohort F: 2 years	Injury: defined as one that (1) occurred as a result of participation in an organized high school basketball practice or competition, (2) required medical attention by an Athletic Trainer or a physician, and (3) resulted in restriction of the student-athlete’s participation for 1 or more days beyond the day of the injury.	- Male: 1.83 injuries per 1000 AE (43.2 ankle/foot; 12.8% head/face/neck; 10.6% knee) - Female: 2.08 injuries per 1000 AE (35.9 ankle/foot; 18.2% knee; 14.2% head/face/neck)
Brumitt [19]	N: 71 G: Male A: 20.2 C: USA D: Cohort F: Off season	Injury: any muscle, joint, or bone problem/injury (mechanism of injury: noncontact) of the low back or the lower extremity that occurred either during practice or competition that required the athlete to be removed from that day’s event or to miss a subsequent practice or competition.	- Total: 3.9 per 1000 AE (72% ankle/foot) - Initial time loss noncontact injuries: 2.7 per 1000 AE - Subsequent time loss injuries: 1.2 per 1000 AE
Curtis [20]	N: 230 G: 141 male; 89 female A: 20 C: USA D: cohort F: 1 year	Ankle injury: injury to the ankle ligaments (registered by athletic trainer).	Cushioned column shoes: 1.33 injuries per 1000 exposures Noncushioned column shoes: 1.96 injuries per 1000 exposure	Cushioned column shoes vs. noncushioned column shoes Relative risk (95% CI) = 1.47 (0.32–6.86)
De Vries [21]	N: 90 G: 42 male; 48 female A: 26–29 C: The Netherlands D: cohort F: 4 years	Patellar tendinopathy: having pain at the inferior pole of the patella and/or a physician or physical therapist had diagnosed the knee problems as patellar tendinopathy (self-reported).		- Gender (ref = female): odds ratio (95% CI) = 0.8(0.2–2.7) - Age: odds ratio (95% CI) = 1.0(0.8–1.2) - Height (5 cm increase): odds ratio (95% CI) = 1.0(0.8–1.4) - Weight (5 kg increase): odds ratio (95% CI) = 1.2(0.9–1.5) - BMI: odds ratio (95% CI) = 1.02 (1.0–1.5) - Years playing: odds ratio (95% CI) = 1.1(1.0–1.3) - Hours training per week (5 h increase): odds ratio(95% CI) = 0.1 (0.0–10.4) - Playing surface (ref = vinyl/rubber): Concrete: odds ratio (95% CI) = 6.3 (0.4–110.3) - Training increase compared to last year (ref = no): odds ratio(95% CI) = 0.8 (0.2–3.0) - Other sports (ref = no): odds ratio (95% CI) = 2.6(0.8–9.3) - Hours other sports: odds ratio (95% CI) = 1.8(0.8–4.0)
Dick [22]	N: ? G: male A: ? C: USA D: Cohort F: 16 years	Injury: one that (1) occurred as a result of participation in an organized intercollegiate practice or competition and (2) required medical attention by a team certified athletic trainer or physician and (3) resulted in restriction of the student-athlete’s participation or performance for 1 or more calendar days beyond the day of injury.	- Game: Total: 9.9 injuries per 1000 AE (26.2% ankle) In season: 10.1 injuries per 1000 AE Post season: 6.4 injuries per 1000 AE - Practice: Total: 4.3 injuries per 1000 AE (26.8% ankle) Preseason: 7.5 injuries per 1000 AE In season: 2.8 injuries per 1000 AE Post season: 1.5 injuries per 1000 AE
Fletcher [23]	N: ? G: male; female A: 13–19 C: USA D: Cohort F: 5 years	Injury: event that (1) occurred as a result of participation in an organized high school basketball practice or competition, (2) required medical attention by an athletic trainer or a physician, and (3) resulted in restriction of the student-athlete’s participation for 1 or more days after the date of injury.	- Total: 1.68 injuries per 1000 AE (32.6% ankle; 16.5% head/face; 15% knee) - Male: 1.57 injuries per 1000 AE - Female: 1.82 injuries per 1000 AE
Huffman [24]	N: ? G: male; female A: ? C: USA D: cohort F: 2 years	Injury: occurred during an organized high school practice or competition, required medical attention by an athletic trainer or a physician, and resulted in restriction of the athlete’s participation in practice or competition for 1 or more days (registered by athletic trainer).	- Male: Eye: 1.65 injuries per 100,000 AE Dental: 0.47 injuries per 100,000 AE Neck and cervical: 0.71 injuries per 100,000 AE - Female: Eye: 1.40 injuries per 100,000 AE Dental: 1.12 injuries per 100,000 AE Neck and cervical: 1.40 injuries per 100,000 AE
Junge [25]	N: ? G: male; female A: 10–12 C: Denmark D: cohort F: 3 years	Traumatic knee injury: resulting from a specific, identifiable event (registered by parents and clinician). Overuse knee injury: caused by repeated trauma without a single, recognizable event responsible for that injury (registered by parents and clinician).	- Traumatic injury: Period prevalence: 0.97 injuries per 1000 sport-participation units - Overuse injury: Period prevalence: 6.31 injuries per 1000 sport-participation units
Knowles [26]	N: 1,745 G: 1,031male; 714female A: ? C: USA D: cohort F: 4 years	Injury: resulted from participation in a high school sport and either limited the athlete’s full participation in the sport the day following the injury or required medical attention by a medical professional (registered by trained school personnel)	- Male: 2.32 injuries per 1000 AE - Female: 1.28 injuries per 1000 AE
Kuzuhara [27]	N: 95 G: 56 male; 39 female A: 11 C: Japan D: cohort F: 1 year	Injury: any event, including trauma, overuse, or internal disease, related to sports, that occurred during a regular practice or game, and led to a player missing any practice or game subsequent to the injury, and caused the player to seek medical care from a physician or alternative medical specialist (registered by physician or alternative medical specialist).	- Overall: Total: 3.83 injuries per 1000 h of exposure Games: 12.92 injuries per 1000 h of exposure Practice: 3.13 injuries per 1000 h of exposure - Head and neck: Games: 4.70 injuries per 1000 h of exposure Practice: 0.50 injuries per 1000 h of exposure - Upper limb: Games: 3.52 injuries per 1000 h of exposure Practice: 1.50 injuries per 1000 h of exposure - Trunk and back: Practice: 0.05 injuries per 1000 h of exposure - Lower limb: Games: 2.94 injuries per 1000 h of exposure Practice: 0.77 injuries per 1000 h of exposure
