Epidemiology of injury and illness across all the competitive cycling disciplines: a systematic review and meta-analysis

Abstract

Introduction

The Union Cycliste Internationale, the world cycling governing body, has highlighted its mission to promote and support research in cycling epidemiology and medicine, especially for the benefit of lesser-known disciplines within its 2030 Agenda. This systematic review aimed to comprehensively analyse and synthesise the existing literature about cycling-related injuries and illnesses across all competitive disciplines.

Design

Systematic review and meta-analysis of epidemiological studies.

Data sources

PubMed (Medline), Embase and SPORTDiscus were searched in August 2024.

Eligibility criteria

All epidemiological studies including information on injuries and/or illness in competitive cycling from inception to August 2024.

Results

Forty-four studies were included across the competitive cycling disciplines of Mountain Bike (MTB), road, track, bicycle motocross (BMX) and para cycling. No studies were found for cyclocross, gravel, indoor, trials and esports cycling disciplines. BMX had the highest injury incidence overall, with 4.59 injuries per 365 days (95% CI 4.34 to 4.85), while para-cyclists had the highest illness prevalence at 3.89 illnesses per 365 days (95% CI 3.42 to 4.36). The most common injuries were upper limb injuries, particularly in BMX (65.21% (95% CI 54.50 to 74.65)), followed by road cycling (48.32% (95% CI 46.51 to 50.14)) and track cycling (44.18% (95% CI 43.28 to 45.09)). In road cycling, 14.36% (95% CI 13.12 to 15.69) of injuries were head, neck or face-related. Skin lacerations, abrasions and lesions were the most common injury types across disciplines, with bone fractures being the second most common. Despite a higher prevalence of head, neck and face injuries, road cycling had the lowest incidence of concussions (4.68% (95% CI 4.04 to 5.42)). Gastrointestinal illness had the highest prevalence overall in road cycling (17.90% (95% CI 14.4 to 22.0)), while respiratory illness was most prevalent in Track (31.82% (95% CI 22.29 to 42.61)) and MTB (31.82% (95% CI 22.29 to 42.61)).

Conclusion

This systematic review highlights the high prevalence of upper limb injuries across competitive cycling disciplines, particularly in BMX, road and track cycling. Skin lacerations, abrasions and bone fractures were the most common injury types. These findings help inform future injury and illness surveillance studies in cycling disciplines where high-quality epidemiological data are currently lacking and support the development of future injury and illness prevention strategies.

PROSPERO registration number

CRD42024502703.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Injury and illness epidemiology research within cycling has lacked methodological guidance and, therefore, homogeneity around the definitions and reporting standards of injury and illness.

• It remains unknown if the profile of injury and illness differs between competitive cycling disciplines.

• The lack of high-quality, systematic injury and illness data across cycling disciplines limit the development of effective prevention measures and evidence-based strategies.

WHAT ARE THE NEW FINDINGS

• Road cycling and mountain biking are the most represented disciplines within epidemiological studies, while track cycling, bicycle motocross and para cycling are poorly represented outside of the Olympics, European Championships and Paralympics.

• The upper limb is the most commonly acute injured region across disciplines, with the lower limb being the most common location for overuse injuries.

• Further injury/illness epidemiology studies need to be undertaken in all cycling disciplines to understand the risks of the sport and to develop and implement prevention programmes. These studies are particularly needed in cyclocross, gravel, indoor, trials and esports but have no studies to date.

• Prevention measures, such as site-specific Dyneema material, should be discussed with clothing manufacturers to help protect the skin from abrasion-type injuries when crashing. Similarly, educational content on preventing skin lesions (eg, saddle sores) and signs/symptoms of concussion, along with information on their prevalence, could be valuable in reducing these types of injuries.

Introduction

Cycling is a globally popular and diverse sport, encompassing multiple disciplines, each with distinct physiological demands and associated injury and illness risks.¹ The Union Cycliste Internationale (UCI), which is the world governing body for cycling, has over 1 million members across five continents and holds over 33 000 events per year.² Within the UCI Agenda 2030, the importance of promoting injury and illness surveillance to inform prevention strategies is highlighted, with a commitment to: promote and support research in cycling epidemiology and medicine, especially for the benefit of lesser-known disciplines.² Across all injury prevention frameworks, injury surveillance forms the foundation for developing effective, evidence-based injury prevention strategies.³,⁶ Therefore, emphasising the importance of epidemiological research to understand injury profiles and support athlete health and safety.⁷

Despite the recognised importance of epidemiological research, current literature in cycling remains limited and lacks methodological consistency.⁸ Three systematic reviews have been conducted on injuries and illnesses in cycling, with two focusing on mountain biking^{9 10} and one on road cycling, encompassing all levels of cyclists.¹¹ Due to the diverse nature of competitive cycling disciplines, including variations in demands, environments and peloton sizes, the incidence and cause of injury and illness may differ considerably between them. Over the past three Olympic cycles, Bicycle Motocross (BMX), encompassing both racing and freestyle disciplines, has recorded one of the highest proportions of athletes’ injuries across all Olympic sports.¹² Conversely, road cycling was among the lowest.¹² Arguably, expressing injury rates as the proportion of athletes injured may underestimate the true injury risk, particularly when comparing disciplines with vastly different field sizes, such as BMX racing (typically 24 riders) versus road racing (often exceeding 130 riders).⁸ Despite these apparent differences, the overall injury and illness profiles across cycling disciplines remain poorly defined. It is also unclear which disciplines are well represented in the existing literature and which remain under-researched. Addressing this gap is essential to inform future research directions, particularly in line with the UCI Agenda 2030’s call to prioritise lesser-known disciplines.²

Therefore, this systematic review aims to comprehensively analyse and synthesise the existing literature on cycling-related injuries and illnesses across all competitive disciplines. While individual studies often focus on specific disciplines or injury types, this review will collate and critically evaluate the broader body of research, including evidence from para cycling, to provide an inclusive overview of injury and illness patterns in competitive cycling.

