Multilevel Predictors of Colorectal Cancer Screening Use in California

Background:

Screening can detect colorectal cancer (CRC) early, yet its uptake needs to be improved. Social determinants of health (SDOH) may be linked to CRC screening use but are not well understood.

Objectives:

To examine geographic variation in CRC screening and the extent to which multilevel SDOH explain its use in California, the most populous and racially/ethnically diverse state in the United States.

Study Design:

Analysis of individual and neighborhood data on 20,626 adult respondents aged >50 years from the 2005 California Health Interview Survey.

Methods:

We used multilevel logistic regression models to estimate the effects of individual characteristics and area-level segregation, socioeconomic status (SES), and healthcare resources at 2 different geographic levels on CRC screening use.

Results:

We confirmed that individual-level factors (eg, race/ethnicity, income, insurance) were strong predictors and found that area-level healthcare resources were associated with CRC screening. Primary care shortage in the Medical Service Study Area was associated with CRC screening for any modality (odds ratio [OR] = 0.89; 95% confidence interval [CI], 0.80-1.00). County-level HMO penetration (OR = 1.85; 95% CI, 1.47-2.33) and primary care shortage (OR = 0.73; 95% CI, 0.53-0.99) were associated with CRC screening with flexible sigmoidoscopy.

Conclusions:

Contextual factors including locality, primary care resources, and HMO membership are important determinants of CRC screening uptake; SES and segregation did not explain variation in screening behavior. More studies of contextual factors and varying geographic scales are needed to further elucidate their impact on CRC screening uptake.

Am J Manag Care. 2013;19(3):205-216We studied contextual factors and found that locality, availability of primary care, and HMO membership influenced use of colorectal cancer (CRC) screening in California.

• Medical Service Study Areas are useful geographic units for understanding CRC screening use in California.

• There was less use of screening by flexible sigmoidoscopy in counties with a shortage of primary care resources and low HMO penetration.

• Multilevel analyses offer a more complete picture of CRC screening use and the opportunity to prioritize interventions for appropriate subgroups and geographic areas.

Colorectal cancer (CRC), the second-leading cause of cancer death in the United States, resulted in an estimated 51,370 deaths in the United States in 2010.¹ Screening reduces CRC burden by identifying early-stage disease and/or removing precancerous polyps.^2-4 Although CRC mortality rates could be reduced by as much as 50% with complete adherence to screening guidelines, only one-half of United States adults 50 years and older report being up-todate with CRC screening.^2,5 Evidence-based guidelines for screening individuals at average risk for CRC recommend annual fecal occult blood testing (FOBT), flexible sigmoidoscopy (FSIG) every 5 years, or colonoscopy every 10 years.⁴

Use of CRC screening has been shown to vary by individual race/ethnicity, sex, education, income, marital status, insurance coverage, and immigration status.^2,6-16 Other individual characteristics associated with CRC screening uptake include perceived risk, intention for future screening, information-seeking patterns, perceived medical discrimination, life stressors, social support, and healthcare context.^2,6,14,16-21 Of these, having higher education, having higher income, being married, having health insurance and a usual source of care, seeing a physician, receiving a recommendation from a physician to obtain screening, not reporting medical discrimination experiences, and perceiving oneself to be at risk for CRC are most consistently associated with increased CRC screening use.

A few prior studies have assessed how contextual factors influence access to CRC screening.^18,22-25 Prior multilevel analyses^26-28 used poverty rates or rural/urban status as proxy measures of fundamental factors and environmental resources available to communities and the individuals within those communities,^29,30 but findings are not consistent. One recent study examined deprivation and general practitioner density per 100,000 residents, and showed that only the deprivation index was associated with CRC screening use.³¹

To more closely examine which contextual factors might influence CRC screening use, we drew on the World Health Organization social determinants of health framework. This framework underscores the importance of the contexts in which individuals live, work, socialize, and access healthcare for health behaviors and outcomes.³² The World Health Organization framework matched our broad conceptualization of social determinants of health because, unlike most other social determinants of health frameworks, it includes healthcare access. Social determinants of health can shape the opportunities and resources that affect health-related behaviors of individuals, including participation in cancer screening.^9,32,33 Healthcare resources can be conceptualized both at the area level as the local supply of healthcare services and also at the individual level as a measure of insurance or a usual source of care. We used a multilevel framework to study the influences of socioeconomic status (SES), racism, and healthcare resources on CRC screening uptake at individual and area levels with population-based data from the California Health Interview Survey (CHIS). We selected data from California, the largest and most racially/ethnically diverse state in the United States with a large HMO market share that varies by geography. In 1994, California’s dominant HMO, Kaiser Permanente, initiated organized CRC screening using FSIG and FOBT and in 2004, using fecal immunochemical tests.^34,35 Other HMOs also implemented interventions to increase use of CRC screening during this time.^36,37

