Journal List > J Korean Med Sci > v.39(4) > 1516086178

TOOLS
Kim, Hong, Ryu, Lee, Cho, Yu, Lee, and Kim: Factors Affecting Adherence to National Colorectal Cancer Screening: A 12-Year Longitudinal Study Using Multi-Institutional Pooled Data in Korea
Original Article
Public Health & Preventive Medicine

Journal of Korean Medical Science 2024; 39(4): e36.

Published online: 11 January 2024

DOI: https://doi.org/10.3346/jkms.2024.39.e36

Factors Affecting Adherence to National Colorectal Cancer Screening: A 12-Year Longitudinal Study Using Multi-Institutional Pooled Data in Korea

Dae Sung Kim1, *, Jeeyoung Hong2, *, Kihyun Ryu1, Sang Hyuk Lee1, Hwanhyi Cho1, Jehyeong Yu3, Jieun Lee3, Jong-Yeup Kim4

1Division of Gastroenterology, Department of Internal Medicine, Konyang University College of Medicine, Daejeon, Korea.

2Biomedical Research Institute, Konyang University Medical Center, Daejeon, Korea.

3Healthcare Data Science Center, Konyang University Hospital, Daejeon, Korea.

4Department of Biomedical Informatics, Konyang University College of Medicine, Daejeon, Korea.

Address for Correspondence: Jong-Yeup Kim, MD, PhD. Department of Biomedical Informatics, Konyang University College of Medicine, 158 Gwanjeodong-ro, Seo-gu, Daejeon 35365, Republic of Korea. jykim@kyuh.ac.kr

Equal contributor:

*Dae Sung Kim and Jeeyoung Hong contributed equally to this work.

Received 19 June 2023       Accepted 23 November 2023

© 2024 The Korean Academy of Medical Sciences.

The Korean Academy of Medical Sciences

https://creativecommons.org/licenses/by-nc/4.0/
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Consistent uptake of colorectal cancer (CRC) screening is important to reduce the incidence and mortality from advanced-stage CRC and increase the survival rate of the patients. We conducted a longitudinal study to determine the factors affecting CRC screening compliance in Korean adults using individual-level linked data from the Korean National Health and Nutrition Examination Survey, Korean National Health Insurance Service, and Korean Health Insurance Review and Assessment Service.

Methods

We selected 3,464 adults aged 50–79 years as the study population and followed them for 12 years (January 2007–December 2018). The outcome variable was the level of adherence to CRC screening, categorized as nonadherent, intermittently adherent, and consistently adherent. An ordinal logistic regression model was designed to determine the socioeconomic factors, family history of CRC, and medical conditions that could facilitate the consistent uptake of CRC screening.

Results

The results showed a significant and positive association between consistent uptake of CRC screening and the 100–150% income category (odds ratio [OR], 1.710; 95% confidence interval [CI], 1.401–2.088); clerical, sales and service job category (OR, 1.962; 95% CI, 1.582–2.433); residency at medium-sized cities (OR, 1.295; 95% CI, 1.094–1.532); high-school graduation (OR, 1.440; 95% CI, 1.210–1.713); married status (OR, 2.281; 95% CI, 1.946–2.674); use of employment-based national health insurance (OR, 1.820; 95% CI, 1.261–2.626); use of private insurance (OR, 2.259; 95% CI, 1.970–2.589); no disability (OR, 1.428; 95% CI, 1.175–1.737); family history of CRC (OR, 2.027; 95% CI, 1.514–2.714); and history of colorectal neoplasm (OR, 1.216; 95% CI; 1.039–1.422).

Conclusion

The lack of regular participation in CRC screening programs in the Republic of Korea was found to be an issue that requires attention. Policies on CRC screening must place increased emphasis on strengthening educational and public relations initiatives.

Graphical Abstract

jkms-39-e36-abf001.jpg

Keywords: Colorectal Cancer, Cancer Screening, Secondary Prevention, Follow-up Studies

INTRODUCTION

Colorectal cancer (CRC) is the third most common cancer in terms of incidence and the second leading cause of cancer-related deaths worldwide.1 In 2020, the Republic of Korea (hereafter, Korea) had the third-highest CRC incidence and mortality rates in the world, at 54.3 and 17.4 per 100,000 people, respectively.23 In most cases, premalignant or malignant lesions are asymptomatic for a prolonged period; therefore, the stage at diagnosis is critical to the prognosis.4 CRC screening plays a crucial role in detecting precancerous lesions or early-stage CRC,5 and reducing CRC mortality by up to 25%.6
To address the growing burden of cancer, the World Health Organization has recommended the establishment of National Cancer Control Plans that focus on prevention, early detection, diagnosis, treatment, and palliative care.7 In 1999, the Korean government launched the National Cancer Screening Program (NCSP) as a component of the comprehensive 10-year National Cancer Control Plan initiated in 1996.8 As per this program, CRC screening is conducted for individuals aged 50 years and older, primarily via an annual fecal occult blood test (FOBT). The NCSP is a population-based, organized CRC screening program that provides annual FOBTs for adults aged 50 years and older. Since 2004, the NCSP has also offered colonoscopy or a double-contrast barium enema test to individuals who show positive FOBT results.9 Although the CRC screening rate in Korea has increased from 25.7% in 2012 to 40.3% in 2021, it remains the lowest compared with the screening rates for cancers of the stomach, liver, lung, breast, and uterine cervix in 2021.10 This is a concerning issue, given that CRC is the third most common cancer globally and domestically, and early detection through CRC screening is critical for successful treatment and improved outcomes. Another salient issue is whether CRC screening can be sustained over time. Vernon conducted a comprehensive review of several studies on adherence to FOBT and found that it was challenging to achieve an annual compliance rate of over 50% among the general population.11
Understanding the factors that influence adherence to CRC screening over a long-term period is crucial for enhancing the regular uptake rate of CRC screening and improving CRC prognosis. Previous research conducted in Korea on the uptake of CRC screening and its determinants revealed that numerous factors, such as age, income, educational level, residential area, occupation type, marital status, and family history of CRC, have contributed to gradients in the CRC screening rate.1213141516 These studies utilized data from the Korean National Health and Nutrition Examination Survey (KNHANES) and Korea National Cancer Screening Survey, employed a cross-sectional design involving self-reported questionnaires, and captured the rate of CRC screening uptake at a single time point. It is challenging to obtain information on the regular uptake of CRC screening over a long period. The prevalence of benign tumors or precancerous lesions in the colorectal area can be a crucial determinant of the need for CRC screening. However, obtaining this information from cross-sectional survey data is nearly impossible. Therefore, to investigate the various factors influencing the regular uptake of CRC screening over an extended period, we utilized individual-level linked data from three public health databases for a period of 12 years: the Korean Disease Control Agency (KDCA), the Korean National Health Insurance Service (KNHIS), and the Korean Health Insurance Review and Assessment Service (KHIRA).

