Which overuse injuries do amateur road cyclists get?
The knee leads: of 1,879 injuries reported by Cape Town Cycle Tour entrants, knee 32%, shoulder 11%, lower back 8%; 40% interfered with riding, 44% lasted over a year.
Main caveat: Self-reported diagnoses chosen from a drop-down list on a pre-race medical form; no clinical confirmation of site, tissue or diagnosis.
du Toit et al. (University of Pretoria, the IOC Research Centre in Pretoria and the South African Medical Research Council) described the overuse injuries reported by entrants to the 109 km Cape Town Cycle Tour from 2016 to 2020.
Entrants completed an online medical screening questionnaire at registration (compulsory in 2016, voluntary afterwards); 62,758 entries consented to research use, a crude response of 35.9% of an estimated 175,000 entries.
A gradual-onset injury counted if it interfered with cycling or needed treatment, medication or medical advice; 1,879 were reported, a period prevalence of 2.95%.
Of these injuries, 47.4% were in the lower limb, 20.1% the upper limb, 10.0% the hip, groin or pelvis and 7.8% the lower back; by site the knee led (32.1%), then the shoulder (10.6%), lower back (7.8%) and hip or buttock muscles (5.2%).
The commonest diagnoses were anterior knee pain or patellofemoral pain (17.2%), shoulder pain (10.6%), ITB friction syndrome (10.3%) and low back pain (7.7%). Two in five injuries (40.4%) were severe enough to interfere with training or competing, and 43.9% had lasted more than 12 months — most often in the lower back (58.8%), which was also most often severe (52.6%).
Rest (45.9%), physiotherapy (43.0%), stretching (33.2%) and strength exercises (33.1%) were the commonest treatments; a bike set-up change was listed for 15.1% and surgery for 19.6%.
Our reading: the base's first amateur-cyclist picture, and it confirms the knee as the first site to plan for; it puts the shoulder second, a site the base has no row for, and shows that once an overuse problem is bad enough to report it often lasts. It cannot say what caused any of them.
Protocol
Design: descriptive cross-sectional study within the SAFER research programme; ethics: University of Pretoria Faculty of Health Sciences (REC 431/2015 and 749/2019).
Event and sample: 109 km Cape Town Cycle Tour, 2016-2020; entries capped at about 35,000 a year (estimated 175,000 over 5 years). n = 62758 entries consented to research use (crude response 35.9%: 2016 62.5%, 2017 24.4%, 2018 27.2%, 2019 25.0%, 2020 40.0%). The questionnaire was compulsory in 2016 and voluntary in 2017-2020. Individuals entering in more than one year could not be identified, and a rider could report up to 2 injuries each year.
Who (Table 1, weighted): 49,484 entries from men (78.9%) and 13,274 from women (21.2%); over 50 years: 29.7% of all entries men and 5.4% women (35.1% together).
Injury question: a chronic (no accident) cycling injury of muscles, tendons, bones, ligaments or joints, quoted as asked "in your cycling career"; results are reported as injuries in the previous 12 months. Counted if severe enough to interfere with cycling or require treatment, medication or medical advice; then drop-down questions on site, tissue, severity grade, duration, a list of common diagnoses and treatments.
Severity grades (from a running-injury study): I symptoms only after exercise; II during exercise, not interfering; III during exercise, may interfere with training or competition; IV too painful to train or compete. Grades III-IV = more severe.
Duration categories: 0-3, 4-12 and more than 12 months (the longest option).
Treatments: multiple choice, asked only in 2017-2020 (a subgroup of 40,638 entries with 1,348 injuries; 3,149 treatments).
Analysis: frequencies and percentages of all injuries, weighted by sex and age so that 2017-2020 match the 2016 profile; no statistical modelling (SAS 9.4).
Funding and interests: IOC Research Center of South Africa (partial funding) and the South African Medical Research Council (partial funding, statistical analysis); the authors report no potential conflict of interest. Two authors are affiliated with the SAMRC Biostatistics Unit; the corresponding author directs the Sport, Exercise Medicine and Lifestyle Institute in Pretoria. No industry ties are printed.