METHODS

We studied CRC screening use among California residents nested within Medical Service Study Areas (MSSAs) and counties, 2 geographic levels important for healthcare resource delivery in California. Medical Service Study Areas are subcity and subcounty administrative areas identifying “rational service areas” for a variety of healthcare resources and medically underserved designations.³⁸ California designates health professional shortage areas at both MSSA and county levels. The California Department of Health and Human Services organizes its health services delivery (eg, Medi-Cal, Child Health and Disability Prevention Program, cancer screenings) at the county level.

Study Sample and Individual-Level Data

Individual-level data were obtained from the 2005 CHIS adult component, a random-digit dial telephone survey of 43,020 adults conducted in English, Spanish, Mandarin, Cantonese, Vietnamese, and Korean.³⁹ The overall response rate was 29.5% for households and 26.9% for adults. The CHIS response rates are comparable to other telephone-administered population surveys, including the California Behavioral Risk Factor Surveillance System.³⁹ More detailed information on CHIS is available at www.chis.ucla.edu. Our study sample included 20,626 adults. We excluded respondents younger than 50 years (n = 21,014), older than 84 years (n = 1016), with a prior CRC diagnosis (n = 329), or with a proxy interview (n = 141). Sample size for FSIG analyses varied due to skip patterns and resulting missing responses. The MSSA and county data were linked to CHIS respondents using census tract and county variables.

Colorectal Cancer Screening. To evaluate whether individuals were up-to-date with CRC screening, we examined 2 outcomes: (1) up-to-date screening by any modality (ie, FOBT in the past year, FSIG in the past 5 years, or colonoscopy in the past 10 years); and (2) up-to-date screening by FSIG (FSIG within the past 5 years vs not up-to-date by any modality).

To measure up-to-date screening we used the following questions from CHIS:

• Have you ever had a sigmoidoscopy or colonoscopy? These are exams in which a healthcare professional inserts a tube into the rectum to look for signs of cancer or other problems. [If yes:] How long ago did you have your most recent exam? Was your most recent exam a sigmoidoscopy, a colonoscopy, or something else?
• Have you ever done a blood stool test, using a home test kit? [If yes:] How long ago did you do your most recent home blood stool test?

Individual-Level Characteristics. Individual characteristics include race/ethnicity (non-Hispanic White, Latino, African American, American Indian/Alaska Native, Asian American and Pacific Islander, multiracial); age (50-64 years, 65-84 years); sex (male, female); marital status (married/living with a partner, widowed/divorced/separated/never married); educational attainment (less than a bachelor’s degree, at least a bachelor’s degree); income (less than 300% of the federal poverty level, at least 300% of the federal poverty level); insurance coverage (yes, no), place of birth (born in the United States, not born in the United States); and language spoken at home (English only, other language alone or in combination with English). Stressors measured include neighborhood safety, food security, and racial/ethnic discrimination. Perceived neighborhood safety compared safe all/most of the time with safe some/none of the time (responses were missing for nonowners/nonrenters). Food security, which assessed food availability at the household level, compared secure with insecure. Self-reported experiences of any racial/ethnic discrimination (both general and in healthcare) compared yes with no. By testing for the effects of stress-related variables (eg, racial/ethnic discrimination, food insecurity, and acculturation), which have not been extensively examined in previous studies of cancer screening, this study may inform the literature on individual-level stress-related predictors of CRC screening use.^10,11,17

Medical Service Study Area and County-Level Data

Medical Service Study Area data were obtained from the California Office of Statewide Health Planning and Development, the Spatial Impact Factors data set, and the 2005 CHIS.^38,40,41 County data were obtained from the Spatial Impact Factors data set and the 2004 Area Resource File.^40,42 We examined 4 contextual factors at the county and MSSA levels, capturing fundamental causes (SES and racial/ethnic residential segregation) and healthcare resources (primary care shortage and HMO penetration).

Areas were categorized as having a primary care shortage by using the criteria developed by the Health Resources and Services Administration.⁴² HMO penetration was categorized as whether the proportion of the county population enrolled in HMO plans was above or below the median level for the state.⁴³ At the MSSA level, we used 2005 CHIS data on enrollment in an HMO to measure HMO penetration. Based on 2 items that ask about Medicare coverage provided through an HMO and (for non-Medicare users) whether the main health plan is an HMO, we estimated the percentage of MSSA residents who were enrolled in an HMO.