Methods

Protocol and registration design

This study was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.¹³ The protocol for this systematic review was registered on the PROSPERO international prospective register for systematic reviews with registration number CRD42024502703.¹⁴

Search strategy

A computerised, systematic literature search was conducted in the electronic databases MEDLINE, Embase (both via Ovid), and Scopus was completed in August 2024. In line with recommendations for systematic reviews¹⁵ on measurement properties, a hand search of the reference lists of included studies and relevant reviews for additional sources and a grey literature search in Google Scholar were conducted. The search strategy was reviewed and approved by a medical librarian at Queen’s University Belfast. The search strategy was based on keywords broken down into those related to cycling and those related to injury and illness. Using Boolean operators (AND, OR), we combined the cycling and injury/illness keywords. For example: (‘Competitive cycling’ OR ‘Road cycling’ OR ‘Elite cycling’). Using the term ‘AND’, we combined cycling keywords and injury/illness keywords. Example (‘Sports injuries’ OR ‘Musculoskeletal injuries’) AND (‘Elite Cycling’ OR ‘Competitive cycling’) (see online supplemental material 1).

Criteria for inclusion of studies

This systematic review included any design (eg, observational, epidemiological, interventional) in which injuries and/or illnesses in competitive cycling disciplines were a primary outcome of investigation. To enable the identification of the most frequently occurring injuries and therein derive injury prevention priorities, studies which reported only on a specific type of injury (eg, stress fractures, concussion, external iliac endofibrosis), illness (gastrointestinal, respiratory) or body part (eg, head injuries) were excluded. Studies in recreational cycling events (ie, not competition based) were also excluded.

Cycling is an umbrella term used for many individual disciplines.¹⁴ It is acknowledged that each discipline may have further subdisciplines; however, for clarity, the broad disciplines as categorised by the UCI were used for grouping. Investigation of all types of injuries and/or illnesses sustained during training and/or competition among competitive athletes across all disciplines was included in this review. A competition was defined in line with the Cambridge dictionary as ‘an organised event in which people try to win a prize by being the best, fastest, etc’.¹⁶

Study inclusion criteria were framed according to PICO:

• Population (P)—competitive cyclists. Competitive cyclists are those who are partaking in competition at any level.

  • As this study focused on competitive cyclists, we divided this population into three different levels modified from Heron et al¹⁷

1. An amateur is a cyclist, described as one who practices cycling for non-economic reasons and participates in official domestic competitions.

2. An elite cyclist is a cyclist who competes at the national/international level but does not receive a regular salary or income for their involvement in the sport.

3. A professional cyclist is a cyclist who competes at the national/international level and receives a regular salary or income for their involvement in the sport.

• Intervention (I)—not applicable.

• Comparator (C)—not applicable.

• Outcome (O)—the outcome of interest in the studies will be injury/illness prevalence, incidence, sex, type, location, aetiology, mode of onset and exposure rate, where available.

Study selection and data extraction

References exported from each database were imported onto the AI systematic review software ‘Rayyan’ (Cambridge, Massachusetts), where duplicates were removed and made available for screening. Independent screening by two reviewers (TF and NH) in a two-step process: title/abstract screening, followed by full-text review (blinded for review). In addition, citation tracking was also used to identify potentially eligible studies. Reviewers used the predefined inclusion and exclusion criteria to determine study eligibility. A third reviewer (DP) was in place to resolve disagreements among these reviewers and to facilitate consensus. Once a study consensus was reached, references were transferred to the Mendeley reference manager for data extraction and write-up.

Data extraction

Systematic extraction of relevant data elements, including the following information listed in online supplemental material 2, onto a purpose-built Microsoft Excel sheet was completed by the first reviewer (TF). Double verification of extracted data was completed by the second reviewer to ensure accuracy and completeness (NH). Data were extracted from results, tables, figures and available supplementary material was accessed for raw injury/illness data, where indicated. If multiple studies originated from the same dataset, they were consolidated to avoid duplication of data. The most comprehensive or recent publication was used as the primary source, with supplementary information drawn from related studies where appropriate. The author of one study was contacted for further information about discipline-specific injury/illness data.¹⁸ We did not contact the authors of all primary studies to clarify eligibility.

Methodological quality assessment

Studies were independently assessed by two reviewers using the National Institute of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies to determine study quality.¹⁹ The proportion of questions labelled yes was calculated, with the higher score indicating a lower risk of bias. After discussion, the authors (TF and NH) came to a consensus on the risk of bias scoring in all cases (online supplemental material 3). It was deemed that the quality appraisal would provide sufficient insight into the methodological rigour of the included studies, making a certainty analysis unnecessary.

Synthesis of results

Findings from the included studies were synthesised to present the proportions of injury and illness by anatomical location, type and aetiology, with 95% CIs using R statistics (V.4.1.1; R Core Team 2014). The variables of injury/illness types and definitions, event category, level of participation, diagnostic categories (ie, medical professionals, self-reported, etc) and injury/illness prevalence and incidence were included. Proportions were calculated from the extracted data for each anatomical region and injury/illness type. Injury regions were categorised into seven broad anatomical regions: upper limb, lower limb, head, face and neck, torso and back, pelvis, hip and dental. A weighted average and 95% CI for each discipline were then calculated for injury and illness.

In line with the 2021 IOC consensus statement for the comparison between disciplines, a meta-analysis was completed on disciplines with two or more studies. Only studies that demonstrated definitional, population-level and reporting homogeneity, assessed using the PICOS framework, were included. Included studies involved professional or elite-level cyclists, used IOC-aligned definitions of injury and illness, reported total event duration (days) and had injury/illness diagnosed by a medical professional, thereby ensuring consistent and comparable reporting standards.⁸ Results from the para cycling discipline (track and road) were combined in line with the most recent Paralympic Games study.^{20 21}

R statistics (V.3.1.2; R Core Team 2014) were used to calculate injury per person-years around each discipline in line with the IOC consensus statement for the comparison of injury rates between disciplines.⁸ The R package ‘metafor’ was used to complete a meta-analysis on each discipline. To assess heterogeneity between studies, the I^² statistic was employed. I^² quantifies the proportion of total variation across studies that is attributable to heterogeneity rather than chance. An I^² value greater than 50% was considered indicative of high heterogeneity.²² Because heterogeneity between studies was expected, a random-effects model with the Hartung-Knapp-Sidik-Jonkman adjustment was used for all analyses.²³

Certainty assessment

The certainty of evidence was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.²⁴ Evidence was classified as high, moderate, low or very low, reflecting the level of confidence in the estimated effect. Downgrading decisions were based on key criteria, including risk of bias, inconsistency, indirectness, imprecision and potential publication bias.

