Sage Journals: Discover world-class research

Abstract

Background: Rural populations are diagnosed with cancer at different rate and stages than nonrural populations, and race/ethnicity as well as the area-level income exacerbates the differences. The purpose of this analysis was to explore cancer screening rates across persistent poverty rural counties, with emphasis on nonwhite populations. Methods: The 2008 Behavioral Risk Factor Surveillance System was used, combined with data from the Area Resource File (analytic n = 309 937 unweighted, 196 344 347 weighted). Unadjusted analysis estimated screening rates for breast, cervical, and colorectal cancer. Multivariate analysis estimated the odds of screening, controlling for individual and county-level effects.Results: Rural residents, particularly those in persistent poverty counties, were less likely to be screened than urban residents. More African Americans in persistent poverty rural counties reported not having mammography screening (18.3%) compared to 15.9% of urban African Americans. Hispanics had low screening rates across all service types. Multivariate analysis continued to find disparities in screening rates, after controlling for individual and county-level factors. African Americans in persistent poverty rural counties were more likely to be screened for both breast cancer (odds ratio, 1.44; 95% confidence interval, 1.12-1.85) and cervical cancer (1.46; 1.07-1.99) when compared with urban whites. Conclusions: Disparities in cancer screening rates exist across not only race/ethnicity but also county type. These disparities cannot be fully explained by either individual or county-level effects. Programs have been successful in improving screening rates for African American women and should be expanded to target other vulnerable women as well as other services such as colorectal cancer screening.

Keywords

cancer screening rural disparities persistent poverty

The delivery of preventive services has been shown to reduce morbidity and mortality due to cancer.^1-5 Thus, recommendations, based on strong scientific evidence and potential for public health benefit, exist for many cancer screenings, aimed toward reducing premature mortality due to cancer.⁶ Unfortunately, rural populations are diagnosed with cancer at different rates than nonrural populations,^7-9 and recent downward trends in breast cancer incidence are not shared by rural residents.¹⁰ This may be because of lower screening rates evident among rural residents.^11-14 Provider availability is closely linked with a usual source of care, and subsequently, preventive service delivery^15-19 and is hampered by low health insurance coverage rates.^20-22 In addition, many rural counties lack adequate proportions of health care providers and are often classified as health professional shortage areas.²³

Individual income also plays a significant role in service delivery, particularly in rural areas.²⁴ For some rural residents, health care costs may be higher because of increased transportation needs²⁵ yet may remain lower overall compared to urban residents.^26-28 In addition, the poverty level of the area is associated with decreased screening rates, particularly in rural areas.²⁹ Poverty levels in some counties remain consistently high enough to earn a persistently poor category or designation.³⁰ Persistent poverty counties are more likely to lack basic necessities such as public services, quality food supplies, education, and health care services.³¹ We hypothesized that rural residents in persistent poverty counties were even less likely to receive recommended cancer screenings than residents in other rural or urban counties, controlling for other factors related to service delivery. Therefore, this analysis sought to investigate the impact of individual characteristics, area effects, and rurality on preventive service utilization.

Methods

Population

Data were drawn from the 2008 Behavioral Risk Factor Surveillance System (BRFSS), conducted by the Centers for Disease Control and Prevention, which has been shown to be valid and reliable in its assessments of outcomes and behaviors.^32,33 The analysis was delimited to adults with complete information for the factors of interest. Because preventive services are age and sex specific, the number of unweighted observations used varies (see Table 1).

Table 1.

Demographic and County Characteristics of the Study Population by Level of Rurality, 2008 Behavioral Risk Factor Surveillance System