Areas were categorized as having fewer than 20% or at least 20% of their residents living in poverty. Segregation was measured using a diversity index. The diversity index estimated the probability of an individual from one group interacting with an individual from another group within a defined area, with scores ranging from 0 (complete segregation) to 1 (complete integration).⁴⁴ We recoded the diversity index to indicate areas above and below the median values in California. We chose the diversity index as the measure of segregation because it can be computed for more than 2 racial/ethnic groups; thus, it does not require that the analyses be stratified by racial/ethnic group. Metropolitan Statistical Area/Primary Metropolitan Statistical Area, counties, and Census Place are common levels of geography at which residential segregation is measured. For this study, we constructed a segregation measure for subcounty MSSAs in addition to counties because we were interested in evaluating whether MSSAs are a meaningful geographic level for assessing the effects of contextual factors on health behaviors such as CRC screening.

Analysis

We used descriptive statistics to characterize our study sample, ArcGIS (http://www.esri.com/software/arcgis/) to map the geographic variation in MSSA and county-level CRC screening use, and logistic regression models to estimate the association between CRC screening use and respondent characteristics. Individual-level descriptive analyses were weighted and adjusted for the design effect of CHIS 2005 using SAS-callable SUDAAN with jackknife replicate weights.⁴⁵ To reduce the likelihood of false-positive (type 1) error due to multiple comparisons, the standard critical P value of .05 was adjusted using Bonferroni’s correction. We applied the adjusted critical P value of .00125 (.05/[2 outcomes X 20 characteristics]).

To evaluate the effects of individual characteristics as well as the effects of area-level segregation, SES, and healthcare resources at the county and MSSA levels on individuallevel CRC screening use (by any modality and by FSIG), we constructed a series of multilevel logistic models using the gllamm function in Stata version 9 (StataCorp LP, College Station, Texas). As MSSAs are groups of census tracts and do not span county boundaries, we built models of individuals nested within MSSAs, which were nested within counties. We first developed an unconditional or “empty” model, which partitioned the variation in CRC screening use by each level (ie, individual, MSSA, and county), to assess the influence of contextual factors at different geographic scales on CRC screening use. We then added individual-level variables to assess how much of the variation could be explained by individual-level factors. Finally, we added MSSA and county characteristics to assess the effects of specific MSSA and county factors while controlling for the individual-level determinants of CRC screening use.

All variables were binary, consistent with the recommendation of Rabe-Hesketh and Skrondal.⁴⁶ Because the CHIS sampling design oversamples in some counties, individual CHIS sample weights were first normalized and then scaled for the MSSA and county levels using the probabilityweighted iteratively generalized least squares weight-scaling method.^47,48 The resulting scaled weights were then included in the multilevel model.

We present the area-level random effects (the between-area variation) using the measures previously suggested in the literature.^49-52 The interclass correlation provided the between-area variance. The median odds ratio (MOR) adjusted the between-area variance to use the same scale as the individual-level odds ratio estimates. The MOR can be interpreted as the increased risk associated with an individual moving to an area with a higher risk. If the MOR was close to 1.0, there was little variation between areas; larger MOR values (2.0 or higher) indicated greater between-area variation. We report the proportional change in variance (PCV) for models 2 and 3 to show how much of the residual arealevel variance from the previous models was explained by the additional explanatory variables. The interclass correlation, MOR, PCV, and interval odds ratios were derived from model output following methods described by Larsen and colleagues (2000 and 2005).^50,51 For ease of interpretation, we focus on MORs and PCVs in the Results section, but include the full set of output in the table of model results. For the area-level fixed effects (the within-area contextual effects), we report the interval odds ratio because it offers information on the magnitude of the fixed estimates in relation to its importance in explaining the residual area variance.

RESULTS

Characteristics of the Study Sample and Unadjusted Odds of CRC Screening

Table 1

presents the study sample’s characteristics. Column 3 shows that 60% of adults aged 50 to 84 years reported being up-to-date with any CRC screening modality. Among our sample, 38% were current for CRC screening by colonoscopy, 13% by FSIG, and 10% by FOBT. The majority of respondents were non-Hispanic white, between 50 and 64 years of age, married/living with a partner, insured, born in the United States, only spoke English at home, felt safe in their neighborhood all or most of the time, and reported no experiences of discrimination. They also had less than a bachelor’s degree, income at least 300% of the federal poverty level, and food security. Most lived in MSSAs and counties with greaterthan-the-median segregation levels, high HMO penetration, and without primary care shortages, and where less than 20% of the residents were living in poverty.