METHODS

Data source and study population

In 2019, the Korean National Evidence-based Healthcare Collaborating Agency (KNECA) initiated a healthcare data linkage project that integrated individual-level data from multiple sources, including the KDCA, KNHIS, KHIRA, and the Korean National Cancer Center. In our study, we used individual-level data from three sources: 1) KNHANES data from the KDCA, 2) information on CRC screening uptake and insurance from the KNHIS, and 3) claims data from the KHIRA. We received data from the KDCA on 50,404 respondents to the KNHANES from 2007–2012 and linked it to CRC screening and insurance data from the KNHIS and claims data from the KHIRA at the individual level.17 Each participant with malignant neoplasms (Korean Standard Classification of Diseases [KCD] codes C00–C96) from the KHIRA database between 2007 and 2012 was selected by KNECA and matched by sex and age to 10 KNHANES participants without malignant neoplasms. A total of 17,637 of 50,404 participants in the 2007–2012 KNHANES were included: cancer group (n = 1,603) and non-cancer group (n = 16,034). A total of 3,464 from 17,637 participants in the 2007–2012 KNHANES aged 50–79 years were selected as the study population and included in the final analysis. Three exclusion criteria were applied: 1) age < 50 or ≥ 80 years at index-date (January 1, 2007) (n = 13,616), 2) death between 2007 and 2012 (n = 323), and 3) diagnosed with CRC (KCD codes C18–C20) between 2007 and 2012 (n = 234). The process of data linking and sample selection is shown in Fig. 1.
Fig. 1

Flow diagram of the recruitment process of the study participants.

KNHANES = Korean National Health and Nutrition Examination Survey, KNHIS = Korean National Health Insurance Service, KHIRA = Korean Health Insurance Review and Assessment Service.
jkms-39-e36-g001
The KNHANES contains participants’ unique socioeconomic information, such as household income, education, marital status, and private health insurance status, that is impossible to obtain from the KNHIS and KHIRA. The KNHIS provides unique and important information about participants having undergone a public CRC screening service, treated as an outcome variable, disability condition, and family history of CRC. The KHIRA records data on factors associated with CRC screening such as inflammatory bowel diseases (IBDs), e.g., ulcerative colitis and Crohn’s disease, benign neoplasms of the colorectal area, gastrointestinal bleeding, and changes in bowel habits. The process for deriving predictors and outcome variables from linked data sources including the KNHANES, the KNHIS, and the KHIRA, was elaborated in Supplementary Table 1.

Measures of main interest

Outcome variable

The outcome variable was the level of adherence to CRC screening, categorized into three ordinal groups: nonadherent, intermittently adherent, and consistently adherent. Adults who underwent CRC screening were operationally defined as those included in the 2007–2018 CRC screening data table obtained from the KNHIS database. If either the FOBT or colonoscopy was performed, CRC screening was considered complete for that year. The follow-up period for this study commenced on January 1, 2007, and concluded on December 31, 2018, with censoring performed for deaths and new CRC diagnoses. To account for the biennial uptake of CRC screening in Korea that was converted to annual uptake by the Korean Ministry of Health and Welfare in 2012,18 we tracked CRC screening data throughout the follow-up period. Specifically, we ensured that participants born in odd-numbered years underwent CRC screening in 9 years (i.e., 2007, 2009, 2011, 2013, 2014, 2015, 2016, 2017, and 2018), and those born in even-numbered years underwent screening in 9 years (i.e., 2008, 2010, 2012, 2013, 2014, 2015, 2016, 2017, and 2018). Since we excluded participants from the KNHANES who died between 2007 and 2012, we were able to guarantee the assessment of at least three CRC screenings. The participants were then categorized into three adherence groups based on their CRC screening history: nonadherent (no prior CRC screening), consistently adherent (participation in at least half of the CRC screening assessments), and intermittently adherent (did not meet the criteria for nonadherence or consistent adherence).

Medical conditions of affecting adherence to CRC screening

We generated four variables, namely gastrointestinal bleeding, changes in bowel habits, neoplasm of the colorectal area, and IBD, using claims data from the KHIRA. The four variables were based on the following KCD codes for which the patients required medical services: 1) gastrointestinal bleeding (KCD codes K55.21, K57.31, K57.53, K57.91, K62.5, K92.1, and K92.2), 2) changes in bowel habits (KCD code R19.4), 3) neoplasm of the colorectal area (carcinoma in situ of lower intestine with KCD codes D01.0, D01.1, D01.2, D01.3, and D01.4, benign neoplasm of colorectal area with KCD codes D12, D12.0, D12.1, D12.2, D12.3, D12.4, D12.5, D12.6, D12.7, D12.8, and D12.9), and 4) IBD (Crohn’s disease with KCD codes K50, K50.0, K50.1, K50.8, and K50.9 and ulcerative colitis with KCD codes K51, K51.0, K51.2, K51.3, K51.4, K51.5, K51.8, and K51.9).

Demographic data of the study sample

The KNHANES encompasses demographic data including sex, age, place of residence, and marital status. Age, as of January 1, 2007, was categorized into three groups: 50–59, 60–69, and 70–79 years. We compared the adherence to CRC screening in three distinct regions: metropolitan cities, small-to-medium-sized cities, and rural areas. The relative poverty line was used as an indicator of household income, defined as 50% of the median household income by household size provided annually by the Korea Statistics Office.19 Household income by household size was categorized in the KNHANES data as less than 50% (below the relative poverty line), 50% to < 100%, 100% to < 150%, 150% to < 200%, and ≥ 200%. Marital status was classified as unmarried, divorced (including separated), or married. Based on the International Standard Classification of Education,20 education levels were classified into the following categories: elementary school graduates or less, middle school graduates, high school graduates, and college graduates or higher.
The occupational categories in the KNHANES dataset were classified according to the Korean standard classification of occupations21 and grouped into three categories: manual workers (service workers; sales workers; skilled agricultural, forestry, and fishery workers; craft and related trade workers; equipment, machine operating, and assembling workers; and unskilled workers), non-manual workers (managers, professionals, and clerks), and others (armed forces, housewives, and students). Furthermore, the KNHANES dataset comprises information on current employment status (employed or not), shift work (performed or not), employee status (self-employed or employed by others), occupation position (regular, temporary, or daily), and working hours. Following the classification used in previous studies,222324 we categorized occupational classes into six types: high-level professionals and managers; low-level professionals; clerical, sales, and service workers; manual workers; self-employed entrepreneurs and farmers; and individuals who have never worked or fall into other categories. The KNHIS compiles disability registry data indicating the presence or severity of physical or mental disabilities based on the Korean Disabled Persons Welfare Act.25 The study population was categorized into disabled and non-disabled individuals. The National Health Insurance (NHI) in Korea provides coverage to approximately 97% of the country’s total population, whereas Medicaid is available for those in the lowest 3% income range. Health insurance types were categorized as Medicaid, self-employed NHI, or employment-based NHI. We extracted information on whether individuals received services from local public health organizations between January 1, 2007, and the endpoint for the KHIRA data. Local public health organizations include public health clinics, centers, and units and mother-and-child health organizations.26 The medical history of patients with any type of malignant cancer (KCD codes C00–C96), except for CRC (KCD codes C18–C20), was obtained from the KCD list of the KHIRA as well as from the self-reported cancer survey data in both the KNHANES and the KNHIS cancer screening data tables. Additionally, the family history of patients who underwent CRC was extracted from self-reported information in the KNHIS cancer screening data table.

