Fax: (212) 717-3056
Carcinoma detection at the breast examination center of Harlem
Version of Record online: 28 JUN 2002
Copyright © 2002 American Cancer Society
Volume 95, Issue 1, pages 8–14, 1 July 2002
How to Cite
Liberman, L., Freeman, H. P., Chandra, S., Stein, A. L., Mccord, C., Godfrey, D. and Dershaw, D. D. (2002), Carcinoma detection at the breast examination center of Harlem. Cancer, 95: 8–14. doi: 10.1002/cncr.10640
- Issue online: 28 JUN 2002
- Version of Record online: 28 JUN 2002
- Manuscript Accepted: 4 FEB 2002
- Manuscript Revised: 28 JAN 2002
- Manuscript Received: 2 JAN 2002
- breast carcinoma;
- breast neoplasms
Breast carcinoma is one of the leading causes of excess mortality rates in Harlem, an inner-city neighborhood with the highest mortality rates and worst life expectancy in New York City. This study reports the results of a breast carcinoma screening and diagnostic program in Harlem.
Retrospective review was performed of a database of 49,750 visits to the Breast Examination Center of Harlem from 1995 to 2000. During this period, 181 breast carcinomas were diagnosed in 178 women. The medical records of these 178 women were reviewed to determine the method of detection, stage, and treatment.
Among these women, 89% were black or Hispanic, 45% had no medical insurance, and 38% had incomes below federal poverty guidelines. Breast carcinoma stage, known for 167 carcinomas, was Stage 0 in 38 (23%), Stage I in 38 (23%), Stage II in 63 (38%), Stage III in 24 (14%), and Stage IV in 4 (2%). Fifty-six cases (34%) were minimal breast carcinomas. Of 181 breast carcinomas, 122 (67%) were palpable and 59 (33%) were nonpalpable, detected only by mammography in asymptomatic women. Nonpalpable, as opposed to palpable, breast carcinomas were significantly more likely to be ductal carcinoma in situ (30 of 55 [54%] vs. 8 of 112 [7%], P < 0.0000001) or minimal breast carcinoma (39 of 55 [71%] vs. 17 of 112 [15%], P = 0.0000001) and were more likely to be treated with breast-conserving surgery (47 of 56 [84%] vs. 76 of 110 [69%], P < 0.04).
A breast carcinoma screening and diagnostic program has been established in Harlem, a traditionally underserved area in New York City. Early, curable breast carcinomas were detected but outreach remains a challenge, particularly for the uninsured. Cancer 2002;95:8–14. © 2002 American Cancer Society.
A decade ago, McCord and Freeman1 described the excess mortality rates in Harlem, an inner-city neighborhood in New York City in which 96% of the inhabitants were black and 41% lived below the poverty line. With U.S. white mortality ratios as the standard, the standardized mortality ratios for deaths for people younger than the age of 65 years in Harlem were 2.91 for males and 2.70 for females. The chief causes of excess mortality rates were cardiovascular disease, cirrhosis, homicide, and neoplasms; neoplasms were the second most important cause of death for people younger than the age of 65 years. Of 353 health areas in New York, 54 (with a total population of 650,000) had mortality rates for persons younger than 65 years old that were at least twice the expected rate; all but one of these high-mortality areas were predominantly black or Hispanic. The authors concluded that Harlem and other urban areas with largely black populations have extremely high mortality rates that justify special consideration analogous to that given to natural disaster areas.
Breast carcinoma is one of the leading causes of excess mortality rates among women in Harlem. Until the 1980s, most breast carcinomas in Harlem were diagnosed in hospitals, the largest of which was Harlem Hospital Center. Freeman and Wasfie2 reviewed the findings and outcome for 708 patients with breast carcinoma who were diagnosed, treated, and/or followed between 1964 and 1986 at Harlem Hospital Center; of these women, 94% were black and almost 50% had no medical insurance. Breast carcinoma treatment included surgery in 512 (72%) patients and radiotherapy and/or chemotherapy alone in 94 (13%) patients; 102 (14%) patients refused treatment or died before it could be administered. Among 437 patients with staging data available, breast carcinoma was Stage I in 27 (6%), Stage II in 195 (45%), Stage III in 171 (39%), and Stage IV in 44 (10%). Of 512 patients who had surgery, a mastectomy was performed on 479 (94%) patients and breast-conserving surgery was performed on 33 (6%) patients. Among patients who had surgery, 10-year survival rates were 54%, 35%, 18%, and 0% for Stages I, II, III, and IV, respectively. For all patients, the observed 5-year survival rate was 39%.