Columns 4 and 5 of Table 1 show unadjusted odds for being up-to-date with CRC screening by any modality and by FSIG. Respondents were at increased odds of being up-to-date with CRC screening by any modality and by FSIG if they were older, non-Hispanic white/African American, married, more educated, of higher income, insured, born in the United States, only spoke English at home, lived in a safe neighborhood, and had food security. For any modality, those who were male or reported no experiences of discrimination were at increased odds of being up-to-date. Regardless of modality, at both geographic levels respondents living in areas with higher poverty rates or primary care shortages had decreased odds of being up-to-date with CRC screening. For FSIG, less HMO penetration was associated with lower odds of being up-to-date. Higher MSSA segregation was associated with lower odds of being up-to-date by any modality.

Figure

The shows the geographic variation in CRC screening use by modality and geographic levels. Median unadjusted rates of CRC screening use were similar for each geographic level: any modality was 61% for MSSA and 62% for county; FSIG was 20% for both levels.

Multilevel Assessment of CRC Screening

Table 2

shows multilevel logistic regression results for CRC screening use by any modality and by FSIG. Model 1 provides estimates of the between-area variation without any explanatory variables. In model 2, individual-level variables were added to determine their effects and to assess how much additional variation they explain. In model 3, MSSA and county variables were added to estimate their effects while controlling for individual-level characteristics and to assess whether they explain any additional variation.

By Any Modality. In model 1, the MORs are 1.43 and 1.18 for MSSAs and counties, respectively. These MORs indicate that the between-area variation was modest at the MSSA and small at the county level.⁵⁰ Addition of individual characteristics in model 2 compared with model 1 further reduced the between-area variation for both MSSAs and counties as shown by the PCV: 21% for MSSA and 67% for county. Addition of area-level factors in model 3 did not change the between-area variation for either geographic level (PCV is 0% for both MSSA and county).

Results for individual-level characteristics in model 2 indicate that Latinos and Asian American and Pacific Islanders had statistically significantly lower odds of being up-to-date with CRC screening by any modality compared with whites. Those who were older, married/living with a partner, more educated, had higher income, and were insured had statistically significantly higher odds of being up-to-date.

In model 3, a statistically significant decrease in CRC screening odds was observed for MSSAs with a primary care shortage. The CRC screening uptake associated with individual characteristics observed in model 2 also held in model 3.

By Flexible Sigmoidoscopy. In model 1, the MORs are 1.55 and 1.37 for MSSAs and counties, respectively. These MORs indicate that the between-area variation is modest at the MSSA and county level.⁵⁰ Addition of individual characteristics in model 2 compared with model 1 further reduced the between-area variation at the MSSA level (PCV = 29%) and resulted in more between-area variance at the county level (PVC = —45%). Addition of area-level factors in model 3 compared with model 2 did not explain additional between-area variance at the MSSA level (PCV = 0%), but did explain additional between-area variance at the county level (PVC = 31%).

Results for individual-level characteristics in model 2 indicate that African Americans were at significantly increased odds of being up-to-date with screening by FSIG compared with non-Hispanic whites, and American Indians/Alaska Natives and Asian American and Pacific Islanders were at significantly decreased odds of being up-to-date by FSIG. Those who were older, were married/living with a partner, had more education, had higher income, were insured, and did not face food insecurity were at significantly increased odds of being up-to-date with screening by FSIG.

In model 3, a statistically significant increase in CRC screening odds was observed for counties with a higher HMO penetration and a significant decrease in screening odds was observed for counties with a primary care shortage. Most of the CRC screening uptake associations with individual characteristics observed in model 2 also held in model 3 except the association for African Americans was no longer significant in model 3.

DISCUSSION

We used a multilevel framework with 2 geographic levels to examine CRC screening uptake in California. We found that area-level healthcare resources were important in shaping CRC screening use. Specifically, county-level healthcare resources, HMO penetration, and primary care shortages were important contextual factors associated with individual CRC screening use by FSIG. Further, we found that primary care shortage was significant for CRC screening by any modality at the MSSA level but not at the county level, emphasizing the importance of MSSAs for allocating healthcare resources. We also confirmed established associations between individual risk factors and CRC screening use, even when contextual variables were included in the models. Finally, we identified a new individual-level association: food insecurity.