Statistical analysis

Statistical analyses were performed using the R 4.0.0 software (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at P < 0.05. significant. All variables in this study were categorical variables and presented as frequencies and percentages in Table 1. The differences in predictor distributions among the three levels of adherence to CRC screening were examined by univariate analysis with χ2 test (Table 2). As the outcome variable of this study, adherence to CRC screening, was ordinal in nature, ordinary logistic regression27 was used to estimate the adjusted effect of predictors on compliance with CRC screening (Table 3).
Table 1

Distribution of variables in the study population by age groups

jkms-39-e36-i001
Variables No. Age group, yr, No. (%) P value
50–59 60–69 70–79
Sex 0.065
Men 1,576 748 (47.5) 556 (35.3) 272 (17.3)
Women 1,888 844 (44.7) 663 (35.1) 381 (20.2)
Household income/median income < 0.001
< 50% 1,475 382 (25.9) 650 (44.1) 443 (30.0)
50 to < 100% 1,010 538 (53.3) 347 (34.4) 125 (12.4)
100 to < 150% 513 337 (65.7) 125 (24.4) 51 (9.9)
150 to < 200% 239 163 (68.2) 52 (21.8) 24 (10.0)
≥ 200% 227 172 (75.8) 43 (18.9) 12 (5.3)
Occupational class < 0.001
Never worked and others 1,949 688 (35.3) 744 (38.2) 517 (26.5)
Self-employed entrepreneurs and farmers 212 102 (48.1) 82 (38.7) 28 (13.2)
Manual workers 371 117 (31.5) 169 (45.6) 85 (22.9)
Clerical, sales, and service workers 398 293 (73.6) 92 (23.1) 13 (3.3)
Low-level professionals 360 256 (71.1) 97 (26.9) 7 (1.9)
High-level professionals and managers 174 136 (78.2) 33 (19.0) 5 (2.9)
Place of residence < 0.001
Rural area 996 380 (38.2) 364 (36.5) 252 (25.3)
Small-medium cities 1,140 569 (49.9) 388 (34.0) 183 (16.1)
Metropolitan cities 1,328 643 (48.4) 466 (35.1) 219 (16.5)
Education level < 0.001
Elementary school graduate or less 1,744 567 (32.5) 701 (40.2) 476 (27.3)
Middle school graduate 540 306 (56.7) 175 (32.4) 59 (10.9)
High school graduate 706 453 (64.2) 195 (27.6) 58 (8.2)
College graduate or higher 474 266 (56.1) 147 (31.0) 61 (12.9)
Marital status < 0.001
Unmarried, separated or divorced 798 199 (24.9) 287 (36.0) 312 (39.1)
Married 2,666 1,394 (52.3) 931 (34.9) 341 (12.8)
Type of public health insurance < 0.001
Medicaid 121 22 (18.2) 53 (43.8) 46 (38.0)
National health insurance, self-employed 1,212 617 (50.9) 381 (31.4) 214 (17.7)
National health insurance, employed 2,131 953 (44.7) 784 (36.8) 394 (18.5)
Private health insurance < 0.001
Not having 1,780 446 (25.1) 746 (41.9) 588 (33.1)
Having 1,684 1,146 (68.1) 473 (28.1) 65 (3.9)
Disabled participants < 0.001
Non-disabled 3,010 1,454 (48.3) 1,023 (34.0) 533 (17.7)
Disabled 454 137 (30.2) 196 (43.2) 121 (26.7)
Services through local public health organization < 0.001
Not having 3,167 1,539 (48.6) 1,085 (34.3) 543 (17.2)
Having 297 54 (18.2) 133 (44.8) 110 (37.0)
History of all cancer except colorectal cancer < 0.001
Not having 2,605 1,305 (50.1) 870 (33.4) 430 (16.5)
Having 859 288 (33.5) 347 (40.4) 224 (26.1)
Family history of colorectal cancer < 0.001
Not having 3,284 1,494 (45.5) 1,149 (35.0) 641 (19.5)
Having 180 98 (54.4) 67 (37.2) 15 (8.3)
Having gastrointestinal bleeding < 0.001
Not having 3,091 1,459 (47.2) 1,063 (34.4) 569 (18.4)
Having 373 127 (34.0) 161 (43.2) 85 (22.8)
Having bowel habit change < 0.001
Not having 2,295 1,169 (50.9) 747 (32.5) 379 (16.5)
Having 1,169 423 (36.2) 471 (40.3) 275 (23.5)
Having neoplasm of the colorectal area < 0.001
Not having 2,677 1,304 (48.7) 878 (32.8) 495 (18.5)
Having 787 289 (36.7) 340 (43.2) 158 (20.1)
Having inflammatory bowel diseases < 0.001
Not having 3,319 1,550 (46.7) 1,162 (35.0) 607 (18.3)
Having 145 42 (29.0) 58 (40.0) 45 (31.0)
Table 2