Public and patient involvement

This study did not involve public and patient involvement in its design or conduct. However, members of the public, including cycling enthusiasts, athletes and coaches, are included within the dissemination plans of the research.

Equity, diversity and inclusion statement

Our study identified competitive cyclists, inclusive of all genders, socioeconomic backgrounds and cultures, including cross-sectional studies from global populations. The author group comprised medical professionals and epidemiological researchers. One author is female, and three are male, from the United Kingdom, Ireland and France and are not from any marginalised background.

Results

Full-text selection

409 Studies were identified from the initial literature search. After removing duplicates and title, and abstract screening, 56 studies met the full-text review criteria. Of the 56 studies reviewed, 39 were included (online supplemental material 4). The reference lists of the 39 studies identified five articles added to the review figure 1. Thus, the total number of studies included in the review was 44 (online supplemental material 5).

Figure 1. Preferred Reporting Items for Systematic Review and Meta-analysis flow diagram outlining the systematic review process and inclusion of studies.

Quality assessment

The quality assessment of the included observational cohort and cross-sectional studies was conducted using the NIH Quality Assessment Tool. Across the studies, the percentage of applicable questions answered positively (‘yes’) varied from 41% to 83%. This variation highlights heterogeneity in the methodological rigour across the studies, with higher scores suggestive of greater study quality (online supplemental material 3).

Study characteristics

Disciplines represented

Five out of the 10 cycling disciplines were represented across the 44 studies. Four studies represented more than one discipline (MTB, Road, Track, BMX) from data originating from the Olympics 2012,²⁵ 2016²⁶ and 2020/1¹² and then.^{18 18} These reports were broken down into each discipline, allowing for discipline injury pooling. All studies within Paracycling were carried out at the Paralympic Games, combining both track and road disciplines.^{2021 27},³¹ Not all studies reported all outcomes of interest, with variations in the availability of data on injury/illness prevalence, incidence, type, location, aetiology, mode of onset, incidence and exposure rate between studies.

Study design

Within MTB, 15 prospective studies were identified.1218 25 26 32,⁴² Followed by 12 prospective studies in road cycling,1218 25 26 43,⁴⁹ eight in para cycling,^{2021 27},^{31 50} six in BMX^{12 18 25 26 51} and five in track.^{1617 22},²⁶ There were eight retrospective studies in road cycling⁵²,⁵⁹ and three within MTB.⁶⁰,⁶² While MTB has the greatest amount of diversity between its disciplines, each with distinct race formats, terrains and rider profiles. From our review, there were 10 studies in Cross Country and endurance MTB,¹²¹⁶,^{18 23 26} 1 study in enduro,³² 6 studies in DHMTB^{34 40 41 41 60 61} and 1 in 24-hour MTB.³³ From this point onwards, we have combined the data generated from all MTB studies to provide an overall capture of injuries within the broad discipline.

Injury and illness definitions

The definition of a recordable injury and illness varied between studies in terms of methods of diagnosis and reporting standards (online supplemental material 5). Road cycling was the main discipline, which examined overuse injuries across five studies, and the definitions varied. Overuse injuries were defined as any pain or discomfort that was not directly associated with a traumatic event and was different from the normal aches and pains associated with competitive cycling,⁵⁶ any repetitive stress associated with physical activity involved by sports practice either in competition or training sessions, which causes pain, dysfunction, pathology or handicap.⁵⁵ One study, which examined overuse and traumatic injuries, defined overuse injuries as a complaint that was related to an overuse condition while cycling with no trauma.⁶³ One study retrospectively included information about non-traumatic injuries sustained (including all musculoskeletal and all neurological concerns participants may have experienced due to cycling in the preceding year)⁵⁸ and another where cyclists were specifically asked: ‘Do you or did you suffer from any symptoms of a chronic (no accident) cycling injury (muscles, tendons, bones, ligaments or joints) in your cycling career?’.⁵⁹ Overuse injuries were examined in one para cycling study where they were defined as ‘musculoskeletal complaints that occurred without acute trauma or were an aggravation of pre-existing complaints and developed over days, weeks or months without any connection to a causative event’.”⁵⁰

Diagnosis of injury and illness

The diagnosis of injuries and illnesses varied between studies. Injury/illness diagnosis was performed by medical professionals in 13 MTB studies,1218 25 26 32 33 35 38,^{40 60 64} 11 road cycling studies1218 25 26 43,⁴⁹ and 5 BMX,^{12 18 25 26 51} para cycling²⁷,³¹ and track cycling^{12 18 25 26 51} studies. Injuries were self-reported in five MTB studies^{34 36 37 61 62} and three road cycling studies.⁵⁷,⁵⁹ Historical reports of injury were used to collate injury history in one road cycling study⁵⁴; one study did not specifically reference who reported the injuries,⁵⁵ and one retrospectively combined medical attention and time loss injuries.⁵⁶

Level of participation

The participation level of riders varied among studies. Within track,^{12 18 25 26 65} para cycling²⁷,³¹ and BMX,^{12 18 25 26 51} the level of participation was elite/professional. In road cycling, there was a mix of levels across studies varying from underage (<19),⁴⁴ amateur and elite,^{44 47 52 53} professional only,^{43 45 54} elite/professional,^{12 18 25 26 56} combined amateur, elite, professional,^{4849 57},⁵⁹ elite^{55 63} and amateur⁴⁶ only. Similarly, within MTB, there was a mix of levels within studies, varying from elite and professional,^{12 18 25 26} amateur and elite,^{2633 35 37},^{39 60 62} amateur, elite and professional,^{32 61} amateur⁶⁴ and elite⁴⁰ only. Only one study⁶⁵ prospectively monitored athletes across training and competition. Three studies focused on overuse injuries in road cyclists.^{55 56 63}

Sex representation across cycling disciplines

Studies in track cycling,^{12 18 25 26 65} para cycling²⁷,³¹ and BMX^{12 18 25 26 51} included both male and female athletes. In road cycling, there was a mix of studies which included both males and females,^{12 18 25 26 43 44 52 57} males only⁴³⁵⁴,^{56 63} and studies that did not specify sex.^{46 47 53} Similarly, MTB studies included male and female riders,1218 25 26 32 33 36 38,^{41 60} male-only³⁷ and sex not specified.^{34 35}

Injury incident rates

Out of the 44 studies included in this review, 24 presented injury incidence rates (online supplemental material 6). This differed through studies from injuries per 1000 hours of competition,^{1230 32},^{34 37} per 100 hours⁴⁷ per 365 days,^{55 56} per athlete,⁵⁵ per 1000 km,⁵⁵ per comp day,⁵⁵ per 1000 athlete days,^{20 27 55} per 1000 starters⁴⁹ and lifetime prevalence.⁵⁹

Injury locations and injury types

Overall, the most common injury location within competitive cycling was the upper limb, followed by lower limb injuries (table 1). The most common injuries noted across disciplines differed. Within BMX, 65.20% (95% CI 54.50 to 74.60) of injuries were within the upper limb, followed by 48.30% (95% CI 46.50 to 50.10) within road cycling and 44.18% (95% CI 43.20 to 45.00) within track cycling. No study within para cycling referenced the injury-specific injury location or injury type (table 2).

