Mammography Use Helps To Explain Differences in Breast Cancer Stage at Diagnosis between Older Black and White Women

Methods

Data Source

We conducted a retrospective cohort study using the Linked Medicare-Tumor Registry Database [11]. The linked database was jointly created by the National Cancer Institute and the Health Care Financing Administration (HCFA) to enable researchers to conduct cancer-related health services research. The linked database contains cancer information on patients 65 years of age and older from the National Cancer Institute's SEER program that is linked with Medicare enrollment and utilization information from HCFA's Medicare Statistical System for the years 1985 to 1989.

Two Medicare utilization files are available in the linked database. Medical Provider Analysis and Review (MEDPAR) is a 100% utilization file that contains one record for every inpatient hospitalization or stay in a skilled-nursing facility covered under Medicare Part A. The Physicians' Claims file is a 100% utilization file that contains one record for every physician and outpatient claim covered under Medicare Part B. Before 1991, the Physicians' Claims file was available for only 10 states. Data from the SEER and Medicare programs overlap in tumor registries for three areas: Connecticut; metropolitan Atlanta, Georgia and Seattle-Puget Sound, Washington. Specific information on the linkage between SEER and Medicare has been published elsewhere [11]. The match rates for Connecticut, Atlanta, and Seattle are 93.3%, 94.1%, and 91.5%, respectively.

Study Sample

Women were eligible for the study sample (n = 4438) if a first case of primary breast cancer was diagnosed between 1 January 1987 and 31 December 1989 and if they were 67 years of age or older; were black or white; and resided in Connecticut, Atlanta, or Seattle-Puget Sound. Although we selected these areas because physicians' claims were available for all cases, they also represent a geographically diverse population of older women with breast cancer. Women who were enrolled in a health maintenance organization and those with less than 2 full years of Medicare Part B coverage were not eligible for this study because physician claims data (which are required for identifying mammography use) are not available for these women. We limited our final study sample to women who were 67 years of age or older to ensure that all women had at least 2 years of Medicare utilization (claims) information before breast cancer was diagnosed.

Women whose mammography use could not be categorized (21 black women and 271 white women) or whose disease was unstaged (8 black women and 133 white women) were excluded from the study.

Measures

We ascertained the following sociodemographic variables from the SEER file: age at diagnosis, marital status at diagnosis, and SEER area. Age at diagnosis (range, 67 to 100 years) was categorized as 67 to 74 years, 75 to 84 years, or 85 years and older for descriptive purposes but was modeled as a continuous variable. Marital status was defined as married or not married at diagnosis. The SEER area was classified according to the tumor registry of diagnosis: Connecticut, Atlanta, or Seattle. We used 1990 U.S. Census data as an ecologic measure of socioeconomic status. Women were assigned to the median household income of their ZIP code of residence; incomes were grouped as less than $15 000 or at least $15 000.

We obtained information on race from the Medicare beneficiary enrollment file. A comparison of race between Medicare and SEER files demonstrated agreement for 99% of women. Enrollees are classified in Medicare files as black, white, Asian, Native American, Hispanic, or unspecified. Because this analysis focused on black and white women, women of “other” racial groups were not eligible (n = 125).

We computed a modified Charlson Comorbidity Index using Deyo's method of classifying ICD-9-CM (International Classification of Diseases, 9th revision, Clinical Modification) diagnosis codes from inpatient claims [12]. For each woman, we identified all inpatient hospitalizations that began 2 years before diagnosis and ended 1 month after diagnosis. A priori, we extended the observation period to 1 month past the diagnosis because we expected that during the study years 1987 to 1989, most women would have had at least one hospitalization around the time of breast cancer diagnosis. We classified women as 1) nonhospitalized [that is, comorbidity could not be assessed], 2) having no comorbid conditions [Charlson Index, 0], or 3) having one or more comorbid conditions (Charlson Index ≥ 1).