Adherence level with colorectal cancer screening between 2007–2018

jkms-39-e36-i002
Variables No. Adherence level, No. (%) P value
None Intermittent Consistent
Sex 0.066
Men 1,576 362 (23.0) 922 (58.5) 292 (18.5)
Women 1,888 434 (23.0) 1,159 (61.4) 295 (15.6)
Age at year 2007, yr < 0.001
50–59 1,592 255 (16.0) 1,032 (64.8) 305 (19.2)
60–69 1,218 220 (18.1) 753 (61.8) 245 (20.0)
70–79 654 322 (49.2) 296 (45.3) 36 (5.5)
Household income/median income < 0.001
< 50% 1,475 403 (27.3) 860 (58.3) 212 (14.4)
50 to < 100% 1,010 216 (21.4) 611 (60.5) 183 (18.1)
100 to < 150% 513 85 (16.6) 322 (62.8) 106 (20.7)
150 to < 200% 239 49 (20.5) 147 (61.5) 43 (18.0)
≥ 200% 227 46 (20.3) 141 (62.1) 40 (17.6)
Occupational class < 0.001
Never worked and others 1,949 513 (26.3) 1,134 (58.2) 302 (15.5)
Self-employed entrepreneurs and farmers 212 39 (18.4) 131 (61.8) 42 (19.8)
Manual workers 371 90 (24.3) 231 (62.3) 50 (13.5)
Clerical, sales and service workers 398 58 (14.6) 241 (60.6) 99 (24.9)
Low-level professionals 360 70 (19.4) 229 (63.6) 61 (16.9)
High-level professionals and managers 174 28 (16.1) 115 (66.1) 31 (17.8)
Place of residence 0.013
Rural area 996 255 (25.6) 602 (60.4) 139 (14.0)
Small-medium cities 1,140 254 (22.3) 670 (58.8) 216 (18.9)
Metropolitan cities 1,328 290 (21.8) 809 (60.9) 229 (17.3)
Education level < 0.001
Elementary school graduate or less 1,744 446 (25.6) 1,036 (59.4) 262 (15.0)
Middle school graduate 540 103 (19.1) 331 (61.3) 106 (19.6)
High school graduate 706 133 (18.8) 434 (61.5) 139 (19.7)
College graduate or higher 474 115 (24.3) 280 (59.1) 79 (16.7)
Marital status < 0.001
Unmarried, separated or divorced 798 290 (36.3) 423 (53.0) 85 (10.7)
Married 2,666 508 (19.1) 1,657 (62.2) 501 (18.8)
Type of public health insurance 0.002
Medicaid 121 45 (37.2) 58 (47.9) 18 (14.9)
National health insurance, self-employed 1,212 291 (24.0) 721 (59.5) 200 (16.5)
National health insurance, employed 2,131 462 (21.7) 1,302 (61.1) 367 (17.2)
Private health insurance < 0.001
Not having 1,780 546 (30.7) 1,013 (56.9) 221 (12.4)
Having 1,684 251 (14.9) 1,069 (63.5) 364 (21.6)
Disabled person 0.001
Non-disabled 3,010 664 (22.1) 1,823 (60.6) 523 (17.4)
Disabled 454 134 (29.5) 258 (56.8) 62 (13.7)
Services through local public health organization 0.370
Not having 3,167 725 (22.9) 1,900 (60.0) 542 (17.1)
Having 297 74 (24.9) 181 (60.9) 42 (14.1)
History of all cancer except colorectal cancer 0.748
Not having 2,605 607 (23.3) 1,558 (59.8) 440 (16.9)
Having 859 190 (22.1) 524 (61.0) 145 (16.9)
Family history of colorectal cancer < 0.001
Not having 3,284 782 (23.8) 1,964 (59.8) 538 (16.4)
Having 180 18 (10.0) 116 (64.4) 46 (25.6)
Having gastrointestinal hemorrhage 0.455
Not having 3,091 720 (23.3) 1,845 (59.7) 526 (17.0)
Having 373 78 (20.9) 235 (63.0) 60 (16.1)
Having bowel habit change 0.419
Not having 2,295 537 (23.4) 1,361 (59.3) 397 (17.3)
Having 1,169 261 (22.3) 720 (61.6) 188 (16.1)
Having neoplasm of the colorectal area 0.023
Not having 2,677 645 (24.1) 1,590 (59.4) 442 (16.5)
Having 787 153 (19.4) 492 (62.5) 142 (18.0)
Having inflammatory bowel diseases 0.055
Not having 3,319 772 (23.3) 1,980 (59.7) 567 (17.1)
Having 145 25 (17.2) 101 (69.7) 19 (13.1)
Table 3

Adherence with colorectal cancer screening based on multivariate ordinal logistic regression

jkms-39-e36-i003
Variables Odd ratio (95% confidence interval)
Women (vs. Men) 0.916 (0.802–1.046)
Age at year 2007, yr (vs. 50–59)
60–69 0.961 (0.826–1.117)
70–79 0.211 (0.175–0.254)
Household income/median income (vs. < 50% [the poorest])
50 to < 100% 1.356 (1.156–1.590)
100 to < 150% 1.710 (1.401–2.088)
150 to < 200% 1.390 (1.060–1.823)
≥ 200% 1.384 (1.050–1.826)
Occupational class (vs. never worked and others)
Self-employed entrepreneurs and farmers 1.481 (1.117–1.962)
Manual workers 1.010 (0.812–1.257)
Clerical, sales and service workers 1.962 (1.582–2.433)
Low-level professionals 1.309 (1.048–1.635)
High-level professionals and managers 1.493 (1.101–2.024)
Place of residence (vs. rural)
Small-medium cities 1.295 (1.094–1.532)
Metropolitan cities 1.252 (1.065–1.473)
Education level (vs. elementary school graduate or less)
Middle school graduate 1.427 (1.178–1.728)
High school graduate 1.440 (1.210–1.713)
College graduate or higher 1.101 (0.900–1.346)
Marital status (vs. unmarried, separated or divorced) 2.281 (1.946–2.674)
Type of public health insurance (vs. medicaid)
National health insurance, self-employed 1.650 (1.136–2.398)
National health insurance, employed 1.820 (1.261–2.626)
Private health insurance (vs. not having) 2.259 (1.970–2.589)
Disability (vs. disabled) 1.428 (1.175–1.737)
Services through local public health organization (vs. not having) 0.856 (0.677–1.081)
History of all cancer except colorectal cancer (vs. not having) 1.040 (0.893–1.211)
Family history of colorectal cancer (vs. not having) 2.027 (1.514–2.714)
Having gastrointestinal hemorrhage (vs. not having) 1.051 (0.851–1.298)
Having bowel habit change (vs. not having) 0.998 (0.869–1.147)
Having neoplasm of colon or rectum (vs. not having) 1.216 (1.039–1.422)
Having inflammatory bowel diseases (vs. not having) 1.077 (0.781–1.483)

Ethics statement

The study protocol was approved by Konyang University’s Institutional Review Board (IRB) (IRB approval No. 2019-10-022) and the data linkage protocol was approved by the IRB of the KNECA (IRB approval No. NECAIRB20-016-1). The requirement for informed consent was waived by both IRBs because personal information that could identify the individuals was not used in this study.