County HMO penetration and primary care shortage were significant predictors of CRC screening by FSIG, possibly reflecting the lasting effect of HMO screening programs in California in the 1990s and early 2000s.^21,34-37 Our study extends previous research on healthcare market characteristics to show that managed care activity is associated with cancer screening uptake.^23,53-57 HMO penetration has been shown to be associated with CRC screening use by any modality in Asian American populations in California.²³ We found this effect for California’s age-eligible population for screening by FSIG, but neighborhood poverty and other social factors were not associated. Smaller units of geography (eg, census tracts) may be more salient.^58-60 This suggests that understanding the multilevel context that may influence CRC screening and other health behaviors requires great care in matching exposures to the appropriate level of geography.

At the individual level, we found that respondents reporting food insecurity were less likely to be up-to-date with CRC screening, which is consistent with our hypothesis that food insecurity is a competing priority or stressor that may be a barrier to accessing and utilizing CRC screening. This has not been extensively examined in previous studies of cancer screening. Our findings for race/ethnicity, income, education, sex, marital status, and insurance are consistent with previous literature.^{8,12,14,15,19,61-63} In our study, individual-level variables remained consistent, strong predictors even in the presence of contextual factors. Our results also parallel findings from 2 other studies that found individual-level factors explained more variation than contextual factors in CRC screening.^23,30 The null associations for SES and segregation may have been due to the large set of individual-level factors that we were able to account for and include in our models, which may have mediated the associations between these 2 contextual factors and CRC screening use.

Limitations of our study include use of cross-sectional data, which does not allow causal inference. While our study findings are specific to California, the contextual factors we examined are not unique to California and could be studied in a similar fashion in other geographic locations. Although self-report of CRC screening tests may overestimate prevalence of CRC screening use, there is mixed evidence about the extent of this problem.^64-66 A recent study showed that data from the 2007 CHIS overestimated being up-to-date by 6% to 14% across racial/ethnic groups.⁶⁵ Skip patterns in the CHIS survey made analysis of CRC screening measures difficult. As a result of our study, CRC screening measures in CHIS were improved. Finally, adding data on additional resources (eg, endoscopy) as well as other contexts (eg, workplace) would offer a more extensive range of factors that may influence CRC screening use.

Our multilevel analysis showed that contextual factors including locality, availability of primary care, and HMO membership influenced use of CRC screening in California. Our results show that both levels of residential geography influence CRC screening use and suggest that multilevel analysis is a promising line of research for cancer screening. Accounting for contextual factors at different geographic scales can provide a richer characterization of the healthcare resources and market characteristics that influence individual behaviors. Moreover, insights from this study could guide interventions aimed at increasing CRC screening uptake by prioritizing appropriate subgroups and geographic areas. A better understanding of contextual factors of the healthcare and social environment will be important in informing policy decisions for healthcare resource allocation and delivery. Future studies on CRC screening that address multilevel predictors should account for the hierarchical structure of the data by using multilevel analysis or by accounting for clustering within neighborhoods when multilevel analysis is not possible. More experimentation to find the appropriate geographic areas is needed in order to evaluate policy and interventions that address factors to improve screening.Acknowledgments

We would like to acknowledge the Cancer Prevention Fellowship Program, Office of Preventive Oncology, National Cancer Institute for providing financial support for Salma Shariff-Marco in her postdoctoral fellowship. We would also like to acknowledge Tim McNeel, William Waldron, and Jeremy Lyman at Information Management Services, Inc, for their support in merging our data sets, developing programs for the descriptive analysis, and generating the maps for the Figure; and Penny Randall-Levy of the Scientific Consulting Group, Inc, for her help with the references. Finally, we would like to acknowledge Drs Martin L. Brown and Rachel Ballard-Barbash for their thoughtful reviews.

Author Affiliations: From Cancer Prevention Institute of California (SSM), Freemont, CA; Division of Cancer Control and Population Sciences (NB, CNK), National Cancer Institute, Rockville, MD; Westat, Inc (DOS), Rockville, MD.

Funding Source: All authors were employed at the National Cancer Institute.

Author Disclosures: The authors (SS-M, NB, DGS, CNK) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (SS-M, NB, DGS, CNK); acquisition of data (SS-M); analysis and interpretation of data (SS-M, NB, DGS, CNK); drafting of the manuscript (SS-M, NB, DGS, CNK); critical revision of the manuscript for important intellectual content (SS-M, NB, DGS, CNK); statistical analysis (SS-M, DGS); obtaining funding (SS-M); administrative, technical, or logistic support (SS-M, NB); and supervision (SS-M, NB).

Address correspondence to: Salma Shariff-Marco, PhD, MPH, Cancer Prevention Institute of California, 2201 Walnut Ave, Ste 300, Fremont, CA 94538. E-mail: salma.shariff-marco@cpic.org.1. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010 [published correction appears in CA Cancer J Clin. 2011;61(2):133-134]. CA Cancer J Clin. 2010;60(5):277-300.