Table 1. Proportion of acute injuries in each injury location by cycling discipline presented as a percentage [95% CI in parenthesis].

	Area of injury	BMX (%)12 18 25 26	Road (%)1218 25 26 43,45 52	Track (%)12 18 25 26	MTB (%)1218 25 26 32,41 60
Upper limb	Shoulder	65.20 [54.5–74.6)	48.20 [46.51–49.89)	30.70 [19.10–44.20)	44.10 [43.20–45.00)
	Upper arm
	Elbow
	Forearm
	Wrist
	Hand/fingers
	Thumb
Lower limb	Anterior thigh	15.20 [8.80–24.50)	26.56 [25.09–28.08)	15.38 [8.00–28.8)	32.20 [31.40–33.10)
	Posterior thigh
	Knee
	Anterior lower leg
	Posterior lower leg
	Calf/Achilles
	Ankle
	Foot/toes
Head, face and neck	Face (including eyes, ears, nose)	7.60 [3.38–15.50)	15.00 [13.84–16.25)	11.50 [4.43–22.50)	11.50 [10.90–12.10)
	Head
	Neck
Torso and back	Chest	6.50 [2.68–14.20)	7.71 [6.86–8.66)	13.40 [6.80–26.70)	6.21(5.78, 6.66)
	Abdomen
	Thoracic
	Upper back
	Lower back
	Back
	Multiple body locations	–	0.20 [0.08–0.46)	–	4.5 [4.14–4.89)
Pelvic hip	Pelvis	5.40 [2.02–12.8)	1.94 [1.52–2.46)	28.80 [17.6–42.4)	1.06 [0.89–1.27)
	Hip
Dental	Dental	2.00 [0.38–8.38)	0.60 [0.39–0.92)	–	0.22 [0.15–0.33)

BMX, bicycle motocross; MTB, mountain bike.

Table 2. Proportion of acute injury types within each discipline [95% CI in parenthesis].

Type of injury	BMX (%)12 18 25 26 51	Road (%)1218 25 26 43,47 52	Track (%)12 18 25 26	MTB (%)1218 25 26 32,36 38
Skin—laceration/abrasion/skin lesion	35.30 [26.80–44.8)	44.39 [42.77–46.02)	27.60 [17.6–40.3)	65.40 [62.60–68.0)
Bone—acute fracture	6.90 [3.20–13.50)	17.84 [16.63–19.13)	7.69 [2.80–17.70)	8.90 [7.45–10.70)
Bone—stress fracture/stress reaction	–	0.61 [0.40–0.92)	–	–
Bone—contusion	0.86 [0.05–5.41)	0.72 [0.49–1.05)	1.54 [0.08–9.40)	0.01 [0.00–0.39)
Neuro—concussion	6.03 [2.67–12.4)	4.68 [4.04–5.42)	13.8 [6.91–25.1)	5.86 [4.65–7.36)
Muscle—contusion/haematoma	25.00 [17.60–34.0)	18.12 [16.90–19.41)	35.3 [24.2–48.3)	6.05 [4.79–7.54)
Muscle—strain grade 1 or 2	1.72 [0.30–6.71)	0.69 [0.47–1.02)	3.08 [0.53–11.6)	8.38 [6.93–10.1)
Muscle—cramps or spasms	0.86 [0.05–5.41)	0.58 [0.38–0.89)	1.54 [0.08–9.40)	0.01 [0.00–0.39)
Muscle—rupture grade 3	–	0.06 [0.02–0.20)	–	–
Tendon—tendinopathy/tendinosis/tendinitis	2.59 [0.67–7.94)	0.78 [0.54–1.12)	–	–
Tendon—sprain grade 1 or 2	0.86 [0.05–5.41)	–	–	–
Tendon—rupture grade 3	0.86 [0.05–5.41)	–	–	–
Ligament—sprain grade 1 or 2	12.00 [7.00–19.70)	2.41 [1.96–2.96)	3.08 [0.53–11.6)	1.89 [1.22–2.84)
Ligament—rupture grade 3	1.72 [0.30–6.71)	0.08 [0.03–0.24)	–	0.07 [0.00–0.53)
Joint—dislocation/subluxation/instability	2.59 [0.67–7.94)	1.03 [0.74–1.41)	1.54 [0.08–9.40)	0.68 [0.30–1.33)
Joint—arthritis/synovitis/bursitis	–	0.17 [0.08–0.36)	1.54 [0.08–9.40)	0.07 [0.00–0.53)
Joint—lesion of meniscus or cartilage	–	0.08 [0.03–0.24)	1.54 [0.08–9.40)	0.01 [0.00–0.39)
Other injury (assessment note required)	2.59 [0.67–7.94)	6.37 [5.62–7.22)	1.54 [0.08–9.40)	1.60 [1.02–2.55)
Injury without tissue type specified	–	0.61 [0.40–0.92)	–	1.03 [0.59–1.85)

BMX, bicycle motocross; MTB, mountain bike.

Within road cycling, bone fractures accounted for 17.19% (95% CI 15.90 to 18.56) of all injuries seen (table 2). Despite the high incidence of head, neck and face injuries (14.30%—95% CI 13.12 to 15.69), road cycling had the lowest incidence of concussion at 4.19% (95% CI 3.53 to 4.96). Muscle strains (grade 1–2) accounted for 8.38% (95% CI 6.93 to 10.11) of all injuries within MTB and 3.08% (95% CI 0.53 to 11.64) in track.