RESULTS

Table 1 shows the distribution of predictors by age group, categorized as 50–59 years, 60–69 years, and 70–79 years. Among the participants, those aged 50–59 years, the youngest group, had the highest distribution in numerous socioeconomic categories including the highest household income (75.8%), highest occupational class (78.2%), residing in large cities (48.4%), having completed higher education (56.1%), being married (52.3%), having self-employed NHI (50.9%) and employed NHI (44.7%), having private insurance (68.1%), and having a family history of CRC (54.4%) (P < 0.001). The participants aged 60–69 years displayed the highest distribution in the following six categories: receiving services through local public health organizations (44.8%), having a medical history of any type of cancer except CRC (40.4%), being diagnosed with gastrointestinal hemorrhage (43.2%), bowel habit changes (40.3%), neoplasm of the colorectal area (43.2%), and IBD (40.0%).
Table 2 presents the distribution of adherence to CRC screening according to the predictors. All predictors, except sex, experience of visiting a public health center, having a medical history of any type of cancer except CRC, being diagnosed with gastrointestinal bleeding, bowel habit changes, and IBD, were found to be statistically significant (P < 0.05). Most predictor categories were observed in the intermittently adherent group. The following groups showed the lowest distribution of and statistical significance for consistent adherence to CRC screening: age group of 70–79 years (5.5%), lowest income (14.4%), manual workers (13.5%), living in rural area (14.0%), having completed the lowest education level (15.0%), unmarried or separated/divorced (10.7%), Medicaid (14.9%), no private health insurance (12.4%), disabled (13.7%), no family history of CRC (16.4%), and no medical history of colorectal neoplasm (16.5%).
Table 3 shows the results of the multivariate ordinal logistic regression analysis of CRC screening adherence according to the independent variables. The odds of regular screening were the lowest in the 70–79 age group (odds ratio [OR], 0.211; 95% confidence interval [CI], 0.175–0.254). The odds of adherence to regular CRC screening were the highest among those in the 100–150% income group (OR, 1.710; 95% CI, 1.401–2.088); clerical, sales and service workers (OR, 1.962; 95% CI, 1.582–2.433); those residing in medium-sized cities (OR, 1.295; 95% CI, 1.094–1.532); high-school graduates (OR, 1.440; 95% CI, 1.210–1.713), married participants (OR, 2.281; 95% CI, 1.946–2.674); participants with employment-based NHI (OR, 1.820; 95% CI, 1.261–2.626); privately insured participants (OR, 2.259; 95% CI, 1.970–2.589); non-disabled participants (OR, 1.428; 95% CI, 1.175–1.737); participants with a family history of CRC (OR, 2.027; 95% CI, 1.514–2.714); and participants with a colorectal neoplasm (OR, 1.216; 95% CI, 1.039–1.422).