Key Results
Sites and diagnoses (Tables 2 and 3, weighted; % of all 1,879 injuries):
| Region (95% CI) | Leading sites | Leading diagnoses |
|---|---|---|
| Lower limb 47.4% (45.2-49.7) | Knee 32.1%, Achilles 3.9%, foot 2.7%, hamstring 2.7%, calf 2.6% | Anterior knee pain / patellofemoral pain 17.2%; ITB friction syndrome 10.3%; Achilles tendon injury 3.7% |
| Upper limb 20.1% (18.3-21.9) | Shoulder 10.6%, finger 4.4%, wrist 2.6%, elbow 2.1% | Shoulder pain 10.6%; numbness in hand or fingers 3.7% |
| Hip, groin, pelvis 10.0% (8.7-11.3) | Hip or buttock muscles 5.2%, hip 4.0%, groin 0.8% | Hip, glute or buttock muscle injury 4.4%; hip joint pain 4.0%; groin or genital numbness 0.8%; saddle sores 0.8% |
| Lower back 7.8% (6.6-9.0) | Lower back 7.8% | Low back pain 7.7% |
| Head, neck, face 3.2% (2.4-4.0) | Neck 2.8% | Neck pain 2.8% |
| Other 10.4%; trunk or chest 1.1% | — | Other 20.1% |
Severity and duration by region (Table 5 and text; % of that region's injuries):
| Region | More severe (grade III-IV) | Lasting over 12 months |
|---|---|---|
| Lower back | 52.6% | 58.8% |
| Lower limb | 44.1% | 42.2% |
| Hip, groin, pelvis | 38.3% | 52.9% |
| Upper limb | 28.1% | 39.8% |
| All injuries | 40.4% (743 of 1,841 graded: grade III 30.6%, grade IV 9.8%) | 43.9% (825); 4-12 months 34.2%; 0-3 months 21.9% |
Tissue (Table 4): soft tissue 57.0% (tendon 18.9%, muscle 18.5%, ligament 12.7%, nerve 6.9%); bone 18.8% (examples given: bruise or stress fracture); joint 13.3%.
Treatments (Table 7, 1,348 injuries in 2017-2020): rest 45.9%, physiotherapy 43.0%, stretches 33.2%, strength exercises 33.1%, tablets 24.1%, surgery 19.6%, bike set-up 15.1%, cortisone injection 7.0%, new cycling shoes 1.7%.
Numbers that do not add up: the abstract describes percentages as "% entrants", but every table gives the share of reported injuries; the abstract says 43.9% "of cyclists with a GOI" reported symptoms over 12 months, but the figure is a share of injuries (riders could report two, and repeat entrants could not be linked); the abstract gives 40.3% as severe, Table 6 and the Discussion 40.4%; the Results call the over-50s "the majority" at 35.1% and the Discussion says over 60% were over 50, while Table 1 gives 35.1%; 1,879 of 62,758 is 2.99% (our arithmetic) against a printed 2.95% (weighting may explain it); and the quoted screening question asks about the cycling career while results are framed as the previous 12 months.
Mechanisms
Measured: none; the study counts self-reported injuries by site, tissue, severity, duration and treatment.
Authors' interpretation: knee injuries in cyclists have many proposed causes (malalignment, repetitive force, reduced range of motion, muscle imbalance), which this descriptive study did not test; long symptom duration may reflect a definition that leaned to severe injuries, misdiagnosis, untreated fluctuating symptoms, ineffective treatment, sensitised pain, or an older sample.
Inference: the question asked for chronic injuries "no accident", yet bone was reported for 18.8% of injuries and surgery for 19.6%; some reports probably reflect old crash injuries or other conditions (for example a shoulder after a fall) rather than overuse of the bike — so shoulder and bone figures in particular may overstate cycling overuse.
Inference: the 2.95% prevalence is far below surveys that ask about any overuse symptom, because this form counted only injuries that interfered with riding or needed treatment, on a medical screening form at race registration; it is a floor for bothersome injuries, not a rate of aches.
Practical Application
Which overuse injuries do amateur road cyclists get? The knee leads: of 1,879 injuries reported by Cape Town Cycle Tour entrants, knee 32%, shoulder 11%, lower back 8%; 40% interfered with riding, 44% lasted over a year.
Plan for the knee first (B+, descriptive): anterior knee pain (17.2%) and ITB friction syndrome (10.3%) were over a quarter of all reported injuries. Not from this paper; the base's own practice for knee pain on the bike: find the recent load change, then check saddle height measured while pedalling (REF-1261), one change at a time, re-check after about two weeks (SYNTH-73).
Do not wait it out: 43.9% of reported injuries had lasted more than a year, lower-back ones 58.8%, and lower-back injuries were the most often severe (52.6%). Not from this paper; the base's own practice: pain not improving within about two weeks of sensible changes, or recurring, goes to a clinician (SYNTH-73).