We measured mammography use by using Medicare physicians' claims. We identified women who had one or more bilateral mammographies (CPT [Physicians' Current Procedural Terminology] procedure code 76091) within 2 years before breast cancer diagnosis. We classified women as 1) nonusers (n = 833) if they did not have mammography during the 2 years before diagnosis, 2) regular users (n = 733) if they had at least two mammographies within the 2 years before diagnosis that were 10 or more months apart, and 3) peri-diagnosis users (n = 2267) if they had their only mammography or mammographies within 3 months before diagnosis. Women who did not fit into any of these categories were classified as “uncertain” (n = 292) and were excluded from the study. The peri-diagnosis users were a heterogeneous group that included women who underwent screening mammography and were given a diagnosis of breast cancer and women who underwent diagnostic mammography. Therefore, analyses relating previous mammography use to cancer stage at diagnosis considered only nonusers and regular users because these were two distinct groups.

We measured cancer stage at diagnosis using the SEER historical staging system (in situ, localized, regional, distant, or unstaged) because it was available for all women. Stage of disease was dichotomized as early (in situ or localized) or late (regional or distant).

Statistical Analysis

All statistical analyses were done by using SAS statistical software, version 6.11 [13]. Black and white women were compared with respect to sociodemographic factors, comorbidity, stage at diagnosis, and previous mammography use. Chi-square statistics and Student t-tests were performed to identify characteristics that differed significantly between black and white women.

Multivariable logistic regression was used to estimate the adjusted odds of late-stage disease for black women compared with white women [14]. To investigate the extent to which previous mammography use explains the observed black-white difference in stage at diagnosis, we compared simple models with more complex ones and examined changes in the estimated odds ratio for the race-stage association [15]. First, we compared a model that included only race with a model that included race and previous mammography use to determine how much of the excess late-stage disease among black women is explained by differences in previous mammography use. Next, we compared a model that included race, sociodemographic characteristics, and comorbidity information with a model that included these factors and previous mammography use to determine the additional amount of excess late-stage disease among black women that is explained by previous mammography use after adjustment for sociodemographic and comorbidity information. The odds ratio for race and the corresponding 95% CIs were estimated from the β coefficient and SE from the logistic models [14]. We used the following formula to compute the percentage change in the estimated odds ratio to compare our results with those from a previous study [16]:

Percentage change in OR = (Equation 1)

Results

Characteristics of the study sample (n = 4005) are given in Table 1. Overall, 172 (4%) of the women were black. Twenty-two percent of women had not had mammography within the 2 years before breast cancer diagnosis (nonusers), 19% of women had at least two mammographies within the 2 years preceding diagnosis that were at least done 10 months apart (regular users), and 59% had their only mammography within the 3 months before diagnosis (peri-diagnosis users). Late-stage disease was diagnosed in one third (32%) of the women.

Table 1 also shows the characteristics of the study sample by race. Race was confounded with the SEER area of residence. For example, although only 17% of women in our study resided in Atlanta, almost two thirds (66%) of the black women were from Atlanta. Black women were less likely to be married (18% of black women compared with 38% of white women) and were more likely to live in a low-income area (75% compared with 13%). Comorbidity also varied with race: Black women were more likely to have not been hospitalized (27% compared with 19%); however, the black women who were hospitalized were more likely to have at least one comorbid condition (27% compared with 21%) than white women. Age at diagnosis was similar for black and white women.

Black women were over-represented among the nonusers (35% compared with 22%) and underrepresented among the regular users (11% compared with 19%) (Table 1). However, the percentages of black and white peri-diagnosis users were similar (56% and 59%). Late-stage disease was diagnosed more frequently in black women than in white women (39% compared with 32%).