Five studies examined overuse injuries, all focused on road cycling, and the injury profile was different.^{55 56 58 59 63} The lower limb was the most injured region, 48.82% (95% CI 45.38 to 52.27) followed by the torso and back, 19.13% (95% CI 16.56 to 21.99). Muscle strains 20.74% (95% CI 18.00 to 23.79) and tendon-related injuries at 19.95% (95% CI 17.25 to 22.95) were the most common injury types.⁵⁰ In the only para cycling study that specifically distinguished overuse injuries, there was a more balanced distribution between upper limb and lower limb overuse injuries (53% lower/ 47% upper).⁵⁰

Illness prevalence, incidence and aetiology

All studies reporting illnesses did so concurrently with injury data and included only competition-reported illnesses. The findings presented in table 3 show the varied incidence rate of illness injury between able-bodied cyclists and para cyclists. In pooled data across the studies, which examined illness, gastrointestinal illness had the highest prevalence overall (32.43%), followed by respiratory (ear/nose/throat) (21.62%) (table 4). Within para cycling, only two studies^{31 50} presented the illness types among cyclists. The most common illness noted was respiratory, followed by dermatological and gastrointestinal. No study examining para cycling stated the aetiology of illness. Additionally, the reporting of such differences exists with incidence rates differing from per 1000 hours to per 1000 athletes.

Table 3. Illness rates in each discipline with definition of illness for each study.

Study	Discipline	Illness definition	Illness incidence
Soligard et al12	Track	Illnesses as new (pre-existing, not fully rehabilitated conditions were not recorded) or recurring (athletes having returned to full participation after a previous condition) incurred in competition or training during the period between the opening and closing ceremonies of the Olympic Games Tokyo 2020 (23 July to 8 August 2021) receiving medical attention, regardless of the consequences with respect to absence from competition or training.	1.1/1000 hours
	BMX-free style		5.6/1000 hours
	BMX racing		6.3/1000 hours
	Road		0.5/1000 hours
Yanturali et al45	Road	An illness is defined as a physical complaint or presentation not related to injury.	3.01/1000 hours
Killops et al49	Road	‘any cyclist interacting with the medical team and requiring medical assistance or evaluation, taking place from the official start of the event, up to 24hours after the official cut-off time of the event’	2.10 (1.84–2.40) /1000 starters
Derman et al21	Para cycling—road/track	An illness was defined as ‘a complaint or disorder experienced by an athlete, not related to injury. Illnesses include health-related problems in physical (eg, Influenza), psychological (eg, Depression) or social well-being or removal or loss of vital elements (air, water, warmth).	4.7/1000 athlete days
Derman, et al29	Para cycling—road/track	The general definition for reporting an illness was described as ‘any athlete requiring medical attention for an illness regardless of the consequences with regard to absences from training or competition’. A medical illness was specifically defined as ‘any newly acquired illness as well as exacerbations of pre-existing illness that occurred during training and/or competition during the pre-competition or competition periods of the Rio 2016 Summer Paralympic Games’	10.5/1000 athlete days
Schwellnus et al31	Para cycling road	‘Any newly acquired illness as well as exacerbations of pre-existing illness that occurred during training or competition, and during or immediately before the London 2012 Paralympic Games’	Overall14.1/1000 athlete daysFemale IR9.3/1000 athlete daysMale IR16.2/1000 athlete days
	Para cycling track		Overall12.4/1000 athlete daysFemale IR16.5/1000 athlete daysMale IR10.8/1000 athlete days

Table 4. Proportion of illness types within each discipline [95% CI in parenthesis].

Illness type	BMX (%)12 25 26	Road (%)12 18 25 26 45 46	Track (%)12 25 26	MTB (%)25 26 33 35 60 62
Gastro-intestinal	14.20 [8.41–24.3)	17.90 [14.40–22.00)	13.60 [2.70–33.50)	22.70 [14.40–32.80)
Respiratory/ear, nose, throat	–	12.60 [9.70–16.20)	40.90 [19.70–61.40)	31.80 [22.20–42.60)
Dermatologic	42.80 [32.80–54.10)	15.10 [11.90–19.00)	22.70 [7.46–43.70)	13.60 [7.25–22.60)
Allergic/immunological	–	3.30 [1.90–5.50)	–	–
Other	–	6.00 [4.10–8.80)	4.55 [0.11–21.90)	9.09 [4.01–17.10)
Neurological/psychiatric	–	6.50 [4.50–9.40)	9.09 [1.07–28.00)	13.60 [7.25–22.60)
Uro-genital/gynaecological	14.20 [8.41–24.30)	0.30 [0.00–1.40)	9.09 [1.07–28.00)	–
Musculoskeletal	14.20 [8.41–24.30)	–	–	4.55 [1.25–11.20)
Thermoregulatory	–	–	–	–
Cardiovascular	14.2 [8.41–24.30)	16.40 [13.10–20.30)	–	–
Dental	–	–	–	4.55 [1.25–11.20)
Metabolic/endocrinological	–	16.10 [12.80–20.10)	–	–

BMX, bicycle motocross; MTB, mountain bike.

The most common aetiology of illness was environmental—non-exercise related (50% (95% CI 38.27 to 61.73)), followed by infection (32.81% (95% CI 23.12 to 45.38). MTB had the highest prevalence of environmental–non-exercise-related illness (61.9% (95% CI 40.8 to 79.32)), followed by track cycling (50 (95% CI 30.72 to 69.28)). No studies that included para cycling broke down the aetiology of illness.

Meta-analyses of injury and illness per person year

Twelve studies were included in the five discipline-specific meta-analyses on injury and illness. The population level across studies was professional athletes with injury and/or illness diagnoses being performed by medical professionals. All studies aligned with the IOC Consensus Statement reporting standards and were carried out prospectively during a clearly defined competition period. Heterogeneity was high, as indicated by the I² values, all exceeding 50%, suggesting variation between studies within each discipline (online supplemental material 7). There were four studies in BMX (both BMX Freestyle and Racing), track cycling and MTB, with five in road cycling and six in para cycling (three injury/three illness). Per 365 rider days, BMX had 4.59 (95% CI 4.34 to 4.85), track cycling 3.45 (95% CI 2.23 to 4.68), road cycling at 3.68 (95% CI 1.88 to 5.47), MTB at 3.61 (95% CI 2.65 to 4.58) and para cycling at 3.62 (95% CI 3.47 to 3.77) (figure 2). Para cyclists reported 3.89 (95% CI 3.42 to 4.36) illnesses per 365 days (figure 3). From the studies included in the meta-analysis, only three studies^{12 18 26} specified the sex of athletes and the injury/illness rate of each sex. When comparing the rates of injury per discipline per 365 athlete days, the rates differed between disciplines. In BMX, there were 7.46 (95% CI 5.49 to 9.43) injuries per 365 athlete days in males, compared with 4.30 (95% CI 3.14 to 5.47) injuries per 365 days in Female BMX athletes. In road cycling, there were 14.91 (95% CI 2.04 to 27.77) in males and 0.68 (95% CI 0.01 to 1.18) in females. Track cycling is 1.97 (95% CI 0.22 to 3.71) in males and 12.57 (95% CI 0.01 to 35.11) in females. Finally, MTB 1.82 (95% CI 0.61 to 4.42) in males and 4.87 (95% CI 0.51 to 9.22) in females. Comparisons in illness rates between sexes were not carried out as there were not enough data available to complete such comparisons across all disciplines.