	All	Urban	All Rural	Persistent Poverty	Other Rural
Observations (weighted), No.	309 \| 937 (196 \| 344 \| 347)	219 \| 670 (163 \| 260 \| 820)	90 \| 267 (33 \| 083 \| 527)	11 \| 664 (3 \| 836 \| 195)	78 \| 603 (29 \| 247 \| 332)
Overall	100.0	83.2	16.8	2.0	14.8
Race^a,b
White	68.3	65.5	81.9	58.8	85.0
African American	10.2	10.9	6.7	23.7	4.5
Hispanic	14.6	16.3	6.4	10.8	5.9
Other	7.0	7.4	4.9	6.7	4.7
Sex
Male	49.6	49.7	48.9	48.0	49.1
Female	50.4	50.3	51.1	52.0	50.9
Age group, y^a,b
18-24	13.1	13.3	12.4	14.3	12.2
25-34	19.9	20.2	18.7	20.5	18.5
35-44	21.0	21.3	19.1	19.5	19.0
45-54	20.8	20.8	20.9	19.3	21.1
55-64	15.6	15.2	17.5	16.1	17.7
65+	9.5	9.1	11.4	10.3	11.5
Health status^a,b
Excellent/very good/good	85.4	86.0	82.4	76.8	83.1
Fair/poor	14.6	14.0	17.6	23.2	16.9
Employment status^a,b
Employed	65.3	65.9	62.2	57.1	62.9
Unemployed	34.7	34.1	37.8	42.9	37.1
Insurance status^a,b
Insured	84.0	84.5	81.5	74.2	82.5
Not insured	16.0	15.5	18.5	25.8	17.5
Education^a,b
Less than high school	10.5	10.2	11.8	17.0	11.1
High school diploma	54.8	53.0	63.7	62.6	63.8
College or more	34.7	36.8	24.5	20.4	25.0
Marital status^a,b
Married	61.5	61.0	64.1	56.6	65.1
Not married	38.5	39.0	35.9	43.4	34.9
Non-English interview^a,b	7.7	8.3	4.9	2.4	5.2
% Federal poverty level_a,b
<100	15.4	15.1	16.7	26.3	15.4
100-199	15.6	14.7	19.6	19.9	19.6
200-299	16.9	16.1	20.7	17.7	21.1
300+	42.1	44.2	31.5	23.3	32.6
Missing	10.1	9.8	11.4	12.8	11.2
Usual source of care^a,b	79.7	79.6	80.7	76.0	81.3
Delayed care because of cost^a,b	14.8	14.4	16.8	23.7	15.9
Physicians per 1000 residents, quartiles^a,b
<0.48	4.9	3.2	13.3	24.1	11.9
0.76-0.49	15.0	11.5	32.4	38.1	31.6
1.11-0.77	27.2	25.9	33.7	27.6	34.5
>1.11	52.8	59.3	20.6	10.2	22.0
OB/GYN in county^a,b	94.5	97.6	78.9	64.1	80.9
Gastroenterologist in county^a,b	84.5	93.5	40.3	23.6	42.5
% High school or more, aged 25+ y, quartiles^a,b
<71.2	13.2	10.9	24.7	78.7	17.6
71.2-79.1	20.9	19.3	29.2	18.3	30.6
79.2-84.0	30.9	31.5	27.5	1.6	30.9
>84.0	35.0	38.3	18.6	1.4	20.9
% Unemployed, quartiles^a,b
<46.2	15.2	16.0	11.6	0.5	13.0
46.2-50.8	43.2	45.6	31.2	12.0	33.7
50.9-56.2	30.8	30.1	34.3	29.0	35.0
>56.2	10.8	8.3	23.0	58.5	18.3
% White-collar employment, quartiles^a,b
<3.9	5.9	1.5	27.6	47.1	25.0
3.9-5.2	9.6	4.8	33.6	25.7	34.6
5.3-6.8	17.5	15.6	26.9	19.4	27.9
>6.8	66.9	78.1	12.0	7.8	12.5

Data presented as percentages, except where indicated otherwise.

Significantly different than urban, P < .05.

Persistent poverty rural significantly different than other rural, P < .05.

The dependent variables of interest were the self-reported receipt of age- and gender-appropriate preventive services, using US Preventive Services Task Force (USPSTF) recommendations¹⁰: breast cancer screening (women aged 40-75), cervical cancer screening (women aged 18-75), and colorectal cancer screening (adults aged 50-75).

The main independent variable, county type, was derived as a combination of the US Department of Agriculture’s definition of persistent poverty (20% of a county’s population below the poverty level in 1960, 1970, 1980, and 1990)³⁴ and rurality derived from urban influence codes (urban [codes 1 and 2] or rural [codes 3-12]).³⁵ Urban influence codes were used to be consistent with other studies of health services use among rural residents.^36,37 For this analysis, we categorized counties as urban (n = 1090), persistent poverty rural (n = 340), or other rural (n = 1712).

We used several other county-level variables related to cancer screening rates: the number of primary care providers per 1000 residents,³⁸ if a county had an obstetrics/gynecology provider or a gastroenterologist,^11,38 the proportion of adults older than age 25 with at least a high school education (divided into quartiles),³⁹ and proportion working in white-collar occupations (divided into quartiles).³⁹

We relied on Anderson’s behavioral model for health services use for selection of the independent variables.⁴⁰ The predisposing characteristics included race/ethnicity (white, African American, Hispanic, other), sex, age group, and educational attainment (less than high school, high school diploma or some college, college graduate). The enabling resources included employment status, marriage status, language of the interview, income (expressed as a percentage of the federal poverty level), insurance status, a physical exam in the past 5 years, a usual source of medical care, and deferring medical care because of cost. The need factor was self-reported health status (excellent/very good/good vs fair/poor).

Bivariate unadjusted analyses examined the characteristics of the respondents by county type. We next examined the rate of receipt of each of the services by county type. All differences in the distributions across county type were assessed using Wald chi-squares. Multivariate logistic regression analysis was used to determine the factors associated with the receipt of each service. Two models were run for each service; the first controlled for the predisposing, enabling, and need factors described above, whereas the second added the county-level variables. All statistical analyses were conducted using SAS (SAS, SAS Institute, Inc. Cary, NC, USA)-callable SUDAAN (SUDAAN 10.0, 2008, RTI International, Inc., Research Triangle Park, NC, USA), to account for the weighting and complex sample design of the BRFSS.

