Comparison of cause of death between anzdata and the australian national death index

Matthew P Sypek, Kathryn B Dansie, Phil Clayton, Angela C Webster, Stephen McDonald

Research output: Contribution to journalArticlepeer-review

19 Citations (Scopus)


AIM: To understand the differences in how cause of death for patients receiving renal replacement therapy in Australia is recorded in The Australian and New Zealand Dialysis and Transplant Registry (ANZDATA) compared to the National Death Index (NDI).

METHODS: Data linkage was performed between ANZDATA and NDI for all deaths in the period 1980-2013. Cause of death was classified according to ICD-10 chapter. Overall and chapter specific agreement were assessed using the Kappa statistic. Descriptive analysis was used to explore differences where there was disagreement on primary cause of death.

RESULTS: The analysis cohort included 28,675 patients. Ninety five percent of ANZDATA reported deaths fell within +/- 3 days of the date recorded by NDI. Circulatory death was the most common cause of death in both databases (ANZDATA 48%, NDI 32%). Overall agreement at ICD chapter level of primary cause was poor (36%, kappa 0.22). Agreement was best for malignancy (kappa 0.71). When there was disagreement on primary cause of death these were most commonly coded as genitourinary (35%) and endocrine (25.0%) in NDI, and circulatory (39%) and withdrawal (24%) in ANZDATA. Sixty-nine percent of patients had a renal related cause documented as either primary or a contributing cause of death in the NDI.

CONCLUSION: There is poor agreement in primary cause of death between ANZDATA and NDI which is in part explained by the absence of diabetes and renal failure as causes of death in ANZDATA and the absence of 'withdrawal' in NDI. These differences should be appreciated when interpreting epidemiological data on cause of death in the Australian end stage kidney disease population.

Original languageEnglish
Publication statusE-pub ahead of print - 1 Mar 2018


  • Journal Article

Cite this