Knowledge of Colorectal Cancer Screening Guidelines and Intention to Obtain Screening Among Nonadherent Filipino, Hmong, and Korean Americans

Transcription

1 Original Article Knowledge of Colorectal Cancer Screening Guidelines and Intention to Obtain Screening Among Nonadherent Filipino, Hmong, and Korean Americans Janice Y. Tsoh, PhD 1,2 ; Elisa K. Tong, MD 2,3 ; Angela U. Sy, PhD 4 ; Susan L. Stewart, PhD Tung T. Nguyen, MD 2,6 5 ; Ginny L. Gildengorin, PhD 6 ; and BACKGROUND: Nonadherence to colorectal cancer (CRC) screening among Asian Americans is high but not well understood. This study examined correlates of screening intention among Filipino, Hmong, and Korean Americans who were nonadherent to CRC screening. METHODS: Using cross-sectional, preintervention survey data from 504 Asian Americans (115 Filipinos, 185 Hmong, and 204 Koreans) aged years who were enrolled in a multisite cluster randomized controlled trial of lay health educator intervention, we analyzed correlates of self-reported CRC screening nonadherence, which was defined as not being up-to-date for fecal occult blood test, sigmoidoscopy, or colonoscopy. RESULTS: Only 26.8% of participants indicated intention to obtain screening within 6 months (Hmong: 12.4%; Korean: 30.8%; and Filipino: 42.6%; P <.001). Only one third of participants had undergone a prior screening, and a majority did not know that screening is a method of CRC prevention method (61.3%) or had any knowledge of CRC screening guidelines (53.4%). Multivariable analyses revealed that patient provider ethnicity concordance, provider s recommendation of screening, participants prior CRC screening, perceived severity and susceptibility of CRC, and knowledge of guidelines were positively associated with screening intention. Specifically, knowing one or more screening guidelines doubled the odds of screening intention (adjusted odds ratio, 2.38; 95% confidence interval, ). Hmong were less likely to have screening intention than Filipinos, which was unexplained by socio-demographics, health care factors, perceived needs for CRC screening, or knowledge of screening guidelines. CONCLUSION: CRC screening intention among nonadherent Filipino, Hmong, and Korean Americans was low. Targeting knowledge of CRC screening guidelines may be effective strategies for increasing CRC screening intention among nonadherent Asian Americans. Cancer 2018;124: VC 2018 American Cancer Society. KEYWORDS: Asian Americans, cancer screening, colorectal cancer, health knowledge, intention. INTRODUCTION Early detection of colorectal cancer (CRC) reduces CRC-related mortality. 1 The United States Preventive Services Task Force 2 recommends CRC screening for adults aged years. Recommended screening guidelines include annual fecal occult blood test (FOBT), sigmoidoscopy every 5 years, or colonoscopy every 10 years. 2 In 2015, despite improvements in CRC screening adherence in the United States general population, 3 Asian Americans, the fastest-growing racial group in the United States, 4 reported lower CRC screening adherence (49.5%) compared with non-hispanic Whites (65.4%) or African Americans (61.8%). 5 From 2003 to 2011, a significant decline in CRC mortality among non-hispanic whites was observed without a corresponding decline among Asian Americans. 6 Efforts to promote CRC screening among vulnerable groups with low adherence to Corresponding author: Janice Y. Tsoh, PhD, University of California San Francisco, Department Psychiatry, 401 Parnassus Avenue, San Francisco, CA ; janice.tsoh@ucsf.edu 1 Department of Psychiatry, University of California San Francisco, San Francisco, California; 2 Asian American Research Center on Health, San Francisco, California; 3 Department of Internal Medicine, University of California Davis, Davis, California; 4 John A. Burns School of Medicine, University of Hawaii at Manoa, Honolulu, Hawaii; 5 Department of Public Health Sciences, University of California Davis, Davis, California; 6 Division of General Internal Medicine, University of California San Francisco, San Francisco, California We are grateful for the contributions of Moon S. Chen Jr., Angela M. Jo, Min J. Sung, Majorie Kagawa-Singer, Charlene