Search and Selection of Studies
The database search identified 957 articles, with five studies identified through other sources. Following the removal of duplicates and screening for eligibility, 150 studies remained for inclusion in the review. The flow of articles through identification to final inclusion is shown in Fig. 1.
General Characteristics of the Studies
Sports Science and Medicine Topics
The 150 studies included in the review covered eight sport science and medicine topics: biomechanics (n = 15, 10%), fatigue and recovery (n = 9, 6%), injury (n = 45, 30%), match characteristics (n = 24, 16%), nutrition (n = 3, 2%), physical qualities (n = 37, 25%), psychology (n = 13, 9%), and training load (n = 4, 3%) (Fig. 2).
Table 1 demonstrates the rapid growth in studies over recent years, with 65% of studies published between 2011 and 2020. Studies investigating injury range from one study before 1990 (2%) , to 22 studies between 2011 and 2020 (49%). All other topics had over 60% of studies published after 2010. The growth in studies coincides with the increased professionalism of netball and thus the potential increase in funding for research. Additionally, improvements in technology (e.g., inertial movement units) have enhanced the ability to quantify (e.g., external workload) team sports [43, 44], which would explain the increase in match-characteristics research.
Geography of studies
Studies were identified from seven different countries: Australia, Jamaica, Malaysia, New Zealand, Singapore, South Africa and the UK (Fig. 3). The majority of studies were from Australia (n = 81), followed by the UK (n = 29) and New Zealand (n = 16) (Fig. 3), which have high participation rates and semi-/professional leagues. Within the topic of ‘match-characteristics’, 20 out of the 24 studies (83%) were from Australia, which is unsurprising given Australia has the most professional (in terms of funding and spectators) netball competition (Suncorp Super Netball). Nutrition and psychology were the only topics in which studies from Australia did not dominate with 67% (n = 2) and 54% (n = 7) from the UK, but with both still being under-investigated relative to other topics (Fig. 2). The countries of very low representation (Malaysia [n = 4], Singapore [n = 3], Jamaica [n = 2]) investigate injury [32, 45,46,47], physical qualities [48,49,50,51] and biomechanics only .
Fifteen studies investigated biomechanical outcomes related to netball (Supplementary Table S1). Studies used a range of athletes from recreationally trained (n = 3, 20%) to elite level (n = 4, 27%). The majority of studies (80%, n = 12) investigated biomechanical outcomes related to injury risk [8,9,10, 52,53,54,55,56,57,58,59,60]. Movement analysis of the shoulder pass  and shooting  have also been carried out, and the influence of playing surface on landing has been assessed . Differences in shooting action between age-groups have been investigated . Senior-level players extend their elbows and knees simultaneously, while juniors appear to have greater dissociation between the upper and lower peripheries . Furthermore, senior-level players have greater variability of movement between their upper limbs when shooting . A case study (n = 1) identified that during the shoulder pass, the greatest ground reaction forces (GRF) have been observed to coincide with ball release . However, it is unknown whether differences in experience and passing success elicit different passing strategies.
During netball-related landing tasks, well-trained players demonstrated substantial valgus of the knee , with Collings et al.  demonstrating that experience level may not mitigate risk factors associated with poor frontal plane knee control. However, empirical evidence is still required to better understand whether differences in landing technique can modify injury risk.
Seven studies [8, 9, 54, 56, 58, 60] investigated the use of strapping and bracing of joints collectively demonstrating that strapping and bracing may reduce the range of motion but do not influence joint kinetics or kinematics. Yet, there is evidence to suggest that proprioception is improved with ankle bracing  and athletes perceive greater stability in the strapped or braced joint [9, 58]. Furthermore, whilst the joints may not be affected when landing, it is possible that muscle activity is reduced . This information suggests that bracing and taping joints may provide the athlete a perceived benefit rather than actual changes in the kinetic and kinematic outputs and whether these practices cause worthwhile reductions in injury risk is still unknown. Additionally, further investigation is required on whether lower limb movement screening  can help identify future injury risk in netball athletes.