Lanese [28]	N: 382 G: 232 male; 150 female A: ? C: USA D: cohort F: 1 year	Injury: traumatic medical problem due to sports participation and resulting in loss of time from practice or competition (registered by an athletic trainer or student trainer)	- Male: 0.45 injuries per 100 person-hours - Female: 0.48 injuries per 100 person-hours
Leppanen [29]	N: ? G: female A: 12–21 C: Finland D: Cohort F: 2 years	ACL injury: new noncontact magnetic resonance imaging (MRI) confirmed ACL rupture that occurred during a match or scheduled team training.	- Matches 3.4 injuries per1000 h of exposure	- High peak knee flexion angle (per each 10° increase) during jumping: Hazard ratio (HR) (95% CI) = 0.55 (0.34–0.88) - High vertical ground reaction force (per each 100-N increase) during jumping: HR (95% CI) = 1.26 (1.09–1.45)
McGuine [30]	N: 210 G: 119 male; 91 female A: 16 C: USA D: cohort F: 2 years	Ankle sprain injury: trauma that disrupts the structures of the ankle that occurs during a team-sponsored practice or competition session, and causes the athlete to miss the rest of practice or competition or miss the next scheduled team practice or competition (registered by an athletic trainer)	- Total: 1.56 injuries per 1000 exposure - Male: 1.68 injuries per 1000 exposure - Female: 1.44 injuries per 1000 exposure	- Gender: male vs. female Odds ratio = 1.2 (p = 0.606) - Leg dominance: dominant vs. nondominant Odds ratio = 0.5 (p = 0.403) - Body mass index: odds ratio = 1.0 (p = 0.788) - Postural sway: higher vs. lower Odds ratio = 10.2 (p = 0.001)
McKay [31]	N: 10,393 G: 3,421 male; 6,972 female A: ? C: Australia D: cohort F: 2 years	Injury: an action in which the player perceived bodily harm had been sustained, necessitating stoppage of play, substitution or a display of obvious disability (self-reported)	- Total: 18.3 injuries per 1000 participations 24.7 injuries per 1000 h - Male: Elite: 26.9 injuries per 1000 participations Elite: 26.9 injuries per 1000 h Recreational: 14.7 injuries per 1000 participations Recreational: 22.0 injuries per 1000 h - Female: Elite: 23.0 injuries per 1000 participations Elite: 23.0 injuries per 1000 h Recreational: 17.2 injuries per 1000 participations Recreational: 25.7 injuries per 1000 h	- Age: 25–34 yrs: odds ratio (95% CI) = 2.40 (0.88–6.52) 35+ yrs: odds ratio (95% CI) = 1.55 (0.46–5.24) - Height: 170–180 cm: odds ratio (95% CI) = 1.33 (0.50–6.52) 181+ cm: odds ratio (95% CI) = 2.55 (0.90–7.22) - Weight: 65–74 kg: odds ratio (95% CI) = 2.65 (0.94–7.50) 75+ kg: odds ratio (95% CI) = 3.39 (1.29–8.93) - Gender: Female: OR = 0.45 (0.20–1.00) - Training per week: 1 or 2: odds ratio (95% CI) = 1.46 (0.61–3.52) 3 or more: odds ratio (95% CI) = 0.60 (0.16–2.23) - Matches per week: 2 or more: odds ratio (95% CI) = 2.16 (0.97–4.81) - Mechanism: Body contact: odds ratio (95% CI) = 0.61 (0.26–1.40)
Meeuwisse [32]	N: 318 G: male; female A: ? C: Canada D: cohort F: 2 years	Injury: any injury resulting in one or more complete or partial sessions of time loss (registered by team and student therapist)	- Total: Time loss < 7 sessions: 4.09 injuries per 1000 AE Time loss ≥ 7 sessions: 0.85 injuries per 1000 AE - Ankle: Time loss < 7 sessions: 1.01 injuries per 1000 AE Time loss ≥ 7 sessions: 0.21 injuries per 1000 AE - Knee: Time loss < 7 sessions: 0.62 injuries per 1000 AE Time loss ≥ 7 sessions: 0.25 injuries per 1000 AE - Thigh: Time loss < 7 sessions: 0.44 injuries per 1000 AE Time loss ≥ 7 sessions: 0.02 injuries per 1000 AE - Foot: Time loss < 7 sessions: 0.25 injuries per 1000 AE Time loss ≥ 7 sessions: 0.14 injuries per 1000 AE - Hand: Time loss < 7 sessions: 0.18 injuries per 1000 AE Time loss ≥ 7 sessions: 0.07 injuries per 1000 AE - Leg: Time loss < 7 sessions: 0.21 injuries per 1000 AE Time loss ≥ 7 sessions: 0.02 injuries per 1000 AE - Shoulder: Time loss < 7 sessions: 0.16 injuries per 1000 AE Time loss ≥ 7 sessions: 0.07 injuries per 1000 AE
Messina [33]	N: 1,863 G: 973 male; 890 female A: 14–18 C: USA D: cohort F: 1 year	Injury: one that occurred in an organized practice or game, resulted in loss of time from practice or competition, necessitated the consultation of a physician, or involved the head or face (registered by athletic trainer)	- Male: 0.56 injuries per athlete per season 3.2 injuries per 1000 player-hours (32% ankle; 10% knee; 10% hip/thigh) - Female: 0.49 injuries per athlete per season 3.6 injuries per 1000 player-hours (31% ankle; 20% knee)
Mihata [34]	N: G: male; female A: C: USA D: cohort F: 15 years	ACL injury: one that occurs as a result of participation in an organized intercollegiate game or practice, requires medical attention by a team athletic trainer or physician, and results in restriction of the athlete’s participation for one or more days beyond the day of injury (registered by certified and student athletic trainer)	- Male: 0.08 injuries per 1000 AE - Female: 0.28 injuries per 1000 AE
Monfort [35]	N: ? G: male; female A: ? C: USA D: cohort F: 7 years	Injury: noncontact event occurring during an organized practice or competition, requiring medical attention by an Athletic Trainer or a physician, and resulting in student-athletes not participating in practice or competition for at least 1 day.	- Male: Total: 0.61 injuries per 10,000 AE Knee: 0.18 injuries per 10,000 AE Ankle: 0.40 injuries per 10,000 AE - Female: Total: 1.07 injuries per 10,000 AE Knee: 0.41 injuries per 10,000 AE Ankle: 0.52 injuries per 10,000 AE	- Defending vs. ball handling: Total: relative risk (95% CI) = 1.36 (1.08–1.73) Ankle: relative risk (95% CI) = 1.45 (1.01–2.50)