The Breast Examination Center of Harlem is an outpatient facility affiliated with a cancer center in New York City. It was established to combat the high breast carcinoma mortality rates in Harlem by providing ready access to early detection, diagnosis, and treatment. The purpose of this study was twofold: 1) to review our experience with breast carcinomas diagnosed at the Breast Examination Center of Harlem with respect to method of detection, size and stage at diagnosis, and factors associated with detection of early disease and 2) to compare stage and treatment of carcinomas diagnosed at the Breast Examination Center of Harlem to those previously diagnosed at Harlem Hospital.
MATERIALS AND METHODS
The Breast Examination Center of Harlem was established in 1979 as a breast screening and diagnostic facility in Harlem and has been an outreach program of the Memorial Sloan-Kettering Cancer Center (MSKCC) since 1980. Retrospective review was performed of a database of 49,750 visits to the Breast Examination Center of Harlem from January 1995 through August 2000. During this period, 181 breast carcinomas were diagnosed in 178 women. These 178 women constitute the basis of this study.
Staff and Facilities
The Breast Examination Center of Harlem occupies 5200 square feet of space in the State Office Building on 125th Street in New York City. The current on-site staff of the Breast Examination Center includes a director and three additional administrators, six session assistants, four individuals who work in medical records, four nurse practitioners, two mammography technologists, and one patient educator, as well as two on-site patient navigators who help patients requiring follow-up negotiate the complexities of obtaining appropriate care.3 Three physicians specializing in breast surgery each have office hours at the Breast Examination Center one afternoon per week. The Center is open from 8 a.m. to 6 p.m. on weekdays. Facilities include a patient education room with reading materials and videotapes, as well as a children's play area with books and toys to facilitate the process for mothers who do not have alternate child care.
All women visiting the Breast Examination Center have a physical examination performed by a nurse clinician. Symptomatic women also see one of the three physicians specializing in breast surgery. Screening mammography is performed for asymptomatic women who are 40 years and older and for women younger than 40 who are at high risk, such as women with previous breast carcinoma, biopsy-proven diagnosis of atypical ductal hyperplasia or lobular carcinoma in situ, or who had previous chest radiation for Hodgkin disease. Screening mammography is also performed in asymptomatic women younger than 40 who have a first-degree relative who had breast carcinoma before the age of 50; in those women, screening begins at an age that is 10 years younger than the age of the first-degree relative when she was diagnosed with breast carcinoma (but not before age 25 years).4, 5 Mammography is also performed in women with suspicious findings on physical examination.
Three mammography units (LoRad, Danbury, CT) are present on site. The screening mammogram consists of the standard craniocaudal and mediolateral oblique views of each breast. These films are interpreted the next working day by an attending radiologist specializing in breast imaging at MSKCC; if additional studies are needed or if stereotactic or ultrasound-guided biopsies are warranted, they are usually performed at MSKCC. If surgery is needed, it is usually performed by one of three attending surgeons affiliated with the Breast Examination Center of Harlem, although occasionally patients have surgery at MSKCC or choose to be referred elsewhere.
Patient recruitment for the Breast Examination Center of Harlem is accomplished through several routes, including personal visits to church groups and other community organizations. Several breast carcinoma survivors who were diagnosed at the Breast Examination Center of Harlem actively recruit women to visit the Center. In addition, advertisements are made in the local newspapers.
For women who have medical insurance, the insurance is billed for services rendered at the Breast Examination Center. The staff at the Center assists women requiring evaluation in applying for insurance (usually Medicaid) if they lack insurance but qualify. For women who neither have insurance nor qualify for it, services are provided free of charge, supported by the Healthy Women's Partnership of the New York State Department of Health as part of the National Breast and Cervical Cancer Detection Program as well as by MSKCC.
