SEARCH

SEARCH BY CITATION

Keywords:

  • breast carcinoma;
  • cancer prevention and control;
  • cervical carcinoma;
  • colorectal carcinoma;
  • fecal occult blood testing (FOBT);
  • flexible sigmoidoscopy;
  • Papanicolaou (Pap) tests;
  • screening mammography

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

The “Southern Black Belt,” a term used for > 100 years to describe a subregion of the southern U.S., includes counties with high concentrations of African Americans and high levels of poverty and unemployment, and relatively high rates of preventable cancers.

METHODS

The authors analyzed data from a state-based telephone survey of adults age ≥ 18 years to compare the cancer screening patterns of African-American and white men and women in nonmetropolitan counties of this region, and to compare those rates with those of persons in other southern counties and elsewhere in the U.S. The primary study groups were comprised of 2165–5888 women and 1198 men in this region interviewed through the Behavioral Risk Factor Surveillance System. The respondents lived in predominantly rural counties in 11 southern states with sizeable African-American populations (≥ 24.5% of county residents). The main outcome measures were recent use of the Papanicolau (Pap) test, mammography, test for fecal occult blood in the stool (FOBT), and flexible sigmoidoscopy or colonoscopy.

RESULTS

Between 1998–2000, 66.3% (95% confidence interval [95% CI] ± 2.7%) of 1817 African-American women in the region age ≥ 40 years had received a mammogram within the past 2 years, compared with 69.3% (95% CI ± 1.8%) of 3922 white women (P = 0.066). The proportion of African-American and white women who had received a Pap test within the past 3 years was similar (85.7% [95% CI ± 1.9%] vs. 83.4% [95% CI ± 1.5%]; P = 0.068]. In 1997 and 1999, 29.3% of African-American women in these counties reported ever receiving an FOBT, compared with 36.9% in non-Black Belt counties and 42.5% in the remainder of the U.S. Among white women, 37.7% in Black Belt counties, 44.0% in non-Black Belt counties, and 45.3% in the remainder of the U.S. ever received an FOBT. Overall, similar patterns were noted among both men and women with regard to ever-use of FOBT, flexible sigmoidoscopy, or colonoscopy. Screening rates appeared to vary less by race than by region.

CONCLUSIONS

The results of the current study underscore the need for continued efforts to ensure that adults in the nonmetropolitan South receive educational messages, outreach, and provider recommendations concerning the importance of routine cancer screening. Cancer 2002;95:2211–22. Published 2002 by the American Cancer Society.

DOI 10.1002/cncr.10933

The “Southern Black Belt,” a term used for > 100 years by historians, demographers, sociologists, and writers, describes a largely impoverished subregion of the southern U.S. that includes nearly contiguous counties with high concentrations of African Americans.1–4 These counties predominately are rural and economically dependent on agriculture.4, 5 Social problems of the region include poverty, unemployment, low education, and relatively high rates of preventable cancers.2, 3, 6, 7

In studies conducted across the U.S., the poor have been shown to have a higher mortality from breast, cervical, and colorectal carcinoma.8–10 Decreased cancer survival among low-income individuals may be attributed to diminished access to health care, lack of health insurance, later stage of disease at the time of diagnosis, and other factors.8, 10, 11

Several studies have examined the use of cancer screening tests among persons in the U.S.12–20 Nationally, rates of mammography and Papanicolau (Pap) testing have been reported to be similar among African-American and white women in recent years and to have increased over time.19 Rates of colorectal carcinoma screening are low relative to the use of some other screening tests, but are similar among African-Americans and whites nationally.16, 20, 21

Prior studies have suggested that factors that influence cancer screening include cost, lack of health insurance, lack of transportation or child care, and fear of or fatalistic attitudes regarding cancer.22–28 Moreover, previous studies have shown that men and women who have not received routine health care or a provider's recommendation to undergo a cancer screening test are less likely to have been screened.29–31

Geographic area of residence also may be important in that cancer screening tests and other preventive health services may be used less frequently by persons in rural areas than by those in urban areas of the U.S.13, 16, 32, 33 Possible explanations include greater distance to medical facilities, reduced accessibility of services, and fewer providers trained to perform screening.34

Using data from the Behavioral Risk Factor Surveillance System (BRFSS), we examined whether the cancer screening rates of African-American and white men and women in nonmetropolitan counties of the southern Black Belt region of the U.S. demonstrate racial disparities, and we compared these rates with those of African-Americans and whites in other counties of southern states and with those of persons living elsewhere in the U.S. The screening tests were recent mammography, Pap test, fecal occult blood test (FOBT), and flexible sigmoidoscopy or colonoscopy. We also examined whether factors such as education and health insurance coverage help explain the disparities in cancer screening noted among men and women in the southern Black Belt region.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Definition of Black Belt Region

Federal agencies have not yet defined the Southern Black Belt region to our knowledge, although the region has been written about extensively.1–5, 35, 36 To be consistent with previous authors,2 we used a three-step process to identify counties in the Black Belt region in the current analysis (Fig. 1) First, 11 states (Virginia, North Carolina, South Carolina, Georgia, Florida, Alabama, Mississippi, Louisiana, Texas, Arkansas, and Tennessee) were identified that traditionally have been associated with the southern Black Belt region.2 Second, Census 2000 Redistricting Data (Public Law 94–171) were used to identify the percentage of respondents who indicated only one race and that was “black, African American, or Negro” (hereafter referred to as African American).37, 38 These percentages then were used to identify the subset of counties within the southern Black Belt states that had approximately 2 times (or greater) the national average percentage (12.3%) of African-American residents (i.e., counties in which the percentage of the population indicating only 1 race was ≥ 24.5% African American). Third, the Office of Management and Budget 1999 definitions of metropolitan areas were used to restrict further this subset of counties in southern Black Belt states with ≥ 24.5% African Americans to a nonmetropolitan county subset.39 This nonmetropolitan county subset is consistent with maps of the southern Black Belt region developed in 2001 by the Department of Housing and Consumer Economics at the University of Georgia.36

thumbnail image

Figure 1. Southern Black Belt counties of the U.S.