Figure 2. Subdiscipline meta-analysis representing injuries per 365 days. *Bicycle motocross (BMX), mountain biking (MTB).

Figure 3. Subdiscipline meta-analysis representing illnesses per 365 days. *Bicycle motocross (BMX), mountain biking (MTB).

Overall certainty of evidence

The certainty of evidence for injury and illness across cycling disciplines was assessed using the GRADE approach (online supplemental material 8). For injuries, overall certainty was rated as moderate for road cycling, track cycling and para cycling, due to inconsistency between studies. Mountain biking and BMX were rated low certainty, with downgrades due to both inconsistency and imprecision resulting from smaller sample sizes.

For illness, most disciplines also showed moderate certainty (road cycling, MTB and para cycling), while BMX and track cycling were graded low certainty, with downgrades due to inconsistency and imprecision.

Discussion

Overview

This is the first systematic review to examine injuries and illnesses across all competitive cycling disciplines. The key findings of this review were that the 10 broad cycling disciplines, mountain biking (MTB), road, track, BMX and para cycling are represented, while disciplines such as cyclocross, gravel, indoor, trial and esports are not represented within the research. Road and MTB cycling are the most extensively studied, with 20 and 18 studies, respectively, whereas track cycling, BMX and para cycling are under-represented, particularly outside major events such as the Olympics, European Championships and Paralympic Games. The upper limb is the most injured body region, and skin lacerations/abrasions and bone fractures are the predominant injury types across all disciplines. Injury rates per 365 rider days vary, with BMX reporting the highest incidence (4.59 injuries per 365 days), followed by road, MTB, track and para cycling. Gastrointestinal, cardiovascular and respiratory illnesses are the most frequently reported, with environmental factors identified as the leading cause. However, significant gaps persist, particularly regarding the nature and location of injuries and illnesses in para cycling, which remains inadequately documented in the literature.

Cycling as a sport is growing in popularity. Our review highlights the paucity of high-quality, homogeneous injury and illness epidemiology research in cycling. This review supports the UCI Agenda 2030 objectives and provides direction on where high-quality injury and illness epidemiology research within cycling is needed.² The disciplines represented in the research to date are MTB,1218 25 26 32,^{41 60} road,1218 25 26 43,^{49 52} track,^{12 18 25 26 65} BMX^{12 18 25 26 51} and para cycling.^{2021 27},^{31 50} Cyclocross, gravel cycling, indoor cycling, trials and esports have not been represented to date within the research, despite their popularity and arguably each discipline posing a unique injury/illness risk. Thus, injury/illness epidemiology studies are urgently called for in all the cycling disciplines, but specifically in the last five (cyclocross, gravel, indoor, trial and esports).

Studies within cycling which focused on specific injury types (ie, concussion, external iliac endofibrosis) or illnesses (ie, gastrointestinal, campylobacteriosis) were excluded from consideration due to the impact they would have on the over-reporting of specific injuries when data were pooled. Additionally, studies which focused on recreational cycling tours⁶⁶,⁶⁹ were also excluded due to the non-competitive nature of events and the scope of review. Due to the high levels of heterogeneity between studies in each discipline in terms of reporting of injury and definition of injury/illness, we were unable to include all studies in each discipline within the meta-analysis. Therefore, within able-bodied disciplines, the meta-analysis only included studies from the past three Olympic games,^{12 25 26} one multi-discipline European championship study¹⁸ and one in-season prospective road cycling study on professional cyclists.⁴³ Within para cycling, the same approach was taken, including data taken from the past three Paralympic cycles on injury and illness.²⁷,³¹

Injury rates

The reported rates of injuries and definitions differed throughout studies, from the proportions of athletes injured to injuries per 100 hours, per 1000 hours, per 1000 km or 1000 athlete days. Historically, the heterogeneity in injury and illness reporting standards is attributed to the lack of a cycling-specific consensus statement.⁷⁰ However, with the publication of the IOC consensus cycling extension in 2021, the reporting of injuries should be standardised going forward.⁸ Since its publication, only three studies^{12 18 43} have been published, including cycling data; however, only one⁴³ specifically has applied the discipline-specific reporting recommendations of the IOC consensus paper.⁸ The IOC recommends the reporting of injuries within BMX, track sprint events, trials and some MTB events per 100 rides/rounds/runs. Over the past three Olympic cycles,^{12 25 26} the proportional injury prevalence in BMX is among the highest across all Olympic sports. Our meta-analysis of all cycling disciplines follows this trend, showing BMX to have the highest injury rate per 365 days at 4.59 (95% CI 4.24 to 4.94). Conversely, when compared with track cycling at 3.45 (95% CI 1.46 to 5.45), road cycling at 3.68 (95% CI 1.13 to 6.22), MTB at 3.61 (95% CI 2.05 to 5.18) and para cycling at 3.85 (95% CI 3.51 to 4.18) injuries per 365 days, the difference does not appear significant.⁵⁹ It should be acknowledged that all surveillance studies included in our meta-analysis were conducted during a competition period. As most studies are based on competition injuries, there may be an under-reporting of chronic, gradual onset injuries, compared with a predominance of acute, sudden onset injuries.