Data Collection and Analysis
Medical records of women diagnosed with breast carcinoma from 1995 to 2000 were reviewed. Self-reported data regarding race/ethnicity, insurance status, and income were recorded. Income data were interpreted in the context of the poverty guidelines of the U.S. Department of Health and Human Services.6
The method of breast carcinoma detection (mammography, physical examination, or both) was determined by review of patient charts. Time to diagnosis, defined as the time from the first identified abnormality (either at physical examination or mammography) to the tissue diagnosis of carcinoma, was calculated.7
Minimal breast carcinoma was defined as ductal carcinoma in situ (DCIS) or as invasive carcinoma measuring up to 1 cm.8 Stage of breast carcinoma was determined in accordance with the criteria of the American Joint Committee on Cancer.9 Stage and surgical treatment for our patients were compared with data from Harlem Hospital published between 1964 and 1986.2
Data were entered into a computerized spreadsheet (Excel; Microsoft, Redmond, WA) for analysis. Statistical analyses were performed with the chi-square and Fisher exact tests using a computerized statistics program (Epi-Info, Centers for Disease Control, Atlanta, GA).
These 181 breast carcinomas were diagnosed in 178 women of median age 51 years (range, 27–87 years; Table 1). Among 147 women for whom data regarding race/ethnicity were available, 131 (89%) were black or Hispanic (Table 1). Among 163 women for whom insurance status was known, 73 (45%) had no medical insurance (Table 1). Among 111 women for whom income data were available, 42 (38%) had incomes below the federal poverty guidelines.
|Total (%)||Nonpalpable (%)||Palpable (%)|
|Race/ethnicity||n = 147||n = 42||n = 105|
|Black||80 (54.4)||26 (61.9)||54 (51.4)|
|Hispanic||51 (34.7)||13 (30.9)||38 (36.2)|
|White||14 (9.5)||3 (7.1)||11 (10.5)|
|Other||2 (1.4)||0 (0)||2 (1.9)|
|Insurance||n = 163||n = 54||n = 109|
|Private||69 (42.3)||27 (50.0)||42 (38.5)|
|Medicaid||21 (12.9)||8 (14.8)||13 (11.9)|
|None||73 (44.8)||19 (35.2)||54 (49.5)|
|Age (yrs)||n = 181||n = 59||n = 122|
|< 40||26 (14.4)||0 (0)||26 (21.3)|
|40–49||57 (31.5)||16 (27.1)||41 (33.6)|
|50–59||43 (23.8)||17 (28.8)||26 (21.3)|
|60–69||38 (21.0)||17 (28.8)||21 (17.2)|
|70–79||13 (7.2)||6 (10.2)||7 (5.7)|
|80 or older||4 (2.2)||3 (5.1)||1 (0.8)|
Of these 181 breast carcinomas, 122 (67%) were palpable at diagnosis and 59 (33%) were nonpalpable lesions detected by screening mammography. Of 164 carcinomas for which records of mammography and physical examination were available, carcinoma was detected by both mammography and physical examination in 99 (60%) carcinomas, by mammography only in 59 (36%) carcinomas, and by physical examination only in 6 (4%) carcinomas. Diagnosis of 92 (51%) of 181 carcinomas was confirmed by percutaneous needle biopsy under the guidance of palpation (n = 49), ultrasound (n = 22), or stereotaxis (n = 21); for 89 (49%) carcinomas, diagnosis was made by surgical excision. The median time to histologic diagnosis was 34 days (range, 0–298 days).
