1. Age and Racial/Ethnic Disparities in the Diagnosis of Breast Cancer in an Urban Population
Joanne K. Fagan PhD, Denise Fyffe, PhD, Nadine Jenkins, CTR, Charles Cathcart, MD, Margarette Bryan, MD, Mary E. Kelleher, Diane R. Brown, PhD,
University of Medicine and Dentistry of New Jersey Newark, New Jersey
This research was done as apart of our EXPORT Project - Center of Excellence in Health Disparities
Which is a collaboration between the University of Medicine and Dentistry of New Jersey, Jersey City University
And The Cancer Institute of New Jersey

2. Aim
To characterize the newly diagnosed patients with breast cancer from at University Hospital.
The Aim of this project was simply ….

3. Background
Breast cancer accounts for one of every three cancers diagnosed in women (ACS, 2006).
In 2006, an estimated 212,920 women were diagnosed with invasive breast cancer and over 30,970 died from this disease (Smigal, et al, 2006).

4. Breast Cancer in New Jersey
Among females in New Jersey, breast cancer is the most commonly diagnosed cancer across all racial and ethnic groups
For Hispanic females, breast cancer is the leading cause of cancer mortality, while for non-Hispanic White and Black females lung cancer is the leading cause of cancer mortality
Black women have lower incidence rates when compared to White women, (113.3 vs ), but greater mortality rates (34.4 vs. 29.0). 

5. Breast Cancer Disparities
The burden of breast cancer is not equally
distributed.
Minority women are more likely to
Be diagnosed with advanced disease
Have comorbid conditions
Delay initial treatment
Discontinue treatment
Have poorer outcomes
Have shorter survival

6. Breast Cancer in Young Women
According to data from the SEER program, 6.5% of breast tumors occur in women < 40 years of age.
Young women with breast cancer have been shown to:
Have larger tumors
Poorer grade tumors
More lymph node involvement
Fewer hormone receptor positive tumors
More local recurrences
Poorer overall survival

7. Methods Cancer registry data was reviewed at University Hospital.
All women newly diagnosed with breast cancer between 1999 and 2006 were included.
For this analysis we only looked at those patients who were classified as Non-Hispanic Black, Hispanic or Non-Hispanic White
The methods for this project included abstracting cancer registry data from University Hospital for all women newly diagnosed between For this analysis we only looked at those patients who were classified as Non-Hispanic Black, Hispanic or Non-Hispanic White.

8. Demographic Characteristics
Number of Cases
N=522
Age (Mean/SD)
54.5/13.2
Race
Non-Hispanic Black
290 (55.6)
Hispanic
(25.1)
Non-Hispanic White
(19.3)
We had a total of 522 cases. The mean age was 54.5 and our population was approximately 56% Black, 25% Hispanic and 19% White.

9. Racial Distribution of Breast Cancer
This is a very different breakdown than what was estimated for the entire US for 2006, which was 84% white and for
New Jersey from SEER data, which was 89% White, and for Essex county which was 70% White. Granted
the manner in which these proportions were determined were not identical however you see the magnitude of the difference
in our population.

10. Demographic Characteristics
Marital Status
Single
230 (44.1)
Married
(22.4)
Separate/Divorced
99 (19.0)
Widow
Unknown
58 (11.1)
(3.4)
Insurance Status
No insurance
Medicaid/Medicare
Private insurance
160 (30.7)
213 (40.8)
136 (26.1)
13 (2.5)
44% of our population were single and only 26% had private insurance with the remainder having either medicaid/medicare or no insurance.

11. Clinical Characteristic
Insitu
Malignant
105 (20.1)
417 (79.9)
AJCC Stage 1 2 3 4
Unknown
(20.1)
(20.5)
(32.0)
78 (14.9)
(9.8)
(2.7)
Tumor Grade
Grade 1 (Well differentiated)
Grade 2 (Moderately differentiated)
Grade 3 (Poorly differentiated)
Undifferentiated
(27.2)
(40.0)
(0.6)
(22.4)
The majority of the cases were malignant. Approximately 73% of the patients were diagnosed at either stage 0,1 or 2.
The majority of the patients where the tumor grade was known, had Grade 3 or poorly differentiated tumors which means they
tend to grow and spread more rapidly.

