Reducing inequalities in cancer survival for Ontario First Nations: From surveillance to action

Loraine Marrett, Diane Nishri, Amanda Sheppard, Anna Chiarelli, Alethea KewayoshPrevention & Cancer Control, Cancer Care Ontario

Canadian Public Health Association MeetingMay 29, 2014

Objectives

2

• To outline disparities in cancer burden, especially survival, in registered First Nations in Ontario

• To demonstrate the trajectory from surveillance data to research to cancer control action

Outline

• Aboriginal people in Ontario• Cancer burden in Registered First Nations in Ontario• Explaining disparities in survival• Moving from knowledge to action• Going forward

3

Aboriginal People in Ontario

• In Canada,• 1,400,685 Aboriginal people (~4% of total population)

• First Nations (FN): 851,560 (60%)• Métis: 451,795 (32%)• Inuit: 59,445 (4%)• Collectively referred to as ‘FNIM’

• Ontario has 201,100 (21.5%) of Canada’s First Nations people • 62% are “Registered” as Indians under the Indian Act• About 50% live on reserve

Statistics Canada. 2011 National Household Survey. http://www12.statcan.gc.ca/nhs-enm/2011/as-sa/99-011-x/99-011-x2011001-eng.cfm 4

5

Outline

• Aboriginal people in Ontario• Surveillance: Cancer burden in Registered First

Nations in Ontario• Explaining disparities in survival for one cancer• Moving from knowledge to action• Going forward

6

What do we know about cancer burden in Ontario’s FN population?

• Population-wide estimates of cancer burden in Ontario typically come from the Ontario Cancer Registry (OCR)

• The OCR records data on cancer diagnoses for all Ontarians, 1964-2010

• All cases are followed up for vital status and date of death, so survival times from diagnosis can be calculated

• Cannot routinely estimate cancer burden for ethnic/racial subgroups, including FNIM

• Race/ethnicity not systematically recorded in administrative health records upon which the OCR is based

7

Indian Registry System

1968-1991

Ontario Cancer Registry

1968-2001

Ontario Mortality Database

1968-2006

Cancer burden in Registered First Nations in Ontario: Methods

• The ‘Indian Registry System’ (IRS) is maintained by Aboriginal Affairs & Northern Development Canada (AANDC)

• 140,000 Registered First Nations (1968-91)

• Used probabilistic record linkage to link IRS, OCR and Ontario mortality file

8

All cancer incidence, Ontario, 1968-2001,ages 15-74

Females

Age-

stan

dard

ized

rate

/100

,000

Year of diagnosis

Rates age-standardized to the 1991 Canadian population.Horizontal bars around First Nations rates indicate 95% confidence limits.Source: Marrett & Chaudhry, Cancer Causes Control 2003; Marrett et al. unpublished data.

0

50

100

150

200

250

300

350

400

450

1968-1975

1984-1991

1992-1996

1997-2001

1976-1983

0

50

100

150

200

250

300

350

400

450

Males

1976-1983

1968-1975

1984-1991

1992-1996

1997-2001

First Nations

All Ontario

Year of diagnosis• Historically lower cancer incidence rates• Incidence rates rose faster than in Ontario as a

whole

Colorectal cancer incidence, Ontario,1968-2001, ages 15-74

Age -

stan

dard

ized

rate

/100

,000

0

10

20

30

40

50

60

70

80

90

Females

Year of diagnosis

1968-1975

1984-1991

1992-1996

1997-2001

1976-1983

Males

0

10

20

30

40

50

60

70

80

90

1976-1983

1968-1975

1984-1991

1992-1996

1997-2001

Rates age-standardized to the 1991 Canadian populationHorizontal bars around First Nations rates indicate 95% confidence limitsSource: Marrett & Chaudhry, Cancer Causes Control 2003; Marrett et al. unpublished data.

