ORIGINAL ARTICLE
Predictors of Colorectal Cancer Screening: Does Rurality Play a Role? Chinedum O. Ojinnaka, MBBS, MPH; Yong Choi, MPH; Hye-Chung Kum, PhD; & Jane N. Bolin, BSN, JD, PhD Department of Health Policy and Management, Texas A&M Health Science Center School of Public Health, College Station, Texas
Abstract Funding: This project was funded in part by the Cancer Prevention and Research Institute of Texas (grant# PP110176). For further information, contact: Chinedum O. Ojinnaka, MBBS, MPH, Department of Health Policy and Management, Texas A&M Health Science Center School of Public Health, 1266 Adriance Lab Road, College Station, TX 77843; e-mail: [email protected]. doi: 10.1111/jrh.12104
Purpose: The purpose of this study was to explore the associations between sociodemographic factors such as residence, health care access, and colorectal cancer (CRC) screening among residents of Texas. Methods: Using the 2012 Behavioral Risk Factor Surveillance Survey, we performed logistic regression analyses to determine predictors of CRC screening among Texas residents, including rural versus urban differences. Our outcomes of interest were previous (1) CRC screening using any CRC test, (2) fecal occult blood test (FOBT), or (3) endoscopy, as well as up-to-date screening using (4) any CRC test, (5) FOBT, or (6) endoscopy. The independent variable of interest was rural versus urban residence; we controlled for other sociodemographic and health care access variables such as lack of health insurance. Results: Multivariate analysis showed that individuals who were residents of a rural/non-Metropolitan Statistical Area (MSA) location (OR = 0.70, 95% CI = 0.51-0.97) or a suburban county (OR = 0.61, 95% CI = 0.39-0.95) were less likely to report ever having any CRC screening compared to residents of a center city of an MSA. Residents of a rural/non-MSA location were less likely (OR = 0.49, 95% CI = 0.28-0.87) than residents of a center city of an MSA to be up-to-date using FOBT. There was decreased likelihood of ever being screened for CRC among the uninsured (OR = 0.43, 95% CI = 0.31-0.59). Conclusions: Effective development and implementation of strategies to improve screening rates should aim at improving access to health care, taking into account demographic characteristics such as rural versus urban residence. Key words access to care, colorectal cancer, health disparities, policy, rural.
Colorectal cancer (CRC) ranks second in cancer incidence and third in cancer-related mortality in the United States.1 In 2014, it is projected that 136,830 new cases and 50,310 colorectal cancer-related deaths will occur.2 CRC has a significant impact on the economy. In 2010, direct costs for CRC treatment and care were the second most expensive across all types of cancers, resulting in $14.14 billion in expenditures. It is projected that if current treatment cost trends persist by 2020, $17.41 billion will be spent on direct costs for colorectal cancer.3 Furthermore, in their 2008 study, Bradley et al estimated that the national loss in productivity as a result of CRC in 2010 was $12.8 billion.4 Thus, reduction in CRC incidence and mortality would not only positively impact individuals but the US economy as well. 254
CRC results from malignant transformation of outgrowths in the colon and rectum known as polyps; these polyps usually reside in the large intestine for about 10 years before becoming cancerous.5 Therefore, routine screening, detection, and removal of polyps can reduce CRC incidence and mortality.5 The United States Preventive Services Task Force (USPSTF) recommends routine screening for individuals between age 50 and 75 using any of the following tests: (1) annual high-sensitivity fecal occult blood test (FOBT); (2) sigmoidoscopy every 5 years, accompanied by a fecal blood test every 3 years; or (3) screening colonoscopy every 10 years.6,7 Although screening rates have steadily increased,8,9 they remain suboptimal; about 1 in 3 (23 million) ageeligible residents of the United States have never been c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
Predictors of Colorectal Cancer Screening
Ojinnaka et al.