Download figure to PowerPoint

BRFSS Interview Questions

The BRFSS is a state-based telephone survey of adults age ≥ 18 years.40, 41 The BRFSS uses a random-digit-dialing technique and multistage cluster sampling in each participating state to sample noninstitutionalized adults who have telephones.42 A computer-assisted interview is administered by trained interviewers. The interviews include questions regarding general health status, tobacco use, alcohol consumption, demographic and socioeconomic characteristics, screening mammography, Pap tests, colorectal carcinoma screening tests, and other health topics. Each adult female respondent was asked whether she had ever had a mammogram. Those who responded positively were asked how long it had been since their last mammogram. Similar questions were asked concerning the Pap test. Women also were asked whether they had undergone a hysterectomy because Pap tests are not indicated for the majority of women who have no cervix.

With respect to screening for colorectal carcinoma, male and female respondents were asked: “A blood stool test is a test that may use a special kit at home to determine whether the stool contains blood. Have you ever had this test using a home kit?” They also were asked: “A sigmoidoscopy or colonoscopy is when a tube is inserted in the rectum to view the bowel for signs of cancer and other health problems. Have you ever had this exam?” Respondents further were asked how long it had been since they had had their last FOBT, flexible sigmoidoscopy, or colonoscopy.

The estimated response rates (Council of American Survey Research Organizations43) in the 1997 BRFSS, among households of all races and ethnicities in southern states, ranged from 45.7% (Texas) to 76.8% (Georgia). (The numerator denotes the number of completed responses and the denominator is an estimate of the number of households in the sample.) The response rates in the 1998 BRFSS in southern states ranged from 41.2% (Texas) to 65.0% (Arkansas). The response rates in the 1999 BRFSS in southern states ranged from 36.2% (Texas) to 63.4% (Mississippi). Response rates for the 2000 BRFSS were not available at the time of publication. Response rates for southern Black Belt counties also were unavailable.

Survey Sample

The BRFSS survey sample included 1198 men in southern Black Belt counties, 7394 men in non-Black Belt counties of 11 southern states, and 33,469 men elsewhere in the U.S. Similarly, the survey sample included 2165–5888 women in southern Black Belt counties (the numbers varied according to which cancer screening test was being examined and which years were of interest), 12,873–37,473 women in non-Black Belt counties of southern states, and 52,733–156,683 women in the remainder of the U.S. For breast and cervical carcinoma screening, data were pooled for 1998–2000 to increase the size of the sample available for analyses. Data regarding colorectal carcinoma screening were available for 1997 and 1999. The analysis was limited to African-American and white individuals because other racial and ethnic groups were too few in this sample to report screening for them separately. Hispanics were coded according to their race. The ages of 6 men and 62 women in the sample from the Black Belt region were unknown.

Statistical Analysis and Definition of Variables

Analyses of screening mammogram and clinical breast examination use were limited to women who were age ≥ 40 years (n = 5840). Analyses of Pap test use were limited to those women age ≥ 18 years of age who had not undergone a hysterectomy (n = 5888). Analyses of colorectal carcinoma screening were limited to men (n = 1198) and women (n = 2165) who were age ≥ 50 years. Age groups were defined according to current screening guidelines.

U.S. Department of Agriculture 1993 Rural Urban Continuum Codes (to our knowledge the most recent version available) were used to characterize the degree of urbanization of a county and its proximity to a metropolitan area.44 For the current study, Codes 4 and 5 were analyzed as nonmetropolitan, nonrural counties (i.e., suburban areas and small towns) and Codes 6–9 as rural counties (i.e., a completely rural or nonmetropolitan county with an urban population of < 19,999).

The county of residence of each respondent also was classified according to whether the Health Resources Services Administration Area Resource File listed the entire county as a Primary Care Health Professional Shortage Area in 1998.45

Age-adjusted rates of screening test use were calculated for the time period of interest. General linear contrasts were used to test for differences in screening rates and other characteristics after adjusting for age and calendar year.46 The direct method was used to adjust estimates of the proportion of men or women screened for colorectal carcinoma for age and calendar year using the distribution for all men or women in the U.S. who participated in BRFSS surveys (1997 and 1999) as the standard. Similarly, the direct method was used to adjust estimates of the proportion of women screened for breast and cervical carcinoma for age and calendar year using the distribution for all women in the U.S. who participated in BRFSS surveys (1998–2000). All analyses used SUDAAN software to calculate the 95% confidence intervals (95% CI) and to allow for weighting of the estimates.47 The samples were weighted to compensate for the unequal sampling probability resulting from the unique number of telephones per household; the number of unique telephone numbers per primary sampling unit; and poststratification by age, gender, and race. Separate multivariable logistic regression models were created to identify characteristics associated with recent mammogram, Pap test, and FOBT.48 Covariates for categories of educational attainment, rather than household income, were included in the models to avoid problems with colinearity and missing data. Indicator variables for the survey year and age categories were included in all models. Two or more indicator variables were included for categoric variables.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Characteristics of Study Sample

Table 1 compares characteristics of African-American and white, male and female BRFSS respondents in the southern Black Belt region (1997 and 1999) who were age ≥ 50 years. On average, African-American individuals in the region were more likely to be female, less likely to be married, less educated, more likely to report fair or poor general health status, more likely to have seen a physician within the past year, less likely to have health insurance, more likely to reside in a health professional shortage area, and less likely to consume alcohol. African-American individuals also were reported to have a lower household income compared with whites. Similar results were obtained for women age ≥ 40 years and for women age ≥ 18 years who had not undergone a hysterectomy (results not shown).

Table 1. Characteristics of African-American and White Men and Women Age ≥ 50 years in the Black Belt Region of the Southern U.S. Behavioral Risk Factor Surveillance Survey, 1997 and 1999a
 African-American % (95% CI)White % (95% CI)
  • 95% CI: 95% confidence interval; GED: General Education Development certificate, tech: technical.

  • a

    Weighted population estimates adjusted for age and year of survey; men and women who responded don't know or not sure or who refused were excluded.

  • b

    P < 0.01.

  • c

    P < 0.001.

  • d

    P < 0.05, from overall test of differences across groups.