Bivariate associations with late-stage disease among nonusers and regular users (n = 1646) are presented in Table 2. Late-stage disease was significantly more likely to be diagnosed in black women than in white women (odds ratio [OR], 2.49 [95% CI, 1.59 to 3.92]). Late-stage disease was more likely to be diagnosed in women residing in Connecticut (OR, 1.80 [CI, 1.42 to 2.31]) or Atlanta (OR, 1.59 [CI, 1.18 to 2.14]) than in those residing in Seattle. Diagnosis of late-stage disease was also more likely in women who had no comorbid conditions (OR, 1.53 [CI, 1.15 to 2.03]) and women who had at least one comorbid condition (OR, 2.19 [CI, 1.58 to 3.04]) than in women who were not hospitalized. A weak positive association with late stage at diagnosis was observed among women who resided in a low-income area (OR, 1.32 [CI, 1.01 to 1.73]). No significant differences were observed in cancer stage at diagnosis by age at diagnosis or marital status.

A lack of previous mammography use was strongly associated with late-stage disease at diagnosis (Table 2). Late-stage disease was more likely to be diagnosed in nonusers than in regular users (OR, 3.00 [CI, 2.41 to 3.75]).

The crude odds ratios for late-stage disease in a comparison of nonusers with regular users are presented separately for black and white women in Table 3. These analyses were done to determine whether the relation between previous mammography use and cancer stage at diagnosis is significant in black women and in white women. Previous mammography use was strongly associated with stage at diagnosis for both black and white women. Among black women, the odds of diagnosis with late-stage disease was 6.65 (CI, 1.96 to 22.53) in a comparison of nonusers with regular users. Among white women in a similar comparison, the odds of diagnosis with late-stage disease was 2.83 (CI, 2.25 to 3.56).

The crude and adjusted odds ratios for late-stage disease in black compared with white women are presented separately for nonusers and regular users of mammography in Table 4. These analyses were done to determine whether race is related to late-stage disease after adjustment for previous mammography use. Among nonusers, late-stage disease was significantly more likely to be diagnosed in black women than in white women (OR, 2.46 [CI, 1.43 to 4.22]). After adjustment for SEER area, age, marital status, income, and comorbidity, the odds of late-stage disease remained greater for black women (adjusted OR, 2.54 [CI, 1.37 to 4.71]). Among regular users of mammography, however, no important difference in stage at diagnosis was seen between black and white women (adjusted OR, 1.34 [CI, 0.40 to 4.51]).

Table 5 lists the results obtained from logistic regression modeling to adjust the race-stage association for important factors related to late-stage disease. To determine the extent to which previous mammography use explains the black-white difference in cancer stage at diagnosis, we compared the change in the estimated odds ratio from models 1 and 2. Previous mammography use alone significantly reduced the estimated crude odds ratio for late-stage disease in a comparison of black and white women from 2.49 to 2.05 and explained almost 30% of the excess late-stage breast cancer seen in black women.

To determine the extent to which previous mammography use explains the black-white difference in stage at diagnosis after adjustment for the other factors, we compared the change in the estimated odds ratio from models 3 and 4 (Table 5). Model 3 presents the association between race and stage after adjustment for sociodemographic and comorbidity information. Further adjustment for previous mammography use (model 4) reduced the odds ratio from 2.47 to 2.30. After all other factors were considered, previous mammography use explained 12% of the excess late-stage breast cancer seen in black women. When interpreting these odds ratios, the reader should remember that most black women were nonusers.

Discussion

We found that previous mammography use was strongly associated with cancer stage at diagnosis in both black and white women aged 65 years and older. Most important, we found that among women who did not use mammography, late-stage disease was significantly more likely to be diagnosed in black women than in white women. Among regular users of mammography, however, we found no difference in stage at diagnosis between black and white women. These results suggest that previous mammography does help to explain the black-white difference in stage at diagnosis among older women.

We also found that previous mammography use significantly contributed to the black-white difference in stage at diagnosis. Previous mammography use alone explained approximately 30% of the excess late-stage breast cancer among black women. Individual (data not shown) and combined adjustment for sociodemographic characteristics and comorbidity produced only modest changes in the race-stage association; this finding suggests that these factors do not adequately explain the excess of late-stage disease among black women. However, previous mammography use explained an additional 12% of the black-white difference in stage at diagnosis after adjustment for sociodemographic characteristics and comorbidity.

