Breast Cancer Survival among New Mexico Hispanic, American ... · Mexico Hispanic and non-Hispanic white females and New Mexico and Arizona American Indian resident females between

Received 12/1 3/95: accepted 6/28/96.

The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement in

accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

I This research was supported by the New Mexico MEDTEP Research Center for

Ethnic Populations, H507389-02, School of Medicine, University of New Mex-

ico, Albuquerque, New Mexico.

2 To whom requests for reprints should be addressed, at The Lovelace Institutes,

2425 Ridgecrest Drive Southeast, Albuquerque, New Mexico 87108.

3 The abbreviations used are: HMO, health maintenance organization; SEER.

Surveillance, Epidemiology, and End Results; NMTR, New Mexico SEER Tu-

mor Registry.

Vol. 5, 861-866, November 1996 Cancer Epidemiology, Biomarkers & Prevention 86/

Breast Cancer Survival among New Mexico Hispanic, American Indian,

and non-Hispanic White Women (1973-1992)’

Floyd Frost,2 Kristine Tollestrup, William C. Hunt,Frank Gilliland, Charles R. Key, andChristopher E. Urbina

Institute for Health and Population Research, The Lovelace Institutes.

Albuquerque, New Mexico 87108 [F. F.]; Center for Population Health,

University of New Mexico, Albuquerque, New Mexico [K. T., C. E. UI; and

New Mexico Tumor Registry, University of New Mexico Cancer Center,

Albuquerque, New Mexico [W. C. H., F. G., C. R. K.l

Abstract

A study of breast cancer survival was conducted amongNew Mexico Hispanic and non-Hispanic white womenand New Mexico and Arizona American Indian womendiagnosed between 1973 and 1992. The goals were todetermine whether, after adjusting for first treatmentand the extent of disease at diagnosis, American Indianand Hispanic women had poorer survival than non-Hispanic whites and, if survival had improved over time,whether comparable improvements had been made forthe three racial/ethnic groups. Five-year relative survivalrates were calculated, and a Cox proportional hazardsmodel was constructed to compare survival betweenraces/ethnicities, adjusting for first treatment and theextent of disease at diagnosis. Findings indicate thatduring 1983-1992, breast cancer was more commonlydetected at a local stage for all three groups compared to1973-1982. Five-year relative survival improved for non-Hispanic white and American Indian women with local orregional disease, but the improvement was statistically

significant only for non-Hispanic white women and forAmerican Indian women with local disease. Despiteearlier stages at diagnosis, Hispanic females showed lessimprovement in overall or stage-specific survival thannon-Hispanic whites. The Cox model indicated thatAmerican Indian women experienced poorer survivalthan non-Hispanic whites during both time periods.Survival of Hispanic women with breast cancer wascomparable to non-Hispanic whites during 1973-1982 butwas significantly worse during 1983-1992. Thesignificance of this lower survival is amplified byincreasing breast cancer incidence among New MexicoHispanics and American Indians.

Introduction

Lacking an approach to primary prevention of breast cancer,interventions have focused on improving early detection andoptimizing treatment. As part of efforts to monitor improve-

ments in early detection and delivery of treatment in minoritypopulations, studies have examined the stage of disease at

diagnosis and breast cancer survival rates in different racial andethnic populations (1-5). Studies indicate that, compared tonon-Hispanic whites, African Americans, Hispanics, and

American Indians have more advanced breast cancer at the timeof diagnosis (1-6). Low socioeconomic status has been sug-gested as a cause for much of the advanced stage at diagnosis

and poor survival (7-10); however, other studies have ques-tioned these conclusions (1 1, 12).

Although certain minority groups have proportionatelymore cases with advanced stage of disease at diagnosis, previ-

ous studies have not uniformly reported poorer survival forminority women with breast cancer when adjusted for the

extent of the disease at diagnosis. In one hospital-based study,

no evidence was found for reduced survival of African Amer-icans compared to whites (13). An intervention study found that

participation in a HMO3 breast cancer screening program re-suited in the elimination of racial differences in breast cancersurvival rates (14). In New Mexico, for the period 1969-1982,American Indians were found to have poorer survival rates thannon-Hispanic whites, after adjusting for both the stage of dis-ease at the time of diagnosis and whether cancer-directed ther-

apy was provided (4). In contrast to American Indians, NewMexico Hispanics diagnosed between 1969 and 1983 did notshow statistically significant reduced survival compared to non-

Hispanic whites, adjusting for both stage of disease and treat-ment (4).