Stage of breast carcinoma, known for 167 carcinomas, was Stage 0 in 38 (23%) cases, Stage I in 38 (23%) cases, Stage II in 63 (38%) cases, Stage III in 24 (14%) cases, and Stage IV in 4 (2%) cases. Fifty-six (34%) cases were minimal breast carcinomas (Table 2). Treatment of breast carcinoma, known for 166 carcinomas, included therapeutic surgery in 153 carcinomas and chemotherapy without surgery (for locally advanced or metastatic disease) in 13 carcinomas; among 153 carcinomas that were treated surgically, breast-conserving surgery was performed in 123 (80%) and mastectomy in 30 (20%; Table 2).
|Total (n = 181) (%)||Nonpalpable (n = 59) (%)||Palpable (n = 122) (%)||P valuea|
|Stage||n = 167||n = 55||n = 112|
|0||38 (22.8)||30 (54.4)||8 (7.1)||< 0.0000001|
|I||38 (22.8)||18 (32.7)||20 (17.9)||0.05|
|II||63 (37.7)||6 (10.9)||57 (50.9)||0.0000001|
|III||24 (14.4)||1 (1.8)||23 (20.5)||0.003|
|IV||4 (2.4)||0 (0)||4 (3.6)||NS (0.3)|
|Minimal||56 (33.5)||39 (70.9)||17 (15.2)||0.0000001|
|Treatment||n = 166||n = 56||n = 110|
|Breast-conserving surgery||123 (74.1)||47 (83.9)||76 (69.1)||< 0.04|
|Mastectomy||30 (18.1)||9 (16.1)||21 (19.1)||NS (0.79)|
|Chemotherapyb||13 (7.8)||0 (0)||13 (11.8)||< 0.005|
Among 59 nonpalpable, mammographically detected breast carcinomas, 55% were DCIS, 71% were minimal breast carcinoma, and 84% were treated with breast-conserving surgery (Table 2). Among 122 palpable breast carcinomas, 7% were DCIS, 15% were minimal breast carcinoma, and 69% were treated with breast-conserving surgery (Table 2). Nonpalpable, as opposed to palpable, breast carcinomas were significantly more likely to be DCIS or minimal breast carcinoma and to be treated with breast-conserving surgery (Table 2).
Lack of health insurance was more frequent among women with palpable as opposed to nonpalpable breast carcinoma (54 of 109 [50%] vs. 19 of 54 [35%], P = 0.12; Table 1). All breast carcinomas in women younger than 40 years of age were palpable. No significant differences in race/ethnicity were observed between women with palpable versus nonpalpable breast carcinoma (Table 1). Women who had Medicare or private insurance had a higher proportion of minimal breast carcinomas than women who had Medicaid or who lacked health insurance (27 of 64 [42.2%] vs. 25 of 85 [29.4%], P = 0.15).
The stage and treatment of cancers diagnosed at the Breast Examination Center of Harlem were compared with results of breast carcinomas detected, diagnosed, or followed at Harlem Hospital between 1964 and 1986.2 Among our women, there was a significantly higher frequency of early (Stage 0 or I) breast carcinoma, a significantly lower frequency of advanced (Stage III or IV) disease, and a significantly higher frequency of breast-conserving surgery (Table 3).
|Our study (%)||Previous Harlem hospital dataa (%)||P value|
|0||38/167 (22.8)||0/437 (0.0)||< 0.0000001 (OR, 95% CI undefined)|
|I||38/167 (22.8)||27/437 (6.2)||< 0.0000001 (OR, 4.5; 95% CI, 2.5–7.9)|
|II||63/167 (37.7)||195/437 (44.6)||0.15 (OR, 0.8; 95% CI, 0.5–1.1)|
|III||24/167 (14.4)||171/437 (39.1)||< 0.0000001 (OR, 0.3; 95% CI, 0.2–0.4)|
|IV||4/167 (2.4)||44/437 (10.1)||0.003 (OR, 0.2; 95% CI, 0.1–0.6)|
|Breast-conserving||123/153 (80.4)||33/512 (6.4)||< 0.0000001 (OR, 59.5; 95% CI, 33.9–105.1)|
|Mastectomy||30/153 (19.6)||479/512 (93.6)||< 0.0000001 (OR, 0.02; 95% CI, 0.01–0.03)|
In 1990, McCord and Freeman1 reported that the age-adjusted mortality rate from all causes in Harlem was the highest in New York City, more than double that of U.S. whites and was 50% higher than that of U.S. blacks. Freeman and Wasfie2 found that breast carcinoma was a particularly severe problem in Harlem: of 708 breast carcinomas diagnosed, treated, or followed at Harlem Hospital from 1964 to 1986, almost one half were locally advanced (Stage III) or had distant metastases (Stage IV) and the 10-year survival rate for those treated surgically was 27%. These data indicated the urgency of the need to develop preventive and therapeutic measures for breast carcinoma in Harlem that could be delivered directly to the population at highest risk.