Results

Overall, rural residents comprised 16.8% of the sample; residents of persistent poverty rural counties represented 2.0% of the sample (see Table 1). Residents in persistent poverty rural counties were more likely to be nonwhite, younger than 65 years old, in fair or poor health, unemployed, and uninsured; had less than a high school diploma; were not married; spoke English for the interview; deferred care because of cost; lacked a usual source of care; and reported lower incomes than other rural residents. Persistent poverty counties also had a lower primary care physician-to-resident ratio, were less likely to have an obstetrics/gynecology provider or gastroenterologist, and had lower percentages of having a high school diploma, unemployment, and white-collar employment than other rural counties.

The unadjusted analysis showed that service delivery varied significantly by county type (see Table 2). In general, rural residents, particularly persistent poverty rural residents, were less likely to receive services than urban residents. Nearly one-fourth of women in persistent poverty rural counties did not report recommended mammography screening (24.1%) compared to 20.4% of other rural women and 17.2% of urban women. These rates also differed by race/ethnicity; more white women reported not having a mammogram than African American women across all county types. African American women in either persistent poverty or other rural counties had lower screening rates than urban African American women but did not differ from each other.

Table 2.

Percentages of Those Not Obtaining a Service, by Race and County Type, 2008 Behavioral Risk Factor Surveillance System

	All	Urban	All Rural	Persistent Poverty, Rural	Other Rural
Breast cancer screening
All	17.8	17.2	20.8^a	24.1^a,b	20.4^a
White	17.3	16.6	20.3^a	24.4^a,b	20.0^a
African American	16.2	15.9	18.5 ^a	18.3 ^a	18.7 ^a
Hispanic	20.9	20.2	29.0 ^a	36.9 ^a	26.5 ^a
Other	21.9	21.4	25.8	24.3	26.1
Cervical cancer screening
All	14.7	14.4	16.5^a	17.9^a	16.3^a
White	14.5	13.9	16.7^a	17.0^a	16.6^a
African American	12.5	12.3	14.2	17.8^a,b	11.4
Hispanic	14.9	14.9	15.0	19.8	14.0
Other	20.6	20.9	18.3	22.4	17.6
Colorectal cancer screening
All	53.8	53.1	57.0^a	60.6^a,b	56.6^a
White	53.2	52.3	56.6^a	60.6^a,b	56.3^a
African American	50.2	49.5	55.4 ^a	57.2^a	54.3
Hispanic	61.2	60.8	65.9	69.3	64.7
Other	56.8	56.5	59.1	61.2	58.8

Bold indicates significantly different from white, within county type.

Significantly different than urban, P < .05.

Persistent poverty rural significantly different than other rural, P < .05.

More rural women did not receive cervical cancer screening (16.5%), particularly those in persistent poverty rural counties (17.9%), than urban women (14.4%). African American women in urban or other rural counties were more likely to have a screening; there were no race/ethnicity differences within persistent poverty rural counties.

More persistent poverty rural residents (60.6%) did not report a colorectal cancer screening than either other rural residents (56.6%) or urban residents (53.1%). Slightly more African Americans in urban and other rural counties reported a screening, whereas more Hispanics did not report a screening.

The multivariate analysis identified factors that were significantly associated with the receipt of each service (see Table 3). For mammography screening, a significant interaction (P < .05) existed between county type and race/ethnicity; this interaction was tested by including a first-order interactive term in the model and noting the significance of the resulting t test statistic. The first model indicated a significant role of county type that differed by race/ethnicity. White and Hispanic residents living in persistent poverty rural counties were less likely to be screened that urban whites, whereas African Americans in these counties were more likely to be screened. The second model, which included county-level variables, had similar findings, except for a lack of significance among whites. Few of the county-level factors were significant predictors of screening. Report of a physical exam within the past 5 years was the single largest predictor of screening, with an odds ratio (OR) of 5.44 (95% confidence interval [CI], 5.07-5.84; see Table 3).

Table 3.

Adjusted Predictors of Receipt of Preventive Services, by Service Type, 2008 Behavioral Risk Factor Surveillance System

