What were the key findings from the routinely-collected data?

Transcription

1 What were the key findings from the routinely-collected data? Dr Jason Gurney MoH National Health Board Cancer Registry (n=14,096) Hospitalisation and other Databases Mortality Database Notes Review (Upper GI &, n=718) Study Cohort Ethnicity Demographics Tumour Comorbidity Health Care Survival 1

2 2

3 Māori cancer patients are differentto non-māori cancer patients. Age Average Years at Diagnosis Non-Maori Maori Age (Years) 3

4 Age Average Years at Diagnosis Non-Maori Maori Age (Years) 60% 50% 40% 30% 20% 10% 0% Deprivation - All NZ Dep Quintile Non-Maori Maori 4

5 60% 50% 40% 30% 20% 10% 0% Deprivation - All NZ Dep Quintile Non-Maori Maori 60% 50% 40% 30% 20% 10% 0% 60% 50% 40% 30% 20% 10% 0% Deprivation Selected Sites 60% 50% 40% 30% 20% 10% 0% % 50% 40% 30% 20% 10% 0% NZ Dep Quintile NZ Dep Quintile Non-Maori Maori 5

6 Rurality- Proportion Living in Rural Areas Non-Maori Maori 0% 5% 10% 15% 20% 25% 30% Rurality - Proportion Living in Rural Areas Non-Maori Maori 0% 5% 10% 15% 20% 25% 30% 6

7 Combin Stage -Proportion Diagnosed with Advanced Disease (NZCR) 0% 20% 40% 60% Non-Maori Maori Combin Stage -Proportion Diagnosed with Advanced Disease (NZCR) 0% 20% 40% 60% Non-Maori Maori 7

8 Comorbidity is highly-prevalentamong cancer patients but prevalence varies by cancer type. 8

9 Prevalence of Comorbid Conditions - Combined Cancer Sites 0 Sarfati, Gurney, et al. (2013). Asia-Pac J Clin Oncol. Hypertension (Primary) Diabetes (Any) COPD/Asthma Sarfati, Gurney, et al. (2013). Asia-Pac J Clin Oncol Cardiac Arrhythmia

10 Disorder (Moderate/Severe) Hepatitis (Chronic Viral) Renal Disorder Sarfati, Gurney, et al. (2013). Asia-Pac J Clin Oncol Gastro-Intestinal Disorder

11 Māori cancer patients tend to have a greater comorbidity burden. Prevalence (Crude %) 18% 16% 14% 12% 10% 8% 6% 4% 2% Prevalence of Comorbid Conditions - Combined Cancer Sites 0% 11

12 Hypertension (Primary) Non-Māori Māori Hypertension (Primary) Non-Māori Māori

13 Diabetes (Any) Non-Māori Māori Diabetes (Any) Non-Māori Māori

14 Hepatitis (Chronic Viral) Non-Māori Māori Hepatitis (Chronic Viral) Non-Māori Māori

15 4/9/2014 A high comorbidity burden increases likelihood of mortality but the extent of this varies by cancer. 15

16 Blood Disorders Diabetes (With Complications) Gyn Urological Colorectal Upper GI (Condition causes death, even after adjusting for covariates) Congestive Heart Failure Gyn Colorectal Urological Upper GI Hypertension Urological Gyn Colorectal Upper GI Gyn Colorectal Urological Upper GI Adjusted All-Cause Excess Mortality (%) Urological C3 Index Category (Ref) 0 (Ref) Colorectal Gynaecological Upper GI 1 0 (Ref) 1 0 (Ref) 1 0 (Ref) Adjusted All-Cause Excess Mortality (%) 16

17 4/9/2014 Māori have poorer survival outcomes for some cancers compared to non-māori. 17

18 MoH National Health Board Cancer Registry (n=14,096) Hospitalisation Database (+ others for Rx) Mortality Database Notes Review (Upper GI &, n=718) Study Cohort Ethnicity Demographics Tumour Comorbidity Health Care Survival Excess Mortality, Maori vs. Non-Maori - Unadjusted (Worse for Non-Maori) (Worse for Maori) Cancer-Specific Excess Mortality (%) 18

19 Excess Mortality, Maori vs. Non-Maori - Age/Sex/Stage-Adjusted (Worse for Non-Maori) (Worse for Maori) Cancer-Specific Excess Mortality (%) Excess Mortality, Maori vs. Non-Maori - Age/Sex/Stage/Comorbidity-Adjusted (Worse for Non-Maori) (Worse for Maori) Cancer-Specific Excess Mortality (%) 19

20 Does ethnicity influence treatment receipt? Does comorbidityinfluence treatment receipt? 20

21 Gurney, Sarfati, et al. (2013). NZMJ. 21

22 Crude Proportion (%) of Patients Who Received Surgery Receipt of Surgery 1012/1328 patients 1261/1328 patients 590/724 patients 693/724 patients 118/143 patients 136/143 patients MoH Collections Augmented Dataset Crude Proportion (%) of Cancer Patients (Regional ) Receipt of Adjuvant Treatment 208/517 patients 303/517 patients Chemotherapy 267/517 patients 348/517 patients Radiotherapy MoH Collections Augmented Dataset 22

23 Difference in Crude % Treated, MoH vs Augmented Dataset Differential Impact by Ethnicity 20% 19% 15% 16% 14% 13% 11% 8% 7% 9% Maori Non-Maori * Regional stage of disease only MoH National Health Board Cancer Registry (n=14,096) Hospitalisation Database (+ others for Rx) Mortality Database Notes Review (Upper GI &, n=718) Study Cohort Ethnicity Demographics Tumour Comorbidity Health Care Survival 23

24 Upper GI Stage I-III Surgery C3 Index Category Adjusted Odds Ratio: 0.50 ( ) Received Curative Surgery (%) Summary Māori cancer patients are differentto non-māori cancer patients. Younger. More likely to live in deprived areas. More likely to live in rural areas. Tend to have a greater comorbidity burden. Tend to have different comorbidities, e.g. Hepatitis. 24

25 Summary Comorbidity is highly-prevalentamong cancer patients but prevalence varies by cancer type. Summary A high comorbidity burden increases likelihood of mortality. ( and reduceslikelihood of treatment ) 25

26 Summary Our National Collections have immense strengths, and are envied around the world. (Just ask an Australian.) Summary But they also have crucial weaknessesin terms of cancer treatment reporting: Private hospitals not mandated to report. Under-reporting by DHBs.* *MoH. Price of Cancer Report,

27 Summary These weaknesses undermine the ability of the National collections to meaningfully measure sector performance. Acknowledgements Associate Professor Diana Sarfati (PI) James Stanley (biostatistician) Virginia Signal (oncology nurse) And the rest of the C3 research team The Health Research Council of New Zealand 27