Fatigue and Recovery
Nine studies investigated fatigue and recovery within netball (Supplementary Table S2). The majority of studies used elite level cohorts (n = 7, 78%) [63,64,65,66,67,68,69], with the remaining studies using Australian state-level athletes [3, 70]. Two studies (22%) quantified the fatigue response to competition [63, 67], whilst two studies (22%) investigated the effect  and the perceived importance  of various recovery modalities. Four studies (44%) focused on sleep indices, patterns and/or behaviours [3, 65, 66, 69]. Finally, one study investigated the influence of compression garments on netball-specific running performance during a circuit .
Studies investigating the fatigue responses have shown a varying time-course of recovery at 62 h post-match (i.e., not all variables returned to baseline) over a 3-day international tournament  and that subjective mental and physical fatigue were reported to be separate constructs, with limited relationships with performance variables following matches . Sleep and fluid replacements were reported as the perceived most important recovery modalities in 215 netball players . Additionally, Juliff et al.  found that following a netball-specific circuit contrast water therapy and contrast showers improved perception of recovery in comparison to the passive recovery condition, but no difference in physical recovery was reported. However, further research is required on recovery modalities and whether the development of specific physical qualities can positively influence transient (e.g., within match), acute (e.g., following a match) and chronic (e.g., over a season) fatigue and recovery profiles in netball players.
Studies investigating sleep found significant reductions in total sleep time following a match compared to training and rest days , which was consistent with perceived sleep duration and quality . Napping (> 20 min) on the day of performance appears to improve performance  and teams who slept longer reportedly achieve higher tournament positions . Given the importance of sleep and disruption following a match, future research should assess the efficacy and effectiveness of sleep hygiene strategies.
Of the 45 studies to investigate injury in netball, 31 (67%) had an injury epidemiology focus (Supplementary Table S3). These studies involved a wide range of cohorts including populations such as club [42, 71,72,73,74,75,76] and state or national level [32,33,34, 75, 77,78,79,80], whilst 23% (n = 11) of included studies reported netball injuries relative to “general population” patients [45,46,47, 81,82,83,84,85,86,87,88]. Studies also examined netball injury in both junior/youth [72, 87, 89] and senior/open age [77,78,79] athletes. The majority of epidemiological studies collected data over extended periods of a year or longer [32, 45,46,47, 74, 81,82,83,84,85,86,87,88, 90,91,92,93]. However, some studies utilised data representing much shorter periods, e.g., tournament weeks [71,72,73, 76, 94] or days [33, 34, 42, 75, 77, 78, 80].
Various approaches were taken when defining or classifying injury amongst the included studies. Some based classifications on accepted criteria [73, 78], others utilised broad definitions including, for example, aspects such as an injury that occurs whilst participating in sport, that leads to either a reduction in the amount of level of sports activity, or the need for advice or treatment . Finally, narrow definitions such as “trauma to a specific body part resulting in cessation of play” were also utilised .
A common theme of included studies was the reporting of injuries relative to hours of exposure [31, 33, 34, 72, 74, 76, 80, 94] or number of athletes [32, 34, 42, 73, 75, 81, 88, 89, 91,92,93]. Injury incidence was commonly reported between 12 and 14 injuries per 1000 exposure hours [30, 31, 34, 74, 76]. In addition to injury incidence and prevalence, the activity participants were performing (e.g. matches or training) when the injury occurred was also considered [33, 78], as was the specific movement or mechanism (e.g. landing, contact, overuse) involved [31, 32, 45, 71, 75, 77, 85, 88, 94]. Furthermore, the type and/or site of injury was commonly reported [30, 32, 33, 42, 47, 71, 73,74,75,76,77, 79, 80, 82,83,84, 86,87,88,89,90,91] with lower limb injury, particularly to the knee and ankle, common [30, 32, 42, 45, 71, 74, 75, 79, 80, 83, 84, 88, 90, 91, 94]. Whilst a systematic review of the mechanisms of non-contact knee injury has recently been carried out , the volume of injury epidemiology research in netball may warrant a systematic review and meta-analysis to provide researchers and practitioners with a high level of evidence regarding incidence and prevalence in addition to injury mechanism and site. Additionally, given there is no current evidence on the epidemiology and burden of concussion in netball, research in this area is also warranted.