	Mammography		Cervical Cancer Screening
	Model 1	Model 2	Model 1	Model 2
Intercept	0.45 (0.35, 0.58)	0.38 (0.28, 0.51)	0.57 (0.42, 0.78)	0.51 (0.35, 0.75)
County type and race/ethnicity interaction
Urban, white	Reference	Reference	Reference	Reference
Urban, African American	1.29 (1.14, 1.45)	1.25 (1.10, 1.41)	1.47 (1.28, 1.69)	1.45 (1.26, 1.66)
Urban, Hispanic	1.29 (1.09, 1.51)	1.24 (1.05, 1.45)	1.24 (1.04, 1.48)	1.23 (1.03, 1.47)
Urban, Other	0.78 (0.67, 0.91)	0.77 (0.66, 0.90)	0.56 (0.48, 0.66)	0.55 (0.47, 0.65)
Persistent poverty rural, white	0.83 (0.71, 0.96)	0.87 (0.73, 1.04)	1.04 (0.87, 1.26)	1.11 (0.90, 1.38)
Persistent poverty rural, African American	1.39 (1.10, 1.75)	1.44 (1.12, 1.85)	1.37 (1.02, 1.84)	1.46 (1.07, 1.99)
Persistent poverty rural, Hispanic	0.46 (0.22, 0.93)	0.49 (0.25, 0.96)	1.22 (0.66, 2.24)	1.33 (0.07, 2.50)
Persistent poverty rural, other	1.06 (0.67, 1.69)	1.05 (0.66, 1.69)	0.94 (0.54, 1.62)	0.97 (0.56, 1.71)
Other rural, white	0.92 (0.87, 0.98)	0.96 (0.89, 1.04)	0.95 (0.87, 1.03)	0.96 (0.87, 1.06)
Other rural, African American	1.38 (0.96, 1.99)	1.43 (0.99, 2.06)	1.97 (1.35, 2.88)	2.02 (1.37, 2.99)
Other rural, Hispanic	0.91 (0.63, 1.32)	0.95 (0.65, 1.38)	1.37 (0.96, 1.95)	1.39 (0.97, 1.99)
Other rural, other	0.79 (0.61, 1.01)	0.82 (0.64, 1.06)	1.07 (0.76, 1.51)	1.07 (0.76, 1.52)
Age group, y^a,b
18-39			Reference	Reference
40-49	Reference	Reference	1.13 (1.04, 1.24)	1.13 (1.04, 1.24)
50-64	1.71 (1.60, 1.83)	1.71 (1.60, 1.83)	0.88 (0.81, 0.95)	0.88 (0.81, 0.95)
65-75	1.70 (1.56, 1.85)	1.70 (1.56, 1.86)	0.56 (0.51, 0.62)	0.56 (0.50, 0.62)
Excellent/very good/good health	1.14 (1.05, 1.23)	1.14 (1.05, 1.23)	1.13 (1.02, 1.26)	1.13 (1.02, 1.26)
Employed	1.09 (1.02, 1.16)	1.09 (1.02, 1.17)	1.37 (1.26, 1.50)	1.37 (1.26, 1.50)
Insured	1.63 (1.48, 1.80)	1.63 (1.48, 1.81)	1.47 (1.32, 1.65)	1.48 (1.33, 1.65)
Educational attainment
Less than high school	Reference	Reference	Reference	Reference
High school diploma	1.07 (0.94, 1.20)	1.07 (0.95, 1.21)	1.09 (0.94, 1.26)	1.09 (0.94, 1.26)
College	1.34 (1.17, 1.53)	1.34 (1.18, 1.54)	2.02 (1.72, 2.39)	2.01 (1.70, 2.37)
Married	1.18 (1.11, 1.26)	1.19 (1.12, 1.27)	2.29 (2.12, 2.47)	2.30 (2.13, 2.48)
Spoke English for the interview	0.62 (0.51, 0.76)	0.62 (0.51, 0.75)	0.53 (0.43, 0.65)	0.53 (0.43, 0.65)
Income, % of poverty level
<100	0.55 (0.49, 0.61)	0.54 (0.49, 0.61)	0.93 (0.80, 1.07)	0.93 (0.80, 1.07)
100-199	0.64 (0.58, 0.70)	0.64 (0.58, 0.70)	0.81 (0.72, 0.92)	0.81 (0.72, 0.92)
200-299	0.75 (0.70, 0.81)	0.75 (0.70, 0.81)	0.89 (0.80, 0.98)	0.89 (0.80, 0.99)
300+	Reference	Reference	Reference	Reference
Missing	0.86 (0.78, 0.94)	0.86 (0.78, 0.94)	0.59 (0.51, 0.68)	0.59 (0.51, 0.68)
Usual source of care	2.07 (1.89, 2.27)	2.07 (1.88, 2.27)	1.88 (1.70, 2.08)	1.88 (1.71, 2.08)
Delayed care because of cost	0.83 (0.76, 0.91)	0.83 (0.76, 0.91)	1.02 (0.92, 1.13)	1.02 (0.92, 1.13)
Had a physical exam in past 5 years	5.43 (5.06, 5.83)	5.44 (5.07, 5.84)	4.86 (4.46, 5.28)	4.86 (4.47, 5.28)
