Racial and age-related disparities in obtaining screening mammography: results of a statewide database.

ABSTRACT

Background. This study compared mammography rates in Arkansas, based on race and age, to identify any disparities in obtaining mammography.

Methods. The Arkansas Mammography Data Collection Project contained data on 133,549 women, 87,680 of them aged 40 to 64 and 45,869 aged 65 years or older. Race data were available on 64,526 women. Mammography rates were compared by patient age and race.

Results. Women aged 65 years and older were less likely to obtain mammograms than younger women: 21.67% versus 23.30%. Mammograms were more likely to be classified as diagnostic in older women and as screening in younger women. Older African American women were less likely to obtain mammograms than younger women of the same race (7.23% vs 8.02%), while older white women were more likely to obtain mammograms than younger white women (11.64% vs 11.31%). African Americans were less likely to obtain mammograms in both age groups.

Conclusions. Older African American women are less likely to obtain mammograms than white and younger women. The reasons for this disparity warrant further evaluation.

**********

BREAST CANCER is the most commonly diagnosed cancer in women and the second leading cause of cancer death among women in the United States. (1) Screening mammography improves the chance of early detection of breast cancer, leading to a better outcome among screened women. (2-5) The recommended age for screening mammograms varies among public health agencies and countries and ranges from 35 to 50 years. (6) However, it is widely agreed that all women should have routine annual mammograms after age 50. Furthermore, many countries limit the upper age of screening mammography between the ages of 59 and 74, with most using 69 years as the upper limit. (6) The American Cancer Society recommends that screening mammograms begin at age 40, with no upper age limit as long as a woman is otherwise in good health. (7) Mandelblatt et al (8) reported that screening older women with and without comorbid conditions is beneficial. Screening older women is important, especially since the risk of breast cancer increases with age . (9) Breast cancer has the second highest cancer death rate for women aged 60 to 69 and the third highest for women aged 70 and older. (1) Approximately 48% of newly diagnosed breast cancers occur in women 65 and older, and nearly 57% of breast cancer deaths occur in these same women. (10) Despite these facts, older women are less likely to obtain screening mammograms and other cancer screening tests. (11,12) In addition, African American women were reported to have a lower rate of screening mammograms. (13) These studies were limited by relatively small numbers of women or by the inherent limitations of the surveys used to query women or health care providers. We report the results of statewide mammography data compiled by the Arkansas Mammography Data Collection Project for 1997.

METHODS

A comprehensive survey of 112 mammography centers throughout the state of Arkansas was conducted to collect individual patient demographic information and mammography examination data. The data elements collected from the participating centers included age, race (white, African American, other), insurance status (private, Medicare, Medicaid, none), date of mammogram, type of mammogram (screening, diagnostic), and city or zip code of residence. Data were entered into a specially designed Access database. A detailed description of the methodology is described elsewhere. (14)

The women were divided into two age groups. The first group included women aged 40 to 64, and the second group included women aged 65 and older. Mammography rates for each group were calculated using the 1997 estimated state census data. The percentages of mammography type, insurance type, and race were calculated for each group using the Arkansas Mammography Data Collection Project database. The differences between the groups were compared using chi-square methods.

RESULTS

Of the 112 centers contacted, 92 (82%) participated in the project. Data were collected on 157,976 mammograms. Because some women had more than one mammogram, the number of individual women who had mammography in 1997 was 148,586, and 133,549 (90%) of them were aged 40 or older. Of the latter group, 87,680 (66%) were between the ages of 40 to 64, and 45,869 (34%) were 65 years of age or older. The mammography rates were 23.3% for the younger group and 21.7% for the older group (P<.0001). The majority of mammograms were classified as screening, and approximately one third of the mammograms were classified, as diagnostic (Table 1). The type of mammogram was unknown in less than 5% of cases. Older women were more likely to have diagnostic mammograms, whereas the younger women were more likely to have screening mammograms (36% vs 34% and 59.7% vs 61.9%, respectively, P<.0001).

Insurance status analysis revealed that more than 65% of the younger women had private insurance, and 62.8% of the older women had Medicare alone or with additional insurance. The proportion of uninsured young women was more than 50% higher in the younger group than in the older group (7.5% vs 3.2%). The insurance type was not documented in 23.3% of younger women and 20.4% of older women (Table 1).