Mountcastle [36]	N: ? G: male; female A: ? C: USA D: cohort F: 10 years	ACL injury: acute complete tear with laxity and subsequent confirmation of an ACL deficiency by arthroscopy of MRI evaluation (registered by orthopedic surgeon)	- Male: 0.10 injuries per 1000 exposures - Female: 0.29 injuries per 1000 exposures
Owoeye [37]	N: 141 G: 75 male; 66 female A: 15–18 C: Nigerian D: Cohort F: one season	Injury: when an injured player required at least minimum on-field (medical) care such as ice, tape, etc. regardless of whether the player was able to continue or not (that is both medical attention and time-loss injuries were recorded).	- Total: 22.7 injuries per 100 participants (40.6% knee; 21.9% ankle; 9.4% elbow) - Male: 1.1 injuries per 100 participants - Female: 0.9 injuries per 100 participants
Pappas [38]	N: ? G: male; female A: 7–17 C: USA D: cohort F: 6 years	?	- Male 7–11 years: Ankle sprains: 3.4 injuries per 10,000 athlete days Finger sprains: 5.2 injuries per 10,000 athlete days Finger fractures: 4.2 injuries per 10,000 athlete days Knee sprains: 0.9 injuries per 10,000 athlete days Facial lacerations: 1.5 injuries per 10,000 athlete days - Girls 7–11 years: Ankle sprains: 3.8 injuries per 10,000 athlete days Finger sprains: 5.4 injuries per 10,000 athlete days Finger fractures: 4.2 injuries per 10,000 athlete days Knee sprains: 0.7 injuries per 10,000 athlete days Facial lacerations: 0.4 injuries per 10,000 athlete days - Male 12–17 years: Ankle sprains: 26.5 injuries per 10,000 athlete days Finger sprains: 6.5 injuries per 10,000 athlete days Finger fractures: 7.3 injuries per 10,000 athlete days Knee sprains: 3.9 injuries per 10,000 athlete days Facial lacerations: 5.5 injuries per 10,000 athlete days - Girls 12–17 years: Ankle sprains: 23.2 injuries per 10,000 athlete days Finger sprains: 7.9 injuries per 10,000 athlete days Finger fractures: 7.3 injuries per 10,000 athlete days Knee sprains: 5.6 injuries per 10,000 athlete days Facial lacerations: 1.6 injuries per 10,000 athlete days
Parkkari [39]	N: 59 G: male; female A: 14–74 C: Finland D: cohort F: 1 year	Injury: new acute trauma or overuse injury that caused a significant complaint to the subject (registered by trained personnel)	9.1 injuries per 1000 participation hours
Powell [40]	N: 12,914 G: 6,831male; 6,083 female A: ? C: USA D: cohort F: 3 years	Injury: injury that caused cessation of participation in the current game or practice and prevents the player’s return to that session or to the session after the day of onset (registered by certified athletic trainer)	- Male: 28.3 injuries per 100 players (39% ankle; 14% hip/thigh/leg; 11% forearm/wrist/hand; 11% knee; 10% face/scalp) - Female: 28.7 injuries per 100 players (37% ankle; 16% hip/thigh/leg; 10% forearm/wrist/hand; 16% knee)
Rechel [41]	N: ? G: male; female A: ? C: USA D: cohort F: 1 year	Injury: occurred as a result of participation in an organized high school practice or competition, and required medical attention by an athletic trainer or physician, and resulted in restriction of the athlete’s participation for 1 day or more beyond the day on injury (registered by athletic trainer)	- Male: Total: 1.89 injuries per 1000 AE Practice: 1.46 injuries per 1000 AE Competition: 2.98 injuries per 1000 AE - Female: Total: 2.01 injuries per 1000 AE Practice: 1.37 injuries per 1000 AE Competition: 3.60 injuries per 1000 AE
Robinson [42]	N: ? G: male; female A: ? C: USA D: cohort F: 7 years	Shoulder injury: any injury to the proximal humerus, scapula, clavicle, acromio-clavicular joint, and surrounding tendons, ligaments, and musculature, that occurred as a result of participation in an organized practice or competition, required medical attention by an athletic trainer or physician, and resulted in restriction of the athlete’s participation for 1 or more days (registered by athletic trainer)	- Male: Total: 0.50 injuries per 10,000 AE Practice: 0.32 injuries per 10,000 AE Competition: 0.95 injuries per 10,000 AE - Female: Total: 0.61 injuries per 10,000 AE Practice: 0.34 injuries per 10,000 AE Competition: 1.24 injuries per 10,000 AE
Stevenson [43]	N: 193 G: male; 54% female; 46% A: 23 C: Australia D: cohort F: 5 months	Injury: occurs while participating in sport and leads to one of the following consequences: a reduction in the amount or level of sports activity, the need for advice or treatment, and/or adverse economic or social effects (registered by supervised staff)	15.1 injuries per 1000 h of sports participation
Wang [44]	N: 42 G: ? A: 17 C: China D: cohort F: 1 year	Ankle injury: an incident or period occurring in connection with basketball during training or competition that occurs at the ankle, handicaps the player during performance, or completely prevents the player from playing basketball or makes the player look for treatment to continue playing basketball (registered by physiotherapist)		- Postural sway: Mediolateral: odds ratio = 1.220 (p = 0.1) Anteroposterior: odds ratio = 1.216 (p < .001)
Waterman [45]	N: G: male; female A: C: USA D: cohort F: 4 years	Ankle injury (syndesmotic or medial): primary ankle sprains resulting in any time lost to sport and activity that occurred within the study period (registered at sports medicine or orthopedic clinic)	- Male: Syndesmotic: 24.75 injuries per 100,000 AE Medial: 6.19 injuries per 100,000 AE - Female: Syndesmotic: 6.74 injuries per 100,000 AE
Yde [46]	N: 56 G: 27 male; 29 female A: <18 C: Denmark D: cohort F: 1 year	Injury: an incident occurring during a match or training in the club, causing the player to miss at least one match or one training session (registered by players).	Total: 3.0 injuries per 1000 playing hours (43% hand/fingers; 33% ankle) Games: 5.7 injuries per 1000 playing hours Practice: 2.4 injuries per 1000 playing hours