DISCUSSION

Only 16.9% of the participants showed consistent adherence to CRC screening programs during the study (2007–2018). Adherence to cancer screening programs is an important public health issue. We considered that it is important to assess adherence to CRC screening not only at a single point in time, but also over a long period of time. Cross-sectional annual screening rates for CRC in Korea are low compared to other cancers and compared to other countries. In this study, we found that long-term adherence CRC screening is also significantly low. Since the inclusion of CRC screening in the NCSP in 2004, there has been an annual increase in participation rate of 3.3%. In 2017, the participation rate for CRC screening exceeded 40%.10 On the other hand, participation rates for screening for all other cancer types, except CRC, surpassed 40% in 2011.9 Despite a steady increase in CRC screening rates in Korea, these rates remain lower than those observed in other countries. In the United States, the CRC screening rate for individuals over 50 years of age was approximately 63% in 2015, and 60.3% of the participants were screened by endoscopy.28 Several European countries have implemented population-based organized CRC screening programs, with screening rates for fecal immunochemical tests exceeding 50% in countries such as the United Kingdom (55.4%), Finland (66.6%), the Netherlands (71.3%), Portugal (62.8%), and Sweden (60.2%).2930
Despite the dearth of evidence from long-term observational studies regarding routine participation rates in CRC screening, there are a few possible explanations for the < 50% CRC screening participation rates reported annually by the Korean government and < 20% distribution in the category of consistent adherence to CRC screening in this study. The first explanation is the high accessibility of colonoscopy. The NCSP does not offer endoscopy as the primary screening modality, allowing individuals to opt out of the NCSP and undergo private screening with colonoscopy at their own expense. Colonoscopy is a highly accessible and cost-effective option for CRC screening in Korea as many experienced endoscopists perform colonoscopies. This makes it relatively easy for individuals to undergo colonoscopy for CRC screening, provided that they are willing to cover the expenses.31 Another possible explanation is the lack of continuity of care. The strength of the primary care system in Korea has been reported to be the weakest among the Organization for Economic Cooperation and Development countries.32 The weakness of the primary care system can be partially attributed to the referral system lacking a gatekeeping function.33 Lasser et al.34 reported that general practitioners and family doctors significantly increased CRC screening rates through a gatekeeping function. Korea’s NCSP has an open certification system for cancer screening units without a mandatory gatekeeping function, meaning that many types of medical facilities, including primary care facilities, hospitals, and screening facilities, can be certified. Additionally, invitees from the NCSP for CRC screening can visit any certified screening unit. Therefore, the invitees experience a lack of continuity in CRC screening due to the lack of a close relationship with the physician at the screening unit, and this could negatively affect the rates of regular participation in CRC screening. The third possible explanation stems from the potential inconveniences associated with FOBT. Suh et al.9 suggested that the low participation rate in CRC screening could be attributed to the process of screening via FOBT, which is used as the primary screening test in the Korean NCSP. Individuals invited for CRC screening are required to collect stool samples at home and then visit a screening unit to submit the samples. This procedure typically involves two separate visits to the screening unit: the first to pick up the container, and the second to submit the sample. This process may present a challenge for sustained participation in CRC screening among the participants of this study.
As shown in Tables 2 and 3, this study provides evidence of socioeconomic inequalities in consistent adherence to CRC screening in Korea, and various socioeconomic factors have been identified as barriers to or facilitators of consistent participation in CRC screening. The lowest rates of consistent adherence to CRC screening were significantly distributed in groups with low socioeconomic status (SES) that included those aged 70–79 years, those residing in a rural area, unmarried participants (including those who were separated/divorced), those who were Medicaid beneficiaries, and those with a disability. A previous study35 explained the low participation rate in CRC screening among individuals of low SES. The study had identified several barriers to CRC screening among underserved populations, including lack of time to visit CRC screening facilities, insufficient knowledge about CRC screening, physical disabilities, and lack of accessibility to transportation services. This study revealed an intriguing finding of a reverse U-shaped pattern in the rates of consistent adherence to CRC screening among predictors with four or more categories, including household income, occupation, and education. The pattern observed in this study indicates that the rates of consistent adherence to CRC screening increased from the most underserved group to the less underserved group, and subsequently decreased up to the least underserved group. Cho et al.36 argued that Korean adults prefer colonoscopy over FOBT as a primary screening method owing to its accuracy and ability to provide therapeutic options and that individuals with high income levels tended to prefer colonoscopy. Moreover, similar to the findings of our study, Cho et al.36 demonstrated that the participation rates in CRC screening were lower among NHI recipients of high-income status than among those with low income status; this could be attributed to the possibility that NHIS recipients of high-income status opted to undergo colonoscopy through a private screening program instead of participating in the NCSP. Considering that income, education, and occupation are highly interrelated socioeconomic indicators, previous studies indicating that individuals with high income levels are more likely to undergo colonoscopies can be extrapolated to education and occupation. Two previous studies1237 have shown a significant positive correlation between private health insurance and regular participation in CRC screening. Individuals enrolled in private health insurance plans that cover cancer care services are more likely to continue CRC screening. Consequently, if patients are diagnosed with CRC, the financial burden of subsequent diagnosis and treatment costs would be reduced. Our study identified family history of CRC as a facilitator of consistent adherence to CRC screening. A previous study38 suggested that individuals without a family history of CRC may perceive themselves as being less susceptible to CRC, which can lead to low participation in CRC screening.
In addition to socioeconomic factors, this study examined several medical conditions that may affect CRC screening. We examined the relationships between CRC screening and gastrointestinal bleeding, changes in bowel habits, colorectal neoplasms, and IBD. Of the four factors, colorectal neoplasms were significantly associated with consistent adherence to CRC screening (OR, 1.216; 95% CI, 1.039–1.422). Colorectal neoplasms, represented by colorectal adenomas, are associated with CRC development. Most CRCs are thought to develop through a stepwise progression known as the adenoma-carcinoma sequence, in which normal colon tissue progresses from colorectal adenoma (a premalignant lesion) to colon cancer.39 After endoscopic resection of a colorectal adenoma, doctors typically explain to patients that adenomas are precancerous lesions and educate them on the importance of early detection and treatment to prevent CRC. Therefore, people who have been treated for colorectal neoplasms will be more likely to undergo cancer screening owing to their increased knowledge of CRC prevention than those who do not receive such treatments. A 2021 nationwide study conducted by the Korean National Cancer Center found that 80.3% of the participants believed that cancer is a preventable disease.40 This study demonstrates a high level of knowledge and interest in cancer among Koreans, and we expect a high level of interest in CRC among Koreans.
Conversely, participants with IBD were less likely to consistently adhere to CRC screening than those without IBD (13.1% vs. 17.1%, P = 0.055), suggesting that patients with IBD tended to be less likely to be screened for CRC, although statistical significance was not demonstrated. Because colonoscopy plays an important role in the differential diagnosis of IBD and is effective in assessing the response to treatment, patients with IBD frequently undergo colonoscopy.4142 In addition, IBD is associated with a high risk of dysplasia or malignancy and requires periodic surveillance colonoscopies.43 The American Society for Gastrointestinal Endoscopy and European Society of Gastrointestinal Endoscopy recommend that surveillance colonoscopy be performed 8–10 years after the onset of IBD.4445 Even in the remission period, surveillance colonoscopy should be performed every 1–5 years, depending on the patient’s individual risk factors, resulting in the need for frequent colonoscopies. Therefore, it can be expected that patients who have already had a colonoscopy for clinical purposes are unlikely to undergo FOBT-based CRC screening. There was no significant difference in the rate of consistent CRC screening between participants with a history of gastrointestinal bleeding or bowel habit changes and those without. Colonoscopy is recommended for diagnosing and treating gastrointestinal bleeding.46 However, periodic follow-up endoscopy is not recommended after full recovery from gastrointestinal bleeding; therefore, it can be assumed that this may not have affected the long-term CRC screening behavior. Changes in bowel habits can be a symptom of CRC; therefore, diagnostic colonoscopy is indicated.47 As with gastrointestinal bleeding, once a colonoscopy is performed to differentiate the cause of a change in bowel habits, follow-up examinations are not usually performed multiple times; therefore, it can be assumed that long-term examination behavior is not affected.
This study has several limitations. First, the findings of this study may not be representative enough to extrapolate to the entire population of Korea. The KCDA conducted the KNHANES with a total of 50,404 participants between 2007 and 2012 and produced representative health outcomes for the entire Korean population. Unfortunately, we received combined KNHIS and KHIRA data for only 17,637 KNHANES participants, accounting for approximately one-third of the 50,404 participants in the 2007–2012 KNHANES. Therefore, the findings of this study cannot be generalized to the entire Korean population. Second, similar to the limitations of other national data-based big data studies, our study was designed based on KCD codes, which leaves a lot of room for physician subjectivity in determining the codes, and therefore may not perfectly reflect the actual situation of the study subjects. This inconsistency has the potential to negatively impact the high reliability of the study. Third, despite the availability of both population-based and opportunistic CRC screening in Korea, the database employed in this study was not capable of capturing opportunistic screenings conducted outside the NCSP that involved undergoing a colonoscopy as the first screening test in CRC screening, skipping the FOBT. Therefore, it is impossible to determine regular participation in the overall CRC screening using data from public and private screening data. Fourth, it was difficult to identify the reasons for nonparticipation in CRC screening using the data obtained in this study. There is a need to explore the influence of psychological factors such as discomfort, concern about complications, and anxiety about the procedure. Fifth, this study is based on the KCD code and is a big data study linking multi-year data from multiple institutions. Therefore, due to the structural characteristics of the study, it was not possible to identify and evaluate in detail the timing of diagnosis and recovery of comorbid medical conditions that may affect CRC screening uptake. Sixth, this study was conducted as part of multi-institutional data linkage project of the KNECA, which involved a data link process and 1:10 matching of cancer patients to non-cancer patients. Therefore, there is the potential for selection bias that occurred before the data was available to the researchers. And, we excluded deaths and subjects diagnosed with CRC during the period 2007–2012. This was done to ensure a minimum of three assessments to assess consistent uptake of CRC screening, but this exclusion could also introduce selection bias.
Despite these limitations, this study is valuable because it examined the level of regular participation in CRC screening and various influencing factors through 12 years of retrospective follow-up observation by linking individual-level data from three public health databases. Additionally, the study examined whether a history of one of the four types of colorectal diseases, in addition to socioeconomic factors, influenced regular participation in CRC screening. Although the NCSP is designed to ensure CRC screening for all Koreans aged 50 years and older, the high accessibility of colonoscopy in Korea limits the ability of studies to identify the overall regular participation rates in CRC screening and its determinants using only NCSP data. Therefore, a longitudinal survey, such as a panel survey, is necessary to produce nationally representative estimates of regular participation in CRC and the influencing factors.

