# Reliability of the Dynamic Foot Index for Observational Assessment of Foot Motion During Gait

## TL;DR

The Dynamic Foot Index (DFI) is a low-cost observational tool for assessing foot motion during walking, showing moderate to substantial reliability when used by the same evaluator.

## Contribution

The study introduces the DFI as a clinically feasible observational tool with validated intra- and inter-rater reliability in healthy adults.

## Key findings

- Intra-rater reliability of the DFI ranged from moderate to substantial (weighted Kappa = 0.54–0.80).
- Inter-rater reliability varied from fair to substantial (weighted Kappa = 0.28–0.68), with lower agreement during midstance.
- Absolute agreement ranged from 65% to 92% across different gait subphases and limbs.

## Abstract

Reliable clinical tools for observational assessment of dynamic foot motion during gait are limited. Although three‐dimensional motion analysis is considered the reference standard, its cost and complexity restrict routine clinical use. The Dynamic Foot Index (DFI) is a simple, low‐cost observational tool designed to assess foot motion during gait.

This methodological study investigated the intra‐ and inter‐rater reliability of the Dynamic Foot Index (DFI) in healthy adults. Forty‐nine participants (98 feet) walked barefoot on a treadmill at a self‐selected speed while being recorded from a posterior view using a smartphone camera. Two experienced physical therapists independently analyzed video recordings of one complete stride per limb, classifying foot motion across three stance‐phase subphases (loading response, midstance, and push‐off) using a three‐point ordinal scale. Intra‐rater reliability was assessed after a 30‐day interval. Weighted Kappa coefficients with 95% confidence intervals and absolute percentage agreement were calculated.

Intra‐rater reliability ranged from moderate to substantial across gait subphases (weighted Kappa = 0.54–0.80), with higher agreement observed during loading response and push‐off. Inter‐rater reliability ranged from fair to substantial (weighted Kappa = 0.28–0.68), with lower values during midstance. Absolute agreement ranged from 65% to 92%, depending on the subphase and limb assessed.

The Dynamic Foot Index (DFI) demonstrated moderate to substantial intra‐rater reliability and variable inter‐rater reliability across stance‐phase subphases in healthy adults. These findings support the use of the Dynamic Foot Index (DFI) as a clinically feasible observational tool for longitudinal assessment of foot motion during gait, particularly when applied by the same evaluator.