N: number of participants; G: gender; A: age; C: country where study was conducted; D: design; F: follow-up period; h: hours; ?: unknown; AE: athlete exposures

Figure 1. Van Mechelen ‘Sequence of Prevention’.

Several risk factors have been found to have a significant association with injuries in basketball. For instance, noncontact injuries occur more often when defending instead of ball handling with a relative risk (95% confidence interval (CI)) of 1.36 (1.08 – 1.73) for all injuries and a relative risk of 1.45 (1.01–2.50) for ankle injuries [35]. Another risk factor, mentioned by multiple studies is postural sway (horizontal movement of the center of gravity while standing) [30,44]. McGuine et al. showed that the odds of sustaining an ankle injury is 10.2 (p = 0.001) for players with a high postural sway. Additionally Wang et al. showed that for ankle injuries anteroposterior postural sway is a risk factor with an OR of 1.216 (p < 0.001) [30,44]. Besides aforementioned risk factors, one study also showed that high vertical ground reaction force during jumping (per each 100-M increase) is associated with injuries with a HR (95% CI) of 1.26 (1–09–1.45) [29]. The same study also presented that a high peak knee flexion angle during jumping (per each 10° increase) decreases the risk on an injury with a HR (95% CI) of 0.55 (0.34–0.88) [29]. Also a weight of > 75 kg increases the odds of sustaining an injury with an OR (95% CI) of 3.39 (1.29–8.93) [31]. For detailed information on the etiology and risk factors of injuries see Table 1 and Figure 1.