12. Clinical Characteristic
Size
< 1 cm cm
> 3 cm
Unknown
(9.8)
167 (32.0)
168 (32.2)
136 (26.1)
No. of lymph nodes examined 1
2-4
> 5
186 (35.6)
(5.7)
71 (13.6)
204 (39.1)
(5.9)
No. of lymph nodes positive
Not Applicable
181 (34.7)
(7.7)
(7.3)
(4.9)
When looking at the size of tumores
Only 10% were < 1 cm in size, and for 26% of the patients the tumor size
was unknown. However, the number of unknowns is relative to the number
of insitu tumors and stage IV tumors, since these cases are not typically or
consistently given a tumor size.
36% of the patients had no lymph nodes examined. However, if lymph nodes
were examined the majority of people had > 4 examined.
Of those who had lymph nodes examined, 35% had none positive and 7% had
more than 4 positive.

13. Demographic Characteristics by Race
Black
Hispanic
White
Age (Mean/SD)
54.9 (13.3)
52.4 (11.9)*
56.1 (14.3)
Marital Status
Single
Married
Separate/
Divorced
Widow
142 (50.9)*
41 (14.7)
55 (19.7)
50 (39.1)
41 (32.0)
33 (25.8)
(3.1)
38 (39.2)
35 (36.1)
(11.3)
13 (13.4)
Insurance Status
No insurance
Medicaid/
Medicare
Private insurance
78 (27.6)
(48.4)
68 (24.0)
63 (49.6)*
38 (29.9)
26 (20.5)
19 (19.2)
38 (38.4)
42 (42.4)
* p<0.05
Now we are just going to look at these same variables stratified by race.
Hispanic patients were younger than Black patients and significantly younger than White patients.
Black patients were significantly more likely to be single (1.6 times more likely).
When we look at insurance status, Black and Hispanic patients were more likely not to have private insurance
and Hispanic women were significantly more likely to have no insurance at all compared to white patients with an OR of 4.1.

14. Clinical Characteristic by Race
Black
Hispanic
White
Insitu
Malignant
59 (20.3)
(79.7)
26 (19.8)
(80.2)
20 (19.8)
81 (80.2)
AJCC Stage 1 2 3 4
(20.7)
(18.9)
(32.6)
(15.8)
(11.9)
26 (20.2)
33 (25.6)
46 (35.7)
18 (14.0)*
(4.7)
(21.3)
(29.8)
(16.0)
(11.7)
Tumor Grade
Grade 1
Grade 2
Grade 3
Undiff
25 (10.9)
76 (33.2)
125 (54.6)
(1.3)
17 (16.2)
37 (35.2)
51 (48.6)
(0.0)
9 (12.6)
29 (40.8)
33 (46.5)
(0.0)
* p<0.05
In looking at some of the clinical characteristics we found the proportion of malignant tumors was pretty consistent across all racial/ethnic groups.
However, Hispanic women were less likely to have cancer that was stage 3 or 4 when compared to the White patients and significantly less than the Black patients. DUE TO THE N.
Black patients were more likely to have tumors that were grade 3 when compared to the Hispanic and White patients.

15. Clinical Characteristic by Race
Black
Hispanic
White
Size
< 1 cm cm
> 3 cm
26 (11.9)
97 (44.5)
95 (43.6)
19 (18.8)
40 (39.6)
42 (41.6)
(9.0)
30 (44.8)
31 (46.3)
No. of lymph nodes examined 1
2-4
> 5
(37.0)
(5.8)
35 (12.7)
123 (44.6)
43 (34.4)
(8.8)
(15.2)
(41.6)
(45.6)
(3.3)
(18.9)
(32.2)
No. of lymph nodes positive
99 (56.3)
25 (14.2)
27 (15.3)
(61.4)
(9.6)
(22.9)
(6.0)
31 (60.8)
7 (13.7)
6 (11.8)
When looking at the size of the tumor the groups look similar except the Hispanic women seem to have more
Tumors < 1 cm in size.
Black and Hispanic women were more likely to have a greater number of lymph nodes examined.
When you look at the data on how many Nodes were positive the groups look similar.

16. Clinical Characteristic by Race
Black
Hispanic
White
Alive
Yes No
(76.9)
67 (23.2)
(93.1)
(6.9)*
85 (84.2)
16 (15.8)
Advanced disease
79 (27.7)
(72.3)
24 (18.6)
(81.4)*
26 (27.7)
68 (72.3)
ERPR Status
-/-
+/-
-/+
+/+
70 (41.2)
20 (11.8)
(0.6)
79 (46.5)
27 (35.1)
(6.5)
(2.6)
43 (55.8)
(28.3)
(20.8)
(0.0)
(50.9)
When assess the vital status of these patients, Hispanic women were
significantly less likely to die than both Black (4 times less likely)
and White women (2.5 times less likely).
We were interested in looking at the differences in regards to the stage of
cancer so we categorized patients as to whether they had advanced disease
or not. Advanced disease included those patients with stage 3 or 4 disease.
As mentioned earlier, the Hispanic women were less likely to have advanced
disease than White women and significantly less likely to have advanced
disease than black women (1.6 times less likely).
The only reason it is significant for Black patients is the numbers
are greater than for White patients.
* p<0.05

17. Proportion of Breast Cancer Patients by Stage and Race
This chart shows the distribution of the diagnosed stage of disease by race.
As I mentioned earlier the Hispanic women did have less cases of
stage 3 and 4 disease.

