Correlation of foot-strike pattern and performance in a mid-size city marathon

Abstract

Introduction: A small number of studies have provided “inrace” foot-strike prevalence of runners in midto long-distance races. The recent studies each classified the foot-strike of less than 300 runners in half-marathon and marathon races. The runner populations varied from all elite runners to mostly “recreational” runners. The prevalence of heel-strike ranged from 74.9% to 93.0%. Correlation with performance, however, has been impossible due to the small sample sizes. Using a seven-fold larger runner population, this study aims to more accurately define the prevalence of heel-strike in a mid-size city marathon and to determine if a correlation between foot-strike and performance exists. Patients/materials and methods: 2112 runners in the 2011 Milwaukee Lakefront Marathon were filmed at the 5.02-mile mark using a Casio EX-ZR100,with a frame rate of 240 frames/s. The camerawassecurelymountedataheightof24 in. andadistanceof12 in. from the side of the road. The camerawas angled at 30degrees from a direct lateral, or perpendicular angle. Foot-strike classification was determined by a 3-member panel utilizing frame-by-frame analysis. Classifications included fore-foot strike, mid-foot strike, heel strike, and split strike (asymmetric between consecutive footstrikes). Results: 1991 runners were classified by foot-strike pattern: 11 fore-foot (0.55%), 101 mid-foot (5.07%), 1865 heel (93.67%), and 14 split (0.70%). The heel-strike prevalence of 93.67%wasmore prominent than in prior studies of both recreational and elite runners. A correlation was demonstrated between foot-strike and performance (as measured by position in race at the 5.02-mile mark), with a p-value <0.0001. Fore-foot strikers were followed by splitstrikers, then mid-foot strikers, then heel-strikers. There was no significant correlation demonstrated between foot-strike and gender (p-value =0.0727). Discussion and conclusions: This study is the largest “in-race” observational study of marathon runners to date. It unequivocally demonstrates: (a) heel-strike prevalence in a predominantly subelite population of runners is significantly greater than previously documented in more elite running populations and (b) foot-strike is correlated with performance (as measured by position in race). Shoe wear (minimalist versus traditional) was not correlated with position in race in non-heel strikers, perhaps placingmore emphasis on foot-strike than shoewear for performance. Further research is necessary to elucidate whether non-heel strike running form produces a faster runner or a faster runner produces a non-heel strike running form.