F. Cuaresma, Penny Lo, May Chee Lo, Ching Wong, and Hy Lam in the parent randomized controlled trial study design, survey instrument development, recruitment, and data collection. We also thank the Asian American Network for Cancer Awareness, Research, and Training (AANCART) interns Vickie Nguyen, Kristine Phung, Mimi Tran, Matthew Jeong, and Filmer Yu for assistance with the project. We also appreciate the efforts of our lay health educators in recruiting participants. The articles in this Supplement were presented at the National Center on Reducing Asian American Cancer Health Disparities (also known as the Asian American Network for Cancer Awareness, Research, and Training AANCART ) meeting held in August The organizational entities that comprise AANCART included the University of California, Davis Comprehensive Cancer Center (Lead); University of California, San Francisco; University of California, Los Angeles; University of Hawaii; Chinese Community Health Organization; and Hmong Women s Heritage Association. This supplement was funded in part through a cooperative agreement grant funded by the National Cancer Institute s Center to Reduce Cancer Health Disparities under grant 3U54 CA The views in this Supplement are those of the authors and do not necessarily reflect the opinions of the American Cancer Society, John Wiley & Sons, Inc., or the National Cancer Institute. DOI: /cncr.31097, Received: August 28, 2017, Accepted: October 10, 2017, Published online March 22, 2018 in Wiley Online Library (wileyonlinelibrary.com) 1560 Cancer April 1, 2018

2 Knowledge and Intention of CRC Screening/Tsoh et al CRCscreening,includingAsianAmericanimmigrantswith limited English proficiency 3,7 and those with lower income or limited health care resources, 3 are urgently needed. To direct intervention efforts effectively, it is critical to understand factors associated with CRC screening nonadherence, particularly factors that are modifiable. This study examined CRC screening intention and its correlates among a diverse group of Filipino, Hmong, and Korean American adults who reported being nonadherent to CRC screening. Specifically, we postulated that knowledge about CRC screening, measured by knowing that screening is a CRC prevention method and knowing the screening guidelines, are associated with screening intention. MATERIALS AND METHODS This study used cross-sectional, preintervention data from a multisite cluster randomized controlled trial testing a lay health educator intervention promoting CRC screening among Asian Americans (ClinicalTrials.gov registration: NCT ). Participant eligibility criteria were: age years; self-identification as Filipino (in Hawaii), Hmong (in Sacramento, CA), or Korean (in Los Angeles, CA); speaking English, Philippine languages (Ilokano and Tagalog), Hmong, or Korean; and having no history of CRC. Study participants were recruited by 82 lay health educators from their social networks. Details of participant recruitment were reported elsewhere. 8,9 Between August 2012 and January 2015, preintervention data were collected from eligible participants who received $20 for completing a survey in their preferred language. Survey instruments in Philippine languages, Hmong, and Korean were pretested with key informants from the targeted communities. 8,9 Surveys were self-administered among Filipinos and Koreans; for Hmong, project staff verbally administered surveys due to the low written literacy in both English and Hmong languages. Because the present study focused on participants who reported CRC screening nonadherence, defined as not being up-to-date for FOBT within the past year, sigmoidoscopy within the past 5 years, or colonoscopy within the past 10 years, the study sample consisted of 504 participants (51.3% of 981 total randomized controlled trial participants) with 115 Filipinos, 185 Hmong, and 204 Koreans. The Institutional Review Boards of the University of California San Francisco and University of Hawaii at Manoa approved all study procedures. Conceptual Framework We integrated constructs from the Theory of Planned Behavior, 10 Health Belief Model, 11 and Andersen s Behavioral Model of Health Services Use. 12 These models have been applied to CRC and other cancer screening research. 