ACKNOWLEDGMENTS

We would like to thank Editage (www.editage.co.kr) for English language editing.

Notes

Funding: This research was supported by a grant from the Korea Health Technology R&D Project of the Korea Health Industry Development Institute (KHIDI) and funded by the Ministry of Health & Welfare, Republic of Korea (Grant No. HI19C1310 and HI22C1518).

Disclosure: The authors have no potential conflicts of interest to disclose.

Author Contributions:

  • Conceptualization: Kim DS, Hong J, Kim JY.

  • Data curation: Kim DS, Hong J, Ryu K, Lee SH, Cho H, Yu J, Lee J, Kim JY.

  • Formal analysis: Hong J, Kim JY.

  • Funding acquisition: Kim JY.

  • Investigation: Kim DS, Hong J, Ryu K, Lee SH, Cho H, Yu J, Lee J.

  • Supervision: Kim JY.

  • Methodology: Kim DS, Hong J, Kim JY.

  • Writing - original draft: Kim DS, Hong J.

  • Writing - review & editing: Kim DS, Hong J, Kim JY.

References

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021; 71(3):209–249. PMID: 33538338.
2. Korean National Cancer Center. Annual report of cancer registry in 2020. Updated 2022. Accessed January 28, 2023. https://ncc.re.kr/cancerStatsView.ncc?bbsnum=618&searchKey=total&searchValue=&pageNum=1 .
3. Statistics Korea. Deaths and death rates by cause (104 item)/by sex/by age (five-year age). Updated 2023. Accessed January 18, 2023. https://kosis.kr/statHtml/statHtml.do?orgId=350&tblId=DT_35007_N009&vw_cd=MT_ETITLE&list_id=350_35007_A002&scrId=&seqNo=&language=en&obj_var_id=&itm_id=&conn_path=A6&path=%252Feng%252Fsearch%252FsearchList.do .
4. Maringe C, Walters S, Rachet B, Butler J, Fields T, Finan P, et al. Stage at diagnosis and colorectal cancer survival in six high-income countries: a population-based study of patients diagnosed during 2000-2007. Acta Oncol. 2013; 52(5):919–932. PMID: 23581611.
5. Kubisch CH, Crispin A, Mansmann U, Göke B, Kolligs FT. Screening for colorectal cancer is associated with lower disease stage: a population-based study. Clin Gastroenterol Hepatol. 2016; 14(11):1612–1618.e3. PMID: 27085763.
6. Hewitson P, Glasziou P, Watson E, Towler B, Irwig L. Cochrane systematic review of colorectal cancer screening using the fecal occult blood test (hemoccult): an update. Am J Gastroenterol. 2008; 103(6):1541–1549. PMID: 18479499.
7. Stewart BW, Wild CP. World Cancer Report 2014. Lyon, France: International Agency for Research on Cancer;2014.
8. Yoo KY. Cancer control activities in the Republic of Korea. Jpn J Clin Oncol. 2008; 38(5):327–333. PMID: 18407932.
9. Suh M, Song S, Cho HN, Park B, Jun JK, Choi E, et al. Trends in participation rates for the National Cancer Screening Program in Korea, 2002-2012. Cancer Res Treat. 2017; 49(3):798–806. PMID: 27857022.
10. Statistics Korea. The National uptake rate of cancer screening. Updated 2023. Accessed January 18, 2023. https://www.index.go.kr/unity/potal/main/EachDtlPageDetail.do?idx_cd=1440 .
11. Vernon SW. Participation in colorectal cancer screening: a review. J Natl Cancer Inst. 1997; 89(19):1406–1422. PMID: 9326910.
12. Suh M, Choi KS, Lee HY, Hahm MI, Lee YY, Jun JK, et al. Socioeconomic disparities in colorectal cancer screening in Korea: a nationwide cross-sectional study. Medicine (Baltimore). 2015; 94(39):e1368. PMID: 26426605.
13. Mai TT, Lee YY, Suh M, Choi E, Lee EY, Ki M, et al. Socioeconomic inequalities in colorectal cancer screening in Korea, 2005-2015: after the introduction of the National Cancer Screening Program. Yonsei Med J. 2018; 59(9):1034–1040. PMID: 30328317.
14. Kim S, Kwon S, Subramanian SV. Has the National Cancer Screening Program reduced income inequalities in screening attendance in South Korea? Cancer Causes Control. 2015; 26(11):1617–1625. PMID: 26335263.
15. Rajaguru V, Kim TH, Shin J, Lee SG. Income disparities in cancer screening: a cross-sectional study of the Korean National Health and Nutrition Examination Survey, 2013-2019. Front Public Health. 2022; 10:820643. PMID: 35372248.
16. Bui NC, Cho HN, Lee YY, Suh M, Park B, Jun JK, et al. Stages of adoption for fecal occult blood test and colonoscopy tests for colorectal cancer screening in Korea. Cancer Res Treat. 2018; 50(2):416–427. PMID: 28494531.
17. Korea Disease Control Agency. Raw data usage manual of the 5th Korea National Health and Nutrition Examination Survey. Updated 2023. Accessed January 20, 2023. https://knhanes.kdca.go.kr/knhanes/sub03/sub03_02_05.do .
18. Ministry of Health & Welfare (KR). Briefing of the National Cancer Screening Program. Updated 2012. Accessed January 24, 2023. http://www.mohw.go.kr/react/jb/sjb030301vw.jsp?PAR_MENU_ID=03&MENU_ID=032901&CONT_SEQ=336966 .
19. Statistics Korea. Monthly average household income according to household size. Updated 2023. Accessed January 3, 2023. https://kosis.kr/statHtml/statHtml.do?orgId=101&tblId=DT_1L9E028 .
20. Organization for Economic Cooperation and Development. Classifying educational programmes: manual for ISCED-97 implementation in OECD countries. Updated 1999. Accessed January 20, 2023. https://www.oecd.org/education/1841854.pdf .
21. Statistics Korea. Korean Standard Classification of Occupations. Updated 2023. Accessed January 20, 2023. http://kssc.kostat.go.kr .
22. Kim M, Chung W, Lim S, Yoon S, Lee J, Kim E, et al. Socioeconomic inequity in self-rated health status and contribution of health behavioral factors in Korea. J Prev Med Public Health. 2010; 43(1):50–61. PMID: 20185983.