Race data were available for 64,526 women. An analysis within the same race revealed that older white women were more likely to obtain mammograms than younger white women (11.6% vs 11.3%, P<.001). However, older African American women were less likely to obtain mammograms than younger women. (7.2% vs 8.0%, P<.001). African American women were less likely to obtain mammograms than white women in both age groups (P<.001) (Table 2).

DISCUSSION

The results of this study clearly show that only a small fraction of women obtained screening mammograms in Arkansas. Even after adjusting for missing data from the 20 nonparticipating centers, less than 30% of the women in Arkansas had a mammogram in 1997. This low rate highlights the need to evaluate current public health interventions and intensify the efforts to increase the number of women utilizing this potentially life-saving test. This is particularly true for older women, who were less likely to have screening mammograms than younger women. These findings reflect a plethora of complicated factors related to the women themselves, to the health care system, and to physicians. (15,16)

A survey of 400 physicians from various specialties revealed that physicians are likely to perform screening examinations on older women less frequently than the American Cancer Society recommends. (17) The heated debate regarding the age to start screening women (ie, 40 vs 50) focused the attention on the lower limits of age, while the upper limit of age was left to the discretion of the physicians. This is one of the largest studies to prove that older women are less likely to have screening mammograms than younger women.

The methods of data collection eliminated many biases related to determining mammography rates by surveying women, such as recall bias, selection bias, and giving socially correct answers. (18) Although data were not obtained from 20 centers, these sites are dispersed across the state and do not represent a special geographic area or demographic distribution. Therefore, the data obtained represent the majority of women screened in Arkansas and reflect the actual mammography practice patterns in the state. Among screened women, older women were more likely to have diagnostic mammograms, which may indicate that physicians may delay obtaining mammograms on older women until there is clinical suspicion.

Although the race-related data were not available on all the patients, it is clear that we had race data on a large sample size. Therefore, it is prudent to calculate the mammography rate by race based on the census data. Knowing that these rates will be lower than the actual rates, it will maintain the differences between race groups.

This analysis of available race data also revealed intriguing results. Older African American women were less likely to obtain mammograms than younger women, while older white women were more likely to obtain mammograms than younger women. When an interracial comparison was done, African American women were less likely to obtain mammograms in both age groups. Older African American women had the lowest mammography rate among all the groups. Therefore, the disparity between younger and older African American women appears to be the major reason for the difference in mammography rate between the two age groups in the whole population. The difference in mammography rates is important, since it may be contributing to the higher mortality rates from breast cancer among African American women. (19,20)

The reasons for disparities between older and younger women and between white and African American women warrant further studies. Determining these reasons will help plan future public health interventions addressing these important issues.

TABLE 1

Type of Mammograms, Insurance Status, and Mammography Rates for the
Study Population

                                                Women Aged
                        Women Aged     [greater than or equal to]65
                        40-64 Years               Years
                       (n = 87,680)            (n = 45,869)
Category                  No. (%)                No. (%)

Mammography type
  Screening *          54,254 (61.88)         27,395 (59.72)
  Diagnostic *         29,862 (34.06)         16,524 (36.02)
  Unknown               3,564 (4.06)           1,950 (4.25)

Insurance
  Medicaid              1,612 (1.84)             178 (0.39)
  Medicaid/private         24 (0.03)              15 (0.03)
  Medicare              1,128 (1.29)          22,547 (49.16)
  Medicare/Medicaid       286 (0.33)             631 (1.38)
  Medicare/private        448 (0.51)           6,240 (13.60)
  Private              57,241 (65.28)          5,423 (11.82)
  Uninsured             6,538 (7.46)           1,489 (3.25)
  Unknown              20,403 (23.27)          9,346 (20.38)

Mammography rate
  Total population *  376,350 (23.30)          9,346 (21.67)

* Differences in age groups are statistically significant (P<.0001).
TABLE 2

Race Data Obtained, Arkansas Race Data, and Mammography Rate by Race

                                                    Women Aged
                            Women Aged     [greater than or equal to]65
                            40-64 Years               Years
                           (n = 40,870)            (n = 23,656)
Category                      No. (%)                No. (%)

Race data obtained

  White                    36,523 (89.36)         21,762 (91.99)
  Black                     3,946 (9.65)           1,711 (7.23)
  Others *                    401 (0.98)             183 (0.77)

Total population

  White                   322,858 (85.78)        186,944 (88.33)
  Black                    49,224 (13.08)         23,652 (11.18)
  Others +                  4,268 (1.13)           1,048 (0.50)

  Total                   376,350                211,644

Mammography rate by race

  White *                  11.31%                 11.64%
  Black *                   8.02%                  7.23%

* Differences in age groups are statistically significant (P<.001).