13 Intention, an immediate antecedent to health behaviors, 10 is an important determinant of screening behaviors. 13 We examined four domains of correlates for screening intention: 1) socio-demographics; 2) health care factors; 3) perceived needs for CRC screening adapted from both Health Belief Model (perceived susceptibility and severity of CRC) and Andersen s Behavioral Model of Health Services Use (general health status and knowledge of CRC); and 4) knowledge of CRC screening. Measures Survey items were adapted from the California Health Interview Survey, a statewide population health survey. 14 Screening intention was assessed by providing a description of each CRC screening test (FOBT, sigmoidoscopy or colonoscopy) and then asking participants their intention to obtain each type of screening test in the next 6 months. 14,15 Sociodemographic variables assessed included: Asian ethnicity, age, sex, birthplace, marital status, education, employment, self-reported household income, years lived in the United States, and spoken English proficiency. Health care factors examined included: health insurance coverage, regular place of health care, whether their primary care doctor was Asian, whether they had seen a doctor in the past year, medical interpreter use, whether a doctor had ever recommended CRC screening, and prior screening for CRC. Perceived needs for CRC screening included: self-rated health, self-reported chronic health conditions, family history of CRC, knowledge of CRC, and concern about getting CRC (perceived severity) or perceived chance of getting CRC (perceived susceptibility). Both perceived susceptibility and severity were relevant factors for CRC screening adherence. 16,17 Knowledge of CRC screening included: answering correctly that getting medical tests to find blood or polyps in the colon can prevent colon cancer and a numeric score (0-4) for correct response to screening guidelines (screening age starting at 50, and recommended testing intervals for FOBT, sigmoidoscopy, and colonoscopy). Statistical Analysis Descriptive statistics were computed for all measures. We conducted bivariate and multivariable analyses of the binary outcome: CRC screening intention in the next 6 months (yes versus no). Generalized linear mixed effects models were used to account for clustering of participants by lay health educator in bivariate and multivariable models. Correlates that attained P <.10 in bivariate analyses were included as covariates in the multivariable regression analyses. Age, sex, English proficiency, and prior screening were included as a Cancer April 1,

3 Original Article priori covariates. Because of the collinearity observed between having seen a physician within the past year and prior screening, we excluded the variable for physician visit. To identify factors associated with screening intention, grouped as domains in the order of modifiability potential, we constructed 3 additive multivariable models. Model 1 tested the associations with sociodemographic and health care factors, which included characteristics that were least modifiable. Model 2 added to Model 1 the perceived needs of screening factors, which were considered potentially modifiable. Model 3 added to Model 2 the knowledge factors, which were assumed to be most modifiable. Because of the bimodal distribution of the screening guideline knowledge score (53.4% scored at 0, 31.8% scored at 1, and 0.2% scored at 4), screening guideline knowledge was dichotomized to 1 1 (1 or more correct responses) versus 0 (none correct). Statistical significance was assessed at the.05 level (2-sided). Statistical analyses were performed using SAS version 9.3. RESULTS Socio-demographics, Health Care Factors, and Perceived Needs of CRC Screening The study sample (n 5 504) included 79.4% females with a mean age of 60.6 years (standard deviation, 7.3 years). All but 3 Filipino participants were immigrants. Table 1 presents the characteristics by each Asian group. A majority had health care access and utilization within the past year. Hmong participants had the