Response to: Comment on "Prevalence and Risk Factors for Diabetic Lower Limb Amputation: A Clinic-Based Case Control Study".

The observations by Bakhtiyari and Mansournia on our study [1] were received with great interest. We believe our study was consistent with a case-control format [2]. In particular, our target population was selected from subjects diagnosed with diabetic foot ulcer (DFU) at the local high-risk foot clinic, effectively fulfilling the criteria of control (DFU without amputations) and case (DFU with amputation), as previously described [3]. The identified risk factors were computed amongst those with and without limb amputations. We do not believe our study was cross-sectional, being retrospective in design, and it was not conducted at a specific point in time rather for a period from January 2011 to December 2013; both are in agreement with the definition of a case-control study [4]. The prevalence of diabetic limb amputation quoted in our study was for our high-risk diabetic foot clinic and not for the general population of North Eastern Australia which was clearly outlined in the title as clinic-based [5]. Interestingly, in line with Bakhtiyari's and Mansournia's observation of a case-control study, our sampling was based on the outcome (amputation) and was known in advance prior to conducting the study [3]. Furthermore, in keeping with a case-control study, we did not undertake propensity matching since cases and controls were outcomes rather than exposures [4]. With respect to our discussion on the prevalence of diabetic limb amputation, comparing clinic-based and non-clinic-based studies, the point was noted but we clearly stated our study was similar to those reported by others [6]. Overall, we do not think the title of our article was confusing since it tallied with the methodology and content of the article.