Specialist in the county		1.22 (1.09, 1.37)		1.09 (0.93, 1.27)
Physicians per 1000
<0.48		0.96 (0.84, 1.09)		0.91 (0.76, 1.08)
0.76-0.49		0.99 (0.91, 1.08)		0.91 (0.82, 1.02)
1.11-0.77		0.93 (0.86, 0.99)		0.93 (0.84, 1.02)
>1.11		Reference		Reference
% High school or more, aged 25+, quartiles
<71.2		1.08 (0.93, 1.25)		0.97 (0.82, 1.16)
71.2-79.1		1.04 (0.95, 1.13)		1.04 (0.93, 1.17)
79.2-84.0		1.08 (1.01, 1.16)		0.96 (0.88, 1.06)
>84.0		Reference		Reference
% Unemployed, quartiles
<46.2		0.89 (0.79, 1.01)		1.09 (0.94, 1.27)
46.2-50.8		0.95 (0.86, 1.05)		1.04 (0.92, 1.17)
50.9-56.2		1.01 (0.92, 1.12)		1.12 (0.99, 1.26)
>56.2		Reference		Reference
% White-collar employment, quartiles
<3.9		0.94 (0.81, 1.09)		1.09 (0.91, 1.31)
3.9-5.2		0.96 (0.86, 1.07)		1.02 (0.88, 1.18)
5.3-6.8		1.04 (0.95, 1.13)		1.04 (0.93, 1.16)
>6.8		Reference		Reference
Hosmer-Lemeshow goodness-of-fit test	0.0295	0.1624	0.0663	0.0526
	Colorectal Cancer Screening
	Model 1	Model 2
Intercept		0.10 (0.08-0.12)
County type
Urban	Reference	Reference
Persistent poverty rural	0.82 (0.75, 0.90)	0.85 (0.76, 0.94)
Other rural	0.91 (0.88, 0.95)	0.96 (0.92, 1.01)
Race
White	Reference	Reference
African American	1.26 (1.17, 1.36)	1.26 (1.17, 1.36)
Hispanic	0.95 (0.86, 1.06)	0.94 (0.85, 1.04)
Other	0.93 (0.84, 1.03)	0.92 (0.83, 1.02)
Male	0.79 (0.69, 0.91)	0.79 (0.69, 0.90)
Age group, y^c
18-39
40-49
50-64	Reference	Reference
65-75	1.19 (1.14, 1.24)	1.18 (1.13, 1.23)
Excellent/very good/good health	0.90 (0.86, 0.95)	0.90 (0.86, 0.95)
Employed	0.86 (0.83, 0.90)	0.86 (0.82, 0.90)
Insured	1.59 (1.46, 1.73)	1.59 (1.45, 1.73)
Educational attainment
Less than high school	Reference	Reference
High school diploma	1.17 (1.08, 1.26)	1.17 (1.08, 1.26)
College	1.27 (1.17, 1.38)	1.26 (1.15, 1.37)
Married	1.14 (1.09, 1.19)	1.14 (1.10, 1.19)
Spoke English for the interview	0.94 (0.84, 1.06)	0.96 (0.86, 1.08)
Income, % of poverty level
<100	0.72 (0.66, 0.78)	0.72 (0.66, 0.78)
100-199	0.85 (0.79, 0.91)	0.85 (0.80, 0.92)
200-299	0.90 (0.86, 0.95)	0.91 (0.87, 0.96)
300+	Reference	Reference
Missing	0.85 (0.80, 0.90)	0.85 (0.81, 0.91)
Usual source of care	1.77 (1.63, 1.92)	1.77 (1.63, 1.92)
Delayed care because of cost	0.88 (0.82, 0.95)	0.88 (0.82, 0.95)
Had a physical exam in past 5 years	3.47 (3.24, 3.71)	3.46 (3.23, 3.70)
Specialist in the county		1.04 (0.99, 1.10)
Physicians per 1000
<0.48		0.96 (0.88, 1.04)
0.76-0.49		0.99 (0.93, 1.04)
1.11-0.77		0.97 (0.92, 1.01)
>1.11		Reference
% High school or more, aged 25+, quartiles
<71.2		1.09 (0.99, 1.19)
71.2-79.1		0.99 (0.94, 1.06)
79.2-84.0		0.98 (0.94, 1.03)
>84.0		Reference
% Unemployed, quartiles
<46.2		1.10 (1.02, 1.19)
46.2-50.8		1.06 (0.99, 1.12)
50.9-56.2		1.00 (0.94, 1.06)
>56.2		Reference
% White-collar employment, quartiles
<3.9		0.97 (0.88, 1.06)
3.9-5.2		0.95 (0.89, 1.03)
5.3-6.8		1.00 (0.95, 1.06)
>6.8		Reference
Hosmer-Lemeshow goodness-of-fit test	<.0001	<.0001