screened for colorectal cancer.10 Previous studies have demonstrated that demographic factors such as race,11-13 ethnicity,11,14,15 age,15 educational attainment,15,16 gender,15,16 marital status,15 and income13 are associated with CRC screening rates. Inadequate access to health care such as last physician visit more than 1 year,15 lack of health insurance,15 or not having a personal doctor13 has also been associated with reduced likelihood of CRC screening. This study explored the associations between demographic characteristics, health care access, and previous CRC screening among Texas residents. We also explored the associations between demographic characteristics, health care access, and being up-to-date with CRC screening using FOBT or endoscopy (sigmoidoscopy or colonoscopy). State variations in CRC screening rates have been demonstrated,17 and the state of Texas ranks 41st (bottom 20%) in CRC screening guidelines adherence compared to the rest of the United States and the District of Columbia.13 It is projected that about 9,760 new cases of CRC and 3,430 CRC-related deaths will occur in the state of Texas in 2014.2,17 Therefore, it is imperative that innovative strategies to improve CRC screening rates among Texans be developed and implemented. This is particularly important because the CDC estimates that CRC mortality would decrease by about 60% if all age-eligible individuals adhere to screening guidelines.1 Furthermore, the unique socioeconomic, cultural, racial, and ethnic characteristics across states could also imply that different groups could be at higher risk for inadequate CRC screening rates within each state. Understanding state-specific predictors of CRC screening could assist and help guide health policy planners in developing state-level policies to help reduce or eliminate screening disparities and, potentially, disparities in CRC incidence and mortality. Although predictors of CRC screening have been previously published, to our knowledge, no study has explored important predictors of CRC screening among residents of Texas using data representative of state demographics. Thus, this study could enable health care providers, public health personnel, and policy makers to understand variations in CRC screening guidelines adherence across various groups and identify groups at risk for inadequate CRC screening, thereby allowing them to develop strategies directed at these groups.
Methods Data and Participants The 2012 Behavioral Risk Factor Surveillance Survey (BRFSS) data were used for this analysis. The BRFSS, hosted by the CDC, is an annual telephone-based survey
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
that collects state-level data on health behaviors of residents of all 50 states, the District of Columbia, American Samoa, Palau, Puerto Rico, the US Virgin Islands, and Guam.18 Analysis was restricted to respondents who were residents of the state of Texas. In line with the USPSTF CRC screening guidelines, respondents below age 50 and above age 75 were excluded. The study was exempt by the Texas A&M Institutional Review Board because the BRFSS data are deidentified public use data sets.
Measures We identified 6 outcomes of interest, namely: previous CRC screening using (1) any of the screening tests (FOBT or endoscopy), (2) FOBT, or (3) endoscopy; and up-to-date CRC screening using (4) any CRC test, (5) FOBT (among those who reported ever having FOBT), or (6) endoscopy (among those who reported ever having had endoscopy). Respondents who reported having a colonoscopy within the past 10 years, or a sigmoidoscopy and FOBT within the past 5 and 3 years, respectively, were considered to be up-to-date using endoscopy. Respondents whose last FOBT was within the previous year were considered to be up-to-date with screening using FOBT. Respondents who reported both a colonoscopy and a blood stool test within the past year were assumed to have had a diagnostic colonoscopy following an abnormal FOBT; they were therefore coded as being within screening guidelines for FOBT. Respondents who reported FOBT who had been categorized as being upto-date for endoscopy were excluded from the FOBT adherence model. All outcome variables were dichotomous variables coded as 0 = no, 1 = yes. Independent variables of interest were chosen based on previous literature.13,15,19 The primary independent variable of interest was residence. The original BRFSS classification of residence was (1) within a Metropolitan Statistical Area (MSA), (2) outside the central city of an MSA but still inside the county containing the center city, (3) inside a suburban county of the MSA, and (4) not in an MSA. These variables were renamed (1) center city of MSA, (2) close to center city of MSA, (3) suburban county, and (4) rural/non-MSA. Other covariates were race (white, black, other), ethnicity (Hispanic, non-Hispanic), educational attainment (some college/college graduate, some high school/high school graduate, less than high school), gender (female, male), age in years (50-64, 65-75), marital status (married, divorced/separated, widowed, never married, a member of an unmarried couple), income, last physician visit, health insurance or plan (yes, no), and having a personal doctor (yes, no).