highest rate of prior CRC screening (50.3%), and Koreans had the lowest rate (17.2%). Over half of the Filipino and Hmong participants and one third of the Korean participants reported being worried about getting CRC. Few (12.1%) participants reported perceiving a high chance of getting CRC, which was similar across the 3 groups. Knowledge of CRC Screening and Guidelines Table 1 shows that only 38.7% of participants knew that CRC screening is a CRC prevention method, ranging from 7% among Hmong to 67.8% among Filipinos. Knowledge of screening guidelines was low across all groups, with only 28.2% knowing that screening started at age 50 and 18.1% knowing that FOBT was recommended annually. Very few (5.2%) knew the screening interval recommendations for sigmoidoscopy or colonoscopy. Half (53.4%) of participants had no knowledge of these guidelines. CRC Screening Intention Only 26.8% reported screening intention within 6 months. Screening intention in descending order by group were Filipinos (42.6%), Koreans, (30.8%), and Hmong (12.4%). Table 2 shows screening intention by sample characteristics. Multivariable Regression Analyses for CRC Screening Intention Table 3 shows the significant correlates (P <.05) of screening intention in 3 additive multivariable regression models. Model 1, which included socio-demographics and health care factors, revealed that ethnicity, marital status, income, health care provider ethnicity, and health care provider recommendation were significant correlates for screening intention. With perceived needs added (Model 2), marital status was no longer significant, and perceived severity and susceptibility were additional significant correlates. Model 3, the final model, revealed that knowing one or more screening guidelines doubled the odds of screening intention. In addition, having an Asian health care provider, having undergone prior screening, having received a health care provider s recommendation of CRC screening, being worried about getting CRC, and perceiving a high chance of having CRC were positively associated with screening intention. Individuals with undisclosed income were less likely to report screening intention. Hmong reported lower screening intention when compared with Filipinos, but Koreans were not different from Hmong or Filipinos. DISCUSSION In this sample of Filipino, Hmong, and Korean Americans who reported nonadherence to CRC screening, 73% reported no intention to undergo screening within 6 months. More than half did not know that CRC screening is a CRC prevention method or had any awareness of CRC screening guidelines. In multivariable analyses, we found that knowing at least one CRC screening guideline doubled the odds of screening intention, but knowing that CRC screening is a CRCpreventionmethodwasnotassociatedwithintention. Prior studies assessed CRC knowledge by measuring awareness of CRC, names of screening tests, CRC risks, and CRC symptoms, which yielded mixed findings regarding the association of knowledge with intention or screening adherence Educational interventions with Filipinos, 20 Hmong, 8 and Koreans 9 are effective in increasing guideline knowledge and in CRC screening, 8,20 with knowledge of both CRC and its screening mediating the intervention effects on screening receipt among Hmong and Filipinos. 8,20 Our findings confirm the importance of increasing knowledge of CRC screening guidelines, a highly 1562 Cancer April 1, 2018

4 TABLE 1. Socio-demographics, Health Care Factors, Perceived Needs, and Knowledge Among Asian Americans Reporting CRC Screening Nonadherence Filipino (n 5 115) Hmong (n 5 185) Korean (n 5 204) Entire Sample (N 5 504) P a Socio-demographics and acculturation Age, y (60.9) 131 (70.8) 145 (71.1) 346 (68.7) (39.1) 54 (29.2) 59 (28.9) 158 (31.3) Sex.19 Men 24 (21.9) 46 (24.9) 34 (17.7) 104 (21.6) Women 91 (79.1) 139 (75.1) 170 (83.3) 400 (79.4) Married 66 (57.4) 119 (64.3) 147 (72.1) 332 (65.9).11 Born in United States 3 (2.6) 0 (0.0) 0 (0.0) 3 (0.6).59 Years lived in United States y 24 (20.9) 34 (18.4) 19 (9.3) 77 (15.3) >10 y b 91 (79.1) 151 (81.6) 185 (90.7) 427 (84.7) Education <.001 Less than high school graduate 31 (27.4) 172 (95.6) 26 (12.8) 229 (46.1) High school graduate 23 (20.4) 5 (2.8) 54 (26.5) 82 (16.5) Some college/technical school 31 (27.4) 1 (0.5) 42 (20.6) 74 (14.9) College graduate 28 (24.8) 2 (1.1) 82 (40.2) 112 (22.5) Missing Employment status <.001 Unemployed/student/homemaker/other 13 (11.3) 115 (62.2) 73 (35.8) 201 (39.9) Employed 71 (61.7) 19 (10.3) 89 (43.6) 179 (35.5) Retired 31 (27.0) 51 (27.6) 42 (20.6) 124 (24.6) Annual household income <.001 <$20, (28.7) 103 (55.7) 67 (32.8) 203 (40.3) $20, (50.4) 6 (3.2) 114 (55.9) 178 (35.3) Don t know/did not provide 24 (20.9) 76 (41.1) 23 (11.3) 123 (24.2) English-speaking skills <.01 Fair/poor/none 33 (28.7) 177 (95.7) 190 (93.1) 400 (79.4) Good/fluent 82 (71.3) 8 (4.3) 14 (6.9) 104 (20.6) Health care factors Has health insurance 101 (87.8) 171 (92.9) 122 (61.0) 394 (79.0) <.001 Has regular place for healthcare 102 (88.7) 170 (91.9) 103 (50.5) 375 (74.4) <.001 Saw doctor in past 12 months 99 (86.1) 142 (76.7) 128 (62.8) 369 (73.2) <.001 Medical interpreter use <.001 Sometimes, often/ always 8 (7.0) 117 (65.3) 77 (37.9) 202 (40.4) Never/rarely 107 (93.0) 65 (35.7) 126 (62.1) 298 (59.6) Missing Health care provider ethnicity.03 Asian 71 (61.7) 71 (38.4) 122 (59.8) 264 (52.4) Non-Asian 31 (27.0) 89 (48.1) (2.9) 126 (25.0) No primary care provider 13 (11.3) 25 (13.5) 76 (37.3) 114 (22.6) Doctor recommended CRC screening 24 (21.2) 15 (8.6) 13 (6.4) 52 (10.6).02 Ever been screened for CRC 42 (36.5) 93 (50.3) 35 (17.2) 170 (33.7) <.001 Perceived needs for CRC screening Self-rated health status <0.001 Excellent/very good/good 95 (84.8) 94 (528) 105 (52.2) 294 (59.9) Fair/poor 17 (15.2) 84 (47.2) 96 (47.8) 197 (40.1) Missing Has chronic health conditions c.21 None 38 (33.0) 58 (31.4) 79 (38.7) 175 (34.7) 1 condition 38 (33.0) 57 (30.8) 70 (34.3) 165 (32.7) 2 or more conditions 39 (33.9) 70 (37.8) 55 (27.0) 164 (32.5) Ever heard of CRC 81 (70.4) 77 (41.6) 154 (75.5) 312 (61.9) <.001 Has family history of CRC 2 (1.7) 2 (1.1) 18 (8.8) 22 (4.4) <.001 Worried about getting CRC 72 (63.2) 112 (60.5) 68 (33.3) 252 (50.1) <.001 Perceived high chance of getting CRC 19 (16.5) 23 (12.4) 19 (9.3) 61 (12.1).16 Knowledge of screening as a CRC prevention method 78 (67.8) 13 (7.0) 104 (51.0) 195 (38.7) <.001 Knowledge of CRC screening guidelines (answered correctly) Guideline 1: start screening age (45.2) 19 (10.3) 71 (34.8) 142 (28.2) <.001 Guideline 2: FOBT testing every year 26 (22.6) 16 (8.7) 49 (24.0) 91 (18.1) <.001 Guideline 3: sigmoidoscopy every 5 years 11 (9.6) 3 (1.6) 52 (25.5) 66 (13.1) <.001 Guideline 4: colonoscopy every 10 years 15 (13.0) 0 (0.0) 11 (5.4) 26 (5.2) <.001 Total correct out of the 4 CRC screening guideline questions < (38.3) 154 (83.2) 71 (34.8) 269 (53.4) 1 71 (61.7) 31 (16.8) 133 (65.2) 235 (46.6) All data are presented as n (%). Unless otherwise indicated, missing observations were excluded from percentage computation. Percentages may not add up to 100% due to rounding. a P values accounted for clustering of participants by lay health educators in the comparisons across ethnic groups. b Years in the United States for >10 years includes the 3 Filipino participants who were born in the United States. c Chronic health conditions included self-reported high blood pressure, high cholesterol, diabetes, or cancer.

5 Original Article TABLE 2. CRC Screening Intention by Socio-demographics, Health Care Factors, Perceived Needs, and Knowledge Among Asian Americans Reporting CRC Screening Nonadherence (N 5 504) CRC Screening Intention Within 6 Months (n 5 135) No Intention of CRC Screening (n 5 369) P a Socio-demographics and acculturation Ethnicity <.001 Filipino 49 (42.6) 66 (57.4) Hmong 23 (12.4) 162 (87.6) Korean 63 (30.8) 141 (69.1) Age, y (27.5) 251 (72.5) (25.3) 118 (74.7) Sex.51 Men 32 (30.8) 72 (69.2) Women 103 (25.8) 297 (74.3).33 Marital status.07 Married 98 (29.5) 234 (70.5) Not married 37 (21.5) 135 (78.5) Years lived in United States y 22 (28.6) 55 (71.4) >10 y b 113 (26.5) 314 (73.5) Education.01 Less than high school graduate 42 (18.3) 187 (81.7) High school graduate 30 (36.6) 52 (63.4) Some college/technical school 21 (28.4) 53 (71.6) College graduate 40 (35.7) 72 (64.3) Missing 0 7 