The Breast Examination Center of Harlem was created in response to this need. The patient population seen at the Breast Examination Center was predominantly minority women, with 45% of women in our study having no medical insurance. Among carcinomas diagnosed at the Breast Examination Center during the study period, 46% were Stage 0 or Stage I and were associated with an excellent prognosis. Nonpalpable, mammographically detected breast carcinomas, which accounted for approximately one third of breast carcinomas diagnosed during the study period, were particularly likely to be in the early stage and therefore were potentially curable. Among nonpalpable breast carcinomas detected by screening mammography, 54% were DCIS, 71% were minimal breast carcinomas, and 84% were treated with breast-conserving surgery. We found a significantly higher proportion of early stage breast carcinomas, a significantly lower proportion of locally advanced or metastatic tumors, and a significantly higher frequency of breast conservation than in the previous report of breast carcinomas at Harlem Hospital diagnosed between 1964 and 1986.2 The proportion of breast carcinomas that were Stage 0 or I increased from 6% to 46%; the proportion of locally advanced or metastatic tumors decreased from 49% to 17%; and breast-conserving surgery increased from 6% to 80%. The difference between the inpatient setting of Harlem Hospital and the outpatient setting of the Breast Examination Center and the temporal differences between the studies likely contributed to the difference in results, but do not fully account for the observed changes.
Since the 1970s, there has been an interval increase in the use of screening mammography (in part due to increased awareness of breast carcinoma), progress in mammographic technique, and more widespread use of breast-conserving surgery nationwide. In the United States from 1974 to 1985, 48% of patients with breast carcinoma presented with disease localized to the breast; from 1989 to 1995, the percentage of cases localized to the breast was 51% for blacks and 63% for whites.10, 11 The improvements in stage of breast carcinoma at diagnosis in Harlem exceeded the general improvements in the United States during that time period, reflecting the dire situation in Harlem in previous years and the marked increase in early screen-detected carcinomas in our program. The Breast Examination Center of Harlem brought improvements in screening and diagnosis in an outpatient setting directly to women in Harlem, allowing them to benefit from advances in early detection and treatment.
Although some have suggested that the poor prognosis of breast carcinoma in black women is due to aggressive tumor biology,12, 13 published data show poor outcome is at least in part related to the later stage at diagnosis.14 Dignam et al.15 found that black and white patients with localized breast carcinoma had similar outcomes and benefited equally from systemic therapy. Cella et al.16 found that race (white vs. nonwhite) was not a significant predictor of survival time, but income and education were. Other factors that may contribute to the late stage at presentation in minority women include underutilization of mammography17, 18 and other ambulatory services19 and the attitudes or beliefs about breast carcinoma biology and treatment.20 These factors should be addressed with improved access to care, outreach, and education.
Although no statistically significant relationship between insurance status and stage at diagnosis was observed among our patients, previous studies have emphasized the importance of insurance status in determining stage at presentation and outcome in patients with breast carcinoma. Lannin et al.20 found that uninsured, as opposed to insured, women were significantly more likely to present with later stage breast carcinoma. Roetzheim et al.21 found that breast carcinoma mortality rates were higher for patients who had Medicaid or lacked health insurance compared with patients who had commercial fee-for-service insurance, correlating with more advanced stage at diagnosis. Unfortunately, the last decade has seen an increase in poverty and in the proportion of women and men who lack medical insurance.22 The impact of insurance status on breast carcinoma outcome is particularly important in light of the fact that racial minorities have been and continue to be far more likely than white Americans to lack health care coverage.22
We detected early-stage breast carcinomas with excellent prognosis in a low-income, predominantly minority population. These results refute the conclusion of Jones et al.,23 who found that mammography use protected against later stage diagnosis in whites, but not in blacks. Their study was limited by lack of detail about the quality of the mammography provided and the follow-up of screen-detected abnormalities. Negotiation through the complexities of follow-up care is daunting, particularly for individuals of limited resources, education, or command of the English language. Our patient navigators,3 individuals who help women with abnormal findings obtain appropriate follow-up (including additional imaging studies, biopsy, or treatment), are essential to our program. A mammogram alone, no matter how good, cannot reduce breast carcinoma mortality rates. Reduction in breast carcinoma mortality rates requires that the carcinoma not only be detected and diagnosed, but also treated.