2. American Cancer Society. Colorectal Cancer Facts & Figures, 2008-2010. Atlanta GA: American Cancer Society; 2008. http://www.cancer.org/research/cancerfactsfigures/colorectalcancerfactsfigures/colorectal-cancer-facts--figures-2008-2010. Accessed February 5, 2013.

3. Levin B, Lieberman DA, McFarland B, et al; American Cancer Society Colorectal Cancer Advisory Group; US Multi-Society Task Force; American College of Radiology Colon Cancer Committee. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology. 2008;134(5):1570-1595.

4. Agency for Healthcare Research and Quality. Prevention and Chronic Care Program. U.S. Preventive Services Task Force. http://www.ahrq.gov/clinic/uspstfix.htm. Published 2009. Accessed October 13, 2009.

5. Smith RA, Cokkinides V, Eyre HJ. Cancer screening in the United States, 2007: a review of current guidelines, practices, and prospects. CA Cancer J Clin. 2007;57(2):90-104.

6. Rakowski W, Breslau ES. Perspectives on behavioral and social science research on cancer screening. Cancer. 2004;101(5)(suppl):1118-1130.

7. Ananthakrishnan AN, Schellhase KG, Sparapani RA, Laud PW, Neuner JM. Disparities in colon cancer screening in the Medicare population. Arch Intern Med. 2007;167(3):258-264.

8. van Jaarsveld CH, Miles A, Edwards R, Wardle J. Marriage and cancer prevention: does marital status and inviting both spouses together influence colorectal cancer screening participation? J Med Screen. 2006;13(4):172-176.

9. Link BG, Northridge ME, Phelan JC, Ganz ML. Social epidemiology and the fundamental cause concept: on the structuring of effective cancer screens by socioeconomic status. Milbank Q. 1998;76(3):375-402.

10. Etzioni DA, Ponce NA, Babey SH, et al. A population-based study of colorectal cancer test use: results from the 2001 California Health Interview Survey. Cancer. 2004;101(11):2523-2532.

11. Jerant AF, Fenton JJ, Franks P. Determinants of racial/ethnic colorectal cancer screening disparities. Arch Intern Med. 2008;168(12):1317-1324.

12. Shapiro JA, Seeff LC, Thompson TD, Nadel MR, Klabunde CN, Vernon SW. Colorectal cancer test use from the 2005 National Health Interview Survey. Cancer Epidemiol Biomarkers Prev. 2008;17(7):1623-1630.

13. Trivers KF, Shaw KM, Sabatino SA, Shapiro JA, Coates RJ. Trends in colorectal cancer screening disparities. Am J Prev Med. 2008;35(3):185-193.

14. Hiatt RA, Klabunde C, Breen N, Swan J, Ballard-Barbash R. Cancer screening practices from National Health Interview Surveys: past, present and future. J Natl Cancer Inst. 2002;94(24):1837-1846.

15. Meissner HI, Breen N, Klabunde CN, Vernon SW. Patterns of colorectal cancer screening uptake among men and women in the United States. Cancer Epidemiol Biomarkers Prev. 2006;15(2):389-394.

16. Holden DJ, Jonas DE, Porterfield DS, Reuland D, Harris R. Systematic review: enhancing the use and quality of colorectal cancer screening. Ann Intern Med. 2010;152(10):668-676.

17. Crawley LM, Ahn DK, Winkleby MA. Perceived medical discrimination and cancer screening behaviors of racial and ethnic minority adults. Cancer Epidemiol Biomarkers Prev. 2008;17(8):1937-1944.

18. O’Malley AS, Forrest CB, Feng S, Mandelblatt J. Disparities despite coverage: gaps in colorectal cancer screening among Medicare beneficiaries. Arch Intern Med. 2005;165(18):2129-2135.

19. Zapka JG, Puleo E, Vickers-Lahti M, Luckmann R. Healthcare system factors and colorectal cancer screening. Am J Prev Med. 2002;23(1):28-35.

20. Weitzman ER, Zapka J, Estabrook B, Goins KV. Risk and reluctance: understanding impediments to colorectal cancer screening. Prev Med. 2001;32(6):502-513.

21. Farmer MM, Bastani R, Kwan L, Belman M, Ganz PA. Predictors of colorectal cancer screening from patients enrolled in a managed care health plan. Cancer. 2008;112(6):1230-1238.