18. Proportion of Breast Cancer Patients Below the Age of 40 by Stage and Race
However, when we looked at just those women who were <40 years of age we see a different picture.
Approximately 29% of the Black women and 38% of the Hispanic women <40 have either stage 3 or 4 disease
Where as 25% of the White women have stage 3 or 4 disease.

19. Proportion of Breast Cancer Patients by Grade and Race
When we look at the grade of tumor by race, again we see very similar
distributions with the Hispanic and Black women, with slightly more, poorly
differentiated tumors.

20. Proportion of Breast Cancer Patients Below the Age of 40 by Grade and Race
However, when we look at only those women <40 years of age the distribution again shifts. The Hispanic and Black women have a
Greater proportion with grade 3.

21. Proportion of Breast Cancer Patients by Size and Race
Looking at size of tumor, again the distributions are very similar when looking at the entire population.

22. Proportion of Breast Cancer Patients Below the Age of 40 by Size and Race
However, for the younger women the distribution again shifts with the
Hispanic and Black women <40 having more tumors >2.9 cm in size
compared to the White women <40.

23. Proportion of Breast Cancer Patients by ER/PR status and Race
If we look at Estrogen Receptor and Progesterone Receptor status the
Hispanic and Black patients have more ER-/PR- tumors compared to the
White patients.
Only had this data on 300 or 57% of the population

24. Proportion of Breast Cancer Patients Below the Age of 40 by ERPR status and Race
If we just look at those under 40 we note that only 14% of the white patients are ER- and PR-
Where as 58% and 59% of the Hispanic and Black patients are ER- and PR-, respectively.

25. Proportion of Breast Cancer Patients by Status and Race
When we look at proportion of deaths, Black patients are more likely to die than White patients
and Hispanic patients are the least likely to die.

26. Proportion of Breast Cancer Patients Below the Age of 40 by Status and Race
However, for the younger women the distribution again looks different with the Hispanic and Black women <40
both being more likely to die than the White women <40. There are no White women <40 who have died where as
approximately 16% and 13% of the Black and Hispanic women < 40 have died.

27. Conclusions
The proportion of Black, Hispanic and White women who were diagnosed with advanced stage disease varied (28%, 19% and 28% respectively).
Our population had 11.7% of the breast tumors occur in women age <40.
Our conclusions are that …..

28. Conclusions
Younger (<40) Black and Hispanic women were more likely to be diagnosed with advanced stage disease than White women (29%, 38% and 25% respectively).
Younger (<40) Black and Hispanic women were more likely to be diagnosed with a higher grade tumor, larger sized tumor, have more ER and PR negative tumors and die than White women.

29. Limitations Analysis based on cancer registry data
Limited by the variables available
In a validation study of charts from
looking at size and ER/PR status, Kappa statistic of 0.86 and 0.89 respectively
Misclassification of race/ethnicity
Race/ethnicity taken from medical record and may involve subjective appraisals by hospital personnel .
Small sample size
Although these findings are intriguing it is important to note the limitations of this
research. This analysis was based on cancer registry data which is limited by the
variables available. We did do a validation study of charts comparing
tumor size and ER/PR status and discovered excellent agreement between the
data reported in the registry and our chart review, with a Kappa statistic of 0.86
for tumor size and 0.89 for ER/PR status.
There is also the possibility of a misclassification of race/ethnicity.
Race and ethnicity in many instances is taken from medical records
and may involve subjective appraisals by hospital personnel
Another limitation is that our numbers were small and our cell size
was even smaller when we stratified by age.

30. Future Research
Approaching the New Jersey Department of Health and Senior Services to expand our look at the registry data from all of New Jersey for
Continuing to do patient chart reviews to validate registry data and to obtain more information on tumor biology and socioeconomic status.
What are our plans for the future:
To approach the New Jersey Department of Health and Senior Services to
expand our look to include all newly diagnosed cases of breast cancer in
New Jersey. This way we would be able to look at geographical variations
in our findings.
And to continue trying to do patient chart reviews to further validate the
registry data and to obtain more information on tumor biology
and socioeconomic status.