The age groups for mammography screening were 40 to 50 (referent level), 50 to 65, and 65 to 75

The age groups for cervical cancer screening were 18 to 40 (referent level), 40 to 50, and 50 to 65 to 75

The age groups for colorectal cancer screening were 50 to 64 (referent level) and 65 to 75.

Cervical cancer screening results were similar to breast cancer screening. The first model indicated African American women were more likely to be screened than urban whites but did not differ across county type. The second model found little change in the estimated odds; African American women in either persistent poverty rural or urban counties had lower odds of screening than other rural African Americans but were not different from each other. The added county variables also did not contribute significantly to the odds of screening. Similar to breast cancer screening, a physical exam within the past 5 years was the single largest predictor of screening (OR, 4.86; 95% CI, 4.47-5.28).

For colorectal cancer screening, no evidence of a significant interaction between county type and race/ethnicity was found. The first model indicated a significant effect by county type; residents in persistent poverty and other rural counties were less likely to be screened than urban residents. The second model, which included additional county-level variables, also showed decreased odds for screening for persistent poverty rural residents, whereas those in other rural counties were not significantly different from urban residents. These county-level variables were not significantly associated with the odds of being screened. Report of a physical exam within the past 5 years was the single largest predictor of screening, with an odds ratio of 3.46 (95% CI, 3.23-3.70).

Discussion

The purpose of this analysis was to examine the hypothesis that residents of persistent poverty rural counties were less likely to obtain recommended cancer screenings. We found that residing in a persistent poverty rural county was a contributor to service use, even after controlling for factors related to service delivery.

African American women in persistent poverty rural counties were more likely to report mammography screenings than those in other counties, whereas the reverse trend was found among Hispanic women. For cervical cancer screening, African American women in persistent poverty rural counties were less likely to be screened than those in other rural counties. Among Hispanic women, no differences in cervical cancer screening were found in rural counties. These differences highlight the significant interaction between race/ethnicity and county type and the need to be aware of such interactions when developing targeted and tailored interventions to improve cancer-preventive service utilization and behaviors.

Also, it is important to note that the unadjusted rates of mammography and cervical cancer screenings for African American women in persistent poverty rural counties were lower than urban white women, yet the adjusted rates were higher. This may be explained by lower rates of usual source of care and health insurance and the higher rates of delayed care found among persistent poverty rural residents (see Table 1). Evidence also indicates that African Americans were less likely to be insured and have a usual source of care and thus tended to report higher rates of delayed care compared with whites.⁴¹ Therefore, adjusting for these 3 factors in the multivariate analyses could have led to the improved likelihood of being screened among African Americans in persistent poverty rural counties.

These results are interesting, given the apparent success of mammography screening across county types among African American women. These results contradict findings from the 2007 National Healthcare Disparities Report, which found that African American women were less likely to receive a mammogram.⁴¹ Our study does confirm other study findings, some of which used the BRFSS and others that used primary data collection methods.^42,43 It is unclear why the results from these studies may differ, but further research would be warranted to identify more accurately the service delivery rates experienced by nonwhite women. Given the number of programs and services specifically targeting underserved and minority women, especially those with lower incomes, these results are, however, within the realm of credulity.^44-50

For colorectal cancer screenings, those in persistent poverty rural counties and nonwhites except for African Americans were at higher risk for not receiving services. Programs such as the Centers for Disease Control and Prevention’s Screen for Life program are raising awareness of the need for screenings, yet much still needs to be done to improve access to such services and to aide low-income or uninsured populations’ ability to pay for such services. Given the success of mammography and cervical cancer screening, it would be useful to use these programs and interventions as models for other screenings.^49,51 Although more attention is being paid to colorectal cancer screenings, particularly among nonwhites, more needs to be done to ensure it reaches low-income populations and areas.

This analysis, although finding significant differences across county types and race/ethnicity, was not able to take into account all known factors to affect health care service delivery. Patient perceptions and preferences for cancer screening were not captured. Studies indicate race/ethnicity, gender, age, and even geographic location all affect health care–seeking attitudes.^52-54 Intangible factors, such as racism or discrimination, may also play a role in service delivery, particular for minority and low-income populations.⁵⁵ Knowledge regarding these factors would be helpful in designing successful interventions aimed toward increased service delivery.

This study does have several limitations. First the cross-sectional design of the study limits our ability to examine causal relationships among the factors of interest. Also, the data are derived from a landline-based phone survey, and minority, young, and low-income populations are disproportionately represented in nonlandline telephone households.⁵⁶ Also, self-reported information regarding health care utilization has the potential for recall bias, with respondents reporting service delivery unable to be verified because of a lack of claims or clinical data.⁵⁷

In conclusion, this analysis found a significant interaction between county type and race/ethnicity. The fact that African American women were more likely to be screened for breast and cervical cancer than white women, even in persistent poverty counties, indicates successful implementation of targeted screening programs. However, the lower odds among whites and Hispanics indicate such programs should be expanded to improve cancer-preventive care delivery among all women. The low screening rates for colorectal cancer screening require further attention, across both race/ethnicity and type of county.

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Maciosek

Coffield

Edwards

Flottemesch

Goodman

Solberg

. Priorities among effective clinical preventive services: results of a systematic review and analysis. Am J Prev Med. 2006;31(1):52-61.

Maciosek

Solberg

Coffield

Edwards

Goodman

. Colorectal cancer screening: health impact and cost effectiveness. Am J Prev Med. 2006;31(1):80-89.

Mandel

Bond

Church

. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. N Engl J Med. 1993;328(19):1365-1371.

Mandelblatt

Saha

Teutsch

. The cost-effectiveness of screening mammography beyond age 65 years: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med. 2003;139(10):835-842.

Mandelblatt

Cronin

Bailey

. Effects of mammography screening under different screening schedules: model estimates of potential benefits and harms. Ann Intern Med. 2009;151(10):738-747.