Age (yrs)  
 50–6970.6 (67.0–74.3)66.7 (64.4–68.9)
 ≥ 7029.4 (25.7–33.0)33.3 (31.1–35.6)
Genderb  
 Male38.1 (34.0–42.2)44.8 (42.4–47.2)
 Female61.9 (57.8–66.0)55.2 (52.8–57.6)
Marital statusc  
 Currently married48.2 (44.3–52.2)71.0 (69.1–73.0)
 Divorced or separated16.8 (14.1–19.5)9.4 (8.1–10.7)
 Widowed27.8 (24.7–31.0)17.4 (15.9–19.0)
 Never married or living as unmarried couple7.2 (5.0–9.3)2.1 (1.5–2.7)
Educational attainmentc  
 < High school graduate54.5 (50.7–58.2)24.4 (22.5–26.4)
 High school graduate/GED28.6 (25.0–32.1)37.0 (34.8–39.3)
 Some college/tech school8.2 (6.2–10.3)21.5 (19.4–23.6)
 College graduate8.7 (6.5–11.0)17.0 (15.2–18.8)
Household incomec  
 < $15,00045.6 (41.1–50.1)20.0 (18.0–22.0)
 $15,000–$34,99943.3 (38.6–47.9)43.1 (40.4–45.8)
 $35,000–$49,9996.6 (4.4–8.8)16.2 (14.2–18.2)
 ≥ $50,0004.5 (2.7–6.4)20.7 (18.5–22.8)
Employment statusc  
 Currently employed32.0 (28.9–35.0)36.9 (35.0–38.8)
 Homemaker or retired50.3 (47.7–52.9)54.2 (52.4–55.9)
 Unemployed2.8 (1.6–3.9)1.4 (0.9–1.9)
 Unable to work15.0 (12.3–17.7)7.5 (6.2–8.7)
No. of persons in householdc  
 127.7 (24.6–30.8)22.4 (20.7–24.0)
 238.1 (34.2–42.0)58.5 (56.2–60.8)
 ≥ 334.2 (30.1–38.4)19.2 (17.1–21.2)
General health statusc  
 Good to excellent52.5 (48.6–56.4)70.5 (68.3–72.6)
 Fair or poor47.5 (43.6–51.4)29.5 (27.4–31.7)
Saw a physician within past yearc  
 Yes87.2 (84.5–89.9)78.2 (76.3–80.2)
 No12.8 (10.1–15.5)21.8 (19.8–23.7)
Any health insurance coveragec  
 Yes82.8 (79.9–85.7)91.9 (90.7–93.2)
 No17.2 (14.3–20.1)8.1 (6.8–9.3)
Resident of health professional shortage aread  
 Yes44.7 (40.7–48.8)40.0 (37.4–42.6)
 No55.3 (51.2–59.3)60.0 (57.4–62.6)
Area of residence  
 Rural78.2 (74.8–81.5)77.2 (75.0–79.4)
 Suburban area/small town21.8 (18.5–25.2)22.8 (20.6–25.0)
Current cigarette smoking  
 Yes16.6 (13.7–19.6)17.4 (15.6–19.3)
 No83.4 (80.4–86.3)82.6 (80.7–84.4)
Current alcohol consumptionc  
 Yes16.0 (12.9–19.1)25.4 (23.3–27.5)
 No84.0 (80.9–87.1)74.6 (72.5–76.7)

Comparisons of Cancer Screening Rates in Southern Black Belt Counties With Other Regions of the U.S.

Table 2 shows breast, cervical, and colorectal carcinoma screening rates among African-American and white women in southern Black Belt region counties, in other counties of southern states, and in the remainder of the U.S. African-American women were not less likely to have had a mammogram within the past 2 years or a Pap test within the past 3 years compared with white women in southern Black Belt region counties (Table 2). African-American and white women in southern Black Belt counties were less likely to have had a recent mammogram than were African-American and white women elsewhere in the U.S. (P < 0.001), but similar geographic differences were not noted for recent Pap tests (Table 2). Receipt of colorectal carcinoma screening varied both by race and region. With regard to the ever-use of FOBT among women, African-American women in both Black Belt and non-Black Belt counties were less likely to have received the test than were white women. Both African-American and white women in Black Belt counties also were less likely to have ever received an FOBT than were women in non-Black Belt counties and across the U.S. This same pattern of recent use of sigmoidoscopy was observed among women (Table 2).

Table 2. Percentage of African-American and White Women in Black Belt Counties, in Non-Black Belt Counties of Southern States, and in the Remainder of the U.S. who Received a Mammogram, Clinical Breast Examination, or Pap Test in 1998–2000, or a FOBT, Sigmoidoscopy, or Colonoscopy in 1997 or 1999. Behavioral Risk Factor Surveillance Systema
 Black Belt countiesNon-Black Belt counties of southern statesRemainder of U.S.
African-AmericanWhiteAfrican-AmericanWhiteAfrican-AmericanWhite
No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)
  • Pap: Papanicolaou; FOBT: fecal occult blood test; 95% CI: 95% confidence. No. indicates the total number of subjects within strata who did or did not undergo the cancer screening test.

  • a

    Weighted population estimates adjusted for age and year using the distribution for all women who participated in Behavioral Risk Factor Surveillance System surveys in the U.S. in 1998–2000 (for adjustment of breast and cervical carcinoma screening rates) or 1997 and 1999 (for adjustment of colorectal carcinoma screening rates). Women who responded don't know or not sure or who refused were excluded.

  • b–f

    Women age ≥ 40 years.

  • b–f, c–f, c

    P < 0.001, comparing African-American women in Black Belt counties with African-American women in non-Black Belt counties of southern states.

  • b–f, d, c–f

    P < 0.001, comparing white women in Black Belt counties with white women in non-Black Belt counties of southern states.

  • b–f, e, c–f

    P < 0.001, comparing African-American women in Black Belt counties with African-American women in the remainder of the U.S.

  • b–f, f, c–f

    P < 0.001, comparing white women in Black Belt counties with white women in the remainder of the U.S.

  • g

    Women age ≥ 18 years who had not undergone a hysterectomy.

  • h–j, h

    Women age ≥ 50 years.

  • h–j

    P < 0.01, comparing African-American women in Black Belt counties with white women in Black Belt counties.

  • h–j

    P < 0.01, comparing African-American women in Black Belt counties with African-American women in non-Black Belt counties of southern states.

