An audit of Australia and New Zealand dialysis and transplant registry (ANZDATA) haemodialysis patient data

Abstract

Aim: To audit haemodialysis patient data submitted to ANZDATA. Background: Most clinical registries in Australia, including ANZDATA, do not audit entered data. Inaccurate data will bias registry analysis. Many renal units rely on nursing staff to collect and enter data for ANZDATA. Methods: A retrospective audit of individual haemodialysis patient data recorded by ANZDATA at 31/12/2009 was completed by consultant nephrology staff in a blinded fashion. Original data had been recorded by nursing staff. Patients received treatment at a public hospital, 2 affiliated satellite dialysis units, and 3 private haemodialysis units. Results: 51 audits were completed of a total 134 patients (38%) undertaking haemodialysis in 2009. Primary renal disease was incorrect in 24%, mainly due to incorrect glomerulonephritis or diabetes code. Creatinine at first dialysis after allowing a 10% error margin was incorrect in 31%. Most errors occurred with peripheral vascular disease at entry (42%) and current (52%), ischaemic heart disease at entry (24%) and current (28%), smoking status (29%), chronic lung disease current (24%), diabetes at entry (16%) and current (18%), access at first dialysis (23%), and revision of vascular access (28%). Cancer was incorrect in 30%. When cancer was reported as present, only 40% had accurate and complete data. Remaining data was >80% correct. Height and weight at entry was difficult to audit retrospectively. Conclusions: Nurses were good at data collection except primary renal disease and comorbidities. Education of nurses as well as medical staff checking comorbidity and primary renal disease data for haemodialysis patients may improve data quality. Making comorbidity data objective or event based may improve data consistency. A larger audit should be completed.<br />