255
Predictors of Colorectal Cancer Screening
Ojinnaka et al.
Table 1 Proportion of Missing Variables Variable Outcome variables Ever had CRC screening Ever had FOBT Ever had endoscopy Compliance using FOBT/endoscopy Compliance using FOBT Compliance using endoscopy Independent variables Residence Race Ethnicity Educational attainment Gender Marital status Income Age Checkup Personal doctor Health plan
% Missing
3.37 3.82 3.33 2.95 3.47 7.27 14.86 7.01 0.97 0.17 0.00 0.28 11.54 0.00 1.01 0.38 0.42
Statistical Analysis Table 1 shows the proportion of missing data for each variable; missing data ranged from 0% for gender to 11.54% for income. Descriptive statistics are presented as population-weighted percentages of the independent variables by residence. Proportion of respondents who reported previous CRC screening by our 6 outcomes of interest are also presented. Missing data for all variables (outcome and predictor variables) were addressed using multiple imputation by chained equations under a missing at random (MAR) assumption.20-22 Although the BRFSS data set contained imputed values for age and race, imputed by mean and mode substitution, respectively, we did not use the imputed variables in our imputation models or analyses to ensure uniformity. Twenty data sets were imputed. Bivariate and multivariate logistic regressions were then conducted using the multiple imputation data. Estimates were combined to obtain overall estimates and confidence intervals.20-22 Analyses were adjusted for the BRFSS 2012 complex survey design. All statistical tests were 2-sided, and findings were considered statistically significant at P < .05. All analyses were conducted using Stata 13.23
Results Descriptive Statistics Most respondents (49.3%) were residents of a center city of an MSA (Table 2). About 83% of these respon256
dents were white while 11.4% were black. With regard to rural/non-MSA residence, 90.4% were white and 5.5% were black. Approximately 21% of respondents were of Hispanic ethnicity. With regard to health care access, 24.3% of all respondents had their last checkup more than 1 year ago, 10.3% did not have a personal doctor, and 12.8% did not have a health plan. Using any of the recommended tests, 71.4% of respondents reported previous CRC screening. About 34.0% reported previous screening using FOBT while 65.2% of respondents reported previous screening using endoscopy (Table 3). Almost 92% of those who reported previous endoscopy were up-to-date compared to 28.5% of those reporting previous FOBT. Compared to those living in the center city of an MSA, rural/non-MSA residents were less likely to have ever been screened for CRC using any test (73.2% vs 68.6%), FOBT (37.6% vs 31.7%), or endoscopy (65.9% vs 62.2%); they also had a lower proportion of those who were up-to-date on screening using FOBT (32.0% vs 17.2%). Compared to whites, blacks had higher proportions of those who reported CRC screening using any of the recommended tests (68.6% vs 74.5%), FOBT (31.1% vs 39.7%), endoscopy (63% vs 66.1%), and up-to-date screening using FOBT (25.5% vs 43.3%). However, whites had a slightly greater proportion of those reporting being up-to-date using any test (87.3% vs 87.0%) or endoscopy (91.5% vs 90.7%). Compared to non-Hispanics, Hispanics had lower proportions of those reporting CRC screening using any test (72.8% vs 49.6%), FOBT (35.0% vs 16.9%), or endoscopy (66.9% vs 43.6%), and being up-to-date with screening using endoscopy (87.8% vs 91.8%). However, Hispanics had a greater proportion of those who were up-to-date using FOBT compared to non-Hispanics (41.7% vs 26.5%).