The ethnically and racially diverse population of NewMexico and eastern Arizona, the availability of a high-quality

regional SEER tumor registry data spanning a 20-year period,and the high fraction of low-income residents without adequatehealth insurance coverage make this area well suited for a studyof breast cancer survival rates in different racial and ethnicgroups. A study of breast cancer survival rates in non-Hispanicwhite, Hispanic, and American Indian women was completedusing New Mexico SEER tumor registry data from 1973 to

1992. The purposes ofthis study were to determine: (a) whetherbreast cancer survival has improved over time; (b) whether

breast cancer survival rates for Hispanic and American Indianwomen are lower than for non-Hispanic white women, afteradjusting for the stage of disease at diagnosis and whether

cancer-directed surgery, radiation, and chemotherapy were re-ceived; and (c) whether progress has been made toward relative

Association for Cancer Research. by guest on October 3, 2020. Copyright 1996 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

https://bloodcancerdiscov.aacrjournals.org

862 Breast Cancer Survival among New Mexico Women

equality in survival from female breast cancer among the threeracial/ethnic groups.

Subjects and Methods

Breast cancer incidence, survival, demographic, and diagnosticdata for women were obtained from the NMTR maintained bythe University of New Mexico. Demographic and diagnostic

data, used to predict survival, were restricted to information thatwas reliably recorded for the entire study period (1973-1992).

Demographic information obtained from the SEER tumor reg-istry included age, race/ethnicity, marital status, and whether

the woman resided in predominantly urban Bernalillo County.Diagnostic information included the year of diagnosis, sum-

mary stage of disease at diagnosis, and number of days betweendiagnosis and first treatment. Treatment data included whethercancer-directed surgery, radiation, or chemotherapy had beenreceived. Treatment was coded negative if it was known not tohave been received or if it was unknown whether or not it wasreceived. Radiation therapy was coded as given if it was re-

ceived before, after, or both before and after surgery. Chemo-therapy was coded as given if single or multiple agents wereused or if chemotherapy was provided but the agents were notspecified (15). Outcome information included whether thewoman was alive or dead at the time of last follow-up, the

number of months of known survival, whether the woman haddied, and the cause of death. A review of the dates of last

contact for women not known to have died showed that the

follow-up rates were similar for the three racial/ethnic groups.Information on whether the breast cancer was first identifiedfrom death records or at autopsy, and whether a prior tumor(malignant or in situ) had been diagnosed was also obtained foreach case from the SEER tumor registry.

The NMTR information on the race and ethnicity forwomen diagnosed with breast cancer, which is obtained from

hospital records, is comparable to the United States Bureau of

the Census racial/ethnic identifiers. When no ethnicity infor-mation was available, Hispanics were identified using the 1980

census list of Spanish surnames (16-18). American Indianswere identified primarily by statements in the hospital records,

but place of residence (e.g. , pueblo or reservation) and treatinghospital (e.g., Indian Health Service hospitals) were also usedas secondary identifiers. Non-Hispanic whites were identifiedby specific designation in records or by exclusion from otherracial/ethnic groups.

The vital status and cause-of-death information is obtainedby the NMTR annually from the patient’s primary physician.Registry records are linked periodically with death records fromthe New Mexico Bureau of Vital Statistics and with the Na-tional Center for Health Statistics National Death Index.

Survival analyses were restricted to non-Hispanic white,Hispanic, and American Indian women diagnosed with malig-nant breast cancer between January 1, 1973, and December 31,1992. Only women who were New Mexico residents or Amer-ican Indian Arizona residents at the time of diagnosis and werenot first identified through death records or at autopsy wereincluded in the study. The study was further restricted towomen who had not been diagnosed previously with a malig-

nant or in situ cancer of any type and who had a known stageof disease at diagnosis. Persons with a previous cancer diag-nosis were excluded because the presence of another cancermay bias survival rate comparisons between racial/ethnicgroups.

Relative survival rates expressed as a percentage werecalculated using the 1980 United States female age-race/eth-

nicity-specific mortality rates (19). This calculation comparedobserved 5-year survival rates for all New Mexico non-Hispanic white and Hispanic females and New Mexico and

Arizona American Indian females with breast cancer to the

expected survival based on 1980 age-race/ethnicity-specificmortality rates for United States females (20).

