REF-1313 · Buist et al., 2010 · The American Journal of Sports Medicine · Paper · PubMed

I'm a fit cyclist; do I need to start running like a beginner?

Probably: male novices from sports without axial load had about twice the injury hazard (HR 2.05, 1.03-4.11; comparator not stated). One cohort, men only, not cyclists. B.

Main caveat: Read from the abstracts. Definitions (non-axial sport, previous injury, running experience), units, the full models and funding were not available.
limited evidenceMeasured on non-athletes and recreational people

Buist and colleagues looked for predictors of running injury in 532 novice runners (226 men, 306 women) preparing for a recreational 4-mile event in a systematic training programme — the participants of the GRONORUN trial, which had randomised them to a graded 13-week or a standard 8-week programme (REF-925; injuries 20.8% vs 20.3%).

After a baseline questionnaire and an orthopaedic examination, they were followed for 13 weeks; an injury was self-reported running-related pain of the lower extremity or back that restricted running for at least a week. Twenty-one percent had at least one injury.

In men, BMI (HR 1.15, 95% CI 1.05-1.26), an injury in the past year (HR 2.7, 1.36-5.55) and previous participation in sports without axial load (HR 2.05, 1.03-4.11) predicted injury; in women, only navicular drop remained (HR 0.85, 0.75-0.97). Type A behaviour and hip and ankle range of motion did not affect risk.

The companion BJSM paper followed 629 novice and recreational runners for 8 weeks before the same kind of event with a looser definition (restriction of running for at least a day): 25.9% were injured, 30.1 injuries per 1000 h; men were more prone than women (HR 1.4, 1.0-2.0), and no previous running experience was the most important risk factor in men (HR 2.6, 1.2-5.5) and women (HR 2.1, 1.2-3.7).

Our reading: what a novice has done before matters — no running at all, or in men only non-impact sport, roughly doubled the hazard — which gives the base's "a fit cyclist who starts running is a beginner below the hips" its first cited support, from a cohort of men in mixed non-axial sports.

Protocol

AJSM cohort: n = 532 novice runners (226 men, 306 women) enrolled in the GRONORUN trial's systematic training programmes for a 4-mile (6.4-km; the abstract says 6.7-km) event; baseline questionnaire and orthopaedic examination (including navicular drop and hip and ankle range of motion); followed during the 13-week training period; injury = self-reported running-related musculoskeletal pain of the lower extremity or back causing restriction of running for at least 1 week; multivariable Cox regression separately for men and women; the authors' "Level of evidence, 2".
BJSM cohort (companion): 629 novice and recreational runners training 8 weeks for a 4-mile recreational event; potential risk factors measured prospectively; injury = musculoskeletal pain of the lower limb or back restricting running for at least 1 day; multivariable Cox regression; incidence per 1000 h. Whether its runners overlap with GRONORUN is not stated in the abstract.
Definitions not in the abstracts: which sports counted as "without axial load" and what they were compared with (no sport, or sport with axial load; the same group's GRONORUN 2 design paper, REF-1307, uses three categories, non-axial meaning sport "like swimming and cycling"), how previous injury and running experience were asked, the unit of the BMI and navicular-drop hazard ratios.
Funding and interests: not given in the abstracts (full texts not read); PubMed indexes the AJSM paper as "Research Support, Non-U.S. Gov't". The same group's GRONORUN 2 trial was funded by ZonMw (REF-1307, design paper); that is not evidence for these papers.

Key Results

PredictorHazard ratio (95% CI)Cohort
Men: BMI (per unit, our reading)1.15 (1.05-1.26)AJSM, 226 men
Men: injury in the past year2.7 (1.36-5.55)AJSM
Men: previous sport without axial load2.05 (1.03-4.11)AJSM
Women: navicular drop0.85 (0.75-0.97)AJSM, 306 women
Type A behaviour; hip and ankle range of motionno effect on riskAJSM
Men vs women1.4 (1.0-2.0)BJSM, 629 runners
No previous running experience, men2.6 (1.2-5.5)BJSM
No previous running experience, women2.1 (1.2-3.7)BJSM
Incidence: AJSM 21% with at least one injury in 13 weeks (restriction ≥ 1 week); BJSM 25.9% in 8 weeks (restriction ≥ 1 day), 30.1 (95% CI 25.4-34.7) injuries per 1000 h of running.
Navicular drop's direction: a hazard ratio below 1 means more navicular drop went with less injury in women — not the direction a "flat foot = risk" view would predict; the unit is not in the abstract.
Not an independent sample: the AJSM cohort is the REF-925 trial population; its injuries are the same events as the trial's 20.8% and 20.3%.

Mechanisms

Measured: associations of baseline characteristics with time to first injury.
Authors' interpretation: male and female novice runners have different risk profiles; higher BMI, previous injury and previous sport without axial loading are important predictors in men; novices may benefit most from preventive interventions.
Inference: sports without axial loading (swimming and cycling are the obvious examples; the abstract names none) build the aerobic engine without loading bone and tendon in impact, so the legs start running as a novice's whatever the fitness; this is consistent with the bone time course under new impact loading (REF-1315) and with low bone density in road cyclists (REF-896), but the abstract offers no mechanism.
Inference: "no running experience" and "non-axial background" overlap; the abstracts do not show whether they were modelled together.

Practical Application

I'm a fit cyclist; do I need to start running like a beginner? Probably: male novices from sports without axial load had about twice the injury hazard (HR 2.05, 1.03-4.11; comparator not stated). One cohort, men only, not cyclists. B.
Treat no running history as a risk marker (B, from the abstract; aerobic fitness was not measured, so it was not compared): in 629 novice and recreational runners, no previous running experience went with HR 2.6 in men and 2.1 in women; in 930 novices REF-926 found no link with running experience (P = .30). Not from this paper; the base's own practice: beginner run-walk build and small single-run steps (SYNTH-75, REF-1102).
A returner with a running history is in a better position than a true novice (inference from the same contrast; recent running experience also went with about half the odds of quitting in REF-1311, an association); a slower ramp for a returner is still practice (SYNTH-36).
Men with a higher BMI or a recent injury: the slowest ramp (B for the association, men only; the ramp is practice); in women this cohort found no useful predictor.
Do not count this cohort as new trial evidence: it is the REF-925 population; the trial answers "does a graded programme help" (no), this paper "who gets hurt".

Limitations

Read from the abstracts. Definitions (non-axial sport, previous injury, running experience), units, the full models and funding were not available.
Same participants as an existing entry. The AJSM cohort is REF-925's trial population, pooled across two programmes of different length (13 and 8 weeks).
Small strata. A 21% injury rate split by sex; the non-axial hazard ratio's interval starts at 1.03.
Self-reported injuries with different definitions in the two papers (at least a week vs at least a day of restriction).
Short events (4 miles) and short follow-up (8-13 weeks); the findings in men may not hold in women and vice versa.
Funding and interests: not in the abstracts.
How this entry was checked and changed
Why this level

(from the abstracts) — large prospective cohorts (rubric B+), held at B because both papers were read from the abstract only (the full texts are not open; definitions, units and comparators are unknown), the AJSM cohort is the same 532 runners as REF-925 and not an independent sample, the sex-specific models split a 21% injury rate into small strata, and the non-axial-sport finding is one of several predictors tested, in men only