  • k

    P < 0.001, comparing African-American women in Black Belt counties with white women in Black Belt counties.

  • l–n

    P < 0.05, comparing African-American women in Black Belt counties with white women in Black Belt counties.

  • l–n

    P < 0.01, comparing African-American women in Black Belt counties with African-American women in the remainder of the U.S.

  • l–n

    P < 0.05, comparing white women in Black Belt counties with white women in the remainder of the U.S.

Mammogram within past 2 yrsb–f181766.3 (± 2.7%)392269.3 (± 1.8%)481074.2 (± 1.6%)27,21974.6 (± 0.6%)710577.7 (± 1.7%)120,85075.5 (± 0.4%)
Pap test within past 3 yearsg227085.7 (± 1.9%)352483.4 (± 1.5%)709785.9 (± 1.3%)30,07984.8 (± 0.6%)10,69486.6 (± 1.3%)144,78184.7 (± 0.3%)
Ever had an FOBTdefh–j63529.3 (± 4.3%)149937.7 (± 3.0%)165836.9 (± 3.0%)11,02944.0 (± 1.2%)264142.5 (± 3.2%)48,56145.3 (± 0.8%)
FOBT within past yearc–fh62513.5 (± 3.2%)147015.6 (± 2.2%)163721.0 (± 2.5%)10,90722.2 (± 1.0%)261825.2 (± 2.8%)48,05922.1 (± 0.6%)
Ever underwent a sigmoidoscopy or colonoscopycehk62427.1 (± 4.1%)148238.8 (± 3.0%)163637.2 (± 3.0%)10,95441.3 (± 1.2%)261437.7 (± 3.1%)48,36041.0 (± 0.7%)
Underwent a sigmoidoscopy or colonoscopy within past 5 yearschl–n62121.4 (± 3.8%)146326.2 (± 2.7%)162030.5 (± 2.9%)10,85728.7 (± 1.1%)259128.3 (± 2.8%)48,07929.5 (± 0.7%)

African-American men age ≥ 50 years were less likely to have had a recent FOBT than were white men in southern Black Belt counties (P = 0.015), African-American men in other counties of southern states (P < 0.001), and African-American men elsewhere in the U.S. (P < 0.001) (Table 3). African-American men also were less likely to have undergone a recent sigmoidoscopy or colonoscopy compared with African-American men elsewhere in the U.S. (P = 0.029). Among white men, rates of recent colorectal carcinoma screening were lower in southern Black Belt counties compared with other counties of southern states (P < 0.001) and in the remainder of the U.S. (P < 0.001).

Table 3. Percentage of African-American and White Men, aged ≥ 50 Years, in Black Belt Counties, in non-Black Belt Counties of Southern States and in the Remainder of the U.S. Who Had Undergone a FOBT, Sigmoidoscopy, or Colonoscopy. Behavioral Risk Factor Surveillance System, 1997 and 1999a
 Black Belt countiesNon-Black Belt counties of southern statesRemainder of U.S.
African-AmericanWhiteAfrican-AmericanWhiteAfrican-AmericanWhite
No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)
  • FOBT: fecal occult blood test; 95% CI: 95% confidence interval. No. indicates the total number of subjects within strata who did or did not undergo the cancer screening test.

  • a

    Weighted population estimates adjusted for age and year using the distribution for all men who participated in Behavioral Risk Factor Surveillance Systems surveys in the U.S. in 1997 and 1999.

  • b–d, b–f, b

    P < 0.001, comparing white men in Black Belt counties with white men in non-Black Belt counties of southern states.

  • b–d, b–f

    P < 0.001, comparing African-American men in Black Belt counties with African-American men in the remainder of the U.S.

  • b–d, b–f, d

    P < 0.001, comparing white men in Black Belt counties with white men in the remainder of the U.S.

  • b–f

    P < 0.05, comparing African-American men in Black Belt counties with white men in Black Belt counties.

  • b–f

    P < 0.01, comparing African-American men in Black Belt counties with African-American men in non-Black Belt counties of southern states.

  • g

    P < 0.05, comparing African-American men in Black Belt counties with African-American men in the remainder of the U.S.

Ever had an FOBTb–d26321.9 (± 5.8%)91627.8 (± 3.4%)77527.5 (± 3.8%)653337.7 (± 1.5%)139935.8 (± 4.1%)31,25437.2 (± 0.9%)
FOBT within past yearb–f2607.7 (± 3.7%)91113.3 (± 2.6%)76915.4 (± 3.0%)648418.9 (± 1.2%)139321.6 (± 3.5%)31,06019.2 (± 0.7%)
Ever undergone a sigmoidoscopy or colonoscopybd26637.5 (± 7.0%)90637.4 (± 3.7%)77139.3 (± 4.3%)649446.4 (± 1.5%)138344.9 (± 4.3%)31,08247.3 (± 0.9%)
Sigmoidoscopy or colon-oscopy within past 5 yearsbdg26531.3 (± 6.7%)90027.9 (± 3.4%)75833.0 (± 4.1%)645135.0 (± 1.5%)137540.2 (± 4.3%)30,94237.3 (± 0.9%)

Multivariate logistic regression analyses (results not shown) suggested that the odds of recent mammography among women in the southern Black Belt region was similar to that of other women elsewhere in the U.S. after adjustment for other predictors of screening (adjusted odds ratio [OR] = 0.96; 95% CI, 0.87–1.05) and that the odds of a recent Pap test among women in the southern Black Belt was slightly higher than that of other women elsewhere in the U.S. after adjustment for other predictors of screening (adjusted OR = 1.13; 95% CI, 1.00–1.28). Multivariate logistic regression analyses also confirmed that the odds of FOBT within the past year among men and women in the southern Black Belt was lower than that of other men and women elsewhere in the U.S. after adjustment for other predictors of screening (adjusted OR = 0.79; 95% CI, 0.69–0.91) (P < 0.004) (results not shown).

Factors Associated with Cancer Screening in the Southern Black Belt

Stratum specific rates of recent mammography (adjusted for age and calendar year) for women in the Black Belt region (results not shown) demonstrated associations between not having had a mammogram within the past 2 years and older age, unmarried status, lower educational level, lower household income, unemployed status, fair or poor health status, not having seen a physician within the past year, lack of health insurance coverage, residence in a health professional shortage area, rural residence, cigarette smoking, and lack of adherence to cervical carcinoma screening guidelines. Stratum specific Pap test rates for women in the Black Belt region demonstrated similar associations (results not shown).