Injury locations

The location of injury across all cycling disciplines differs from those seen in field-based sports and track and field.⁷¹,⁷⁴ Within field-based sports and athletic track sports, there is a higher prevalence of lower limb injuries noted. Conversely, within cycling, the upper limb is the most injured area. Our findings within MTB and road cycling are in line with the recent systematic reviews by Braybrook et al⁹ and Buchholtz et al,¹⁰ who examined MTB cycling injuries and Rooney et al,¹¹ who examined road cycling injuries. Our review findings show that road cyclists are at the highest risk of head injuries (15% (95% CI 13.84 to 16.25)), followed by para cyclists and then track and MTB athletes. With the high prevalence of head injuries seen across able-bodied cycling disciplines, arguably, a similar trend will be seen among para cyclists. Our review shows that the prevalence of head injuries in para cycling is unknown, and the screening for such injuries has been previously discussed. Despite BMX athletes having the highest injury rate per 365 days, they present with the lowest number of head/neck/face injuries. However, the upper limb accounts for 65.21% (95% CI 54.50 to 74.65) of all injuries noted within BMX. This is likely linked with track jumps and freestyle elements and protective behaviour when falling. Overuse injuries were examined by five road cycling studies, with the lower limb being the most common injured region (48.82% (95% CI 45.38 to 52.27)).^{55 56 58 59 63}

Injury types

The injury types and patterns seen within this review are like those seen in previous systematic reviews in MTB^{9 10} and road cycling.¹¹ The most prevalent injury noted across BMX, road and MTB was skin lacerations/abrasions/lesions. Within track cycling, there was a higher prevalence of muscle contusions/haematomas. This is possibly explained by the banked nature of the track, meaning when riders crash, they slide down the bank with other riders colliding with them. There were no studies which solely focused on training-related injuries/illnesses, where one may expect the risk of lacerations and abrasions to be slightly lower. However, skin lesions may remain high as saddle sores are a common complaint among cyclists, and this injury would fall under the category of skin lesions.^{75 76}

Bone fractures were the second most common injury noted among competitive cyclists. Our findings show the highest prevalence of bone fractures is among road cyclists (17.84%–95% CI 16.63 to 19.13). Road cycling places a greater emphasis on power-to-weight ratios and, as a result, can potentially be linked with a greater proportion of athletes at risk of relative energy deficiency (RED) syndrome.⁷⁷ REDs is a multifaceted syndrome that can affect both male and female athletes, caused by RED, which impacts key physiological functions, including bone health, psychological health, immunity and menstrual function in the female athlete, which can, in turn, lead to increased fracture risk. Thus, practitioners should acknowledge the high fracture prevalence in cycling, encourage weight-bearing/resistance training (particularly during the off-season periods) and use validated screening tools referenced within the latest IOC consensus update on REDs.⁷⁷

Despite competitive road cyclists’ high prevalence of head and neck injuries, they had the lowest prevalence of concussion at 4.68% (95% CI 4.04 to 5.42). Conversely, track cycling was found to have the highest prevalence of concussion, accounting for 13.85% (95% CI 6.91 to 25.17) of all injuries. It is important to note that studies focusing solely on concussion were excluded from this review, which may have led to an under-representation of concussion prevalence across disciplines in this review. The diagnosis of concussion within cycling is an ongoing challenge, with a need for discipline-specific concussion protocols to enable the assessment of concussion postimpact and the monitoring of symptoms postincident. The proposed RoadsIde head Injury assessment (RIDE) framework, adapted from World Rugby for road cycling by Heron et al, may be a starting point for developing a screening framework for each discipline.⁷⁸ The UCI published a Harrogate concussion consensus agreement in 2021⁷⁹ along with pocket cards⁸⁰ to aid in the screening of concussions. However, this statement needs to be updated as there has been the publication of the sixth IOC Concussion in Sport consensus statement and the new Sports Concussion Assessment Tool version 6.⁸¹ Furthermore, combined with the high prevalence of neck injuries and the high prevalence of concussions noted across studies, this raises a point for several debates. Concussion among para-athletes has been a topic of recent debate, which has led to a position statement on the topic.⁸² Our review shows concussion rates among para-cyclists are unknown, despite return to sport times being shown to be over three times longer in para-cyclists.⁸³ More needs to be done to improve our understanding of the para-cyclists’ injury profile, along with screening and prevention of such head injuries. For example, baseline concussion screening, concussion education and neck strengthening exercises would collectively provide a first step in improving injury screening and preventing head and neck injuries, including concussions.^{84 85}

The most common overuse injuries identified were muscle strains and tendon complaints, which align with the high training volumes typically associated with road cycling disciplines.^{55 56 58 59 63} In the single para cycling study included, the injury profile was more balanced between the upper and lower limbs. This is likely because hand cyclists spend prolonged periods using their upper limbs to power their hand cycles, which increases their susceptibility to overuse injuries in these areas.⁵⁰

Illness rates and aetiology

Our findings of illness types agree with the narrative expectations of the expert authorship group that published the IOC Cycling Extension consensus in 2021.⁸ The consensus paper expected a high prevalence of gastrointestinal and respiratory illness, which is confirmed by the findings of our review. The expression of illness varied across the studies included in this review. However, with the recent IOC consensus on cycling and specific guidelines for reporting injuries and illnesses, future research should show less definitional inconsistency.

The main illness seen within cycling was gastrointestinal, and the main aetiology of illness was attributed to environmental, non-exercise-related factors. Cycling is a worldwide sport, with competitive racing calendars requiring riders to travel across the globe. Consequently, the changing competitive environments, food types, food quality and water quality will all potentially contribute to gastrointestinal illness risk. Our meta-analysis on illness rates among cyclists included data from the past three Olympic games, which have been held across three separate continents (Europe, London, the United Kingdom in 2012, South America, Rio de Janeiro, Brazil in 2016 and Asia—Tokyo, Japan in 2020 (1)). This highlights the influences in cultural, environmental and time zone changes encountered by cyclists that may contribute to illness risk. Furthermore, Stuart et al found within MTB that when mud on trails is greatest, illness risk significantly increases, presumably linked with participants being splashed in the face or swallowing muddy water.⁸⁶ While this would also apply to other endurance disciplines which take place outdoors, such as road cycling, it has not yet been researched. Cycling teams across all disciplines should look to implement illness prevention strategies in keeping with recent research findings.^{87 88}

The proportions of illness varied across disciplines, with respiratory illness being highest in track (31.82% (95% CI 22.29 to 42.61)), MTB (31.82% (95% CI 22.29 to 42.61)). The high prevalence of respiratory illness seen on track may be linked to the heated, drier air of the indoor velodrome, which is favoured for faster environmental conditions. However, it can be more aggressive on the airways and lead to ‘track cough’ or asthma flare-ups (exercise-induced bronchospasm).⁸⁹ An important consideration was the higher rates of cardiovascular illness seen in road cycling. Mandatory cardiac screening is part of all professional cycling teams’ preseason medical and a UCI requirement. Our sample included large-scale mixed-participation races such as the Cape Town Cycle Tour,⁴⁹ where high rates of cardiovascular illness were documented. This highlights the importance of cardiovascular screening for all competitive athletes and not just those at a professional level.