23. Yoon TH, Moon OR, Lee SY, Jeong BG, Lee SJ, Kim NS, et al. Differences in health behaviors among the social strata in Korea. Korean J Prev Med. 2000; 33(4):469–476.
24. Aranda E, Franck JE, Ringa V, Sassenou J, Coeuret-Pellicer M, Rigal L, et al. Social inequalities in participation in cancer screening: does the mode of data collection matter? The CONSTANCES cohort. Eur J Public Health. 2021; 31(3):602–608. PMID: 34233352.
25. Korea Ministry of Government Legislation. Act on Welfare of Persons with Disabilities. Updated 2023. Accessed January 22, 2023. https://elaw.klri.re.kr/kor_service/lawView.do?hseq=57347&lang=ENG .
26. Korea Ministry of Government Legislation. Regional Public Health Act. Updated 2023. Accessed January 22, 2023. https://elaw.klri.re.kr/kor_service/lawView.do?hseq=49575&lang=ENG .
27. Wurm MJ, Rathouz PJ, Hanlon BM. Regularized ordinal regression and the ordinalNet R Package. J Stat Softw. 2021; 99(6):1–42.
28. American Cancer Society. Colorectal cancer facts & figures 2017-2019. Updated 2017. Accessed January 27, 2023. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2017-2019.pdf .
29. Segnan N, Patnick J, Von Karsa L. European Guidelines for Quality Assurance in Colorectal Cancer Screening and Diagnosis. Luxembourg: Publications Office of the European Union;2010.
30. International Agency for Research on Cancer. Colorectal cancer screening: IARC handbooks of cancer prevention, volume 17. Updated 2019. Accessed February 2, 2023. https://publications.iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Colorectal-Cancer-Screening-2019 .
31. Choi JH, Cha JM, Yoon JY, Kwak MS, Jeon JW, Shin HP. The current capacity and quality of colonoscopy in Korea. Intest Res. 2019; 17(1):119–126. PMID: 30301340.
32. Ahn SH. Assessment of primary care level in Korea and comparison with the developed countries. J Korean Acad Fam Med. 2001; 22(4):483–497.
33. Organization for Economic Cooperation and Development. OECD Health Care Quality Review: Korea. Paris, France: Organization for Economic Cooperation and Development Publishing;2012.
34. Lasser KE, Ayanian JZ, Fletcher RH, Good MJ. Barriers to colorectal cancer screening in community health centers: a qualitative study. BMC Fam Pract. 2008; 9(1):15. PMID: 18304342.
35. Lee YY, Jun JK, Suh M, Park BY, Kim Y, Choi KS. Barriers to cancer screening among medical aid program recipients in the Republic of Korea: a qualitative study. Asian Pac J Cancer Prev. 2014; 15(2):589–594. PMID: 24568462.
36. Cho YH, Kim DH, Cha JM, Jeen YT, Moon JS, Kim JO, et al. Patients’ preferences for primary colorectal cancer screening: a survey of the National Colorectal Cancer Screening Program in Korea. Gut Liver. 2017; 11(6):821–827. PMID: 28750489.
37. Asgary R, Garland V, Jakubowski A, Sckell B. Colorectal cancer screening among the homeless population of New York City shelter-based clinics. Am J Public Health. 2014; 104(7):1307–1313. PMID: 24832144.
38. Jones RM, Devers KJ, Kuzel AJ, Woolf SH. Patient-reported barriers to colorectal cancer screening: a mixed-methods analysis. Am J Prev Med. 2010; 38(5):508–516. PMID: 20409499.
39. Staton CA, Chetwood AS, Cameron IC, Cross SS, Brown NJ, Reed MW. The angiogenic switch occurs at the adenoma stage of the adenoma carcinoma sequence in colorectal cancer. Gut. 2007; 56(10):1426–1432. PMID: 17566019.
40. Oh JK, Park E, Kim B, Choi YJ, Yun EH, Lim MK, et al. Awareness of and practice toward cancer prevention recommendations: results of the Korean National Cancer Prevention Awareness and Practice Survey in 2021. Epidemiol Health. 2022; 44:e2022068. PMID: 36049500.
41. Spiceland CM, Lodhia N. Endoscopy in inflammatory bowel disease: role in diagnosis, management, and treatment. World J Gastroenterol. 2018; 24(35):4014–4020. PMID: 30254405.
42. Jung SA. Differential diagnosis of inflammatory bowel disease: what is the role of colonoscopy? Clin Endosc. 2012; 45(3):254–262. PMID: 22977813.
43. Kim KO, Chiorean MV. Advanced neoplasia detection using chromoendoscopy and white light colonoscopy for surveillance in patients with inflammatory bowel disease. Intest Res. 2020; 18(4):438–446. PMID: 33131232.
44. Murthy SK, Feuerstein JD, Nguyen GC, Velayos FS. AGA clinical practice update on endoscopic surveillance and management of colorectal dysplasia in inflammatory bowel diseases: expert review. Gastroenterology. 2021; 161(3):1043–1051.e4. PMID: 34416977.
45. Dekker E, Nass KJ, Iacucci M, Murino A, Sabino J, Bugajski M, et al. Performance measures for colonoscopy in inflammatory bowel disease patients: European Society of Gastrointestinal Endoscopy (ESGE) Quality Improvement Initiative. Endoscopy. 2022; 54(9):904–915. PMID: 35913069.
46. Triantafyllou K, Gkolfakis P, Gralnek IM, Oakland K, Manes G, Radaelli F, et al. Diagnosis and management of acute lower gastrointestinal bleeding: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2021; 53(8):850–868. PMID: 34062566.
47. Herrod P, Boyd-Carson H, Doleman B, Blackwell J, Hardy E, Harper F, et al. Safe investigation of isolated change in bowel habit with a flexible sigmoidoscopy? A systematic review and meta-analysis. Ann R Coll Surg Engl. 2019; 101(6):379–386. PMID: 30855983.

SUPPLEMENTARY MATERIAL

Supplementary Table 1

Predictors and outcome extracted from linked data of the KNHANES, the KNHIS and the KHIRA
jkms-39-e36-s001.doc
TOOLS
Similar articles