First NationsAll Ontario

Year of diagnosis

Lung cancer incidence, Ontario, 1968-2001, ages 15-74

Age -

stan

dard

ized

rate

/100

,000

0

10

20

30

40

50

60

70

80

90

Females

Year of diagnosis

1968-1975

1984-1991

1992-1996

1997-2001

1976-1983

Males

0

10

20

30

40

50

60

70

80

90

1976-1983

1968-1975

1984-1991

1992-1996

1997-2001

Rates age-standardized to the 1991 Canadian populationHorizontal bars around First Nations rates indicate 95% confidence limitsSource: Marrett & Chaudhry, Cancer Causes Control 2003; Marrett et al. unpublished data.

First NationsAll Ontario

Year of diagnosis

High smoking rates in off- and especially on-reserve FN

Five-year age-standardized observed survival (%) by cancer site and sex comparing the First Nations population to other Ontarians for cancers diagnosedbetween 1992-2001 with follow-up to December 31st, 2007 and censoring at age 75. Source: Nishri, Sheppard, Withrow & Marrett. Int J Cancer (in press)

Male Female Male Female Female Female MaleColorectal Lung Cervix Breast Prostate

0

20

40

60

80

100

First NationsOther Ontar-ians

Five

-yea

r obs

erve

d su

rviv

al (%

)

Other On-tarians

5-year Cancer Survival in Registered First Nations vs Other Ontarians, 1992-2001

Survival significantly poorer for all cancers except female lung cancer

Outline

• Aboriginal people in Ontario• Cancer burden in Registered First Nations in Ontario• Research: Explaining disparities in survival• Moving from knowledge to action• Going forward

13

• Purpose: To further examine survival disparity between Ontario FN and non-FN women diagnosed with breast cancer between 1995-2004

• Reviewed charts for stage and other important prognostic factors (comorbidity, treatment, smoking, etc) not in the OCR

14

Key findings• FN women 50% more likely to be diagnosed at stages II+• % with appropriate treatment-for-stage and -age same• Only stage 1 survival worse for FN women

• More comorbidity, especially diabetes, in FN women explained this

• No other prognostic factors significantly different • Caveats • Full assessment of impact of distance not possible since

matched on cancer centre• No data on socioeconomic status• Only breast cancer

15

Outline

• Aboriginal people in Ontario• Cancer survival in Registered First Nations in Ontario• Explaining disparities in survival• Cancer control: Moving from knowledge to action• Going forward

16

Aboriginal Cancer Strategy (ACS II) 2012-15

• Cancer Care Ontario’s roadmap to address FNIM cancer control needs in Ontario

• Led by CCO’s Aboriginal Cancer Control Unit

• ACS III is under development

Vision“To improve the performance of the cancer system with and for FNIM peoples in Ontario in a way that honours the Aboriginal Path of Well-being”

“The ACS II sets out a clear plan for reducing risk & preventing cancer…”

ACS II: Strategic Priorities

18

Outline

• Aboriginal people in Ontario• Cancer burden in Registered First Nations in Ontario• Explaining disparities in survival• Moving from knowledge to action• Going forward…more to be done…

19

Updating and completing the picture: continuing surveillance• Linkage of Indian Registry System 1991-present &

Ontario Cancer Registry to estimate current cancer burden (collaboration with the Chiefs of Ontario and ICES)

• Access data from linking 1991 Canadian Census (long form) and the Canadian Cancer Registry to estimate cancer burden in FN and Métis, nationally (and large provinces)

• Novel ways to identify non-Status FN, Métis and Inuit are necessary in order to complete the FNIM picture

20

Conclusions

• Having cancer surveillance data for Ontario FN was foundational to establishing the strategic priorities and specific priority activities and securing funding to implement

• ACSI I focuses on both system level change as well as programmatic activities

• Ongoing surveillance and research is important to monitor and understand changes and to assess and adjust strategy

21

Thank you!Questions?