+ Includes Asian, Asian/Pacific Islanders, and Native Americans.

Acknowledgments. We thank Indu Soora, MPH, CHES, for help in the data collection process and Harry S. Mohrmann, MS, for technical assistance.

References

(1.) Greenlee RT, Murray T, Bolden S, et al: Cancer Statistics 2000. CA Cancer J Clin 2000; 50:7-33

(2.) Lenner P, Jonsson H: Excess mortality from breast cancer in relation to mammography screening in northern Sweden. J Med Screen 1997; 4:6-9

(3.) Cooper GS, Yuan Z, Bowlin SJ, et al: An ecological study of the effectiveness of mammography in reducing breast cancer mortality. Am J Public Health 1998; 88:281-284

(4.) Alexander FE, Anderson TJ, Brown HK, et al: Fourteen years of follow-up from Edinburgh randomized trial of breast cancer screening. Lancet 1999; 353:1903-1908

(5.) UK Trial of Early Detection of Breast Cancer Group: Sixteen-year mortality from breast cancer in the UK Trial of Early Detection of Breast Cancer. Lancet 1999; 353:1909-1914

(6.) Shapiro S, Coleman EA, Broeders M, et al: Breast cancer screening programmes in 22 countries; current policies, administration and guidelines. Int J JEpidemiol 1998; 27:735-742

(7.) Smith RA, Mettlin CJ, Davis KJ, et al: American Cancer Society guidelines for the early detection of cancer. CA Cancer J Clin 2000; 50:34-49

(8.) Mandelblatt JS, Wheat ME, Monane M, et al: Breast cancer screening for elderly women with and without comorbid conditions, a decision analysis model. Ann Intern Med 1992; 116:722-730

(9.) Lickly HL: Primary breast cancer in the elderly. Can J Surg 1997; 40:341-351

(10.) Caplan LS: Disparities in breast cancer screening: is it ethical. Public Health Rev 1997; 25:31-41

(11.) Mandelblatt JS, Gold K, O'Malley AS, et al: Breast and cervix cancer screening among multiethnic women: role of age, health, and source of care, Prev Med 1999; 28:418-425

(12.) Burack RC, George J, Gurrey JG: Mammography use among women as a function of age and patient involvement in decision-making. J Am Ceriatr Soc 2000; 48:1817-1821

(13.) Burns RB, McCarthy EP, Freund KM, et al: Black women receive less mammography even with similar use of primary care. Ann Intern Med 1996; 125:173-182

(14.) Jazieh AR, Soora I: Mammography utilization pattern throughout the state of Arkansas: a challenge for the Future. J Community Health 2001; 26:249-255

(15.) Cohen H: Breast cancer screening in older women: the geriatrician/internist perspective. J Gerontol 1992; 47:134-136

(16.) Zapka JG: A qualitative study about breast cancer screening in older women: implications for research. J Gerontol 1992; 47:93-100

(17.) Weisman CS, Celentano DD, Teitelbaum MA, et al: Cancer screening services in the elderly. Public Health Rep 1989; 104:209-214

(18.) May DS, Torntell AE: Mammography use by elderly women: a methodological comparison of two national data sources. Am J Epidemiol 1998; 8:439-444

(19.) Chevarley F, White E: Recent trends in breast cancer mortality among white and black US women. Am J Public Health 1997; 87:775-781

(20.) McCarthy EP, Bums RB, Coughlin SS, et al: Mammography use helps to explain differences in breast cancer stage at diagnosis between older black and white women. Ann Intern Med 1998; 128:729-736

RELATED ARTICLE: KEY POINTS

* Older women are less likely to get screening mammograms than younger women.

* Older African Americans are the least likely group to get screening.

* Statewide databases are important to determine disparities and identify needs.

From the Department of Internal Medicine, University of Arkansas for A Medical Sciences, Little Rock.

Supported by Grant No. 231 115 1150 11099 from the Arkansas Department of Health, Little Rock.

Reprint requests to Abdul Rahman Jazieh, MD, MPH, University of Cincinnati, Barrett Center for Cancer, 234 Goodman Ave. ML 0501, Cincinnati, OH 45267-0501.