Unlocking the multidimensionality of plantar pressure measurements for the evaluation of footwear in people with diabetes

The offloading effectiveness of custom-made footwear for people with diabetes is assessed using plantar pressure measurements. While such pressure data is multidimensional, it is mostly analyzed using a scalar − maximum peak plantar pressure (PMax). We aimed to investigate the associations between multiple peak plantar pressure parameters for footwear assessment and determine whether this assessment depends on the chosen parameter. In-shoe plantar pressure was measured in 77 participants with diabetes, peripheral neuropathy, and a recent ulcer or amputation history, while walking in their own custom-made footwear. Six peak plantar pressure parameters were extracted, both scalar (i.e. Pmax, time integral and gradient) and multidimensional (i.e. time curve, map and time map). Footwear was ranked from highest to lowest outcome for each parameter and associations with Pmax were compared using Spearman’s rank correlation coefficient. A footwear comparison within subjects using Fleiss’ Kappa analysis determined the agreement between parameters using two pairs of footwear of each participant. The rank correlation coefficient was moderate to strong between PMax and the other scalar parameters (ρ = 0.46–0.70), and negligible to weak between PMax and the multidimensional parameters (ρ = 0.03–0.25). Percentage agreement between parameters for the within-subject footwear comparison was poor (47.5 %, κ = 0.0652). We conclude that the association and agreement between in-shoe peak pressure parameters is low and the assessment of offloading effectiveness depends on the chosen parameter. This is the first step in unlocking the potential of a multidimensional approach in plantar pressure analysis, possibly changing how we evaluate footwear offloading effectiveness.