What are the Benefits and Risks Associated with Changing Foot Strike Pattern During Running? A Systematic Review and Meta-analysis of Injury, Running Economy, and Biomechanics

Sports Med. 2020 May;50(5):885-917. doi: 10.1007/s40279-019-01238-y.


Background: Running participation continues to increase. The ideal strike pattern during running is a controversial topic. Many coaches and therapists promote non-rearfoot strike (NRFS) running with a belief that it can treat and prevent injury, and improve running economy.

Objective: The aims of this review were to synthesise the evidence comparing NRFS with rearfoot strike (RFS) running patterns in relation to injury and running economy (primary aim), and biomechanics (secondary aim).

Design: Systematic review and meta-analysis. Consideration was given to within participant, between participant, retrospective, and prospective study designs.

Data sources: MEDLINE, EMBASE, CINAHL, and SPORTDiscus.

Results: Fifty-three studies were included. Limited evidence indicated that NRFS running is retrospectively associated with lower reported rates of mild (standard mean difference (SMD), 95% CI 3.25, 2.37-4.12), moderate (3.65, 2.71-4.59) and severe (0.93, 0.32-1.55) repetitive stress injury. Studies prospectively comparing injury risk between strike patterns are lacking. Limited evidence indicated that running economy did not differ between habitual RFS and habitual NRFS runners at slow (10.8-11.0 km/h), moderate (12.6-13.5 km/h), and fast (14.0-15.0 km/h) speeds, and was reduced in the immediate term when an NRFS-running pattern was imposed on habitual RFS runners at slow (10.8 km/h; SMD = - 1.67, - 2.82 to - 0.52) and moderate (12.6 km/h; - 1.26, - 2.42 to - 0.10) speeds. Key biomechanical findings, consistently including both comparison between habitual strike patterns and following immediate transition from RFS to NRFS running, indicated that NRFS running was associated with lower average and peak vertical loading rate (limited-moderate evidence; SMDs = 0.72-2.15); lower knee flexion range of motion (moderate-strong evidence; SMDs = 0.76-0.88); reduced patellofemoral joint stress (limited evidence; SMDs = 0.63-0.68); and greater peak internal ankle plantar flexor moment (limited evidence; SMDs = 0.73-1.33).

Conclusion: The relationship between strike pattern and injury risk could not be determined, as current evidence is limited to retrospective findings. Considering the lack of evidence to support any improvements in running economy, combined with the associated shift in loading profile (i.e., greater ankle and plantarflexor loading) found in this review, changing strike pattern cannot be recommended for an uninjured RFS runner.

Prospero registration: CRD42015024523.

Publication types

  • Meta-Analysis
  • Systematic Review

MeSH terms

  • Biomechanical Phenomena
  • Gait*
  • Humans
  • Risk Assessment
  • Running*