Preventive interventions and related effectiveness

Four studies regarding injury preventive interventions were included [11,47–49]. It is worth mentioning that all four studies showed significant effects in the prevention of injuries in basketball. The preventive measures and their effects sizes are presented in Table 2 and Figure 1.

Table 2. Musculoskeletal injuries among recreational basketball players: preventive intervention and related effectiveness.

Reference	Participation and design	Injury definition	Preventive intervention	Outcome
Aerts [47]	N: 183 G: male, female A: 24–27 C: Belgium D: RCT F: 1 season	Acute injury: sport injury with sudden, direct cause/onset with at least minimum medical care (e.g. ice, tape) and missing 1 or more training or game session. Overuse injury: physical discomfort with insidious onset with pain and/or stiffness of musculoskeletal system and present during and/or after sport activity.	Jump-Landing-Technique Intervention Program imbedded with the warming-up including exercises that need to be performed twice a week for 3 months (5–10 min each time).	- Acute injury: Intervention: 3.0 injuries per 1000 h exposure Control: 4.0 injuries per 1000 h exposure HR (95% CI): 0.37 (0.12–1.1) - Overuse injury: Intervention: 1.0 injuries per 1000 h exposure Control: 1.4 injuries per 1000 h exposure HR (95% CI): 0.47 (0.09–2.56) - Lower extremity injury: Intervention: 3.6 injuries per 1000 h exposure Control: 5.4 injuries per 1000 h exposure HR (95% CI): 0.40 (0.16–0.99)
Eils [48]	N: 232 G: male, female A: 23–26 C: Germany D: RCT F: 1 season	Injury: event that forced the subject to terminate the ongoing basketball activity and, in addition, prevented further participation in the next scheduled basketball activity (registered by coach, physiotherapist or player)	Multi-station proprioceptive exercise program integrated within the regular training routines consisting in 6 stations of exercises that need to be performed once a week (20 min each time) during the season (incremental intensity).	Intervention: 1.53 injuries per 1000 sports participation Control: 4.31 injuries per 1000 sports participation Odds ratio (95% CI): 0.355 (0.151–0.835)
Emery [49]	N: 926 G: male, female A: 12–18 C: Canada D: cRCT F: 1 season (18 weeks)	Injury: occurring during basketball resulting in medical attention beyond the team therapist assessment and/or the inability to complete the current session and/or participate in a subsequent basketball session (registered by team therapist or team manager)	Sport-specific balance training (5 min) within the warm-up routine (aerobic, static stretch and dynamic stretch) that was performed 5 times a week, and home exercise program (20 min) with wobble board.	- All injury: Intervention: 3.3 injuries per 1000 players hours Control: 4.03 injuries per 1000 players hours Relative risk (95% CI): 0.8 (0.57–1.11) - All acute injury: Intervention: 2.77 injuries per 1000 players hours Control: 43.83 injuries per 1000 players hours Relative risk (95% CI): 0.71 (0.5–0.99) - Lower extremity injury: Intervention: 2.69 injuries per 1000 players hours Control: 3.18 injuries per 1000 players hours Relative risk (95% CI): 0.83 (0.57–1.19) - Ankle injury: Intervention: 1.57 injuries per 1000 players hours Control: 2.46 injuries per 1000 players hours Relative risk (95% CI): 0.71 (0.45–1.13)
Longo [11]	N: 121 G: male A: 14–15 C: Italy D: cRCT F: 1 season (9 months)	Injury: traumatic event associated with basketball and resulting in absence from at least one scheduled practice session or match (registered by coach).	11+ injury prevention program (FIFA11+) including 15 different exercises with the focus on neuromuscular effect to the lower extremities and to the core (running, stretching, strength, plyometrics, balance).	- All injuries: Intervention: 0.95 injuries per 1000 AE Control: 2.16 injuries per 1000 AE Odds ratio (95% CI): 0.316 (0.165–0.603) - Match injuries: Intervention: 0.81 injuries per 1000 AE Control: 1.4 injuries per 1000 AE Odds ratio (95% CI): 0.477 (0.236–0.965) - Training injuries: Intervention: 0.14 injuries per 1000 AE Control: 0.76 injuries per 1000 AE Odds ratio (95% CI): 0.175 (0.049–0.626) - Lower extremity injuries: Intervention: 0.68 injuries per 1000 AE Control: 1.4 injuries per 1000 AE Odds ratio (95% CI): 0.404 (0.194–0.839)