Risk and Prevention
The remaining 14 injury studies included in the review had an injury risk, influence, or prevention focus (Supplementary Table S4). Junior (n = 4, 29%)[89, 96,97,98] through to open age players (n = 8, 57%)[10, 35, 99,100,101,102,103,104] were investigated and a variety of athlete performance levels were found within the studies including, school or club [96,97,98,99,100,101, 103, 105], inter-district [99,100,101], state or elite [102, 104, 106] and international . Although a range of athlete cohorts have been used in netball injury and prevention research, there may be value in extending this work, particularly in higher-level athletes as this group appears underrepresented in current studies.
A common purpose of these studies was the analysis of injury risk factors. These included prior injury  and the role of physical capacity (including movement quality) and anthropometry [89, 98, 101,102,103,104]. Other studies examined the impact of injury prevention strategies [35, 96, 97] and mechanisms/movement patterns involved in specific injuries (e.g. anterior cruciate ligament (ACL)). Further work analysed the impact of netball on balance and postural sway . Despite the large number of injury epidemiology studies (Supplementary Table S3) there is relatively little work examining injury risk and prevention strategies in netball and these could be the focus of future research.
Given the range of purposes of the studies included in this section, it is not surprising that a variety of data were collected. The most common related to injury history [35, 98, 99, 101,102,103,104,105] whilst balance (e.g. STAR excursion test) was also commonly assessed [99, 101, 105]. Some studies measured physical capacity (e.g. vertical jump) [96, 101, 104]; others reported somatotype and anthropometric variables [98, 102, 103]. In addition, performance on movement screens  and mobility measures  were utilised along with subjective questionnaires relating to the landing technique .
The results of studies assessing injury risk and the value of prevention programs were varied. For example, no difference in balance between participants with and without previous ankle sprain was demonstrated in one study  whilst the odds of ankle sprain were four times higher when STAR excursion posterior and medial direction was < 77.5% leg length , and a large proportion of knee injuries were associated with knee valgus . Similarly, studies examining the role of somatotype in injury risk produced equivocal results [102, 103].
The findings regarding the role of physical capacity and/or movement skill in injury are also inconsistent. For example, whilst a functional stability program reduced injury occurrence in international level players  and a 6-week program improved landing mechanics associated with ACL injury , one study found no difference in movement competency, jump performance or ankle dorsiflexion in injured players . In contrast, increased postural sway was evident on the preferred leg in those previously injured . Paradoxically, there is also some suggestion that greater jumping ability and anaerobic fitness were associated with greater injury risk . In comparison to other sports, there is very limited knowledge regarding the role of physical capacity (e.g. strength, aerobic capacity) and other factors (e.g. injury history) in injury risk in netball and this warrants further work. Furthermore, there may be a benefit from studies assessing injury burden (e.g. time lost for specific categories of injury), recurrence rate and the impact of specific rehabilitation protocols.
The details of the 24 studies investigating match-play in netball are displayed in Supplementary Table S5. Of those studies, 63% (n = 15) have examined match activity profiles [11, 12, 14, 15, 27, 107,108,109,110,111,112,113,114,115,116], whilst seven have focussed on technical and tactical aspects [13, 28, 29, 117,118,119,120], and two have examined performance outcomes [121, 122]. A large proportion (n = 17, 71%) of the studies examining aspects of match play involved elite-level competition [12, 15, 27,28,29, 107,108,109,110,111,112, 117,118,119,120,121,122] and all (except one study ) evaluated female netballers.
A range of metrics have been reported when examining the activity profiles of netball, including speed and distance, and non-locomotor activities (e.g., jumping). These were obtained using various approaches such as notational analysis, computerised analysis systems (e.g. Dartfish) , Inertial Measurement Units (IMU; accelerometers, gyroscopes, magnetometers) , Global Positioning Systems (GPS)  and Local Positional Systems (LPS) . The accelerometery derived metric PlayerLoad™ was used to quantify external load in nine of the activity profile studies [11, 14, 15, 110, 111, 113,114,115,116]. The predominance of technologies such as IMU and LPS to quantify the activity profile of netball is likely a function of many studies being conducted using high-level athletes participating in indoor matches, which precludes the use of technologies such as GPS commonly used in outdoor team sports.