Previous CRC Screening On bivariate analysis, residence had no statistically significant association with ever being screened for CRC (Table 4). However, on multivariate analysis, individuals who were residents of a suburban county of an MSA (OR = 0.61, 95% CI = 0.39-0.95) and a rural/non-MSA (OR = 0.70, 95% CI = 0.51-0.97) were less likely to report ever having a CRC screening compared to residents of a center city of an MSA. Respondents aged 65-75 were more likely (OR = 2.28, 95% CI = 1.78-2.93) to report being screened for CRC compared to those between ages 50 and 64. Hispanics were also less likely to report ever being screened (OR = 0.73, 95% CI = 0.54-0.99) compared to non-Hispanics. Those who had never been married were less likely (OR = 0.49, 95% CI = 0.30-0.80) to report ever being screened compared to married respondents. With regard to educational attainment, those c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
Marital status
Gender
Educational attainment
Ethnicity
Total sample Race
a
Married Divorced/separated Widowed Never married A member of an unmarried couple Total
Female Male
Some college/college graduate Some high school/high school graduate Less than high school 6.1
7.8
64.2 16.9 8.1 8.3 2.4
73.6 12.4 7.0 6.1 0.9
57.8 42.2
28.9
28.9
56.7 43.3
65.1
83.0 17.0
85.0 7.9 7.1
8.8
Outside the CC of an MSA but Inside the County Containing the Center City (%)
63.3
73.6 26.4
78.8 16.1 5.1
White Black Other
Non-Hispanic Hispanic
28.0
49.3
Characteristics
In the Center City of an MSA (%)
Table 2 Descriptive Statistics of Select Characteristics of Respondents
66.9 18.3 10.0 2.6 2.2
59.5 40.6
1.9
48.1
50.0
81.7 18.4
89.6 6.4 4.0
13.9
Inside a Suburban County of the MSA (%)
69.5 13.1 11.0 5.2 1.2
54.2 45.8
6.4
47.3
46.2
85.4 14.6
90.4 5.5 4.2
100
Rural/Non-MSA (%)
67.8 15.3 8.4 6.8 1.8
56.9 43.1
6.6
33.1
60.3
78.6 21.4
83.2 11.4 5.5
Total
(Continued)
.0377
.7388
Predictors of Colorectal Cancer Screening: Does Rurality Play a Role? Chinedum O. Ojinnaka, MBBS, MPH; Yong Choi, MPH; Hye-Chung Kum, PhD; & Jane N. Bolin, BSN, JD, PhD Department of Health Policy and Management, Texas A&M Health Science Center School of Public Health, College Station, Texas
Abstract Funding: This project was funded in part by the Cancer Prevention and Research Institute of Texas (grant# PP110176). For further information, contact: Chinedum O. Ojinnaka, MBBS, MPH, Department of Health Policy and Management, Texas A&M Health Science Center School of Public Health, 1266 Adriance Lab Road, College Station, TX 77843; e-mail: [email protected]. doi: 10.1111/jrh.12104
Purpose: The purpose of this study was to explore the associations between sociodemographic factors such as residence, health care access, and colorectal cancer (CRC) screening among residents of Texas. Methods: Using the 2012 Behavioral Risk Factor Surveillance Survey, we performed logistic regression analyses to determine predictors of CRC screening among Texas residents, including rural versus urban differences. Our outcomes of interest were previous (1) CRC screening using any CRC test, (2) fecal occult blood test (FOBT), or (3) endoscopy, as well as up-to-date screening using (4) any CRC test, (5) FOBT, or (6) endoscopy. The independent variable of interest was rural versus urban residence; we controlled for other sociodemographic and health care access variables such as lack of health insurance. Results: Multivariate analysis showed that individuals who were residents of a rural/non-Metropolitan Statistical Area (MSA) location (OR = 0.70, 95% CI = 0.51-0.97) or a suburban county (OR = 0.61, 95% CI = 0.39-0.95) were less likely to report ever having any CRC screening compared to residents of a center city of an MSA. Residents of a rural/non-MSA location were less likely (OR = 0.49, 95% CI = 0.28-0.87) than residents of a center city of an MSA to be up-to-date using FOBT. There was decreased likelihood of ever being screened for CRC among the uninsured (OR = 0.43, 95% CI = 0.31-0.59). Conclusions: Effective development and implementation of strategies to improve screening rates should aim at improving access to health care, taking into account demographic characteristics such as rural versus urban residence. Key words access to care, colorectal cancer, health disparities, policy, rural.