N: number of participants; G: gender; A: age; C: country where study was conducted; D: design; F: follow-up period; RCT: randomized controlled trial; AE: athlete exposures; HR: hazard ratio; min.: minutes.

Aerts (2013) researched a Jump-Landing-Technique intervention program imbedded in the warming-up, including exercises that need to be performed twice a week for a period of 3 months [47]. Even though the effects of this intervention program did not achieve a statistically significant effect for acute and overuse injuries (HR (95% CI) of 0.37 (0.12–1.1) and 0.47 (0.09–2.56) respectively), it succeeded in the prevention of lower extremity injuries [47]. The onset of injuries in the control and intervention group was 5.4 and 3.6 injuries per 1000 h exposure respectively, with a HR (95%) of 0.40 (0.16–0.99) [47].

Eils (2010) applied a successful multi-station proprioceptive exercise program in their prevention study. This was integrated within the regular training routines consisting in six stations of exercises that needed to be performed once a week during the season [48]. The intervention program showed a significant effect with an OR (95% CI) of 0.34 (0.15–0.84) as the control group had 4.31 injuries per 1000 sports participation and the intervention group 1.53 injuries per 1000 sports participation [48].

Another injury preventive measure was evaluated by Emery (2007) [49]. The intervention consisted of a home exercise program with a wobble board and sport-specific balance training within the warming-up routine (aerobic, static stretch and dynamic stretch) that was performed five times a week [49]. Their intervention program managed to achieve a significant decrease for all acute injuries along a relative risk (95% CI) of 0.71 (0.5–0.99) with control group sustaining 43.83 injuries per 1000 players hours and the intervention group sustaining 2.77 injuries per 1000 players hours [49].

Longo (2012) studied the FIFA11+ intervention program among basketball players, including 15 different exercises with the focus on neuromuscular effect to the lower extremities and to the core (running, stretching, strength, plyometrics, balance) [11]. The FIFA11+ intervention program showed a significant decrease in the odds for all injuries, match injuries, training injuries and lower extremity injuries with ORs (95% CI) of respectively 0.32 (0.17–0.60), 0.48 (0.24–0.97), 0.18 (0.05–0.63), and 0.40 (0.19–0.84) [11].

Discussion

The aim of this study was to present a systematic overview of the incidence, etiology as well as an insight into the effect of preventive measures on basketball-specific injuries among recreational basketball players. A total of 34 studies were found on the incidence and etiology of basketball injuries among recreational basketball players. Results of our review showed that especially ankle, knee and shoulder injuries are common among recreational basketball players. Injuries in the head/neck/face are also often reported. Of the aforementioned 34 studies only eight studies described the risk factors among recreational basketball players. Four articles were retrieved from the literature about preventive measures in recreational basketball. The incidence of musculoskeletal injuries among recreational basketball players ranged from 0.0047 injuries per 1000 AE for dental injuries to 10.1 injuries per 1000 AE for overall injuries during match play. Significant risk factors for injuries were defending, postural sway, high vertical ground reaction force during jumping and weight >75 kg. All prevention studies have shown to have a significant effect on reducing the risk of injury ranging from an odds ratio of 0.175 (95% CI: 0.049–0.626) for training injuries and a relative risk of 0.83 (95% CI: 0.57–1.19) achieved with FIFA 11+ prevention exercises and sport-specific balance training, relatively.

The four steps of van Mechelen’s ‘sequence of prevention’ model remain widely used when it comes to descriptive epidemiology, etiology and prevention of sports injuries [2,50]. Our results suggest that ankle and knee injuries were the most common injuries among recreational basketball players. Therefore, one might expect that the etiology of these common injuries might have been thoroughly explored. However, our results also emphasizes the lack of high-quality basketball-specific studies on the etiology of these basketball-specific injuries as well as the lack of basketball and injury-specific prevention programs concerning recreational basketball players. Two biomechanical risk factors during jumping (high vertical ground reaction force; high peak knee flexion angle) were found to lead to an increased risk for ACL injuries. Consequently, this suggests that technique during jumping/landing should be taken into consideration when it comes to AL+CL injury prevention [51].

