Increasing uptake of colorectal cancer screening in Korea: a population-based study

Worldwide, colorectal cancer (CRC) is the fourth most commonly diagnosed cancer in men and third in women [1]. About 1 million new cases of CRC were diagnosed in 2002 (9.4% of all cancer diagnoses worldwide) [1]. However, at least a 25-fold variation in the occurrence of CRC has been reported worldwide. Western countries have the highest incidence rates [2]. Recently, CRC incidence rates have been increasing rapidly in countries where the overall risk was once low (especially in Asia), whereas rates in high-risk countries are either increasing gradually, stabilizing (North and West Europe), or declining over time (North America) [3]. In Korea, incidence rates have risen markedly over the past few years (e.g., annual percent change, APC = 7.3% per year between 1999 and 2005) [4, 5]. From 2003 to 2005, CRC was the third most commonly diagnosed cancer in Korea (12.0% of all cancer diagnoses) [5].

Based on evidence that screening reduces CRC incidence and mortality [6‐9], national guidelines in several countries now recommend regular CRC screening for average-risk persons aged ≥50 years using one or more of the following options: annual fecal occult blood test (FOBT), flexible sigmoidscopy every 5 years, a combination of FOBT (or fecal immunochemical test) and flexible sigmoidoscopy (FSIG), colonoscopy every 10 years, and/or double-contrast barium enema (DCBE) every 5 years [10, 11]. Although the incidence of CRC is increasing rapidly in Asian countries, national guidelines for CRC screening are nonexistent in most Asian countries [12]. Recently, the Asia Pacific Working Group on Colorectal Cancer reached a consensus to develop guidelines for CRC screening and recommend FOBT, FISG, and colonoscopy as the best options [13].

In 2002, the National Cancer Center in Korea, and Korean Society of Coloproctology developed guideline for CRC screening in average-risk adults [14]. This guideline recommends that average-risk adults begin screening for CRC at age 50, having either (a) a colonoscopy every 5-10 years or (b) DCBE combined with FSIG every 5-10 years. In 2004, the Korean government also introduced nationwide CRC screening as part of the National Cancer Screening Programme (NCSP) for low-income groups. Due to capacity related to performing colonoscopy screenings on every age-eligible person, the NCSP provides an annual FOBT (immunochemical) test as the initial mass screening method for men and women ≥50 years instead of colonoscopy [15]. The NCSP provides further investigation with a colonoscopy or DCBE for any low-income individual with a positive FOBT result. In 2004, Medical Aid recipients and National Health Insurance (NHI) beneficiaries within the lower 30% income bracket were eligible for free-of-charge FOBT screening services for CRC under NCSP. In 2005, the NCSP expanded its target population to the 50% income bracket. Apart from these organized CRC screening programmes, FOBT, DCBE, and colonoscopy testing are conducted in outpatient clinics or private health-assessment centers for opportunistic screening. However, in these cases individuals must pay all procedure-related costs.

Despite the evidence that screening can reduce the incidence of and mortality from CRC, uptake of CRC screening has been lower than for other mass cancer screening interventions [16, 17]. Globally, only half the eligible population undergoes CRC screening [18‐22]. In particular, screening for CRC is not commonly practiced in most Asian countries, and CRC screening behaviors in Asian populations remains largely unknown. Although national CRC screening programms have been offered since 2004 in Korea, no studies have reported the extent to which various CRC screening procedures have been used by the general population. In addition, no studies have analyzed whether usage of various types of screening procedures has changed over time since the introduction of the national CRC screening programme.

This study assessed the rates of CRC screening in Korea by screening modalities using a population-based survey. In particular, it examined trends in CRC screening rates after the introduction of the NCSP using four independent cross-sectional, nationally representative datasets collected during 2005-2008, and identified the factors associated with uptake of CRC screening by test modality.

Table 1 presents descriptive data about the study participants. The demographic distribution of the sample was almost homogenous among survey years, with the exception of household income and health insurance status. A larger proportion of 2008 survey respondents reported higher income compared to 2005 survey respondents.

Table 2 presents the proportion of subjects who reported CRC testing. The proportion reporting FOBT testing within the previous year increased significantly from 7.2% in 2005 to 21.3% in 2008 (p < 0.001). Rates of endoscopy use also increased from 18.0% in 2005 to 20.5% in 2008, albeit not significant (p = 0.060). Overall, CRC screening within the recommended time interval increased significantly from 22.9% in 2005 to 36.6% in 2008 (p < 0.001).

Tables 3 show changes in FOBT and endoscopy uses between 2005 and 2008, comparing 2005 rates with 2008 rates as relative risks. The results indicate an increased proportion of FOBT use from 2005 to 2008 in all study population. In particular, increases in FOBT testing were highest among those with lower income status (relative difference percentage = 511.9%) and women (relative difference percentage = 266.1%).

The results from the full logistic regression models for FOBT (model 1), endoscopy (model 2), and any test (model 3) are shown in Table 4. Rates of FOBT testing and any CRC testing were significantly increased in 2008 compared to 2005 at the adjusted α of 0.017. With regard FOBT test, those who aged 60-69 years, and reported their health status as "excellent/very good" were more likely to have undergone FOBT testing. Endoscopies were significantly more likely among those aged 60-69 years, those with higher income, and non-smokers. Women were significantly less likely to have had an endoscopy test. Overall, male, individuals aged 60-69 years, those with higher income, and non-smokers were more likely to have undergone up-to-date CRC screening.