Statistical analyses were conducted using SPSS-PC+(version 4.0, Chicago, IL). Proportional hazard survival anal-

ysis was completed with EGRET software (1993 version;Seattle, WA) using the Cox proportional hazards model (21).Because efforts to improve early detection through mammog-raphy screening have been increasing during recent years in

New Mexico, separate relative and Cox proportional hazardssurvival analyses were performed for the periods 1973-1982

and 1983-1992.

Independent variables in the Cox proportional hazards

model included: race, ethnicity, stage of disease, marital status,time between diagnosis and first treatment, whether the caseresided in predominantly urban Bernalillo County, and type oftherapy. Dummy variables were created for race, ethnicity,stage of disease, marital status, and type of therapy. The end

points included death from breast cancer and death from any

cause. Analysis of survival from all causes of death was done

to determine whether comorbidity or changes in misclassifica-tion of cause of death may have affected the breast cancersurvival analysis. Survival was measured in months since di-agnosis. The initial Cox proportional hazards model was con-

structed using age, year of diagnosis, stage of disease, race/

ethnicity, and type oftreatment. Other predictive variables suchas marital status, time between diagnosis and first treatment,and whether the case resided in urban Bernalillo County wereentered stepwise to determine whether any of these predictedduration of survival, given the first set of factors included in theequation. Separate analyses were done for the periods I 973-1982 and 1983-1992 using the same predictive variables.

Results

A total of 1 1,769 breast cancers was diagnosed among NewMexico Hispanic and non-Hispanic white females and NewMexico and Arizona American Indian resident females between

1973 and 1992. This included 450 American Indians, 2519Hispanics, and 8800 non-Hispanic whites. Excluding cases firstidentified through death records or at autopsy (2.6% of non-Hispanic white cancers, 2.3% of Hispanic cancers, and 3.6% ofAmerican Indian cancers), cases with an unknown stage of

disease at diagnosis (2.2% of non-Hispanic white cancers, I .9%of Hispanic cancers, and 1 .7% of American Indian cancers),cases with in situ breast cancer (8. 1% of non-Hispanic whitecancers, 7.2% of Hispanic cancers, and 5. 1% of AmericanIndian cancers), and women with a prior cancer (12.3% ofnon-Hispanic white cancers, 9.4% of Hispanic cancers, and

6.0% of American Indian cancers) reduced the total to 9414(Table 1). Of these, 426 were American Indian, 2068 were

Hispanic, and 6920 were non-Hispanic white. The stages ofdisease at diagnosis for these 9414 cases were as follows: 5302with local disease, 3417 with regional disease, and 695 withdistant metastases.

Differences are apparent between the racial/ethnic groupsin the age and stage of disease at diagnosis. Hispanic andAmerican Indian women diagnosed with breast cancer wereyounger than non-Hispanic whites (Fig. 1). Forty-four % of

American Indian breast cancers were found in women underage 50 compared to 35% for Hispanics and 22% for non-Hispanic whites (P < 0.0001). For both 1973-1982 and 1983-



<40 40-49 50-59 60.69 70+ <40 40.49 50.59 60-69 70+

Cancer Epidemiology, Biomarkers & Prevention 863

Table 1 Breast cancers by race/eth nicity, pe nod of diag nosis, and st age of disease

Total no. cancersLocal Regional Distant

No. % No. % No. %

1973-1982

Non-Hispanic white 2702 1503 56 982 36 2 17 8

Hispanic 757 377 50 300 40 80 11

American Indian 160 62 39 75 47 23 14

All races 3619

1983-1992

Non-Hispanic white 4218 2527 60 1436 34 255 6

Hispanic 1311 695 53 513 39 103 8

Americanlndian 266 138 52 111 42 17 6

All races 5795

Total (1973-1992)

Non-Hispanic white 6920 4030 58 2418 35 472 7

Hispanic 2068 1072 52 813 39 183 9

American Indian 426 200 47 186 44 40 9

All races 9414

Fig. 1. Age at diagnosis by race/ethnicity. fl, non-

Hispanic white; #{149},Hispanic; [1, American Indian.

50

40

30

20

10

0

1992, Hispanic and American Indian women had more ad-vanced disease at the time of diagnosis than non-Hispanicwhites (Table 1; P < 0.001). American Indians were diagnosed

at a more advanced stage of disease than were Hispanics (P <0.02). Compared to the period 1973-1982, non-Hispanic white,Hispanic, and American Indian women diagnosed during 1983-1992 were diagnosed at a less advanced stage of disease (P <

0.001; Table 1).

