Worldwide, colorectal cancer (CRC) is the third most commonly diagnosed cancer in men and second in women, with an estimated 1.4 million cases and 693900 deaths occurring in 2012.1 While CRC incidence rates are highest in Western countries, such as Australia/New Zealand, Europe, and Northern America,1 CRC incidence rates have recently been increasing in regions historically at low risk of CRC.2 In Korea, CRC incidence rates have risen markedly over the past few years, with 5.0% and 3.7% increases in annual percentage changes (APC) in men and women, respectively, from 1999 to 2013.3 In 2014, CRC had the third highest incidence rate among all major cancers in both genders.3

In general, socioeconomic status (SES) has been shown to account for increases in the prevalence of various diseases, especially non-communicable disease.45 Regarding cancer, previous studies have indicated that cancer deaths are more common among individuals of low SES.678 However, these individuals are less likely to undergo screening. In developed countries, including Korea, Japan, and the United States, individuals of lower SES have commonly been found to undergo less screening than those of high SES.910111213

In Korea, the National Cancer Screening Program (NCSP) was implemented by the Ministry of Health and Welfare to provide free screening services for major cancers, starting with gastric and breast cancer in 1999.14 Nationwide CRC screening was first initiated in 2004 as part of the NCSP for low-income groups. Due to a limited capacity to perform colonoscopy screening for every age eligible participant, the NCSP offers a fecal occult blood test (FOBT) annually as the initial mass screening method for men and women ≥50 years.15 For individuals with a positive FOBT result, the NCSP covers further investigation with a colonoscopy or double-contrast barium enema. In 2004, Medical Aid recipients and National Health Insurance (NHI) beneficiaries within the lower 30% income bracket were eligible for free-of-charge CRC screening services under the NCSP. In 2005, the target population was expanded to the 50% bracket. Apart from these nationwide screening programs, CRC screening tests are available at outpatient clinics or private health centers as opportunistic screening options at full out-of-pocket payment from recipients.16

After implementation of CRC screening in the NCSP, screening rates have increased.1617 Notwithstanding, even though overall screening rates have increased, whether inequalities in the uptake of screening are present, specifically in CRC, remains unclear. Previous studies focusing on SES disparity or inequality in CRC screening have applied conventional logistic regression or the concentration index to report inequality outcomes.181920 Inequality indicates for the degree of association between rates for a health indicator and the distribution of the population among ordered groups (e.g., education and income). On the other hand, health disparity is defined as a particular type of health difference that is closely linked with social, economic, and/or environmental disadvantages.21 In this study, we utilized the slope index of inequality (SII) and relative index of inequality (RII) to describe inequalities in CRC screening. Compared to traditional approaches, these measures have two ideal properties: First, because they are regression-based, they give an inequality measure across the full range of SESs, not just a comparison of the two most extreme groups as in conventional logistic regression. Second, they both take into account changes in the inequality by SES groups over time.

This study was conducted to evaluate CRC screening rates in Korea and trends therein from 2005 to 2015 for individual screening modalities using data collected via a population-based survey. Further, we assessed inequalities in CRC screening in Korea using SII and RII according to education and income levels.

This study attempted to measure education- and income-related inequalities in the use of CRC screening services by estimating absolute and relative inequalities and by evaluating changes in observed inequalities over the past decade. Using the cross-sectional KNCSS database, we found that CRC screening rates annually increased within the 11-year period from 2005–2015. Persons of younger age, lower household income, lower education, and male sex experienced more substantial increases in screening rates, with higher APC values. Our results indicated educational inequalities existed in CRC screening in Korea during the studied period and favored those who had higher education, with significant pooled estimate values in both absolute and relative indices. Regarding household income inequality, although significant results were noted in two separate study years, the overall pooled estimates of RII were not significant. Further, no apparent trend was observed in income inequality among survey years.