Colorectal cancer (CRC) ranks second in cancer incidence and third in cancer-related mortality in the United States.1 In 2014, it is projected that 136,830 new cases and 50,310 colorectal cancer-related deaths will occur.2 CRC has a significant impact on the economy. In 2010, direct costs for CRC treatment and care were the second most expensive across all types of cancers, resulting in $14.14 billion in expenditures. It is projected that if current treatment cost trends persist by 2020, $17.41 billion will be spent on direct costs for colorectal cancer.3 Furthermore, in their 2008 study, Bradley et al estimated that the national loss in productivity as a result of CRC in 2010 was $12.8 billion.4 Thus, reduction in CRC incidence and mortality would not only positively impact individuals but the US economy as well. 254
CRC results from malignant transformation of outgrowths in the colon and rectum known as polyps; these polyps usually reside in the large intestine for about 10 years before becoming cancerous.5 Therefore, routine screening, detection, and removal of polyps can reduce CRC incidence and mortality.5 The United States Preventive Services Task Force (USPSTF) recommends routine screening for individuals between age 50 and 75 using any of the following tests: (1) annual high-sensitivity fecal occult blood test (FOBT); (2) sigmoidoscopy every 5 years, accompanied by a fecal blood test every 3 years; or (3) screening colonoscopy every 10 years.6,7 Although screening rates have steadily increased,8,9 they remain suboptimal; about 1 in 3 (23 million) ageeligible residents of the United States have never been c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
Predictors of Colorectal Cancer Screening
Ojinnaka et al.
screened for colorectal cancer.10 Previous studies have demonstrated that demographic factors such as race,11-13 ethnicity,11,14,15 age,15 educational attainment,15,16 gender,15,16 marital status,15 and income13 are associated with CRC screening rates. Inadequate access to health care such as last physician visit more than 1 year,15 lack of health insurance,15 or not having a personal doctor13 has also been associated with reduced likelihood of CRC screening. This study explored the associations between demographic characteristics, health care access, and previous CRC screening among Texas residents. We also explored the associations between demographic characteristics, health care access, and being up-to-date with CRC screening using FOBT or endoscopy (sigmoidoscopy or colonoscopy). State variations in CRC screening rates have been demonstrated,17 and the state of Texas ranks 41st (bottom 20%) in CRC screening guidelines adherence compared to the rest of the United States and the District of Columbia.13 It is projected that about 9,760 new cases of CRC and 3,430 CRC-related deaths will occur in the state of Texas in 2014.2,17 Therefore, it is imperative that innovative strategies to improve CRC screening rates among Texans be developed and implemented. This is particularly important because the CDC estimates that CRC mortality would decrease by about 60% if all age-eligible individuals adhere to screening guidelines.1 Furthermore, the unique socioeconomic, cultural, racial, and ethnic characteristics across states could also imply that different groups could be at higher risk for inadequate CRC screening rates within each state. Understanding state-specific predictors of CRC screening could assist and help guide health policy planners in developing state-level policies to help reduce or eliminate screening disparities and, potentially, disparities in CRC incidence and mortality. Although predictors of CRC screening have been previously published, to our knowledge, no study has explored important predictors of CRC screening among residents of Texas using data representative of state demographics. Thus, this study could enable health care providers, public health personnel, and policy makers to understand variations in CRC screening guidelines adherence across various groups and identify groups at risk for inadequate CRC screening, thereby allowing them to develop strategies directed at these groups.