As our meta-analysis included only professional/elite-level athletes, this group has been seen to have a high prevalence of upper respiratory symptoms, particularly during competition.⁹⁰ A notable finding of this study meta-analysis was the illness rate among para cyclists of 3.89 (95% CI 3.42 to 4.36), which is higher than all other competitive disciplines represented. Our review shows that within para cyclists, respiratory illness is as high as 70% (95% CI 60.09 to 80.52). High illness rates have been seen within prospective para athletics, swimmers and powerlifting studies; however, only one study that examined cyclists agrees with our findings.^{50 91} The para athlete may have greater illness susceptibility linked with their disability/disorder⁹² or a visually impaired rider may have an increased reliance on tactile feedback and, thus, a greater risk of gastrointestinal/respiratory illness.⁷⁷

Within para cycling, the profile of illness is poorly understood despite the higher rates of illness seen compared with able-bodied disciplines.⁵⁰ Thus, emphasises the requirement for specialist knowledge in managing illness profiles unique to athletes with disabilities. Additionally, there was a high prevalence of dermatological illness seen within track cycling. Crashes on the track can lead to widespread skin abrasions (‘track burn’) and wood splinters. The need to ensure thorough debridement of the wound post-crash and removal of all splinters is fundamental to reducing the risk of dermatological infection. As cyclists undertaking high training volumes may lower their ability to fight infection and consequently be more susceptible to wound infection, removal of all splinters is fundamental to reducing the risk of dermatological infection.⁹³

Unlike injuries, illnesses in sports are not defined by whether they occur in competition or training. This challenges the completeness of the illness profile presented within this review, as all the studies that have examined illness have done so within a competition or included competition and a narrow precompetition training window. There is often a delayed onset of symptoms due to the incubation time of bacterial and viral infections. Therefore, the true prevalence of illness may be under-reported. Further prospective studies within competition and training on cycling are warranted to capture a true understanding of illness prevalence within cyclists.

Clinical recommendations

Clinicians should be aware of the high prevalence of upper limb injuries and the prevalence of skin (lacerations/abrasions/lesions) injuries in cycling. Prevention measures such as the use of Dyneema fibre around key abrasion sites should be discussed with clothing manufacturers to help protect the skin from abrasion-type injuries when crashing.^{94 95} Similarly, educational content around skin lesions (saddle sore) prevalence and prevention may be valuable in preventing such injuries.^{76 96} With the high prevalence of head injuries noted and the low incidence of concussions diagnosed across different disciplines, further discussion is warranted around the detection of concussions in racing and the monitoring window to accommodate the often-transient presentation of such symptoms that will determine the removal of the rider from competition. Physiotherapists and sports doctors should consider adding strength and conditioning and include neck strengthening to help reduce neck-related injuries and improve bone mineral density, thus decreasing fracture risk.⁸⁴ Within para cycling, specific concussion protocols are warranted. A particular emphasis should be placed on research around para-athlete health injuries, injury and illness surveillance. Finally, continued education reinforcing hand hygiene, particularly in certain high-risk groups such as visually impaired athletes with the increased reliance on tactile feedback, may aid in the reduction of respiratory and gastrointestinal illnesses.

Limitations

This systematic review and meta-analysis should be considered in light of its limitations. First, the focus on competitive cycling studies means that recreational cyclists, who do not participate in competitions, may not be adequately represented in the findings. Additionally, a limitation arises from the variability in the reporting definitions and methodologies for injury and illness across studies. Reporting biases, particularly the under-reporting of injuries not assessed by medical professionals, alongside gaps in data for specific disciplines, such as para cycling, further limit the generalisability of the results. These limitations have been previously acknowledged in the IOC extension for the cycling consensus paper.⁸ Future studies exploring cycling injury and illness epidemiology and the epidemiology of specific pathologies in cycling should consider the IOC consensus statement extension in their methodology. Combining cycling disciplines within broad areas such as MTB disciplines, enduro, downhill and cross country is not without limitations. Each subdiscipline has unique demands, injury patterns and risk factors, which may be obscured when grouped, potentially limiting the specificity and applicability of the findings to individual disciplines. Including injury/illness classification systems, such as the Orchard Sports Injury Classification System, within injury and illness surveillance research in cycling is an important consideration to get an accurate injury diagnosis, understanding of the specific injuries seen within cycling.⁹⁷ Finally, the meta-analyses conducted in this study are not without limitation, particularly in terms of the number of studies included, and that it only includes professional cyclists. However, this preliminary analysis serves as a foundation for future research, to expand the number of studies in each discipline and improve consistency across studies.

Conclusion

This is the first systematic review to examine injury and illness across all competitive cycling disciplines. The main finding was that only 5 out of 10 cycling disciplines are represented within the literature, with them being MTB, road, track, BMX and para cycling. Across all able-bodied disciplines, the most injured region was the upper limb, followed by the lower limb. Within para cyclists, the lower limb was the most injured region, closely followed by the upper limb. The most common injury found was skin lacerations/abrasions/lesions, followed by bone fractures. Furthermore, the most common illness within able-bodied disciplines was gastrointestinal, followed by respiratory illnesses. Para cyclists expressed the highest risk of illness when compared with all other disciplines; however, there is a lack of evidence on the type of injuries and illnesses seen within para cyclists. These areas should be targeted for prevention within cycling teams, including developing clothing to minimise the risk of abrasions, introducing general weight-bearing muscle-strengthening exercises into routine cycling training and illness prevention measures. This review confirms the paucity of high-quality, homogeneous injury and illness epidemiology research in cycling. The findings of this review support the UCI Agenda 2030 objectives, calling for more high-quality injury and illness epidemiology research within cycling.² Future research should focus on female cyclists and para athletes, along with focusing on the lesser-known disciplines highlighted within this review.

Data availability statement

Data are available upon reasonable request.