For instance, the shoulder Injury Prevention Programme, developed by the Oslo Sports Trauma Research Center, is an example of a sport- and injury-specific prevention program [52]. The aim of the prevention program was to increase glenohumeral internal rotation, external rotation strength and scapular muscle strength, as well as improve kinetic chain and thoracic mobility [53]. This prevention program showed to be very successful in the prevention of shoulder injuries among elite handball players. Even though the competition level of our population is recreational and not elite, such an approach might also be explored among basketball players in order to prevent shoulder injuries as those are often reported. However, one must keep in mind that mechanisms of shoulder injuries among handball players might differ from basketball players and different preventive exercises might be needed. Nevertheless, it is a successful example of an injury and sport-specific prevention program.

The results of our review present that the few available basketball-specific intervention programs among recreational players have shown to be successful for prevention of especially the lower extremity injuries. Consequently, one might hypothesize that research on injury- and basketball-specific preventive strategies and the successful implementation of these strategies will lead to a significant decrease of injury occurrence.

Methodological aspect

All included studies scored a moderate or low risk of bias. Most of the studies were scored moderate. We choose to use studies scored with a low risk of bias as well as moderate scored studies for the results of this review as a consequence of a lack of high-quality studies.

Moreover, it was hard to represent the findings as one result, since the outcome measures were described in various ways. For instance, Beynnon et al. (2005) presented their injury rate in number of injuries per 1000 participation-days, Agel et al. (2007) in number of injuries per 1000 AE, and Lanese et al. (1990) in number of injuries per 100 person-hours. This inconsistency in reporting epidemiological data makes it difficult to compare studies with each other, and does not allow to perform meta-analyses. This aspect should be taken into consideration in future studies.

We included only articles written in English. Although relevant studies might be available in other languages, we believe that this potential limitation did not affect significantly our findings. We did not include cross-sectional or retrospective designed studies, which might be a potential limitation. We only included studies with a prospective design in order to formulate valid answers to our research questions while maintaining the highest scientific quality. We are confident that our review presents an accurate overview of the available scientific literature related to the incidence, etiology and prevention of basketball-specific injuries among recreation basketball players with regards to a highly sensitive search strategy as well as the screening of references of included studies and/or retrieved reviews.

Implications for practice and further research

The results of our review presents that there is lacking scientific literature on especially the etiology of basketball-specific injuries among recreational basketball players. Studies focusing only on ankle, knee and shoulder injuries are available, whereas studies on injuries of other specific regions of the body, such as the head/neck/face region or the fingers, lack. One might logically assume that these kind of injury-specific studies give a major insight on the incidence as well as the etiology of that specific injury. This might enable the pathway to developing successful basketball- and injury-specific preventive strategies. Next to the development of preventive strategies, research on implementation of these strategies is just as important to achieve a significant reduction in the occurrence of injuries among recreation basketball players. Furthermore, future injury surveillance studies should use one consistent way of describing injury rates in order to compare the outcome measures of different studies with each other. Contrary to football, a consensus statement on injury definitions and data collection procedures in studies of basketball is lacking [54]. Such a consensus statement would be a valuable tool and should be developed in order to maintain consistency in future studies of basketball.

Conclusion

The scientific literature giving an insight on the etiology as well as the prevention of basketball-specific injuries among recreational basketball players is lacking. Thus high quality and not only basketball-specific but also injury-specific studies among recreational basketball players are needed. This will not only give an insight in the etiology of basketball-specific injuries but also will facilitate the development of preventive strategies.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. Peer reviewers on this manuscript have no relevant financial relationships to disclose.

Additional information

Funding

None.

Appendix 1: Search strategies

Medline via Pubmed

#1 = Wounds and Injuries[Mesh] OR injur*[tiab]

#2 = Basketball[Mesh] OR basketball[tiab]

#3 = Cohort studies[mesh:noexp] OR longitudinal studies[mesh:noexp] OR follow-up studies[mesh:noexp] OR prospective studies[mesh:noexp] OR cohort[TIAB] OR longitudinal[TIAB] OR prospective[TIAB] NOT retrospective[TIAB] NOT retrospective studies[mesh:noexp]

#4 (first and second research question) = #1 AND #2 AND #3 (Filters: Humans; English)

#5 (third and fourth research question) = #1 AND #2 (Filters: Randomized Controlled Trial; Humans; English)

SPORTDiscus via EBSCOhost.