US Preventive Services Task Force. Guide to Clinical Preventive Services, 2009. http://www.ahrq.gov/clinic/USpstfix.htm . Accessed March 15, 2010.

Sariego

. Patterns of breast cancer presentation in the United States: does geography matter? Am Surg. 2009;75(7):545-549.

Wingo

Tucker

Jamison

. Cancer in Appalachia, 2001-2003. Cancer. 2008;112(1):181-192.

Benard

Coughlin

Thompson

Richardson

. Cervical cancer incidence in the United States by area of residence, 1998 2001. Obstet Gynecol. 2007;110(3):681-686.

10.

Hausauer

Keegan

Chang

Glaser

Howe

Clarke

. Recent trends in breast cancer incidence in US white women by county-level urban/rural and poverty status. BMC Med. 2009;7:31.

11.

Bennett

Bellinger

Probst

. Receipt of cancer screening services among rural minorities: the dually disadvantaged. J Rural Health. 2010;58(10):1896-1902.

12.

Hueston

Hubbard

. Preventive services for rural and urban African American adults. Arch Fam Med. 2000;9(3):263-266.

13.

Larson

Fleishman

. Rural-urban differences in usual source of care and ambulatory service use analyses of national data using Urban Influence Codes. Med Care. 2003;41(7)(suppl):III65-III74.

14.

Zhang

Tao

Irwin

. Utilization of preventive medical services in the United States: a comparison between rural and urban populations. J Rural Health. 2000;16(4):349-356.

15.

Gornick

Eggers

Riley

. Associations of race, education, and patterns of preventive service use with stage of cancer at time of diagnosis. Health Serv Res. 2004;39(5):1403-1427.

16.

Kang-Kim

Betancourt

Ayanian

Zaslavsky

Yucel

Weissman

. Access to care and use of preventive services by Hispanics: state-based variations from 1991 to 2004. Med Care. 2008;46(5):507-515.

17.

Sambamoorthi

McAlpine

. Racial, ethnic, socioeconomic, and access disparities in the use of preventive services among women. Prev Med. 2003;37(5):475-484.

18.

Corbie-Smith

Flagg

Doyle

O’Brien

. Influence of usual source of care on differences by race/ethnicity in receipt of preventive services. J Gen Intern Med. 2002;17(6):458-464.

19.

Cornelius

Smith

Simpson

. What factors hinder women of color from obtaining preventive health care? Am J Public Health. 2002;92(4):535-539.

20.

Ayanian

Weissman

Schneider

Ginsburg

Zaslavsky

. Unmet health needs of uninsured adults in the United States. JAMA. 2000;284(16):2061-2069.

21.

Davidson

Bastani

Nakazono

Carreon

. Role of community risk factors and resources on breast carcinoma stage at diagnosis. Cancer. 2005;103(5):922-930.

22.

Rodríguez

Ward

Pérez-Stable

. Breast and cervical cancer screening: impact of health insurance status, ethnicity, and nativity of Latinas. Ann Fam Med. 2005;3(3):235-241.

23.

Health Resources and Services Administration. Shortage Designation: Health Professional Shortage Areas (HPSAs). http://bhpr.hrsa.gov/shortage/hpsadesignation.htm . Accessed March 15, 2010.

24.

Hueston

Hubbard

. Preventive services for rural and urban African American adults. Arch Fam Med. 2000; 9(3):263-266.

25.

Probst

Laditka

Wang

Johnson

. Effects of residence and race on burden of travel for care: cross sectional analysis of the 2001 US National Household Travel Survey. BMC Health Serv Res. 2007;7:40.

26.

Jolliffe

. The Cost of Living and the Geographic Distribution of Poverty. Washington, DC: US Department of Agriculture, Economic Research Service; 2006. http://www.ers.usda.gov/publications/err26/err26.pdf . Accessed March 15, 2010.

27.

Kurre

. Is The cost of living less in rural areas? Int Reg Sci Rev. 2003;26(1):86-116.

28.

Nord

. Does it cost less to live in rural areas? evidence from new data on food security and hunger. Rural Sociol. 2000;65(1):104-125.

29.

Schootman

Jeffe

Baker

Walker

. Effect of area poverty rate on cancer screening across US communities. J Epidemiol Community Health. 2006;60(3):202-207.

30.

Miller

Weber

. Persistent Poverty Across the Rural-Urban Continuum. Working Paper 18910. Oregon State University, Rural Poverty Research Center (RPRC). 2003. http://ideas.repec.org/p/ags/osruwp/18910.html . Accessed March 15, 2010.

31.

US Department of Agriculture, Economic Research Service. Understanding Rural America. 2003. http://www.ers.usda.gov/publications/aib710/aib710l.htm . Accessed March 15, 2010.

32.

Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System 2008 Summary Data Quality Report. 2009. http://www.cdc.gov/brfss/technical_infodata/quality.htm . Accessed March 9, 2011.

33.

Nelson

Holtzman

Bolen

Stanwyck

Mack

. Reliability and validity of measures from the Behavioral Risk Factor Surveillance System (BRFSS). Soz Praventivmed. 2001;46(suppl 1):S03-S42.