Our results were in line with previous studies in Korea in which an increase in CRC screening uptake was reported. One study of the Korea NCSP database noted an overall uptake rate for CRC screening of 30.1%, with an upward trend from 2006 to 2011.16 Another study using Korea National Health and Nutrition Examination Survey (KNHANES) data reported that CRC screening rates increased from 1998 to 2012, with an average rate of 30.4%.15 Using data from the NCSP, participation rates for CRC screening were again reported to have increased gradually from 7.3% in 2004 to 25.0% in 2012.17 The noted increase in participation in CRC screening was expected due to the implementation of the NCSP in Korea. Continuity in nationwide cancer screening services through the NCSP was deemed to have contributed to the gradual increase in CRC screening rates. In addition, a previous study reported that the increase in CRC screening might be attributable to a substantial increase in FOBT testing, compared to other modalities, especially among women and low-income groups.16 This could be explained by the introduction of annual FOBT testing at no charge as an initial screening tool for low-income individuals beginning in 2004.

To illustrate disparities in screening, we selected education as one indicator of SES. Even though education is a central indicator of SES, it has not received focus in previous studies of screening disparity: the majority of related studies on CRC screening disparity have addressed race, ethnicity, income, and insurance status.19252627 A few studies have previously found education to be associated with screening participation,2829 and results from our analyses showed that education inequality is present in CRC screening. This is in line with previous publications on differences in CRC screening according to education levels. One study suggested that health literacy is an associated factor with educational attainment, and might have mediated the contribution of education to screening inequality, to some extent.20 In the context of CRC screening, previous studies have reported that a low level of health literacy results in more problems in seeking and understanding health-related information or completing screening procedures.303132 In Korea, one reason for this inequality could be a lack of knowledge about cancer screening or a lack of awareness of the existence of the nationwide screening program.33 Further, research should be conducted to investigate the pathway of how low education level influences the uptake of CRC screening.

In this study, results on income inequality in CRC screening were inconsistent throughout the study period, and therefore, no distinct income inequality was found. Although overall findings did not show income inequality for the total study population or each gender separately, significant results were observed in 2007 and 2015 among women participants. Opposite outcomes were also observed in other years with negative values for both absolute and relative measures; however, these were not statistically significant. A previous study reported that higher CRC screening rates were observed among men of higher income status, whereas income disparities were not seen among women.11 There could be many reasons for income inequalities in CRC screening. We suspect that differences in eligible criteria for being tested by FOBT and colonoscopy might contribute to income inequality in CRC screening. While NCSP ensures free-of-charge FOBT testing for all individuals, only those with positive FOBT results are provided free-of-charge endoscopy testing. Furthermore, the NHI only reimburses colonoscopy costs for those with symptoms or colorectal problems. Therefore, income inequalities in CRC might be due to a greater use of endoscopy tests among individuals of higher income status. Indeed, another study reported that low household income was associated with lower screening rates with colonoscopy in urban areas in Korea.18 Meanwhile, one study hypothesized that nationwide expansion of public cancer screening would likely increase both total screening rates and income inequality in the uptake of cancer screening services.34 One plausible explanation mentioned by the authors is that, while income mainly represents an individual's ability to pay for services, utilization of cancer screening services might be influenced by other factors.34 For example, even though national screening services are free to all target individuals, those with low income occupations might face more difficulties in leaving work to receive screening.

Our study has several limitations. The KNCSS survey is heavily reliant on self-reported screening information. Therefore, our screening information may reflect recall bias in describing past screening experiences. However, previous studies have shown that self-reported cancer screening history is reliable and agrees well with medical records.3536 Thus, the impact of recall bias on the outcomes would likely be small. Further, only income and education levels were considered among various socioeconomic factors that can influence health care utilization. Finally, using SII and RII as inequality indices limited our choice of SES variables for analysis, as variables must be able to be hierarchically ordered.

This study provides evidence on changes in education and income inequalities in CRC screening in Korea over an 11-year period. Here, we discovered education inequalities in screening experience for CRC among individuals targeted for screening in the NCSP and these inequalities tended to decrease toward the end of the study period. No apparent trend was noted in income inequalities. The introduction of free CRC screening appears to have been effective in increasing CRC screening rates among low-income groups in Korea. However, around half of adults in Korea still do not participate in CRC screening. It will be important to continue to monitor inequality trends in screening, as well as to investigate factors associated with CRC screening particularly among those with low educational level. Moreover, further studies are needed to thoroughly outline SES inequalities in CRC screening, as well as factors associated therewith.