Stratum specific colorectal carcinoma screening rates for men and women in the Black Belt region (results not shown) demonstrated that not having had an FOBT within the past year was associated with younger age, black race, lower education level, unemployed status, having three or more persons in the household, not having seen a physician within the past year, lack of health insurance coverage, rural residence, and cigarette smoking. A similar pattern was noted for sigmoidoscopy or colonoscopy within the past 5 years except that the associations with race, employment status, lack of health insurance coverage, number of persons in the household, rural residence, and cigarette smoking were not found to be statistically significant. Moreover, women, unmarried persons, and low-income persons were less likely to have been screened for colorectal carcinoma by sigmoidoscopy or colonoscopy (results not shown).

A multivariate analysis was conducted using data from respondents in the Black Belt region to determine whether associations with race, socioeconomic factors, and other factors persisted after adjusting for a variety of variables. According to multivariate analysis, several factors were associated with recent mammography (Table 4). For example, a higher level of education was associated with having had a mammogram within the past 2 years (P < 0.001). Having seen a physician within the past year, health insurance coverage, good or excellent health status, lack of current cigarette smoking, and currently being married all were associated with having had a recent mammogram. No association was observed with black versus white race in multivariate analysis.

Table 4. Multivariate Results for Having Had a Mammogram within the Past 2 Years among Women age ≥ 40 Years in the Black Belt Region of the U.S. Behavioral Risk Factor Surveillance System, 1998–2000a
 Adjusted OR(95% CI)
  • OR: odds ratio; 95% CI: 95% confidence interval; GED: General Education Development certificate; tech: technical.

  • a

    Women who responded don't know or not sure or who refused to answer were excluded.

  • b

    P < 0.001 based on the Wald chi-square test.

  • c

    P < 0.01 based on the Wald chi-square test.

Survey yearb  
 19981.00 
 19990.93(0.77–1.13)
 20001.50(1.24–1.80)
Age (yrs)b  
 40–491.00 
 50–691.57(1.27–1.94)
 ≥ 701.06(0.78–1.44)
Race  
 White1.00 
 Black0.99(0.82–1.17)
Marital statusb  
 Currently married1.00 
 Divorced or separated0.86(0.67–1.10)
 Widowed0.62(0.49–0.80)
 Never married or living as unmarried couple0.55(0.38–0.79)
Educational attainmentb  
 < High school graduate1.00 
 High school graduate/GED1.25(1.02–1.53)
 Some college/tech, school1.56(1.23–1.99)
 College graduate1.81(1.39–2.35)
No. of persons in household  
 3+1.00 
 21.04(0.85–1.28)
 11.28(1.01–1.64)
Employment status  
 Currently employed1.00 
 Homemaker or retired1.06(0.85–1.32)
 Unemployed0.96(0.57–1.62)
 Unable to work1.09(0.80–1.49)
General health statusc  
 Good to excellent1.34(1.12–1.60)
 Fair or Poor1.00 
Current cigarette smokerb  
 Yes0.59(0.48–0.71)
 No1.00 
Saw physician within past yearb  
 Yes6.31(5.15–7.73)
 No1.00 
Any health insurance coverageb  
 Yes2.07(1.67–2.57)
 No1.00 
Area of residence  
 Rural1.00 
 Suburban area/small town1.14(0.94–1.38)
Resident of health professional shortage area  
 Yes1.00 
 No0.91(0.77–1.07)

Multivariate analysis revealed that several factors were associated with a recent Pap test: having seen a physician within the past year, a higher education level, and currently being married (Table 5). An association between recent Pap test and black versus white race also was observed on multivariate analysis (adjusted OR, 1.98; 95% CI, 1.48–2.64).

Table 5. Multivariate Results for Having Had a Pap Test within the Last 3 Years among Women age ≥ 18 Years in the Black Belt Region of the U.S. Behavioral Risk Factor Surveillance System, 1998–2000a
 Adjusted OR(95% CI)
  • Pap: Papanicolaou; OR: odds ratio; 95% CI: 95% confidence interval; GED: General Education Development certificate; tech: technical.

  • a

    Women who responded don't know or not sure or who refused to answer were excluded along with those who had undergone hysterectomy.

  • b

    P < 0.05 based on the Wald chi-square test.

  • c

    P < 0.001 based on the Wald chi-square test.

Survey yearb  
 19981.00 
 19991.25(0.96–1.63)
 20001.38(1.07–1.78)
Age (yrs)c  
 18–291.00 
 30–390.77(0.55–1.08)
 40–490.46(0.32–0.66)
 50–690.34(0.24–0.50)
 ≥ 70 years0.14(0.08–0.24)
Racec  
 White1.00 
 Black1.98(1.48–2.64)
Marital statusc  
 Currently married1.00 
 Divorced or separated0.61(0.43–0.88)
 Widowed0.43(0.29–0.64)
 Never married or living as unmarried couple0.27(0.19–0.38)
Educational attainmentc  
 < High school graduate1.00 
 High school graduate/GED1.51(1.12–2.04)
 Some college/tech school2.13(1.49–3.05)
 College graduate3.04(2.06–4.50)
No. of persons in household  
 3+1.00 
 21.27(0.96–1.68)
 11.44(1.02–2.03)
Employment status  
 Currently employed1.00 
 Homemaker or retired0.96(0.70–1.31)
 Unemployed1.23(0.76–1.99)
 Unable to work0.95(0.59–1.52)
General health status  
 Good to excellent1.16(0.88–1.54)
 Fair or poor1.00 
Current cigarette smoker  
 Yes1.05(0.80–1.37)
 No1.00 
Saw physician within past yearc  
 Yes6.64(5.26–8.38)
 No1.00 
Any health insurance coverage  
 Yes1.11(0.85–1.46)
 No1.00 
Resident of health professional shortage area  
 Yes1.01 
 No1.00(0.80–1.26)
Area of residence  
 Rural1.00 
 Suburban area/small town1.00(0.78–1.30)

Several factors were associated with recent FOBT on multivariate analysis among men and women in the Black Belt region: older age, having seen a physician within the past year, educational level, and lack of cigarette smoking (Table 6). The associations with race and gender were not found to be statistically significant.