The fraction of women receiving cancer-directed therapyincreased between the two time periods (Table 2). In particular,

the fraction receiving surgery increased for all racial/ethnic

groups (P < 0.001). During each time period, a smaller fractionof American Indian women with breast cancer received surgery

than did either Hispanic or non-Hispanic white women withbreast cancer (P < 0.02).

The 5-year relative survival rate for all New Mexicoresidents diagnosed with breast cancer between 1973 and 1982was 74%. Non-Hispanic whites had the highest survival of75%, while Hispanics had a 71% relative survival (Table 3).American Indians had a survival rate of only 47%. Five-yearrelative survival improved substantially between 1973-1982and 1983-1992 for non-Hispanic whites and American Indians.The overall 5-year relative survival rate increased to 80%. For

non-Hispanic whites it was 83%, while for American Indians itimproved to 69% (Table 3). Five-year relative survival forHispanics increased slightly from 71% in 1973-1982 to 74% in1983-1992 (Table 3).

Five-year relative survival from local disease increased in

all three groups during the most recent time period (Table 3).Differences between the racial/ethnic groups were most appar-

ent for women with regional disease. For regional disease,modest improvements in survival between periods (1973-1982versus 1983-1992) for non-Hispanic whites and American In-dians were observed, but the small number of American Indianswith breast cancer resulted in wide confidence intervals for the

relative survival estimates. The estimated 5-year relative sur-vival for Hispanic women with regional disease was unchanged

between 1973-1982 and 1983-1992, whereas for those withdistant disease it declined. The decline in survival for Hispanicwomen was not statistically significant.

Results of the Cox proportional hazards analyses for breast

cancer as the cause of death are given in Table 4. The maritalstatus of the case, a greater than 50-day delay between diag-nosis and first treatment, and whether the case resided inBernalillo County were not consistently predictive of increased

or reduced survival rate during the two time periods (1973-




Table 2 Cancer treatment for breast cancers by rac c/ethnicity and period of

diagnosis

Type of cancer-

directed treatmentl971-l9W�

1983-1992

% cases (95% CI)”

Surgery

Non-Hispanic white 89% (87.9-90.1%) 96% (95.4-96.6%)

Hispanic 89% (86.6-9l.4�T) 96% (94.9-97.1%)

American Indian 81% (74.5-87.5%) 92% (88.9-95.1%)

Chemotherapy

Non-Hispanic white 18% (16.5-19.5% 24% (22.7-25.3%)

Hispanic 24% (20.8-27.2%) 32% (29.5-34.5%)

American Indian 37% (29.0-45.0%) 38% (32.3-43.7%)

Hormone therapy


Hispanic 12% (9.6-14.4%) 30% (27.6-32.4%)

American Indian 20% (13.3-26.7%) 30% (24.7-35.3%)

Radiation therapy


Hispanic 25% (21.8-38.2%) 28% (25.6-30.4%)

American Indian 33% (25.2-40.8%) 21% (16.3-25.7%)

‘, Cl, confidence interval.

1982 and 1983-1992). For this reason, they were not includedin the final survival model. During both time periods, the

hazard ratio for American Indian race was elevated signifi-cantly after adjusting for age, stage of disease, and treatment.Hispanic ethnicity, adjusted for stage of disease at diagnosis

and treatment received, was not associated with an elevatedhazard ratio during the period 1973-1982 (Table 4). However,

during 1983-1992, Hispanic ethnicity was predictive of re-duced survival, compared to non-Hispanic whites. A morerecent year of diagnosis was also predictive of increased sur-vival for this earlier time period but not for the later period. The

overall survival from all causes of death was also shorter forAmerican Indians during both time periods and for Hispanics

during the later time period.

Discussion

This study has found an improvement in the stage at diagnosisof breast cancer during 1983-1992 for non-Hispanic white,

Hispanic, and American Indian women compared to the period1973-1982. The fraction of American Indian women diagnosedwith distant disease during the period 1983-1992 declined from14 to 6%, a percentage comparable to that of non-Hispanicwhites for the same time period. The fraction of women with

breast cancer receiving cancer-directed surgery also increased.

