Interpreting incidence trends for treated end-stage renal disease: implications for evaluating disease control in Australia

BACKGROUND: Five sources of change modify trends in incidence of treated end-stage renal disease (ESRD): (i) demography; (ii) disease control, comprising prevention and treatment of progressive kidney disease; (iii) competing risks, which encompass dying from untreated uraemia or non-renal comorbidity; (iv) lead-time bias; and (v) classification bias. Thus, rising crude incidence of treated ESRD may conceal effective disease control when there has been demographic change, lessening competing risks, or the introduction of bias. METHODS: Age-specific incidences of treated ESRD in Australia were calculated from Australia and New Zealand Dialysis and Transplant (ANZDATA) Registry data by indigenous/non-indigenous status (all causes) and by primary renal disease (non-indigenous only) for two successive decades, 1982-1991 and 1992-2001. RESULTS: We postulate that less competing risks explained much of the increase in treated ESRD in the elderly and Indigenous Australians. The increase in glomerulonephritic ESRD in non-indigenous Australians could be ascribed mainly to immigration from non-European countries. There was no significant change in incidence of treated ESRD in Indigenous or non-indigenous persons aged less than 25 years, in non-indigenous persons aged 25-64 years for ESRD caused by hereditary polycystic disease or hypertension, or in type 1 diabetics aged over 55 years. End-stage renal disease from analgesic nephropathy had declined. The increase in treated ESRD caused by type 2 diabetic nephropathy appeared to be multifactorial. Lead-time/length bias and less competing risks may have concealed a small favourable trend in other primary renal diseases. CONCLUSION: Whether recent disease control measures have had an impact on incidence of treated ESRD is not yet certain, but seems more likely than implied by previous reports.