Our findings differ from those of two previous studies that have examined the association of previous mammography use, race, and stage at diagnosis [16, 17]. Jones and coworkers [16] examined patients with breast cancer who were enrolled in a Connecticut study and found that previous mammography use was associated with stage only in white women and that mammography use explained less than 10% of the black-white difference in stage at diagnosis after adjustment for age. Hunter and coworkers [17] examined data from the Black/White Cancer Surveillance Study and found that previous mammography use was associated with stage at diagnosis only in black women. The investigators did not quantify the explanatory power of mammography use but found that previous mammography use did not appreciably alter the race-stage association.

Four major differences in the design of these studies may contribute to the discordant results. First, our study focused on older women. We studied women 67 years of age and older and imposed no upper age limit. The other studies included 20- to 79-year-old women [16, 17]. Second, we used a more rigorous definition of previous mammography use. To be considered a regular user, women must have had at least two mammographies that were done at least 10 months apart within 2 years before breast cancer diagnosis. The other two studies required at least one screening mammography within 3 years [16] and 6 years [17] before diagnosis. In our study, many of these women would have been classified as nonusers. Third, we used the historical staging system because it was available for all women; the other studies used the more precise TNM (tumor, node, metastasis) staging system. Fourth, we relied on Medicare claims rather than patient self-report to measure mammography use.

Use of Medicare claims data to measure mammography use may have some limitations. First, our study was limited to women enrolled in fee-for-service settings because Medicare data do not capture services provided to health maintenance organization enrollees. Although few women were enrolled in managed care programs during the years of our study, the proportion of Medicare health maintenance organization enrollees increased from 4% to 13% between 1990 and 1997 [11, 18]. It is unclear how mammography use will change as more older women become enrolled in Medicare managed care plans.

Second, Medicare reimbursement policies have changed over time. Medicare began reimbursing providers for biennial screening mammography in 1991 and for annual screening mammography in 1998. These changes in reimbursement raise two issues: 1) Only diagnostic mammography was covered during the years of our study, and 2) subsequent reimbursement for screening mammography has led to greater use among older women.

Although Medicare paid only for diagnostic mammography, studies show that providers were performing screening mammography and billing Medicare under the diagnostic procedure code [7-10]. Nevertheless, we cannot determine whether an individual mammography was done for screening or diagnostic purposes. To address this issue, we considered how women used mammography over time. We defined our measure of previous mammography use to identify two distinct groups: 1) women who had no evidence of mammography use during the 2 years before diagnosis and 2) women who demonstrated a pattern of regular mammography use. Regular users were women who had at least two mammographies that were done at least 10 months apart. We selected the 10-month interval as a clinically reasonable length of time to assume that women were undergoing screening mammography and were not being followed because of a suspicious lump. Although women with a pattern of regular use seemed to be using mammography for screening, we do not know which women had their cancer detected by symptoms and confirmed with diagnostic mammography.

We also cannot determine whether a clinical breast examination was performed. However, data from the 1987 and 1992 National Health Interview Surveys suggest that black and white women were similar with respect to undergoing a clinical breast examination within the previous year [19]. Jones and coworkers [16] found that clinical breast examination did not contribute to the black-white difference in stage at diagnosis. Finally, no evidence suggests that clinical breast examination adds benefit to that obtained from screening with mammography. A meta-analysis of data from mammography trials reported similar reductions in breast cancer mortality with and without clinical breast examination [20].

Several studies show that Medicare reimbursement for biennial screening mammography has generated an overall increase in mammography use among older women, but that this change in policy has not been sufficient to eliminate black-white differences in mammography use [7, 8, 21]. Medicare data from 1993 demonstrate a pronounced lower rate of mammography use in older black women than in older white women (18% compared with 26%) [7]. A recent Connecticut study found that Medicare reimbursement had a limited effect on mammography use in women during the first few years of implementation and that mammography rates continued to be low, particularly among older black women [21].