Employment status.01 Unemployed 41 (20.4) 160 (79.6) Employed 63 (35.2) 116 (64.8) Retired 31 (25.0) 93 (75.0) Annual household income <.001 <$20, (28.1) 146 (71.9) $20, (34.8) 116 (65.2) Don t know/did not provide 16 (13.0) 107 (87.0) English-speaking skills <.01 Fair/poor/none 94 (23.5) 306 (76.5) Good/fluent 41 (39.4) 63 (60.6) Health care factors Has health insurance.93 Yes 107 (27.2) 287 (72.8) No 28 (26.7) 77 (73.3) Has regular place for health care.29 Yes 105 (28.0) 270 (72.0) No 30 (23.3) 99 (76.6) Saw doctor in past 12 months.04 Yes 108 (29.3) 261 (70.7) No 27 (20.0) 108 (80.0) Medical interpreter use.03 Sometimes, often/always 42 (20.8) 160 (79.2) Never/rarely 92 (30.9) 206 (69.1) Missing 1 3 Health care provider ethnicity.03 Asian 85 (32.2) 179 (67.8) Non-Asian 28 (22.2) 98 (77.8) No primary care provider 22 (19.3) 92 (80.7) Doctor recommended CRC <.01 screening Yes 25 (48.1) 27 (51.9) No 108 (24.6) 331 (75.4) Ever been screened for CRC.10 Yes 54 (31.8) 116 (68.2) No 81 (24.3) 253 (75.7) Perceived needs for CRC screening Self-rated health status.59 Excellent, very good/good 81 (27.6) 213 (72.5) Fair/poor 50 (25.4) 147 (74.6) Missing Cancer April 1, 2018

6 Knowledge and Intention of CRC Screening/Tsoh et al TABLE 2. Continued CRC Screening Intention Within 6 Months (n 5 135) No Intention of CRC Screening (n 5 369) P a Has chronic health conditions c.56 None 47 (26.9) 128 (73.1) 1 condition 39 (23.6) 126 (76.4) 2 or more conditions 49 (29.9) 115 (70.1) Ever heard of CRC.04 Yes 94 (30.1) 218 (69.9) No 41 (21.4) 151 (78.7) Has family history of CRC.59 Yes 7 (31.8) 15 (68.2) No 128 (26.6) 354 (73.4) Worried about getting CRC <.01 Yes 83 (32.9) 169 (67.1) No 52 (20.7) 199 (79.3) Perceived high chance of getting <.01 CRC Yes 31 (50.8) 30 (48.2) No 104 (23.5) 339 (76.5) Knowledge of screening as a CRC <.001 prevention method Medical tests to find blood or polyps in the colon can prevent CRC Yes/correct 81 (41.5) 114 (58.5) No/incorrect 54 (17.5) 255 (82.5) Knowledge of CRC screening guidelines Total correct, out of the 4 CRC <.001 screening guideline questions 0 41 (15.2) 228 (84.8) 1 94 (40.0) 141 (60.0) All data are presented as n (%). Unless otherwise indicated, missing observations were excluded from percentage computation. Percentages, across each row, may not add up to be 100.0% due to rounding. a P values accounted for clustering of participants by lay health educators. b Years in the United States for >10 years included the 3 Filipino participants who were born in the United States. c Chronic health conditions included self-reported high blood pressure, high cholesterol, diabetes, or cancer modifiable factor, as part of any strategy to promote CRC screening among nonadherent Asian Americans. In addition to knowledge of screening guidelines, perceived needs as measured by participants worry about getting CRC or perceiving a high chance of getting CRC were associated with higher screening intention. This is consistent with other studies showing associations between perceived severity or susceptibility for CRC and screening adherence. 17,18 This study also revealed that some health care factors are associated with higher screening intention. Participants who had an Asian health care provider were more likely intend to get screening compared with those who had no provider. Ethnically concordant providers may have better insights and understanding of cultural and contextual factors relevant for screening adherence among patients. Both physician recommendation for CRC screening and prior CRC screening, as found in previous studies, 17,18 were positively associated with CRC screening uptake. In contrast to population data showing low CRC screening adherence among individuals with low income, 3 we found no difference between low versus higher income in screening intention. Participants with undisclosed income, however, had a lower odds of screening intention. This finding warrants further investigation, because this group might experience different barriers to intention or adherence. The difference in screening intention between Hmong and Koreans was explained by guideline knowledge, because Model 3 showed no difference between the 2 groups in screening intention when guideline knowledge was included. However, even after adjusting for socio-demographics, health care factors, perceived needs, and knowledge, Hmong remained less likely to intend to get screening compared with Filipinos. Cultural beliefs in cancer or prevention among Hmong, which remain understudied, could be plausible explanations for lack of screening intention or behaviors. Because preventive health care may be a new concept Hmong older adults, they may think that screening or seeking preventive health care is unnecessary without experiencing health symptoms. 