22. Giorgi Rossi P, Federici A, Bartolozzi F, Farchi S, Borgia P, Guasticchi G. Understanding non-compliance to colorectal cancer screening: a case control study, nested in a randomised trial [ISRCTN83029072]. BMC Public Health. 2005;5:139.

23. Ponce NA, Huh S, Bastani R. Do HMO market level factors lead to racial/ethnic disparities in colorectal cancer screening? a comparison between high-risk Asian and Pacific Islander Americans and high-risk whites. Med Care. 2005;43(11):1101-1108.

24. Koka VK, Potti A, Fraiman GN, Hanekom D, Hanley JF. An epidemiological study evaluating the relationship of distance from a tertiary care cancer center to early detection of colorectal carcinoma. Anticancer Res. 2002;22(4):2481-2483.

25. Haas JS, Fitzmaurice G, Brawarsky P, et al. Association of regional variation in primary care physicians’ colorectal cancer screening recommendations with individual use of colorectal cancer screening. Prev Chronic Dis. 2007;4(4):A90.

26. Fukuda Y, Nakamura K, Takano T. Reduced likelihood of cancer screening among women in urban areas with low socioeconomic status: a multilevel analysis in Japan. Public Health. 2005;119(10):875-884.

27. Schootman M, Jeffe DB, Baker EA, Walker MS. Effect of area poverty rate on cancer screening across US communities. J Epidemiol Community Health. 2006;60(3):202-207.

28. Lian M, Schootman M, Yun S. Geographic variation and effect of area-level poverty rate on colorectal cancer screening. BMC Public Health. 2008;8:358.

29. James TM, Greiner KA, Ellerbeck EF, Feng C, Ahluwalia JS. Disparities in colorectal cancer screening: a guideline-based analysis of adherence. Ethn Dis. 2006;16(1):228-233.

30. Coughlin SS, Thompson TD. Colorectal cancer screening practices among men and women in rural and nonrural areas of the United States, 1999. J Rural Health. 2004;20(2):118-124.

31. Pornet C, Dejardin O, Morlais F, Bouvier V, Launoy G. Socioeconomic determinants for compliance to colorectal cancer screening: a multilevel analysis. J Epidemiol Community Health. 2010;64(4):318-324.

32. Commission on the Social Determinants of Health. Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants of Health. Final Report. Geneva, Switzerland: World Health Organization; 2008. http://www.who.int/social_determinants/thecommission/ finalreport/en/index.html. Accessed February 5, 2013.

33. Link BG, Phelan J. Social conditions as fundamental causes of disease. J Health Soc Behav. 1995;Spec No:80-94.

34. Palitz AM, Selby JV, Grossman S, et al. The Colon Cancer Prevention Program (CoCaP): rationale, implementation, and preliminary results. HMO Pract. 1997;11(1):5-12.

35. Levin TR, Jamieson L, Burley DA, Reyes J, Oehrli M, Caldwell C. Organized colorectal cancer screening in integrated health care systems. Epidemiol Rev. 2011;33(1):101-110.

36. Ganz PA, Farmer MM, Belman M, et al. Improving colorectal cancer screening rates in a managed care health plan: recruitment of provider organizations for a randomized effectiveness trial. Cancer Epidemiol Biomarkers Prev. 2003;12(9):824-829.

37. Ganz PA, Farmer MM, Belman MJ, et al. Results of a randomized controlled trial to increase colorectal cancer screening in a managed care health plan. Cancer. 2005;104(10):2072-2083.

38. California Office of Statewide Health Planning and Development. Medical service study areas. http://gis.ca.gov/catalog/BrowseRecord.epl?id=23784. Published 2005. Accessed October 13, 2009.

39. California Health Interview Survey. CHIS 2005 Methodology Series. Report 1— Sample Design. Los Angeles, CA: UCLA Center for Health Policy Research; April 2007.

40. Mobley LR. RTI Spatial Impact Factor Database. Version 1. Developed for the National Cancer Institute under the R21 grant Spatial Impact Factors and Mammography Use, Principal Investigator Lee Mobley. Research Triangle Park, NC: RTI International. 2006.

41. California Health Interview Survey. CHIS 2005 Adult Survey. Los Angeles, CA: UCLA Center for Health Policy Research; 2006.

42. National Center for Health Workforce Analysis. Area Resource File (ARF). Rockville, MD: Bureau of Health Professions, Health Resources and Services Administration; 2004.

43. National Center for Health Workforce Analysis. User Documentation for the Area Resource File (ARF) 2004 Release. Rockville, MD: Bureau of Health Professions, Health Resources and Services Administration; 2004.