Table 6. Multivariate Results for Having Had a FOBT within the Last Year among Men and Women age ≥ 50 Years in the Black Belt Region of the U.S. Behavioral Risk Factor Surveillance System, 1997 and 1999a
 Adjusted OR(95% CI)
  • FOBT: fecal occult blood testing; OR: odds ratio; 95% CI: 95% confidence interval; GED: General Education Development certificate; tech: technical.

  • a

    Persons who responded don't know or not sure or who refused to answer were excluded.

  • b

    P < 0.05 based on the Wald chi-square test.

  • c

    P < 0.01 based on the Wald chi-square test.

  • d

    P < 0.001 based on the Wald chi-square test.

Yearb  
 19971.00 
 19991.33(1.03–1.72)
Age (yrs)b  
 50–691.00 
 ≥ 701.42(1.03–1.96)
Race  
 White1.00 
 Black0.80(0.59–1.09)
Gender  
 Male1.00 
 Female1.12(0.85–1.48)
Marital status  
 Currently married1.00 
 Divorced or separated1.01(0.61–1.68)
 Widowed0.87(0.56–1.34)
 Never married or living as unmarried couple0.38(0.15–0.94)
Educational attainmentb  
 < High school graduate1.00 
 High school graduate/GED1.51(1.09–2.10)
 Some college/tech school0.96(0.64–1.44)
 College graduate1.35(0.89–2.03)
No. of persons in household  
 3+1.00 
 21.51(1.01–2.27)
 11.25(0.73–2.12)
Employment status  
 Currently employed1.00 
 Homemaker or retired1.08(0.77–1.51)
 Unemployed0.41(0.12–1.42)
 Unable to work1.54(0.97–2.45)
General health status  
 Good to excellent0.91(0.68–1.21)
 Fair or poor1.00 
Current cigarette smokerc  
 Yes0.58(0.39–0.86)
 No1.00 
Saw physician within past yeard  
 Yes6.90(3.55–13.40)
 No1.00 
Any health insurance coverage  
 Yes1.26(0.74–2.14)
 No1.00 
Resident of health professional shortage area  
 Yes1.00 
 No0.91(0.70–1.19)
Area of residence  
 Rural1.00 
 Nonrural1.24(0.93–1.66)

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The U.S. Department of Health and Human Services sets nationwide target goals for breast, cervical, and colorectal carcinoma screening.49 A primary goal is to achieve equity in health by eliminating disparities. Year 2010 objectives include increasing to at least 70% the percentage of women age ≥ 40 years who received a mammogram within the preceding 2 years. Year 2010 objectives for the nation also include increasing to at least 97% the percentage of women age ≥ 18 years who have ever received a Pap test and increasing to at least 90% the percentage of women who received a Pap test within the preceding 3 years. The results of the current study suggest that women in nonmetropolitan counties of the southern Black Belt region of the U.S. may be approaching these objectives. The lack of racial disparity in recent Pap test rates among women in the Black Belt region may be attributed in part to the increased opportunity for a Pap test associated with childbearing status and contraceptive methods used for family planning. Women who recently have been pregnant or are contemplating pregnancy are more likely to receive Pap tests in connection with standard prenatal and postnatal care. Similarly, sexually active women of childbearing age who use oral contraceptives are more likely to receive regular Pap tests as part of family planning services administered at the majority of county health departments or by private practitioners.

Colorectal carcinoma screening rates are low relative to other cancer screening tests, and disparities in colorectal carcinoma screening were evident by race and by region, after adjustment for age and calendar year. African-American women in Black Belt counties were less likely to have ever received a FOBT compared with white women, and women in Black Belt counties of either race were less likely than those in non-Black Belt counties and the remainder of the U.S. to have received colorectal carcinoma screening tests. For recent test use, racial differences were less notable, but regional differences persisted. These differences by race are not detected when BRFSS data are analyzed at the national level.20, 21

In the current study, rates of recent mammogram and FOBT varied across categories of residence (rural areas vs. small towns and suburban areas) within the southern Black Belt region, but the associations with rural residence did not persist in multivariate analysis. The greater use of cancer screening services among men and women in small towns and suburban areas compared with rural areas may be explained by the greater availability of medical services in nonrural areas. Men and women living in rural areas may have less access to health practitioners and, consequently, fewer preventive health services.32–34 Physician recommendation, which is known to have a positive influence on the use of cancer screening tests, has been reported to vary significantly between rural and urban individuals.29–32 The disparities in colorectal carcinoma screening may be due in part to unequal physician supply in rural areas. Nevertheless, residence in a health professional shortage area was not found to be associated independently with decreased colorectal carcinoma screening in the current study.

Ensuring that routine cancer screening is available to all is important because differences in cancer rates by race and socioeconomic status have been reported. African-American women are at greater risk of dying of invasive cervical carcinoma than are white women in the U.S.11, 50, 51 African-American women also are more likely to be diagnosed with breast carcinoma at a later stage of disease and have higher mortality rates, although breast carcinoma incidence rates are lower in African-American women compared with white women.50, 51 In recent years, incidence rates for colorectal carcinoma have declined among white men and women in the U.S., but the rates have leveled off for African-American men and women and are higher than those of whites.52

A possible limitation of the current study is response bias because the telephone survey excluded persons living in households without a telephone and response rates were low. Variation in response rates across states could account in part for geographic or regional variation in estimates of cancer screening rates. Thus, low response rates in many of the southern states (and in many states outside the South) is a major limitation of the analysis. BRFSS response rates specifically for persons in the southern Black Belt region currently are unavailable. Response rates by race or socioeconomic status also are unavailable at the current time. With respect to the possible effects of nonresponse on the estimates of cancer screening, men and women who lack a household telephone are more likely to be poor or residents of rural areas, and several studies have demonstrated that cancer screening rates are lower among low-income persons and among those who live in rural areas of the U.S.13, 14, 16 Thus, the estimates of cancer screening observed in the current survey may be biased upward to some extent, and the differences in cancer screening between the southern Black Belt and other geographic regions may have been underestimated.