Early detection and treatment is an important way to reduce mortality from colorectal cancer. Nonetheless, Asians tend to have little understanding about CRC screening. This report is the first to assess CRC screening rates by various screening methods in a Korean population at average risk for the disease. Data from national surveys reveal that CRC screening remains underused in Korea: in 2008, only 36.6% of the Korean population aged ≥50 years had undergone CRC screening within the recommended time interval. In contrast, data from the National Health Interview Survey indicate that in the United States, 44% of adults aged 50 to 64 years had undergone a recommended CRC screening test in 2005 [25].

However, the current analysis revealed a substantial increase in up-to-date CRC screening in the general population from 2005 to 2008. In addition, it showed substantial changes in the proportion of the general population being screened by different screening modalities. In 2005, only 23% of the Korean population was screened by FOBT, endoscopy (colonoscopy and/or sigmoidoscopy), and/or DCBE within guidelines. Screening rates did increase over the 3-year period, yet by 2008 still only 36.6% of respondents had been screened by 'any test' within guidelines. The majority of this increase was explained by a 14% increase in screening FOBT, yet there was only a 2.5% increase in endoscopy. Although endoscopy testing increased only slightly between 2005 and 2008 (relative difference percentage = 14%), FOBT testing rates increased relatively more during the same period (relative difference percentage = 196%). By 2008, FOBT had become the most common CRC screening test modality. In contrast, in the United States the prevalence of CRC screening using endoscopy was more than twice that using FOBT [25].

The large increases in FOBT use can be explained in part by health care policy and public awareness. Since 2004, the Korean government and NHIC have provided free-of-charge annual FOBT testing as an initial screening tool for low-income individuals. The current analysis revealed that the greatest increases in FOBT use were among individuals with low household income (relative difference percent = 512%). Further, FOBT testing increased 266% among women from 2005 to 2008. These results suggest that the substantial increase in FOBT testing among women and low-income groups might have been associated with the introduction of the NCSP. The current analysis also revealed that endoscopy screening rates increased slightly from 2005 to 2008. These relatively small increases in endoscopy use can be attributed in part to the fact that the NHI only reimburses colonoscopy and sigmoidscopy testing for those with symptoms or colorectal problems. Other health or educational policies, e.g., mass-media campaigns might have impacted on the increment of the CRC screening rates.

Consistent with previously published literature regarding CRC screening, we have shown that test use varies by gender, age, income, smoking status [16, 18‐22]. Interestingly, respondents with higher income levels were more likely to have an endoscopy test. Under the Korean health insurance system, an endoscopy test costs almost 10 times as much as an FOBT test. In addition, the NSCP ensures free-of-charge FOBT testing for all individuals, whereas endoscopy testing is only provided free of charge for those whose FOBT results are positive. Furthermore, the NHI only reimburses colonoscopy and sigmoidscopy testing for those with symptoms or colorectal problems. Therefore, endoscopy use varied with household income, possibly suggesting that the cost of endoscopy is a barrier to CRC screening.

We examined trends in screening for CRC by screening modality and explored the factors associated with uptake of CRC screening in a population sample. However, the study's cross-sectional design and reliance on self-reported screening limit the KNCSS. Previous studies have indicated that self-reported FOBT screening data might result in overestimates of the proportion screened [26, 27], and recent research has suggested that over-reporting may be greater for FOBT than for endoscopy [28]. Therefore, the proportion of subjects who were actually up-to-date for CRC testing may be lower than that indicated by the current analysis. Further, the cross-sectional design of the study precludes conclusions about whether any observed associations were causal. Finally, our study sample may not be representative because of low response rate. However, the KNCSS respondents had health characteristics and behaviors that were very similar to those found in other extensively used surveys, such as the Korea National Health and Nutrition Survey.

Our results demonstrate that CRC screening increased after the introduction of the NCSP, according to a population-based survey. Increasing recognition of the efficacy and importance of CRC screening might have influenced CRC screening rates. However, overall screening rates increased less than did FOBT screening rates. The data indicate increases in both FOBT and endoscopy testing between 2005 and 2008, combined with a shift toward greater use of FOBT compared with endoscopy. This shift might be associated with national CRC screening policies. The introduction of free CRC screening appears to have been effective in increasing CRC screening rates among low-income groups in Korea. However, the increases in FOBT test use have implications for public health practice. Although FOBT is simple, safe, most inexpensive, and its routine use was shown to reduce CRC mortality, it is limited by poorer sensitivity, mainly for premalignant lesions. Colonoscopy, on the other hand, is a more accurate technique, yet invasive, carries a risk of bleeding and perforation, requires preparation and premedication, and involves much higher costs. The options for CRC screening tests allow for flexibility but can also render decisions about recommending or choosing a particular test difficult. Each test has its tradeoffs in terms of efficacy, complications, discomfort, time, and cost [20]. More than half of adults in Korea are still not up-to-date with their CRC tests. It will be important to continue to monitor trends in screening, as well as to investigate factors associated with up-to-date CRC screening. More research will also be required to increase CRC screening rates, particularly among those who had not have CRC screening under NCSP.