A new diabetic foot risk assessment tool: DIAFORA

Aims This study aimed to derive a new model to classify subjects with diabetes and active diabetic foot ulcer by their risk of lower extremity amputation. Methods A prospective cohort study was conducted that included all subjects with diabetic foot ulcer attending our Hospital Diabetic Foot Clinic from 2010 to 2013. Variables were collected at baseline. Subjects were followed up until healing, lower extremity amputation, death or for at least 3 months. Logistic regression was used to derive the new model, and the area under the receiver operating characteristic curve was assessed to propose the model with the greatest discrimination. Results A total of 293 participants were included and followed for a median of 91 days. In 23.2% amputation was required, 5.1% died and 3.1% were lost. Our final model included the variables most commonly used in clinical practice for diabetic foot risk assessment (presence of neuropathy, foot deformity, peripheral arterial disease and previous foot complications) in addition to multiple diabetic foot ulcer, infection, gangrene and bone involvement. This model had an area under the receiver operating characteristic curve of 0.91 [95% confidence interval (CI) 0.87–0.95] and as classification of 0.89 (95% CI 0.84–0.93) for lower extremity amputation prediction. The high‐risk group presented a positive likelihood ratio of 5 (95% CI 3–8) and predictive value of 58 (46–71). Only one minor lower extremity amputation occurred in the low‐risk group. Conclusions We propose a new classification: diabetic foot risk assessment (DIAFORA). This classification was equally or more accurate for lower extremity amputation prediction in diabetic foot ulcer patients when compared with the existing ones. Copyright © 2016 John Wiley & Sons, Ltd.