21 Nonetheless, this topic deserves additional research. Of note, although Hmong participants reported the highest rates of prior CRC screening, they had Cancer April 1,

7 Original Article TABLE 3. Multivariable Regression Analyses of CRC Screening Intention Within 6 Months Among Asian Americans Reporting CRC Screening Nonadherence (n 5 504) Model 1 Model 2 (Model 1 1 Perceived Needs) Model 3 (Model 2 1 Knowledge) Socio-demographics Ethnicity (Ref: Hmong) Filipino 9.39 ( ) a 8.05 ( ) a 4.78 ( ) b Korean 6.40 ( ) a 7.32 ( ) a 3.79 ( ) Married (Ref: not married) 1.68 ( ) d 1.62 ( ) 1.55 ( ) Household income (Ref: < $20,000/ year) $20,000 or more 0.61( ) 0.57 ( ) 0.55 ( ) Don t know/did not provide 0.39 ( ) a 0.38 ( ) a 0.43 ( ) c Health care factors Health care provider ethnicity (Ref: No provider) Asian 2.44 ( ) a 2.25 ( ) a 2.44 ( ) a Non-Asian 1.48 ( ) 1.31 ( ) 1.28 ( ) Doctor recommended CRC screening 2.42 ( ) a 2.25 ( ) b 2.04 ( ) d (Ref: No) Ever been screened for CRC (Ref: 1.81 ( ) a 1.64 ( ) 1.65 ( ) d never been screened) Perceived need for CRC screening Ever heard of CRC (Ref: never NA 1.08 ( ) 0.96 ( ) heard) Worried about getting CRC (Ref: NA 2.07 ( ) a 1.84 ( ) c No) Perceived high chance of getting NA 2.23 ( ) a 2.05 ( ) c CRC (Ref: No) Knowledge of screening as a CRC prevention method Medical tests to find blood or polyps NA NA 2.03 ( ) in the colon can prevent CRC (Ref: No/incorrect) Knowledge of CRC screening guidelines Knew one or more guidelines (Ref: knew none of the guidelines) NA NA 2.38 ( ) a Abbreviations: NA, not applicable; Ref, reference group. Data are presented as the adjusted odds ratio (95% confidence interval). All models accounted for clustering of participants by lay health educators and were adjusted by age, sex, education, employment status, English proficiency, and use of medical interpreters, which were not associated with CRC screening intention in the multivariable models (P >.05). Significant adjusted odds ratios (P <.05) appear in bold type. a P <.01. b P c P d P the lowest screening intention. This finding underscores the need to emphasize the importance of regular CRC screening per guideline recommendations. Our study has some limitations. The study sample consisted primarily of immigrants who were enrolled in an educationalinterventiontrial.inaddition,mosthadused health care services within the past year. As such, the findings may not be generalizable to other Filipino, Hmong, and Korean Americans or other Asian American immigrants. In addition, the cross-sectional nature of the data did not allow inference of causation of identified factors leading to changes of intentions; hence the findings are not conclusive as to which factors motivate or lead to screening intention. In conclusion, CRC screening intention among nonadherent Filipino, Hmong, and Korean Americans was low; only 1 in 4 participants intended to get CRC screening within 6 months. Knowing at least one CRC screening guideline recommendation doubled the odds of intending to get screened. Interventions aiming at increasing knowledge of CRC screening guidelines and the perceived needs of screening may be effective strategies for increasing CRC screening intention, and subsequent receipt, among nonadherent Asian Americans. FUNDING SUPPORT Supported in part by grants from the National Cancer Institute, Center to Reduce Cancer Health Disparities (Asian American Network for Cancer Awareness, Research and Training, and grants U54CA and R01CA138778) Cancer April 1, 2018