These trends should predict better breast cancer survival. Infact, for all races, 5-year relative survival for females with

breast cancer increased from 74% in 1973-1982 to 80% in1983-1992 (P < 0.05). The 5-year survival rates of both

non-Hispanic whites and American Indians showed strikingimprovements between 1973-1982 and 1983-1992. The 5-yearstage-specific relative breast cancer survival rates for localdisease also improved for both non-Hispanic whites and Amer-

ican Indians from 1973 to 1992. Improved survival for non-Hispanic white and American Indian women with regionaldisease was not statistically significant. For Hispanics, stage-specific female breast cancer survival improved between 1973-1982 and 1983-1992, but the difference was not statistically

significant.Adjusting for age, stage of disease, and treatment, the Cox

proportional hazards analysis reconfirmed an earlier findingthat prior to 1983, New Mexico American Indians with breast

Table 3 Five-year relative survival from breast cancer death by period, stageof disease. and race/ethnicity

Five-year relative survival

(95% confidence interval)

1973-1982 1983-1992

All stages

Non-Hispanic white 75% (73-77%) 83% (8 1-85%)

Hispanic 71% (68-75%) 74% (7 1-77%)

American Indian 47% (39-55%) 69% (62-76%)

Local disease

Non-Hispanic white 85% (83-88%) 93% (91-95%)

Hispanic 86% (82-90%) 90% (86-94%)

American Indian 65% (52-78%) 89% (80-97%)

Regional disease


Hispanic 63% (57-69%) 63% (58-68%)

American Indian 43% (31-55%) 53% (41-66%)

Distant disease


Hispanic 20% (10-29% 14% (6-22%)

American Indian 9% (0-22%)

‘, Only 17 cases.

cancer had shorter survival times than either Hispanic or non-

Hispanic white females with breast cancer. Despite earlierdetection and fewer untreated women with breast cancer for

each racial/ethnic group, during the period 1983-1992, Amer-ican Indian females with breast cancer continued to have poorersurvival than did non-Hispanic whites. This difference persisted

even after adjusting for the stage of disease at diagnosis, tumorstage, and whether treatment was received. Because pathologylaboratories in New Mexico are generally unaware of the race/ethnicity of cancer cases, it is unlikely that differential staging

occurred by race/ethnicity or that staging of breast cancersamong Hispanics had become worse during the study period.

The widening gap in survival for Hispanic women withbreast cancer relative to non-Hispanic whites is also a concern.

Although 5-year relative survival for Hispanics with local dis-

ease may have improved slightly, survival for Hispanic womenwith regional or distant disease showed little evidence of im-provement. The public health problem represented by the rel-

atively shorter survival for New Mexico Hispanic and Ameri-can Indian women with breast cancer is amplified by anincrease in breast cancer incidence among New Mexico His-

panic and, to a lesser extent, American Indian women. NewMexico age-adjusted Hispanic breast cancer incidence in-creased by a factor of 2.2 between 1969-1972 and 1989-1992.

Between 1968-1972 and 1982-1987, Hispanic female breastcancer mortality increased by a factor of 1.25 (22). For Amer-ican Indians, breast cancer incidence remains lower than for

either Hispanics or non-Hispanic whites, but it increased by a

factor of 1.2 between 1968-1972 and 1989-1992 (22), with acomparable increase in mortality. In addition, because a higher

fraction of southwest Hispanic and American Indian women areyoung, a higher fraction of Hispanic and American Indianbreast cancers are occurring in younger women.

Differences in survival between the racial/ethnic groupsmight be expected to be greatest for women with local diseaseif access to screening mammography varies greatly among

these groups. After the initial screen, screening mammographywould be expected to detect more local disease at an earlierstage. Survival should be higher in groups with better access tomammograms. Differences in the 5-year relative survival forwomen with local disease was small for the three groups



