Dialysis outcomes in Australia & New Zealand Stephen McDonald ANZDATA Registry

Dialysis outcomes What is collected by ANZDATA: What is not collected: ANZSN ASM 2002 Dialysis outcomes What is collected by ANZDATA: Mortality Process indicators Urea clearance Haemoglobin What is not collected: Indicators of health & functional state Mental health Physical health Surrogate markers of cardiovascular disease (LVH)

Acceptance of New Patients by Age ANZSN ASM 2002 Acceptance of New Patients by Age Australia New Zealand New RRT rates by age group 1990-2001

Survival on dialysis – 15-64 yrs ANZSN ASM 2002 Survival on dialysis – 15-64 yrs Kaplan-Meier survival estimates, by year Survival 5 10 15 20 0.00 0.25 0.50 0.75 1.00 1965- 1970- 1975- 1980- 1985- 1990- 1995- 2000- Are mortality rates changing over time Are dialysis practices changing? Trends towards increasing URR, higher flow rates Not longer hours ?membrane flux Time (years) Survival on dialysis, 15-64 years old, at onset of RRT, censored at transplantation

Survival, 65-74 years ANZSN ASM 2002 Kaplan-Meier survival estimates, by year 1.00 0.75 Survival 0.50 1990- 1985- 1995- 0.25 1975- 1980- 1970- 0.00 5 10 Years Survival on dialysis, 65-74 years old, at onset of RRT

Proportion of CAD by age ANZSN ASM 2002 Proportion of CAD by age .8 .6 Proportion with CAD .4 .2 15- 25- 35- 45- 55- 65- 75- 85- Age (years) Proportion of new dialysis patient with definite or suspected coronary artery disease (CAD), 1995-2001

ANZSN ASM 2002 Methods All patients who began RRT in Australia or NZ since 1 October 1991 Commencement of collection of co-morbid data Periods of transplant function excluded Censored at death Cox regression Time dependent covariates for dialysis modality Co-morbidities ascertained at RRT start

Outcomes by age ANZSN ASM 2002 1.00 15- 0.75 25- 35- 0.50 45- 55- 0.25 65- 85- 0.00 75- 1 3 5 7 9 Years Kaplan Meier survival curves by age group for patients treated with dialysis who began RRT treatment in Australia or New Zealand 1 Jan 1991 - 31 Dec 2000. Periods of transplant function excluded.

Comorbid factors and outcome ANZSN ASM 2002 Comorbid factors and outcome 2.5 2 1.5 Hazard ratio 1 .5 DM1 DM2 Aust Ab Maori / PI CAD Cerebrovasc High BP Late referral Lung Male PVD PD Hazard ratios for mortality on dialysis, dialysis only, 1991-present, adjusted for age and gender

Effect of BMI on outcome ANZSN ASM 2002 Effect of BMI on outcome 2 1.5 Hazard ratio 1 <20 20-24.9 25-29.9 30-34.9 35- BMI (kg/m2) Hazard ratios for mortality by BMI at entry, dialysis only, 1991-present, adjusted for age and gender and other comorbidities.

Predicted median survival ANZSN ASM 2002 Predicted median survival 25 centile 50 centile 75 centile 15 Survival (years) 10 5 45 55 65 75 85 Age at RRT start Median [IQR] survival by age. Age modelled as fractional polynomial, adjusted to 0 comorbidities & non-indigenous male

Interactions with time ANZSN ASM 2002 Interactions with time Hazard ratio (death) Time period post RRT start 0-90 day 3-12 months 1-2 year 2-5 years 5+ years .5 1 1.5 2 2.5 Hazard ratios for late referral for mortality on dialysis, 1991-present, adjusted for age and gender and other co-morbidities

Interactions with time - 2 ANZSN ASM 2002 Interactions with time - 2 2.5 2 Hazard ratio (death) 1.5 1 0-90 day 3-12 months 1-2 year 2-5 years 5+ years Time period post RRT start Hazard ratios for coronary artery for mortality on dialysis, 1991-present, adjusted for age and gender and other comorbidities

Interactions of CAD with age ANZSN ASM 2002 Interactions of CAD with age 5 Hazard ratio 2 1 25- 35- 45- 55- 65- 75- 85- Age (years) Hazard ratios for mortality of coronary artery disease by age category, adjusted for gender, dialysis patients

Indigenous outcomes Survival ANZSN ASM 2002 Years Years 1.00 1.00 Non-indigenous 0.75 0.75 Non-indigenous Pacific Islander 0.50 Pacific Islander 0.50 Maori Survival 0.25 0.25 Aboriginal Aboriginal Maori 0.00 0.00 5 10 5 10 Years Years Survival of new ESRD patients, 1991-2001, crude (left) and adjusted to age 40 (right), censored at transplantation. PI includes TSI

Centre effect What is the aim? What is a centre? ANZSN ASM 2002 Centre effect What is the aim? making comparisons? meeting a standard? What is a centre? How do you handle transfers? What do you adjust for ? Role of dialysis dose ? Access

Crude death rates ANZSN ASM 2002 Rate (per 100) 10 20 30 40 Death rates of patients treated with dialysis, 1991-2001, by hospital of initial treatment, for centres with >20 deaths over the period.

Centre effect - 2 ANZSN ASM 2002 50 Rate estimate (per 100 per year) 20 10 5 1 Death rates of patients treated with dialysis, 1991-2001, by hospital of initial treatment, for all centres, sorted by number of new patients over that perio.

ANZSN ASM 2002 Dialysis mortality Dialysis mortality is strongly related to variables which are known at time of RRT start The influence of these varies over time Age is strongly and progressively related to outcome Adverse prognosis of indigenous groups Remains after adjustment for comorbidities

Areas of further investigation ANZSN ASM 2002 Areas of further investigation This work provides the basis for developing and validating a predictive set of equations How to incorporate factors such as dialysis dose and Hb into analyses Vary with time Cause or effect? Centre effect Planned for presentation at D & T 2003

ANZSN ASM 2002 Acknowledgements The ANZDATA Registry is a collaborative effort of the Australian / New Zealand renal community, and could not exist without the dedicated support of the renal units and the individuals therein.

And still older… ANZSN ASM 2002 Survival Years 1 3 5 0.00 0.25 0.50 0.75 1.00 1980- 1985- 1990- 1995- 2000- Kaplan-Meier survival estimates, by year Survival on dialysis, 75+ years old, at onset

Indigenous ESRD incidence ANZSN ASM 2002 Indigenous ESRD incidence New ESRD patient (Australia) by racial origin 1990-2000. For 2000, age standardised incidence rate ratio=7.8 [6.6-9.2]

Interactions with time – 3 ANZSN ASM 2002 Interactions with time – 3 Hazard ratio (death) Time period post RRT start 0-90 day 3-12 months 1-2 year 2-5 year 5+ years .5 1 1.5 2 Hazard ratios for type 2 diabetes for mortality on dialysis, 1991-present, adjusted for age and gender and other comorbidities

Interactions of CAD with age - 2 ANZSN ASM 2002 Interactions of CAD with age - 2 5 2 Hazard ratio 1 .5 25- 35- 45- 55- 65- 75- 85- Age (years) Hazard ratios for coronary artery disease by age category, adjusted for gender, all other co-morbidities and indigenous status