8 Knowledge and Intention of CRC Screening/Tsoh et al CONFLICTS OF INTEREST DISCLOSURES Angela U. Sy has received grants from the National Institutes of Health for work performed outside of the current study. AUTHOR CONTRIBUTIONS Janice Y. Tsoh: Conceptualization, methodology, supervision, formal analysis, writing (original draft, review and editing). Elisa K. Tong: Conceptualization, project administration, and writing (review and editing). Angela U. Sy: Conceptualization, project administration, and writing (review and editing). Susan L. Stewart: Methodology, validation, supervision, formal analysis, and writing (review and editing). Ginny L. Gildengorin: Methodology, validation, formal analysis, and writing (review and editing). Tung T. Nguyen: Conceptualization, writing (review and editing), supervision, and funding acquisition. REFERENCES 1. Edwards BK, Ward E, Kohler BA, et al. Annual report to the nation on the status of cancer, , featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer. 2010;116: U.S. Preventive Services Task Force. Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2016;315: White A, Thompson TD, White MC, et al. Cancer Screening Test Use United States, MMWR Morb Mortal Wkly Rep. 2017;66: Hoeffel EM, Rastogi S, Kim MO, Shahid H. The Asian population: In: U.S. Department of Commerce Economics and Statistics Administration, ed Census Briefs. Washington, DC: U.S. Census Bureau; 2012: c2010br-11.pdf. Published March Accessed August 1, Siegel RL, Miller KD, Fedewa SA, et al. Colorectal cancer statistics, CA Cancer J Clin. 2017;67: Thompson CA, Gomez SL, Hastings KG, et al. The burden of cancer in Asian Americans: a report of national mortality trends by Asian ethnicity. Cancer Epidemiol Biomarkers Prev. 2016;25: Sentell TL, Tsoh JY, Davis T, Davis J, Braun KL. Low health literacy and cancer screening among Chinese Americans in California: a cross-sectional analysis. BMJ Open. 2015;5:e Tong EK, Nguyen TT, Lo P, et al. Lay health educators increase colorectal cancer screening among Hmong Americans: a cluster randomized controlled trial. Cancer. 2017;123: Jo AM, Nguyen TT, Stewart S, et al. Lay health educators and print materials for the promotion of colorectal cancer screening among Korean Americans: a randomized comparative effectiveness study. Cancer. 2017;123: Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. 1991;50: Rosenstock IM. Health belief model and preventive health behavior. Health Educ Monogr. 1974;2: Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36: Kiviniemi MT, Bennett A, Zaiter M, Marshall JR. Individual-level factors in colorectal cancer screening: a review of the literature on the relation of individual-level health behavior constructs and screening behavior. Psychooncology. 2011;20: California Health Interview Survey. CHIS Methodology Series: Report 2 Data Collection Methods. Los Angeles, CA: UCLA Center for Health Policy Research; Rawl SM, Menon U, Champion VL, et al. Do benefits and barriers differ by stage of adoption for colorectal cancer screening? Health Educ Res. 2005;20: Vernon SW, Myers RE, Tilley BC. Development and validation of an instrument to measure factors related to colorectal cancer screening adherence. Cancer Epidemiol Biomarkers Prev. 1997;6: Kim SB. Unraveling the determinants to colorectal cancer screening among Asian Americans: a systematic literature review [published online ahead of print August 4, 2017]. J Racial Ethn Health Disparities. doi: /s Leung DY, Chow KM, Lo SW, So WK, Chan CW. Contributing factors to colorectal cancer screening among Chinese People: a review of quantitative studies. Int J Environ Res Public Health. 2016; 13: Molina Y, Briant KJ, Sanchez JI, O Connell MA, Thompson B. Knowledge and social engagement change in intention to be screened for colorectal cancer [published online ahead of print January 24, 2017]. Ethn Health. doi: / Maxwell AE, Bastani R, Crespi CM, Danao LL, Cayetano RT. Behavioral mediators of colorectal cancer screening in a randomized controlled intervention trial. Prev. Med. 2011;52: Lee HY, Vang S. Barriers to cancer screening in Hmong Americans: the influence of health care accessibility, culture, and cancer literacy. J. Community Health. 2010;35: Cancer April 1,