Our results are encouraging, but further work is needed. Two thirds of carcinomas in our study were palpable at diagnosis, suggesting the need for more outreach to increase screening in asymptomatic women. Screening mammography is underutilized in Harlem. Of Harlem women 50–65 years old who participated in a published survey, 83% reported having had at least one mammogram but only 54% had a mammogram in the past year.24 The median time to histologic diagnosis of approximately 1 month could perhaps be shortened with even more resources devoted to patient navigation. We detected early-stage cancers, but women of lower socioeconomic status (such as those in our patient population) may have worse outcome for comparable stages of detection compared with women of higher socioeconomic status, perhaps due to differences in treatment.16, 25 Long-term follow-up data are therefore needed to determine the outcome of breast carcinomas diagnosed at the Breast Examination Center of Harlem. Assessment of breast carcinoma incidence, stage, treatment, and outcome elsewhere in Harlem is needed to determine breast carcinoma mortality rates in Harlem at the current time and to assess the impact of our program on the surrounding community.
We believe that the Breast Examination Center of Harlem and similar programs7, 26–28 can serve as models for creation of breast carcinoma screening and diagnostic centers in other underserved communities. Early, curable breast carcinomas can be found in women in Harlem, but outreach remains a challenge, particularly for the uninsured. McCord and Freeman1 stated that a major political and financial commitment will be needed to eradicate the root causes of high mortality rates in Harlem: vicious poverty and inadequate access to the basic health care that is the right of all Americans. Risk factors for excess mortality rates in Harlem persist, including high rates of poverty, unemployment, and homelessness, as well as inadequate access to and underutilization of screening services.24 Efforts such as those described here can help to provide life-saving health care to all individuals, regardless of their ability to pay.
The authors gratefully acknowledge the staff of the Breast Examination Center of Harlem for their hard work and devotion, including Joan Baron, Valerie Brown, Judy Cambrelen, Trang Campbell, Karen Canaie, Rebecca Carrasquillo, Loretta Chisolm, Henrietta Clark, Jeralyn Cortez, Milagros Fernandez, Henry Godfrey, Winsome Grant, Jamella Hall, Danielle Jefferson, Marquett Kennely, Ti-Anna Lee, Anita Lloyd, Sandra McDonald, Alexandra Mitnick, Alisa Persaud, Sataram Pillarisetty, Ramona Ramos, Evangelyn Ramsey, Nicole Rhoden, Deborah Roberts, Michelle Torres, Christine Waters-Clayton, and Miriam Williams. The authors also thank David C. Perlman, M.D., for his invaluable assistance.
- 6US Department of Health and Human Services Office of the Assistant Secretary for Planning and Evaluation. Summary historical figures and Federal Register references for the HHS poverty guidelines since 1982. Available from URL: http://aspe os dhhs gov /poverty/figures-fed-reg htm [Accessed December 1, 2001].
- 9AJCC cancer staging manual. Philadelphia: Lippincott-Raven, 1997: 171–180.
- 14Cancer statistics, 2002. CA. 2002; 52: 23–57., , , .
- 15Prognosis among African-American women and white women with lymph node negative breast carcinoma: findings from two randomized clinical trials of the National Surgical Adjuvant Breast and Bowel Project (NSABP). Cancer. 1997; 80: 80–90., , , , .
- 19Racial and ethnic differences in access to and use of health care services, 1977 to 1996. Med Care Res Rev. 2000; 57(suppl 1): 36–54., , .
- 22Race/ethnicity and health insurance status: 1987 and 1996. Med Care Res Rev. 2000; 57(suppl 1): 11–35., .