Cancer Epidemiology, Biomarkers & Prevention 865

Table 4 Cox pro portional hazard analysis

VariableBreast cance r deaths All deaths

Hazard ratio P value Hazard ratio-�

P value

Diagnosed 1973-1982

Age group 1.003 0.824 1.132 <0.001

Hispanic (yes:no) I .050 0.469 0.974 0.625

American Indian (yes:no) 1.533 <0.001 1.424 <0.001

Year of diagnosis 0.979 0.036 0.991 0.249

Regional disease (yes:no) 2.228 <0.001 1.664 <0.001

Distant disease (yes:no) 5.832 <0.001 3.864 <0.001

Surgery (yes:no) 0.347 <0.001 0.367 <0.001

Radiation therapy (yes:no) I .205 0.002 1 . I 26 0.015

Chemotherapy (yes:no) 1.162 0.029 0.997 0.958

Hormone therapy (yes:no) 1.241 0.010 1.197 0.013

Diagnosed 1983-1992

Variable

Age group 1.025 0.157 1.125 <0.001

Hispanic (yes:no) 1.432 <0.001 1.280 <0.001

American Indian (yes:no) 1.695 <0.001 1.621 <0.001

Year of diagnosis 0.980 0.197 0.994 0.627

Regional disease (yes:no) 2.755 <0.001 2.082 <0.001

Distant disease (yes:no) 12.480 <0.001 7.484 <0.001

Surgery (yes:no) 0.328 <0.001 0.353 <0.001

Radiation therapy (yes:no) I .094 0.2 17 1 .033 0.578

Chemotherapy (yes:no) 1.598 <0.001 1.215 0.003

Hormone therapy (yes:no) 1.013 0.863 0.964 0.547

(89-93%), suggesting that the effect of lead-time bias may beminimal.

Earlier detection of breast cancer suggests that each racialgroup should experience better breast cancer survival. It isunclear whether survival from breast cancer continued to im-prove throughout the period 1983-1992. Year of diagnosis

during the period 1983-1992 was not statistically significant inpredicting better survival. It is possible that the maximal effectsof early detection from screening mammography on breast

cancer survival in New Mexico will not be apparent for severalmore years. Use of screening mammography has dramaticallyincreased nationally over the past 3-4 years, and a statewide

program to provide mammography screening to low-incomewomen in New Mexico just began 3 years ago. Therefore,sufficient time may not have not elapsed to detect survival fora large number of women with breast cancer detected throughscreening mammography.

A number of uncontrolled factors may have influenced thefindings of the survival analyses of this study. The cancer-

directed treatment ascertained by the SEER registry only ex-

amined the initial course of therapy, and, therefore, the measureof cancer therapy provided was crude. No information was

available on follow-up care, and underascertainment of canceradjuvant therapy is likely.

The observation that Hispanic and American Indian fe-

males with breast cancer have poorer survival should be ex-

plored further. Improved early detection and increased use ofsurgery would suggest that overall and stage-specific survivalshould have increased for Hispanic females. Comparisons of

breast cancer survival between Hispanic and non-Hispanicwhite female HMO members could be useful in determiningwhether other uncontrolled risk factors or lack of health care

access contribute to poor survival. In a study of a HMO pop-ulation where comparable screening and health care servicesare available to all members, confounding effects of screening,

disease staging, and treatment access could be eliminated. Un-

detected treatment differences may exist between the raciaL’

ethnic groups, because the current study only adjusted forwhether the person received cancer-directed therapy. Follow-up

studies might examine whether the treatment plan was appro-priate and whether therapy was provided in a timely and effec-tive manner. Because most New Mexico American Indianshave access to the Indian Health Service facilities, studies of the

adequacy of treatment for American Indians could be donethrough a single agency. To examine the role of comorbidconditions, which may result in death from causes other than

breast cancer, an analysis of survival from all causes of deathwas conducted and did not alter the findings. However, linkage

between the SEER tumor registry data and the National Centerfor Health Statistics multiple cause-of-death data could be

useful in assessing the role of comorbid conditions. Differencesin the prevalence of other health conditions could explain someof the racial/ethnic differences in breast cancer survival.

References

I . Axtel, L. M.. and Meyers, M. H. Contrasts in survival of Black and White

cancer patients, 1960-73. J. NatI. Cancer Inst.. 60: 1209-1215. 1978.

2. Young. J. L., Ries, L. G., and Pollock, E. S. Cancer patient survival among

ethnic groups in the United States. J. Natl. Cancer Inst.. 73: 341-352, 1984.

3. Le Marchand, L. Ethnic variation in breast cancer survival: a review. Breast

Cancer Res. Treat., 18: S119-S126, 1991.

4. Samet, J. M., Key, C. R., Hunt, W. C., and Goodwin, J. S. Survival of

American Indian and Hispanic cancer patients in New Mexico and Arizona.

1969-82. J. NatI. Cancer Inst., 79: 457-463, 1987.

5. Eley, W. J., Hill, H. A., Chen, V. W., Austin, D. F., Wesley. M. N., Muss. H.B., Greenberg, R. S., Coates, R. J., Correa, P., Redmond, C. K.. Hunter, C. P..Herman, A. A., Kurman, R., Blacklow, R.. Shapiro, S., and Edwards, B. K. Racial

differences in survival from breast cancer: results of the National Cancer Institute

black/white cancer study. J. Am. Med. Assoc., 272: 947-954, 1994.