Self-reported information regarding cancer screening practices may differ from information obtained from the records of healthcare providers. An important bias could have occurred in the current study if the accuracy or completeness of self-reported screening varies substantially by race, socioeconomic status, or geographic region. For example, low-income persons in the rural South may be less likely to participate in telephone surveys or less likely to report their use of cancer screening tests accurately compared with higher-income persons in other regions of the U.S., although data are needed with which to address these questions. An important question regarding self-report is whether diverse groups of respondents may differentially overstate (or understate) cancer screening use. Validation studies have suggested that patients tend to overreport their use of screening and underestimate the time since their last screen.53–55 Despite these limitations, the major findings of the current study are likely to be valid, and they are unlikely to be accounted for completely by selection bias.

The need remains to ensure that medically underserved persons such as those in the southern Black Belt region have access to cancer screening and that they receive educational messages, outreach, and provider recommendations concerning the importance of routine cancer screening. Efforts currently underway in the U.S. include the National Breast and Cervical Cancer Early Detection Program of the U.S. Centers for Disease Control and Prevention (CDC), which provides support to states for breast and cervical carcinoma screening services for medically underserved women.56 Efforts to increase awareness of colorectal carcinoma screening have been intensifying through initiatives such as the designation by Congress of March as National Colorectal Cancer Awareness Month and the CDC-led federal campaign “Screen for Life.” However, additional efforts are needed to reach special populations such as those in the southern Black Belt region and other rural communities in the southern U.S.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Washington BT. Up from slavery. New York: Dover Publications, 1995. [originally published: New York: Doubleday, 1901].
  • 2
    Wimberley RC, Morris LV. The Southern Black Belt. A national perspective. Lexington, KY: University of Kentucky, TVA Rural Studies, 1997.
  • 3
    Morris L, Wimberley RC. The reference book on regional well-being: U.S. regions, the Black Belt, and Appalachia. Location: Mississippi State, Southern Rural Development Center, 1996. Available from URL: http://www.icapp.org/biblio/final.html [accessed April 1, 2002].
  • 4
    LysonTA, FalkWW, editors. Forgotten places. Uneven development and the loss of opportunity in rural America. Lawrence, KS: University Press of Kansas, 2000.
  • 5
    Adu-Nyako K, Thompson A. Food choices: perceptions and purchase behavior of persons in Black Belt states. Greensboro, NC: North Carolina A & T State University, 2000. Available from URL: http://www.ag.ncat.edu/research/projects/2000/food_choices.htm [accessed April 1, 2002].
  • 6
    Pickle LW, Mason TJ, Howard N, et al. Atlas of U.S. cancer mortality among nonwhites: 1950–1980. Washington (DC): U.S. Government Printing Office; 1990 DHHS Pub. No. NIH 90-1582.
  • 7
    Devesa SS, Grauman DJ, Blott WJ, et al. Atlas of cancer mortality in the United States, 1950-94. Bethesda (MD): National Cancer Institute; 1999 NIH Pub. No. 99-4564.
  • 8
    Freeman HP. Cancer in the socioeconomically disadvantaged. CA Cancer Jo Clin. 1989; 39: 266288.
  • 9
    Freeman H. Race, poverty, and cancer [editorial]. J Natl Cancer Inst. 1991; 83: 526527.
  • 10
    Tomatis L. Poverty and cancer [editorial]. Cancer Epidemiol Biomarkers Prev. 1992; 1: 167175.
  • 11
    Weiss HL, Soong S-J, Partridge EE, et al. Cancer incidence among predominantly black, rural-poor populations in southern states. South Med J 1997; 90: 986992.
  • 12
    O'Malley MS, Earp JAL, Harris RP. Race and mammography use in two North Carolina Counties. Am J Public Health. 1997; 87: 782786.
  • 13
    Duelberg SI. Preventive health behavior among black and white women in urban and rural areas. Soc Sci Med. 1992; 34: 191198.
  • 14
    Rakowski W, Pearlman D, Rimer BK, et al. Correlates of mammography among women with low and high socioeconomic resources. Prev Med. 1995; 24: 149158.
  • 15
    Anderson LM, May DS. Has the use of cervical, breast, and colorectal cancer screening increased in the United States? Am J Public Health. 1995; 85: 840842.
  • 16
    Breen N, Kessler L. Changes in the use of screening mammography: evidence from the 1987 and 1990 National Health Interview Surveys. Am J Public Health. 1994; 84: 6267.
  • 17
    Frazier EL, Jiles RB, Mayberry R. Use of screening mammography and clinical breast examinations among Black, Hispanic, and white women. Prev Med. 1996; 25: 118125.
  • 18
    Pearlman DN, Ehrich B, Rakowski W, et al. Breast cancer screening practices among Black, Hispanic, and White women: reassessing differences. Am J Prev Med. 1996; 12: 327337.
  • 19
    Blackman DK, Bennet EM, Miller DS. Trends in self-reported use of mammograms (1989-1997) and Papanicolaou tests (1991–1997)—United States. MMWR morb mortal wkly Rep CDC Surveill Summ. 1999; 48: 122.
  • 20
    Centers for Disease Control and Prevention. Screening for colorectal cancer-United States, 1997. MMWR morb mortal wkly Rep CDC Surveill Summ. 1999; 48: 116121.
  • 21
    Centers for Disease Control and Prevention. Trends in screening for colorectal cancer-United States, 1997 and 1999. MMWR Morb Mortal Wkly Rep CDC Surveill Summ. 2001; 50: 162166.
  • 22
    Brown ML, Potosky AL, Thompson GB, et al. The knowledge and use of screening tests for colorectal and prostate cancer: data from the 1987 National Health Interview Survey. Prev Med. 1990; 19: 562574.
  • 23
    Erwin DO, Spatz TS, Stotts RC, et al. Increasing mammography practice by African American women. Cancer Pract. 1999; 7: 7885.
  • 24