Methods Data and Participants The 2012 Behavioral Risk Factor Surveillance Survey (BRFSS) data were used for this analysis. The BRFSS, hosted by the CDC, is an annual telephone-based survey
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
that collects state-level data on health behaviors of residents of all 50 states, the District of Columbia, American Samoa, Palau, Puerto Rico, the US Virgin Islands, and Guam.18 Analysis was restricted to respondents who were residents of the state of Texas. In line with the USPSTF CRC screening guidelines, respondents below age 50 and above age 75 were excluded. The study was exempt by the Texas A&M Institutional Review Board because the BRFSS data are deidentified public use data sets.
Measures We identified 6 outcomes of interest, namely: previous CRC screening using (1) any of the screening tests (FOBT or endoscopy), (2) FOBT, or (3) endoscopy; and up-to-date CRC screening using (4) any CRC test, (5) FOBT (among those who reported ever having FOBT), or (6) endoscopy (among those who reported ever having had endoscopy). Respondents who reported having a colonoscopy within the past 10 years, or a sigmoidoscopy and FOBT within the past 5 and 3 years, respectively, were considered to be up-to-date using endoscopy. Respondents whose last FOBT was within the previous year were considered to be up-to-date with screening using FOBT. Respondents who reported both a colonoscopy and a blood stool test within the past year were assumed to have had a diagnostic colonoscopy following an abnormal FOBT; they were therefore coded as being within screening guidelines for FOBT. Respondents who reported FOBT who had been categorized as being upto-date for endoscopy were excluded from the FOBT adherence model. All outcome variables were dichotomous variables coded as 0 = no, 1 = yes. Independent variables of interest were chosen based on previous literature.13,15,19 The primary independent variable of interest was residence. The original BRFSS classification of residence was (1) within a Metropolitan Statistical Area (MSA), (2) outside the central city of an MSA but still inside the county containing the center city, (3) inside a suburban county of the MSA, and (4) not in an MSA. These variables were renamed (1) center city of MSA, (2) close to center city of MSA, (3) suburban county, and (4) rural/non-MSA. Other covariates were race (white, black, other), ethnicity (Hispanic, non-Hispanic), educational attainment (some college/college graduate, some high school/high school graduate, less than high school), gender (female, male), age in years (50-64, 65-75), marital status (married, divorced/separated, widowed, never married, a member of an unmarried couple), income, last physician visit, health insurance or plan (yes, no), and having a personal doctor (yes, no).
255
Predictors of Colorectal Cancer Screening
Ojinnaka et al.
Table 1 Proportion of Missing Variables Variable Outcome variables Ever had CRC screening Ever had FOBT Ever had endoscopy Compliance using FOBT/endoscopy Compliance using FOBT Compliance using endoscopy Independent variables Residence Race Ethnicity Educational attainment Gender Marital status Income Age Checkup Personal doctor Health plan
% Missing
3.37 3.82 3.33 2.95 3.47 7.27 14.86 7.01 0.97 0.17 0.00 0.28 11.54 0.00 1.01 0.38 0.42
Statistical Analysis Table 1 shows the proportion of missing data for each variable; missing data ranged from 0% for gender to 11.54% for income. Descriptive statistics are presented as population-weighted percentages of the independent variables by residence. Proportion of respondents who reported previous CRC screening by our 6 outcomes of interest are also presented. Missing data for all variables (outcome and predictor variables) were addressed using multiple imputation by chained equations under a missing at random (MAR) assumption.20-22 Although the BRFSS data set contained imputed values for age and race, imputed by mean and mode substitution, respectively, we did not use the imputed variables in our imputation models or analyses to ensure uniformity. Twenty data sets were imputed. Bivariate and multivariate logistic regressions were then conducted using the multiple imputation data. Estimates were combined to obtain overall estimates and confidence intervals.20-22 Analyses were adjusted for the BRFSS 2012 complex survey design. All statistical tests were 2-sided, and findings were considered statistically significant at P < .05. All analyses were conducted using Stata 13.23