22

Loraine.marrett@cancercare.on.ca

Does stage at diagnosis explain difference in survival?TNM Stage First Nations women

n (%)Non-FN women

n (%)Adjusted OR*

(95% CI)Stage I 95 (34) 284 (44) 1.00

Stage II 130 (46) 244 (38)1.55 (1.16-2.08)

Stage III and IV

58 (20) 122 (19)

Unstageable 4 (<1) 21 (<1)

23

* Adjusted for age of diagnosis, period of diagnosis and cancer centreFN: First NationsSheppard AJ, Chiarelli AM, Marrett LD, Mirea L, Nishri ED, Trudeau ME. 2010. Detection of later stage breast cancer in First Nations women in Ontario, Canada. Can J Public Health.

FN women were more likely than non FN women to be diagnosed with more advanced breast cancer (p<0.01)

Prognostic factors that do not differ between FN and non-FN women

24

Estrogen/progesterone receptor status

First Nations women n (%)

Non-FN womenn (%)

Neither positive 132 (46) 342 (51)

Either positive 155 (54) 329 (49)p>0.05No significant differences

FN: First NationsSheppard AJ, Chiarelli AM, Marrett LD, Nishri ED, Trudeau ME. 2011. Stage at diagnosis and comorbidity influence breast cancer survival in First Nations women in Ontario, Canada. Cancer Epidemiology, Biomarkers & Prevention. 2011.

Stage- and age-appropriate treatment

First Nations women n (%)

Non-FN womenn (%)

No 68 (24) 146 (22)

Yes 215 (76) 504 (78)

Unknown 4 (<1) 21 (<1)

Prevalence of comorbidities, in particular diabetes, does differ between FN and non-FN women

25

Any comorbidity First Nations women n (%)

Non-FN womenn (%)

No 180 (64) 534 (82)

Other 35 (12) 72 (11)

Diabetes 67 (24) 46 (7)

Unknown 5 (<1) 19 (<1)

Significantly different,P<0.01

FN: First NationsSheppard AJ, Chiarelli AM, Marrett LD, Nishri ED, Trudeau ME. 2011. Stage at diagnosis and comorbidity influence breast cancer survival in First Nations women in Ontario, Canada. Cancer Epidemiology, Biomarkers & Prevention. 2011.

First Nations (off-reserve)

Métis Non-Aboriginal First Nations (off-reserve)

Métis Non-Aboriginal

Males Females

0

10

20

30

40

50

60

70

80

90

100

45% 43%

26%

43%35%

18%

Perc

ent

(%)

Percentage of Current Smokers in Ontario, Aged 20+, 2007-2011

Non-traditional Tobacco Use

* *

* *

* Estimate is significantly different from the non-Aboriginal estimateNote: Estimates are age-standardized to the 2006 Ontario Aboriginal identity population.Source: Cancer Care Ontario. Aboriginal Cancer Strategy II – Annual Report: Staying on the path. Toronto, Canada, 2013”

Obesity

First Nations (off-reserve)

Métis Non-Aboriginal First Nations (off-reserve)

Métis Non-Aboriginal

Males Females

0

10

20

30

40

50

60

70

80

90

100

33% 28%19%

26% 26%16%

Perc

ent

(%)

Percentage of population who are obese in Ontario, aged 20+, 2007-2011

* * * *

* Estimate is significantly different from the non-Aboriginal estimateNote: Estimates are age-standardized to the 2006 Ontario Aboriginal identity population.Source: Cancer Care Ontario. Aboriginal Cancer Strategy II – Annual Report: Staying on the path. Toronto, Canada, 2013”

Alcohol consumption

First Nations

(off-reserve)

Métis Non-Aboriginal First Nations

(off-reserve)

Métis Non-Aboriginal

Males Females

0

5

10

15

20

25

30

14% 16%

10% 11% 10% 9%

Perc

ent

(%)

Percentage of population who exceed cancer prevention recommendations for alcohol consumption in Ontario, aged 20+,

2007-2011

* Estimate is significantly different from the non-Aboriginal estimateNote: Estimates are age-standardized to the 2006 Ontario Aboriginal identity population.Measured males who consumed on average >2 drinks per day and females who consumed on average >1 drink per daySource: Cancer Care Ontario. Aboriginal Cancer Strategy II – Annual Report: Staying on the path. Toronto, Canada, 2013”

* *