A consistent theme across studies was the analysis of positional differences [11, 14, 15, 108, 110, 111, 113, 115, 116], with centre-court players displaying higher external loads than both GK and GS [11, 15, 108, 110, 111, 113, 115, 116]. In addition, a small number of studies has compared the activity profiles of different groups (e.g. male vs female, differing standards) [11, 14, 113, 114]. The ongoing development and increasing availability of measurement technologies provide an opportunity to develop a more detailed understanding of netball activity profiles at various performance levels.
The studies examining technical-tactical analysis have quantified a variety of common netball actions such as passing, interceptions, turnovers, and shooting percentage, all including elite level players [28, 29, 117,118,119]. An interesting finding is that experts appear to execute more passes under low levels of defensive pressure than developmental athletes and this resulted in more successful passes being completed by the elite players ; however this was the only study to compare levels of performer. In addition, one study has examined changes in technical-tactical aspects over time  and another study investigated potential strategies regarding the new ‘two-point’ rule in Suncorp Super Netball . Further work examining differences in technical-tactical aspects between levels appears warranted and this may assist with both talent identification and training programme design. Furthermore, more research examining links between game events and performance outcomes (e.g. win vs loss) may provide coaches with important information for both training programme design and tactical decision making.
The impact of factors including travel and fixture scheduling on performance (e.g. win vs loss, points margin) has been examined in two studies [121, 122]. The small volume of work in this topic limits the ability to definitively determine the importance of these aspects to netball performance and should be a focus of future research.
Three studies have investigated nutritional outcomes within netball (Supplementary Table S6) [6, 123, 124]. One study investigated the hydration status of international athletes , showing sweat and fluid intake rates of approximately 250 and 80 ml·h−1, respectively. The other two studies investigated the energy intake, expenditure and appetite of junior athletes at the club  and school  level, identifying alterations in appetite in response to netball exercise. However, this is yet to be investigated in senior or elite-level netball athletes. Furthermore, investigation into nutritional and dietary behaviors of these athletes may assist in the provision of nutritional interventions.
Thirty-seven studies have investigated the testing methods and physical characteristics of netball athletes (Supplementary Table S7), ranging from junior (n = 3) [4, 125, 126] to elite (n = 12) [18, 48, 49, 127,128,129,130,131,132,133,134,135] level. Twenty-two percent (n = 8) of studies specifically investigated the validity and/or the reliability of tests or testing outcomes [133,134,135,136,137,138,139,140] and 43% (n = 16) had a primary emphasis upon testing physical characteristics [16, 18, 20, 50, 125, 129,130,131,132, 141,142,143,144,145,146,147], with an additional four studies quantifying anthropometric characteristics [49, 51, 148, 149]. Finally, 24% (n = 9) of studies investigated the effects of different training interventions on changes in physical characteristics and performance [4, 48, 126,127,128, 150,151,152,153].
The 505 CoD test has been shown to demonstrate acceptable between-day reliability when 1–2 familiarisation sessions are provided . Furthermore, the 30:15 intermittent fitness test (30:15IFT) may be able to detect changes in high-intensity running performance across a training mesocycle . In comparison, the ‘Net-Test’  and Netball Specific Fitness Test , two sport-specific assessments, have demonstrated acceptable reliability and may be able to discriminate between athletes of different playing standards. Similar trends have been shown within the Reactive Agility Test and Planned Agility Tests , although it should be noted that the typical variation between-days for these tests are still unknown. Finally, the GRF produced during a single leg horizontal jump can provide reliable outcomes when assessing balance , while the Netball Movement Screening Tool is reliable when implemented by practitioners with similar levels of experience with the tool . However, it has not been ascertained whether the Netball Movement Screening Tool is a valid method for detecting increased injury risk for netball players. Considering these findings, the reliability of a narrow scope of tests has been assessed. Thus, further information is still required to ascertain the reliability and validity of a range of different tests and screening methods that assess different physical qualities.