Greater mammography use has led to a shift toward earlier stage at diagnosis for all women [22, 23]. However, this shift is not as great for black women as it is for white women, a finding consistent with the persistently lower rate of mammography use among black women [23, 24]. In fact, SEER Public Use data from 1990 to 1992 indicate that late-stage breast cancer continues to be diagnosed more frequently in black women aged 65 years and older than in their white counterparts (34% compared with 28%).

Many factors, including those related to access to health care, the physician, and the patient, have been shown to contribute to the disparity in mammography use among black and white women and may be partly responsible for the persistent racial gap in stage at diagnosis. Blustein [8] found that Medicare enrollees who also had supplemental insurance had greater mammography use than those whose only insurance was Medicare. Even among women with supplemental insurance, black women had lower rates of mammography use [8]. We previously found that greater mammography use was associated with an increasing number of visits to a primary care provider among black and white women but that receipt of primary care was not enough to correct the disparity in mammography use between black and white women [10]. Furthermore, many studies show that a physician's recommendation is the most important determinant of mammography use [25-27]. O'Malley and coworkers [27] found that a physician's recommendation accounted for 60% to 75% of the black-white difference in mammography use. Among older black women in the National Health Interview Survey [6], lack of a physician recommendation was the most commonly cited reason for lack of mammography. Other studies show that patients' knowledge relates to their behaviors and that black women are less knowledgeable about the importance of early detection of breast cancer using mammography [6, 28].

Other potential limitations of our study include misclassification bias, residual confounding, and a small number of black women. We measured socioeconomic status and comorbid illness using proxies because direct measures were not available. Socioeconomic status was measured by assigning each woman to the median income of her residence according to census data. This method has been used in several breast cancer studies [29-33] and may capture unknown characteristics related to a person's neighborhood of residence, such as access to care or environmental exposures [33, 34]. Although we could only estimate comorbidity for women who were hospitalized, information was available for approximately 81% of our sample because many women were hospitalized around the time of their breast cancer diagnosis. Because black women were less likely to be hospitalized, they appeared to be healthier than white women in our study. This finding may in fact be an artifact because among hospitalized women, black women tended to have more comorbid conditions than white women. We chose to present analyses adjusted for comorbidity because analyses that excluded comorbidity variables achieved similar results.

Our study sample was derived from three population-based tumor registries but included relatively few black women because of the demographic characteristics of women receiving Medicare in these states. Among women covered by Medicare in 1990, blacks represented only 1% of women in Washington, 3% of women in Connecticut, and 20% of women in Georgia [10]. Similarly, because of low rates of mammography use among older women in general and black women in particular [10], only a few black women in our sample were regular users of mammography. Nevertheless, their experience was identical to that of while regular users. Late-stage disease was diagnosed in only 20% of regular users, a percentage substantially lower than the percentage of nonusers, regardless of whether they were black or white.

Our study highlights the importance of regular mammography for all older women. We found that among women who were regular users of mammography, there was no black-white difference in cancer stage at diagnosis. This finding suggests that regular mammography may provide the same benefit in older black women and older white women. Our findings also indicate that differences in cancer stage at diagnosis between older black and white women are associated with mammography use. Furthermore, increased regular mammography use may result in a shift toward earlier-stage disease at diagnosis and may narrow the differences in stage at diagnosis between older black and white women.

Presented at the Society of General Internal Medicine annual meeting, May 1996.

Dr. Coughlin: Division of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, 4770 Buford Highway NE (K-55), Atlanta, GA 30341.

Drs. Freund, Ash, and Moskowitz: Section of General Internal Medicine, Evans Department of Medicine, Boston Medical Center, 720 Harrison Avenue, Suite 1108, Boston, MA 02118.

Dr. Rice: Department of Biostatistics, School of Public Health and Tropical Medicine, Tulane University, 1501 Canal Street, New Orleans, LA 70112.

Dr. Marwill: Gillette Medical Department, One Gillette Park, South Boston, MA 02127.

Dr. Shwartz: Operations Management, School of Management, Boston University, 621 Commonwealth Avenue, Boston, MA 02215.