Results Descriptive Statistics Most respondents (49.3%) were residents of a center city of an MSA (Table 2). About 83% of these respon256
dents were white while 11.4% were black. With regard to rural/non-MSA residence, 90.4% were white and 5.5% were black. Approximately 21% of respondents were of Hispanic ethnicity. With regard to health care access, 24.3% of all respondents had their last checkup more than 1 year ago, 10.3% did not have a personal doctor, and 12.8% did not have a health plan. Using any of the recommended tests, 71.4% of respondents reported previous CRC screening. About 34.0% reported previous screening using FOBT while 65.2% of respondents reported previous screening using endoscopy (Table 3). Almost 92% of those who reported previous endoscopy were up-to-date compared to 28.5% of those reporting previous FOBT. Compared to those living in the center city of an MSA, rural/non-MSA residents were less likely to have ever been screened for CRC using any test (73.2% vs 68.6%), FOBT (37.6% vs 31.7%), or endoscopy (65.9% vs 62.2%); they also had a lower proportion of those who were up-to-date on screening using FOBT (32.0% vs 17.2%). Compared to whites, blacks had higher proportions of those who reported CRC screening using any of the recommended tests (68.6% vs 74.5%), FOBT (31.1% vs 39.7%), endoscopy (63% vs 66.1%), and up-to-date screening using FOBT (25.5% vs 43.3%). However, whites had a slightly greater proportion of those reporting being up-to-date using any test (87.3% vs 87.0%) or endoscopy (91.5% vs 90.7%). Compared to non-Hispanics, Hispanics had lower proportions of those reporting CRC screening using any test (72.8% vs 49.6%), FOBT (35.0% vs 16.9%), or endoscopy (66.9% vs 43.6%), and being up-to-date with screening using endoscopy (87.8% vs 91.8%). However, Hispanics had a greater proportion of those who were up-to-date using FOBT compared to non-Hispanics (41.7% vs 26.5%).
Previous CRC Screening On bivariate analysis, residence had no statistically significant association with ever being screened for CRC (Table 4). However, on multivariate analysis, individuals who were residents of a suburban county of an MSA (OR = 0.61, 95% CI = 0.39-0.95) and a rural/non-MSA (OR = 0.70, 95% CI = 0.51-0.97) were less likely to report ever having a CRC screening compared to residents of a center city of an MSA. Respondents aged 65-75 were more likely (OR = 2.28, 95% CI = 1.78-2.93) to report being screened for CRC compared to those between ages 50 and 64. Hispanics were also less likely to report ever being screened (OR = 0.73, 95% CI = 0.54-0.99) compared to non-Hispanics. Those who had never been married were less likely (OR = 0.49, 95% CI = 0.30-0.80) to report ever being screened compared to married respondents. With regard to educational attainment, those c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 254–268
Marital status
Gender
Educational attainment
Ethnicity
Total sample Race
a
Married Divorced/separated Widowed Never married A member of an unmarried couple Total
Female Male
Some college/college graduate Some high school/high school graduate Less than high school 6.1
7.8
64.2 16.9 8.1 8.3 2.4
73.6 12.4 7.0 6.1 0.9
57.8 42.2
28.9
28.9
56.7 43.3
65.1
83.0 17.0
85.0 7.9 7.1
8.8
Outside the CC of an MSA but Inside the County Containing the Center City (%)
63.3
73.6 26.4
78.8 16.1 5.1
White Black Other
Non-Hispanic Hispanic
28.0
49.3
Characteristics
In the Center City of an MSA (%)
Table 2 Descriptive Statistics of Select Characteristics of Respondents
66.9 18.3 10.0 2.6 2.2
59.5 40.6
1.9
48.1
50.0
81.7 18.4
89.6 6.4 4.0
13.9
Inside a Suburban County of the MSA (%)
69.5 13.1 11.0 5.2 1.2
54.2 45.8
6.4
47.3
46.2
85.4 14.6
90.4 5.5 4.2
100
Rural/Non-MSA (%)
67.8 15.3 8.4 6.8 1.8
56.9 43.1
6.6
33.1
60.3
78.6 21.4
83.2 11.4 5.5
Total
(Continued)
.0377
.7388