34.

US Department of Agriculture, Economic Research Service. Understanding Rural America: County Types. February 11, 1997. http://www.ers.usda.gov/publications/aib710/aib710l.htm . Accessed May 15, 2010.

35.

US Department of Agriculture, Economic Research Service. Measuring Rurality: Urban Influence Codes. August 8, 2007. http://www.ers.usda.gov/briefing/rurality/urbaninf/ . Accessed March 2010.

36.

Bennett

Bellinger

Probst

. Receipt of influenza and pneumonia vaccinations: the dual disparity of rural minorities. J Am Geriatr Soc. 2010;58(10):1896-1902.

37.

Bennett

Olatosi

Probst

. Health Disparities: A Rural-Urban Chartbook. Technical report provided to Health Resources and Services Administration/Office of Rural Health Policy. June 2008. http://rhr.sph.sc.edu . Accessed March 15, 2010.

38.

Mueller

Ortega

Parker

Patil

Askenazi

. Health status and access to care among rural minorities. J Health Care Poor Underserved. 1999;10(2):230-249.

39.

Probst

Moore

Glover

Samuels

. Person and place: the compounding effect of race/ethnicity and rurality on health. Am J Public Health. 2004;94:1695-1703.

40.

Anderson

. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36(1):1-10.

41.

Agency for Healthcare Research and Quality. 2007 National Healthcare Disparities Report. Rockville, MD: US Department of Health and Human Services, Agency for Healthcare Research and Quality; February 2008. AHRQ Pub. No. 08-0041.

42.

Coughlin

Leadbetter

Richards

Sabatino

. Contextual analysis of breast and cervical cancer screening and factors associated with health care access among United States women, 2002. Soc Sci Med. 2008;66(2):260-275.

43.

Cui

Peterson

Hargreaves

. Mammography use in the Southern Community Study (United States). J Health Care Poor Underserved. 2007;18(4)(suppl):102-117.

44.

Earp

Altpeter

Mayne

Viadro

O’Malley

. The North Carolina Breast Cancer Screening Program: foundations and design of a model for reaching older, minority, rural women. Breast Cancer Res Treat. 1995;35(1):7-22.

45.

Hiatt

Pasick

Stewart

. Community-based cancer screening for underserved women: design and baseline findings from the Breast and Cervical Cancer Intervention Study. Prev Med. 2001;33(3):190-203.

46.

Mandelblatt

Traxler

Lakin

Kanetsky

Kao

. Targeting breast and cervical cancer screening to elderly poor black women: who will participate? Prev Med. 1993;22(1):20-33.

47.

Margolis

Lurie

McGovern

Tyrrell

Slater

. Increasing breast and cervical cancer screening in low-income women. J Gen Intern Med. 1998;13(8):515-521.

48.

Paskett

Tatum

D’Agostino

. Community-based interventions to improve breast and cervical cancer screening: results of the Forsyth County Cancer Screening (FoCaS) project. Cancer Epidemiol Biomarkers Prev. 1999;8(5):453-459.

49.

South Carolina Department of Health and Environmental Control. Best Chance Network. http://www.scdhec.gov/health/chcdp/cancer/bcn.htm . Accessed July 15, 2010.

50.

Centers for Disease Control and Prevention. National Breast and Cervical Cancer Early Detection Program (NBCCEDP). http://www.cdc.gov/cancer/nbccedp/index.htm . Accessed July 15, 2010.

51.

California Department of Public Health. Cancer Detection Programs: Every Woman Counts. http://www.cdph.ca.gov/programs/cancerdetection/pages/cancerdetectionprogramseverywomancounts.aspx . Accessed July 15, 2010.

52.

Conway-Phillips

Millon-Underwood

. Breast cancer screening behaviors of African American women: a comprehensive review, analysis, and critique of nursing research. ABNF J. 2009;20(4):97-101.

53.

Cronan

Villalta

Gottfried

Vaden

Ribas

Conway

. Predictors of mammography screening among ethnically diverse low-income women. J Womens Health (Larchmt). 2008;17(4):527-537.

54.

Hall

Khoury

Lopez

Lisovicz

Avis-Williams

Mitra

. Breast cancer fatalism: the role of women’s perceptions of the health care system. J Health Care Poor Underserved. 2008;19(4):1321-1335.

55.

Betancourt

Green

Carrillo

Ananeh-Firempong

II . Defining cultural competence: a practical framework for addressing racial/ethnic disparities in health and health care. Public Health Rep. 2003;118(4):293-302.

56.

Blumberg

Luke

Cynamon

. Telephone coverage and health survey estimates: evaluating the need for concern about wireless substitution. Am J Public Health. 2006;96(5):926-931.

57.

Fiscella

Holt

Meldrum

Franks

. Disparities in preventive procedures: comparisons of self-report and Medicare claims data. BMC Health Serv Res. 2006;6:122.

Cancer Screening Delivery in Persistent Poverty Rural Counties

Abstract

Keywords

Methods

Population

Results

Discussion

Footnotes

References