Shoulder and hands exist too: shoulder pain was 10.6% and hand or finger numbness 3.7% of reported injuries; the base has no shoulder entry for cyclists (REF-618 is overhead athletes) and no shoulder row in SYNTH-73, and hand numbness is covered by REF-890 and REF-1040 (inference: part of the shoulder figure may be old crash injuries).
Treatments are a description, not a verdict: rest, physiotherapy, stretching and strengthening were the commonest, and a bike set-up change was used for 15.1% of injuries; no outcome was measured.
Limitations
Self-reported diagnoses chosen from a drop-down list on a pre-race medical form; no clinical confirmation of site, tissue or diagnosis; recall bias over 12 months.
Selection and representativeness. Crude response 35.9% (24.4-62.5% by year); compulsory only in 2016; 78.9% of entries from men and 35.1% over 50 (the authors call for studies in younger riders and in women); one South African mass event.
Entries, not people. The same rider entering several years is counted again; injuries, not injured riders, are the unit, and up to 2 per year could be reported.
Definition. Only injuries that interfered with cycling or needed treatment counted, so mild overuse is missed; duration capped at "more than 12 months"; the screening question's wording (cycling career) and the reported window (previous 12 months) differ.
Reporting inconsistencies. "% entrants" in the abstract versus % of injuries in the tables; 40.3% versus 40.4%; "over 60%" over 50 in the Discussion versus 35.1% in Table 1; 2.95% versus 1,879/62,758 = 2.99% (our arithmetic); "almost 46%" of lower-limb injuries more severe in the Discussion versus 44.1% in the Results, and "almost 35%" treated with medication versus 33%.
Version read. The author manuscript in the University of Pretoria repository (OpenAlex labels it the submitted version); its abstract matches the PubMed abstract, but the typeset article may differ in detail. The tables are images in that PDF and are not in the saved text.
Interests. Funded in part by the IOC Research Center of South Africa and the South African Medical Research Council; no conflicts reported; no industry ties printed.
How this entry was checked and changed
Why this level
large cross-sectional measurement study (rubric: ≥100; n = 62758 consented entries, 1879 injuries), read from the author manuscript in the University of Pretoria repository (text; the seven tables are images and were read from the page images). Descriptive only: injuries and diagnoses are self-reported, the crude response rate was 35.9%, repeat entrants could not be linked across years, and the reporting has inconsistencies (see Limitations)
Full text (checked 2026-10-27)
62,758 consenting race entrants completed the online pre-race medical screening questionnaire during the 2016–2020 Cape Town Cycle Tour, and 1879 reported GOIs in the previous 12 months (full text)
Period prevalence 2.95% (95% CI printed as 2.8.-3.09); crude response rate 35.9%, by year 2016 (62.5%), 2017 (24.4%), 2018 (27.2%), 2019 (25.0%) and 2020 (40.0%) (full text)
The questionnaire was compulsory in 2016 and voluntary in 2017–2020; entrants were capped at 35000 a year (estimated 175 000 over the 5 years) (full text)
Main regions as % of all GOIs: lower limb (47.4%), upper limb (20.1%), hip/groin/pelvis (10.0%), lower back (7.8%); sites: knee (32.1%), shoulder (10.6%), hip/buttock muscles (5.2%) (full text)
Commonest specific GOIs: anterior knee pain/patellofemoral pain (17.2%), shoulder pain (10.6%), ITB friction syndrome (10.3%), lower back pain (7.7%) (full text)
Symptom duration: 44% (825) more than 12 months, 34.2% (642) 4–12 months, 21.9% (411) 0–3 months; over 12 months by region 58.8% lower back, 52.9% hip/groin/pelvis, 42.2% lower limb, 39.8% upper limb (full text)
Severe enough to interfere with training or competing, by region: 52.6% lower back, 44.1% lower limb, 38.3% hip/groin/pelvis, 28.1% upper limb (full text)
Treatments for 1348 GOIs in a subgroup of 40,638 cyclists (2017–2020): rest (45.9%), physiotherapy (43.0%), stretches (33.2%), strength exercises (33.1%); tablets or injections 33%, almost 20% surgery (full text)
Majority male (78.9%) and over 50 years of age (35.1%) in the Results; the Discussion says >60% of participants were >50 years old (full text)
Funding: IOC Research Center of South Africa (partial funding); South African Medical Research Council (partial funding, statistical analysis); no potential conflict of interest reported (full text)