How much running is too much? Identifying high-risk running sessions in a 5200-person cohort study

Key points

1. 01A single session more than 10% longer than your longest run in the past 30 days raised overuse injury risk by 64% (small spike) to 128% (large spike, i.e., more than doubling distance).
2. 02The acute:chronic workload ratio (ACWR), widely used in wearables and research, showed an inverse or null association with injury - higher ACWR was linked to lower injury rates, the opposite of its intended warning signal.
3. 03The week-to-week ratio showed no significant relationship with overuse injury risk at any spike level.
4. 0435% of the 5205 runners sustained a running-related injury over the study period, with 72% of those classified as overuse injuries.
5. 05The single-session paradigm is proposed as a more precise tool for injury prevention than weekly load monitoring approaches.

How it was conducted

Design

Prospective cohort study (Garmin-RUNSAFE Running Health Study), 18-month follow-up

Participants

5205 adult runners, mean age 45.8 years (SD 10.4), 78% male, recruited from 87+ countries

Exposure

Three time-varying training load metrics: single-session distance relative to longest run in prior 30 days; acute:chronic workload ratio (ACWR); week-to-week ratio - each categorised into regression/up to 10%, small spike >10-30%, moderate spike >30-100%, large spike >100%

Primary outcome

First self-reported overuse running-related injury (non-traumatic onset), assessed via weekly questionnaire

Analysis

Multistate Cox regression model with adjusted hazard rate ratios (HRR); adjusted for age, BMI, sex, previous problems, and running experience

Sessions analysed

588 071 total running sessions across all participants

What they found

• Small spike (>10% to 30% above longest run in past 30 days): adjusted HRR 1.64 (95% CI 1.31 to 2.05, p<0.01) for overuse injury.
• Moderate spike (>30% to 100%): adjusted HRR 1.52 (95% CI 1.16 to 2.00, p<0.01) for overuse injury.
• Large spike (>100%, i.e., more than doubling distance): adjusted HRR 2.28 (95% CI 1.50 to 3.48, p<0.01) for overuse injury.
• ACWR large spike (>100%): adjusted HRR 0.75 (95% CI 0.59 to 0.96, p=0.02) - a significant decrease in injury rate, opposite direction to the single-session finding.
• Week-to-week large spike (>100%): adjusted HRR 0.91 (95% CI 0.78 to 1.07, p=0.29) - no significant association.
• By 200 sessions, 30.5% (95% CI 29.0% to 32.0%) had reported an overuse injury and 12.0% (95% CI 11.0% to 13.1%) a traumatic injury.
• 1820 of 5205 runners (35.0%) reported a running-related injury; 1311 (72.0%) were overuse and 509 (28.0%) were traumatic.
• Progression between 1% and 10% (vs. regression or up to 1%): non-significant rate increase of 19% (95% CI -10% to 57%, p=0.22).

Limitations

• Injury outcomes and mechanism (overuse vs. traumatic) were self-reported, introducing potential information bias and possible misclassification between injury types.
• Running distance was used as the sole load metric; intensity, elevation, and other load components were not captured, limiting precision of the exposure measure.
• Subgroup-specific guidance (by age, sex, BMI, or experience level) was not possible due to insufficient injury counts per subgroup, risking sparse data bias.
• The observational design limits causal inference; randomised trials or target trial emulation are needed to confirm that staying below the 10% threshold prevents injuries.

Why it matters

For patients

If you are a recreational runner, keep each new session within 10% of your longest run in the past month to meaningfully cut your overuse injury risk.

For clinicians

Single-session distance relative to a 30-day maximum is a more useful injury-risk signal than the ACWR or week-to-week ratio, and those weekly metrics should not be used in isolation to advise runners on safe progression.

For readers

This is the largest prospective running-injury cohort to date and challenges the validity of the ACWR framework for running, but the observational design means the 10% rule has not yet been tested in a prevention trial.