Physical Characteristics and Anthropometrics
It has been demonstrated that, when compared to sedentary controls, netball athletes are taller and have greater lean body mass [148, 149]. Additionally, athletes at higher playing standards are taller and demonstrate greater sprint and CoD ability [18, 125, 132, 146], while centre-court players have greater fitness and jumping ability than other positional groups [16, 144]. However, it appears that body fat percentage may not be able to discriminate between playing levels [49, 51]. It has been shown that stronger athletes demonstrate greater acceleration (i.e., 5-10 m), CoD, and vertical jump , while the Functional Movement Screen has moderate relationships with trunk stability and CoD ability . Finally, small asymmetries in lower limb stability and vertical hop performance have been shown to occur in club level players , while a high occurrence (63%) of general joint hypermobility has been observed in elite players . From these findings, it is clear that differences in athletes are prevalent. However, a lack of systematic physical profiling has been completed (e.g., a standardised testing battery across age groups and playing standards) which is illustrated in a large number of tests, standardisation protocols, and outcome measures reported. Thus, researchers and practitioners should endeavour to work together to develop valid and reliable testing batteries that provide a comprehensive overview of the netball athlete.
Plyometric and strength training can induce favourable adaptations in measures of isometric strength, power, and CoD, while decreasing potential injury risk factors (e.g., peak landing forces) [48, 126, 127, 151]. Improvements in physical characteristics may be greatest in anaerobic qualities, which may be related to previous training exposure . Additionally, training outcomes may be augmented through the use of blood flow restriction or hypoxia when using low relative intensities during resistance training . However, practitioners must be wary of rapid improvements in physical performance during initial training periods and may require systematic alterations in the training stimulus to promote continued performance improvements . When programming, it may be prudent for practitioners to implement specific training interventions. The NetballSmart Dynamic Warm-up has been shown to improve some performance outcomes (i.e., the vertical jump and prone hold) , the inclusion of barefoot training and backwards running may enhance CoD ability [150, 152] and a combination of core stability, gluteus medius strengthening and proprioceptive exercises may enhance dynamic postural control . These findings indicate that moderate to large improvements in physical performance occur with training, although further research is still required to elucidate the effects of strength and power training in well-trained netball athletes and whether changes in these physical capacities can reduce injury occurrence.
Thirteen studies have investigated an aspect of psychology in netball (Supplementary Table S8). Five studies (38%) focused on motor learning and decision making [21, 154,155,156,157]. Other studies investigated a range of psychological skills or interventions [7, 22, 158,159,160,161,162,163], including communication , anxiety , stressors, , behaviour , team cohesion , and imagery . The cohorts investigated ranged from club to elite and international, with six studies (46%)[154,155,156,157, 159, 160] investigating international level athletes of a range of ages (U17 to open age). Three studies investigated differences between playing standards or levels [155,156,157], while three studies made positional comparisons [22, 155, 162]. However, further research is required on the psychological skills of elite netballers and the effectiveness of mental skills training. Given the high prevalence of injury in netball (Supplementary Table S3), research into the psychological effect (i.e., anxiety and stress) of injuries and the impact of psychological support during injury rehabilitation would also be beneficial.
Motor Learning and Decision Making
Decision-making was found to be a key discriminator between highly skilled and less skilled performers, evident through greater accuracy in a decision-making task [155,156,157], while Richards et al.  demonstrated the ability of a coach to influence an elite team’s decision making process. Positional comparisons demonstrated limited position specificity in perceptual-cognitive skills in decision-making tasks  but GA and WD players have been found to possess greater coping skills compared to other positions .
Four studies have investigated the external load of training in netball [5, 164,165,166] (Supplementary Table S6), with the majority (n = 3) investigating professional level cohorts. Three studies compared the workload of training to competitive match-play [5, 164, 165], two of which investigated specific training dills [5, 164]. One study highlights the need to combine internal and external workloads when monitoring elite netball athletes, suggesting the use of session rating of perceived exertion and CoD as the ideal combination . However, further investigation is required into the training load and training practices of netball across the range of playing standards and age-groups.