Introduction: Diabetes is a major cause of CKD and of mortality in patients on renal replacement therapy (RRT). Auditing the care of patients with diabetes on RRT against published guidelines relies on robust data collection. Objective: This article assesses the completeness of data items collected by the UK Renal Registry (UKRR) that are required to audit the care of patients with diabetes on RRT. Methods: The UKRR receives data on all patients receiving RRT in the UK. Patients with diabetes, diabetes type, and method of renal diagnosis were identified from primary renal disease (PRD) codes and comorbidity data for patients commencing RRT at one of the 57 renal centres in England and Wales between 2010 and 2016. The completeness of demographic and clinical data (blood pressure, cholesterol, glycated haemoglobin [HbA1c], and smoking status) was assessed for the first year of RRT. Results: Ninety-three per cent of all patients on RRT irrespective of diagnosis had a PRD code, but only 28/57 renal centres had comorbidity data completeness ≥70%; 34.9% of patients with diabetic nephropathy (DN) had type 1 diabetes, but this varied between centres (9.2–100%). Overall, 4.2% of DN diagnoses were by biopsy. Data completeness in the first year of RRT for cardiovascular risk factors ranged between 50.0 and 80.0%, with HbA1c data completeness being 63.0%. Of 57 centres, 20 had HbA1c data for ≥70% of patients in the first year of RRT. Conclusions: There is persistent variation between renal centres in the completeness of data collected on patients with diabetes on RRT, impacting on the ability to undertake robust audit. Data linkages and expanded data permissions could see registry data play a key role in ongoing audit and research into patients with diabetes and CKD, provided adequate data can be collected.