6. Richardson. J. L., Langholz, B., Bernstein, L., Burciaga, C.. Danley, K.. andRoss, R. K. Stage and delay in breast cancer diagnosis by race, socioeconomic

status, age and year. Br. J. Cancer 65: 922-926. 1992.

7. Muss, H. B., Hunter, C. P., Wesley, M., Correa. P., Chen, V. W., Greenberg.

R. S., Eley, J. W., Austin, D. F., Kurman, R., and Edwards. B. K. Treatment plansfor black and white women with stage II node-positive breast cancer. Cancer

(Phila.), 70: 2460-2467, 1992.




8. McWhorter, W. P., and Mayer, W. J. Black/white differences in type of initial

breast cancer treatment and implications for survival. Am. J. Public Health, 77:

1515-1517, 1987.

9. Canon, A. G., Semwogerere, A., Lamont, D. W., Hole, D. J., Mallon, E. A.,

George, W. D., and Gillis, G. R. Relation between socioeconomic deprivation and

pathological prognostic factors in women with breast cancer. Br. Med. J., 309:1054-1057, 1994.

10. Ansell, D., Whitman, S., Lipton, R., and Cooper, R. Race, income and

survival from breast cancer at two public hospitals. Cancer (Phila.), 72: 2974-

2978, 1993.

1 1 . Kogevinas, M., Marmot, M. G., Fox, A. J., and Goldblatt, P. 0. Sociocco-

nomic differences in cancer survival. J. Epidemiol. Community Health, 45:

216-219, 1991.

I 2. Klein, W., and Metter, G. Survival of cancer patients by economic status in

a free care setting. Cancer (Phila.), 55: 1552-1555, 1985.

13. Mitra, N. K., Rush, B. F., and Vemer, E. A comparative study of breast

cancer in the black and white populations of two inner-city hospitals. J. Surg.

Oncol., 15: 11-17, 1980.

14. Shapiro. S., Venet, W., Strax, P., Venet, L., and Roeser, R. Prospects for

eliminating racialdifferences in breast cancer survival rates. Am. J. Public Health,

72: 1142-1145, 1982.

15. NIH. The SEER Program Code Manual. NIH Publication No. 92-1999.

Bethesda, MD: NIH, 1992.

16. Buechley, R. W. A reproducible method of counting persons of Spanish

surname. J. Am. Stat. Assoc., 56: 88-97, 1961.

17. Howard, C. A., Samet, J. M., Beuchley, R. W., Schrag, S. D., and Key, C. R.

Survey research in New Mexico Hispanics: some methodological issues. Am. J.

Epidemiol., 11 7: 27-34, 1983.

18. Rosenwaike, I. The status ofdeath statistics for the Hispanic population of the

Southwest. Soc. Sci., 69: 722-736, 1988.

19. Ederer, R., Axtel, L. M., and Cutler, S. J. The relative survival rate: a

statistical methodology. Natl. Cancer Inst. Monogr., 6: 101-121, 1961.

20. NIH. SEER Program Self-Instructional Manual for Cancer Registrars, Book

7, Statistics and Epidemiology for Cancer Registrars. NIH Publication No. 94-

3766. Bethesda, MD: NIH, 1994.

21. Cox, D. R. Regression models and life tables. J. R. Stat. Soc. B., 34: 187-220,

1972.

22. Wiggins, C. L., Becker, T. M., Key, C. R., and Samet, J. M. Cancer mortalityamong New Mexico’s Hispanics, American Indians, and non-Hispanic whites,

1958-87. J. Natl. Cancer Inst., 85: 1670-1678, 1993.



1996;5:861-866. Cancer Epidemiol Biomarkers Prev F Frost, K Tollestrup, W C Hunt, et al. Indian, and non-Hispanic white women (1973-1992).Breast cancer survival among New Mexico Hispanic, American

Updated version

http://cebp.aacrjournals.org/content/5/11/861

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/5/11/861To request permission to re-use all or part of this article, use this link



https://bloodcancerdiscov.aacrjournals.orgcgi/alerts

mailto:[email protected]



Documents

Breast Cancer Survival among New Mexico Hispanic, American ... · Mexico Hispanic and non-Hispanic white females and New Mexico and Arizona American Indian resident females between