    Altpeter M, Earp JAL, Schopler JH. Promoting breast cancer screening in rural, African American communities. The science and art of community health promotion. Health Soc Work. 1998; 23: 104115.
  • 25
    Lemon SMS, Zapka J, Puleo E, et al. Colorectal cancer screening participation: comparisons with mammography and prostate-specific antigen screening. Am J Public Health. 2001; 91: 12641272.
  • 26
    Vernon SW. Participation in colorectal cancer screening: a review. J Natl Cancer Inst. 1997; 89: 14061499.
  • 27
    Peterson SK, Vernon SW. A review of patient and physician adherence to colorectal cancer screening guidelines. Semin Colon Rectal Surg. 2000; 11: 5872.
  • 28
    Strickland J, Strickland DL. Barriers to preventive health services for minority households in the rural south. J Rural Health. 1996; 12: 206217.
  • 29
    Fox SA, Murata PJ, Stein JA, et al. The impact of physician compliance on screening mammography for older women. Arch Intern Med. 1991; 151: 5056.
  • 30
    Rimer BK, Trock B, Engstrom PF, et al. Why do some women get regular mammograms? Am J Prev Med. 1991; 7: 6974.
  • 31
    Zapka JG, Stoddard AM, Costanza ME, et al. Breast cancer screening by mammography: utilization and associated factors. Am J Public Health. 1989; 79: 14991502.
  • 32
    Zhang P, Tao G, Irwin KL. Utilization of preventive medical services in the United States: a comparison between rural and urban populations. J Rural Health. 2000; 16: 349356.
  • 33
    Bryant H, Mah Z. Breast cancer screening attitudes and behaviors of rural and urban women. Prev Med. 1992; 21: 405418.
  • 34
    Carr WP, Maldonado G, Leonard PR, et al. Mammogram utilization among farm women. J Rural Health. 1996; 12 (Suppl): 278290.
  • 35
    McMurray J. Appalachian agency inspires senator's call to tackle ‘Black belt’ poverty. Naples Daily News 2001 Aug 18.
  • 36
    Eversley M. Bid to aid Black Belt builds steam. The Atlanta Journal-Constitution 2002 Jan 27; B1, B5.
  • 37
    GeoLytics, Inc. Census CD: Census 2000 Redistricting Data (Public Law 94-171). East Brunswick, NJ: Geolytics, Inc., 2001. Available from URL: http://www.geolytics.com [accessed April 1, 2002].
  • 38
    U.S. Census Bureau. The black population: 2000. Washington, DC: U.S. Census Bureau, August 2001. Available from URL: www.census.gov/population/www.socdemo/race.html [accessed April 1, 2002].
  • 39
    U.S. Census Bureau Internet. Metropolitan counties in alphabetical order, by state, with Metropolitan Area title, as defined by Office of Management and Budget in June 30, 1999. Washington, DC: U.S. Census Bureau, 1999. Available from URL: http://www.census.gov/population/estimates/metro-city/a99mfips.txt [accessed April 1, 2002].
  • 40
    Centers for Disease Control and Prevention. CDC Surveillance Summaries. MMWR Morb Mortal Wkly Rep CDC Surveill Summ. 1997; 46(SS–3).
  • 41
    Frazier EL, Franks AI, Sanderson LM. Behavioral risk factor data. In: Using chronic disease data: a handbook for public health practitioners. Atlanta: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention 1992: 4-14-17.
  • 42
    Waksberg J. Sampling methods for random digit dialing. J Am Stat Assoc. 1978; 73: 4046.
  • 43
    Frankel LR. The report of the CASRO task force on response rates. In: WisemanF, editor. Improving data quality in a sample survey. Cambridge, MA: Marketing Science Institute, 1983.
  • 44
    U.S. Department of Agriculture Internet. Text file containing the rural-urban continuum codes for all U.S. counties based on the June 1993 definition of metropolitan and nonmetropolitan counties as determined by the Office of Management and Budget (OMB). Washington, DC: U.S. Department of Agriculture Economic Research Service. Available from URL: http://www.ers.usda.gov/briefing/rural/data, [accessed on October 17, 2001].
  • 45
    National Center for Health Workforce Information and Analysis, Bureau of Health Professions. Area Resource File. Rockville, MD: Health Resources and Services Administration, Department of Health and Human Services, 2001.
  • 46
    Draper N, Smith H. Applied regression analysis. 2nd edition. New York: John Wiley & Sons, 1981.
  • 47
    Shah BV, Barnwell BG, Bieler GS. SUDAAN user's manual: software for analysis of correlated data, release 6.40. Research Triangle Park, NC: Research Triangle Institute, 1997.
  • 48
    Hosmer DW, Lemeshow S. Applied logistic regression. New York: John Wiley & Sons, 1989.
  • 49
    U.S. Public Health Service. Healthy People 2010. Volumes I and II. Washington, DC: U.S. Department of Health and Human Services, Public Health Service, 2001. Available from URL: http://www.health.gov/healthypeople/document/html/volume1/03cancer.htm [accessed April 1, 2002].
  • 50
    Baquet CR, Hunter CP. Patterns in minorities and special populations. In: GreenwaldP, KramerBS, WeedDL, editors. Cancer prevention and control. New York: Marcel Dekker Inc. 1995: 2336.
  • 51
    MillerBA, KolonelLN, BernsteinL, et al., editors. Racial/ethnic patterns of cancer in the United States 1988–1992. Bethesda (MD): National Cancer Institute. 1996 NIH Pub. No. 96-4104.
  • 52
    Howe HL, Wingo PA, Thun MJ, et al. Annual report to the nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst. 2001; 93: 824842.
  • 53
    Zapka JG, Bigelow C, Hurley T, et al. Mammography use among sociodemographically diverse women: the accuracy of self-report. Am J Public Health. 1996; 86: 10161021.
  • 54
    Paskett ED, Tatum CM, Mack DW, et al. Validation of self-reported breast and cervical cancer screening tests among low-income minority women. Cancer Epidemiol Biomarkers Prev. 1996; 5: 721726.
  • 55
    Gordon NP, Hiatt RA, Lampert DI, et al. Concordance of self-reported data and medical record audit for six cancer screening procedures. J Natl Cancer Inst. 1993; 85: 566570.
  • 56
    Henson RM, Wyatt SW, Lee NC. The National Breast and Cervical Cancer Early Detection Program: a comprehensive public health response to two major health issues for women. J Public Health Manag Pract. 1996; 2: 3647.