#1 = AB (injur* OR caus* OR epidemiol* OR etiolog* OR etiology* OR mechanism* OR preval* OR inciden* OR occur* OR propor* OR distribut* OR populat* OR risk factor* OR predispose* OR prevent* OR intervent*)

#2 = TI basketball

#3 (all four research questions) = #1 AND #2 (Filters: Academic Journal; English)

Appendix 2: Risk of bias appraisal

Quality in Prognosis Studies (QUIPS)

1. Study participation

• Description of the source population or population of interest

• Description of the baseline study sample

• Adequate description of the study sample recruitment (place, period, sampling strategy)

2. Study attrition

• Adequate response rate for study participants

• Description of attempts to collect information on participants who dropped out

• Reasons for loss to follow-up are provided

3. Prognostic factor (PF) measurement

• A clear definition or description of the PF is provided

• Method of PF measurement is adequately reliable and valid

4. Outcome measurement

• A clear definition of the outcome is provided

• Method of outcome measurement used is adequately reliable and valid

5. Study confounding

• Clear definition/description of the important confounders measured are provided

• Measurement of all important confounders is adequately reliable and valid

• Important potential confounders are accounted for in the analysis

6. Analysis and reporting

• Sufficient presentation of data to assess the adequacy of the analytic strategy

• The statistical analyses are adequately reported (without selective reporting)

Cochrane Collaboration’s tool

1. Sequence generation

• Describe the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

2. Allocation concealment

• Describe the method used to conceal the allocation sequence in sufficient detail to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment.

3. Blinding of participants, personnel and outcome assessors

• Describe all measures used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. Provide any information relating to whether the intended blinding was effective.

4. Incomplete outcome data

• Describe the completeness of outcome data for each main outcome, including attrition and exclusions from the analysis. State whether attrition and exclusions were reported, the numbers in each intervention group (compared with total randomized participants), reasons for attrition/exclusions where reported, and any re-inclusions in analyses performed by the review authors.

5. Selective outcome reporting

• State how the possibility of selective outcome reporting was examined by the review authors, and what was found.

6. Other sources of bias

• State any important concerns about bias not addressed in the other domains in the tool. If particular questions/entries were prespecified in the review’s protocol, responses should be provided for each question/entry.

Appendix 4.

Flowchart of the search procedure according PRISMA

Appendix 3 Results of the risk of bias appraisal.

Study	Participation	Attribution	Prognostic	Outcome	Confounding	Analysis	Total risk of bias
Descriptive epidemiology and etiology of basketball injuries
Agel 2007 [13]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Arendt 1995	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Beynnon 2005 [15]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Beynnon 2014 [16]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Bonza 2009 [17]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Borowski 2008 [18]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Brumitt 2016 [19]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Curtis 2008 [9]	Low	Moderate	Low	Low	High	Low	Moderate
De Vries 2015 [21]	Low	Low	Low	Low	Low	Low	Low
Dick 2007 [22]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Fletcher 2014 [23]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Huffman 2008 [24]	Moderate	Low	N/A	Low	N/A	Low	Low
Junge 2016 [25]	Low	Moderate	N/A	Low	N/A	Low	Low
Knowles 2006 [26]	Moderate	Moderate	Low	Low	Moderate	Low	Moderate
Kuzuhara 2016 [27]	Low	High	Low	Low	Moderate	Low	Moderate
Lanese 1990 [28]	Moderate	High	N/A	Low	N/A	Low	Moderate
Leppänen 2017 [29]	Low	Moderate	Low	Low	Low	Low	Low
McGuine 2000 [30]	Low	Moderate	Low	Low	Moderate	Low	Low
McKay 2001 [31]	Low	Moderate	N/A	Low	N/A	Low	Low
Meeuwisse 2003 [32]	Moderate	Low	Moderate	Low	Moderate	Low	Moderate
Messina 1999 [33]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Mihata 2006 [34]	High	Moderate	N/A	Low	N/A	Low	Moderate
Monfort 2015 [35]	Moderate	Moderate	Moderate	Low	Moderate	Low	Moderate
Mountcastle 2007 [36]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Owoeye 2012 [37]	Moderate	Moderate	N/A	Moderate	N/A	Low	Moderate
Pappas 2011 [38]	Moderate	Moderate	N/A	High	N/A	Low	Moderate
Parkkari 2004 [39]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Powell 2000 [40]	High	High	N/A	Low	N/A	Low	Moderate
Rechel 2008 [41]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Robinson 2014 [42]	Moderate	High	N/A	Low	N/A	Low	Moderate
Longo 2012 [11]	Low	Low	NA	Low	Low	Low	Low
Stevenson 2000 [43]	Low	Low	N/A	Low	N/A	Low	Low
Wang 2006 [44]	Low	Low	Low	Low	Moderate	Low	Low
Waterman 2011 [45]	Moderate	Moderate	N/A	Low	N/A	Low	Moderate
Yde 1990 [46]	Low	Moderate	N/A	Low	N/A	Low	Low
Study	Sequence	Allocation	Blinding	Incomplete	Selective	Other	Total risk of bias
Prevention of basketball injuries
Aerts 2013 [47]	Moderate	Low	NA	Low	Low	Low	Moderate
Eils 2010 [48]	Low	Low	NA	Low	Low	Low	Low
Emery 2007 [49]	Low	Low	NA	Low	Low	Low	Low