Full body gait analysis was used to determine if differences exist in kinematic, kinetic, and temporal-spatial data with and without shoes in able-bodied children. The greatest difference noted between conditions was an increase in stride length with shoes. Minimal changes were seen in kinematics and kinetics with the addition of shoes. Due to the very tight standard deviations of the data, these minimal changes in the magnitude of the curves resulted in statistically significant differences, yet these changes do not appear to be clinically significant. It is believed that this study establishes that barefoot gait analysis is sufficient for most clinical studies, and an additional assessment undertaken while wearing shoes is not necessary.
http://linkinghub.elsevier.com/retrieve/pii/S096663629900005...

Objective: To evaluate the gait patterns of lower limb amputee patients walking with and without shoes and to identify differences in barefoot gait patterns when using different prosthetic feet.
http://journals.lww.com/ajpmr/Abstract/2003/02000/Gait_Patte...

Stress distribution of the foot during mid-stance to push-off in barefoot gait: a 3-D finite element analysis
Objective. To quantify stress distribution of the foot during mid-stance to push-off in barefoot gait using 3-D finite element analysis.

Design. To simulate the foot structure and facilitate later consideration of footwear. Finite element model was generated and loading condition simulating barefoot gait during mid-stance to push-off was used to quantify the stress distributions.
http://www.journals.elsevierhealth.com/periodicals/jclb/arti...

RESEARCH QUESTION: Although gait patterns of children are well described in gait analysis studies for barefoot gait, loading patterns in gait with shoes have not been analysed compared to barefoot gait. METHODS: 30 children (age 1.8 to 4.8 years) were examined. Ground reaction forces were recorded using two series connected Kistler plates, taking the 3D coordinates and the time coordinate. The gait aspect was recorded using a video system. Three shoe types were tested against each other and against barefoot gait. RESULTS: 1399 gait cycles, 739 barefoot, 660 with shoes were usable for calculation. Statistic analyses comparing barefoot gait to gait with shoes showed: A significant increase in ground contact duration, a shifting of maximal load from the rear towards the midfoot area, a significant increased maximal load, a cushioning of the initial impact at primary ground contact. A comparison between the shoes showed no significant influence of a medial subtalar orthotic support. CONCLUSION: Shoes cause a significant change of loading patterns. But there are also significant differences between different shoe types. CLINICAL RELEVANCE: Shoes for children must provide a stabilisation for the hindfoot, and must not restrict motion in the areas of the ankle and the metatatrsophalangeal joints. The amount of load in the different gait phases must be considered for the construction of a dynamic shoe for children.
http://www.ncbi.nlm.nih.gov/pubmed/9823643