44. Maly MT. The neighborhood diversity index: a complementary measure of racial residential settlement. J Urban Aff. 2000;22(1):37-47.

45. California Health Interview Survey. CHIS 2005 Methodology Series. Report 5—Weighting and Variance Estimation. Los Angeles, CA: UCLA Center for Health Policy Research; April 2007.

46. Rabe-Hesketh S, Skrondal A. Multilevel and Longitudinal Modeling Using Stata. College Station, TX: Stata Press; 2005.

47. Pfeffermann D, Skinner CJ, Holmes DJ, Goldstein H, Rasbash J. Weighting for unequal selection probabilities in multilevel models. J R Stat Soc Series B Stat Methodol. 1998;60:23-40.

48. Chantala K, Blanchette D, Suchindran CM. Software to compute sampling weights for multilevel analysis. http://www.cpc.unc.edu/research/tools/data_analysis/ml_sampling_weights. Published April 28, 2011. Accessed September 15, 2011.

49. Merlo J, Chaix B, Ohlsson H, et al. A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena. J Epidemiol Community Health. 2006;60(4):290-297.

50. Larsen K, Merlo J. Appropriate assessment of neighborhood effects on individual health: integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol. 2005;161(1):81-88.

51. Larsen K, Petersen JH, Budtz-Jørgensen E, Endahl L. Interpreting parameters in the logistic regression model with random effects. Biometrics. 2000;56(3):909-914.

52. Merlo J, Yang M, Chaix B, Lynch J, Råstam L. A brief conceptual tutorial on multilevel analysis in social epidemiology: investigating contextual phenomena in different groups of people. J Epidemiol Community Health. 2005;59(9):729-736.

53. Litaker D, Tomolo A. Association of contextual factors and breast cancer screening: finding new targets to promote early detection. J Womens Health (Larchmt). 2007;16(1):36-45.

54. Wholey DR, Burns LR, Lavizzo-Mourey R. Managed care and the delivery of primary care to the elderly and the chronically ill. Health Serv Res. 1998;33(2, pt 2):322-353.

55. Morrisey MA. Competition in hospital and health insurance markets: a review and research agenda. Health Serv Res. 2001;36 (1, pt 2):191-221.

56. Baker LC, Phillips KA, Haas JS, Liang SY, Sonneborn D. The effect of area HMO market share on cancer screening. Health Serv Res. 2004;39(6, pt 1):1751-1772.

57. Phillips KA, Fernyak S, Potosky AL, Schauffler HH, Egorin M. Use of preventive services by managed care enrollees: an updated perspective. Health Aff (Millwood). 2000;19(1):102-116.

58. Massey DS, Denton NA. The dimensions of residential segregation. Soc Forces. 1988;67(2):281-315.

59. Iceland J, Weinberg DH, Steinmetz E. Racial and Ethnic Residential Segregation in the United States: 1980-2000. Washington, DC: US Government Printing Office; 2002.

60. Krieger N, Chen JT, Waterman PD, Soobader MJ, Subramanian SV, Carson R. Geocoding and monitoring of US socioeconomic inequalities in mortality and cancer incidence: does the choice of area-based measure and geographic level matter? the Public Health Disparities Geocoding Project. Am J Epidemiol. 2002;156(5):471-482.

61. Zimmerman RK, Norwalk MP, Tabbara M, Grufferman S. Predictors of colorectal cancer screening in diverse primary care practices. BMC Health Serv Res. 2006;6:116.

62. Peterson NB, Murff HJ, Ness RM, Dittus RS. Colorectal cancer screening among men and women in the United States. J Womens Health (Larchmt). 2007;16(1):57-65.

63. Schenck AP, Klabunde CN, Davis WW. Racial differences in colorectal cancer test use by Medicare consumers. Am J Prev Med. 2006; 30(4):320-326.

64. Bastani R, Glenn BA, Maxwell AE, Ganz PA, Mojica CM, Chang LC. Validation of self-reported colorectal cancer (CRC) screening in a study of ethnically diverse first-degree relatives of CRC cases. Cancer Epidemiol Biomarkers Prev. 2008;17(4):791-798.

65. Palaniappan LP, Maxwell AE, Crespi CM, Wong EC, Shin J, Wang EJ. Population colorectal cancer screening estimates: comparing selfreport to electronic health record data in California. Int J Canc Prev. 2011;4(1).

66. Partin MR, Grill J, Noorbaloochi S, et al. Validation of self-reported colorectal cancer screening behavior from a mixed-mode survey of veterans. Cancer Epidemiol Biomarkers Prev. 2008;17(4):768-776.