The effect of changing stool collection processes on compliance in nationwide organized screening using a fecal occult blood test (FOBT) in Korea: study protocol for a randomized controlled trial

The study will be conducted using a 2 × 2 factorial randomized control trial (RCT) design resulting in four groups. The first factor is the two methods of distributing stool devices (mailing vs. visiting a clinic), and the other is the two stool devices (sampling kit vs. conventional container). Thus, the study will be composed of four groups from two factors (allocation ratio1:1:1:1): (1) sampling kit received by mail; (2) conventional container received by mail; (3) sampling kit received at the clinic; (4) conventional container received at the clinic (control group) (Figure 1).

Individuals aged 50 to 74 years living in Goyang will be recruited from February 2013 to February 2015. The National Cancer Center (NCC) in Goyang has sufficient resources to screen all NCSP recipients living in the Goyang urban area. The study will take place at the NCC Center for Cancer Prevention and Detection. Invitation letters including a brief description of FOBT screening and the purpose of the study will be mailed to target participants, and those who want to participate in the study will provide verbal consent via telephone. After study staff confirms that they meet the study criteria, eligible participants will be randomly allocated to one of four groups. Written informed consent will be obtained from all participants upon visiting the clinic to submit stool specimens within two weeks after receiving a stool device. Exclusion criteria include: a prior diagnosis with any cancer including colorectal cancer; a history of colonoscopy or polypectomy; or a history of a FOBT more recent than one year. Ethical approval has been obtained from the Institutional Review Board at the NCC (NCCNCS-12-683).

There will be four groups: sampling kit received by mail, conventional container received by mail, sampling kit received at the clinic, and conventional container received at the clinic (control group). Sampling kits are being purchased from Eiken Chemical Co., Ltd., Tokyo, Japan. Compared to the currently used conventional container, the new sampling kit is thinner and smaller, with a thin (4.2 cm long) sampling probe attached to the cap that allows easier insertion into the stool. The tip of the sampling probe has a spiral groove to collect stool specimens (Figure 2). The sampling probe with the collected stool (approximately 10 mg) may then be reinserted into the kit. These stool specimens are analyzed in an OC-SENSOR DIANA machine (Eiken Chemical Co., Ltd., Tokyo, Japan) designed for quantitative immunochemical FOBTs and the sampling kits fit this equipment. Stool specimens in conventional containers are transferred to sampling kits for immunochemical FOBTs at the laboratory because the conventional containers do not fit the analyzer. Mailed sampling kits or conventional containers will include an instruction leaflet. All participants who receive stool devices will be instructed in methods of stool collection via telephone or face-to-face by research staff.

The primary outcome will be compliance with FOBT screening; compliance will be defined as the proportion of individual participants who submit stool specimens to the National Cancer Center within two weeks following receipt of collection devices. The secondary outcome will be participant satisfaction and intention to be rescreened in the next screening round as assessed by questionnaire at the clinic. The questionnaire will use a five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = uncertain, 4 = agree, 5 = strongly agree) with three dimensions: satisfaction with each process, from receiving stool devices to submission of specimens at the clinic; overall satisfaction with the entire process; and intention to be rescreened. The tertiary outcome will be the rates of positive FOBT results using sampling kits vs. conventional containers. Positive results are defined as those exceeding 100 ng/mL blood. All participants who undergo FOBT will be informed of the screening results via mail within three weeks, and those whose results are positive will be recommended to undergo colonoscopy. Colonoscopy or further biopsy for abnormal lesions will be performed by highly qualified colonoscopists upon provision of written consent at the NCC. Histopathological results will allow comparison of the detection rates of advanced adenoma or cancer between sampling kits and conventional containers. All outcomes of the study will be explained to respondents upon interview and prior to obtaining written informed consent.

Study participants will be divided into two groups according to the method of receiving the stool-collection device. To determine the sample sizes of two groups, we will establish the null hypothesis (H₀) and alternative hypothesis (H₁) using the difference p ₁ -p ₂ , where p ₁ is the expected compliance rate in the group receiving collection devices by mail and p ₂ is the expected compliance rate in those receiving them at a clinic.

The expected compliance rate of participants visiting a clinic is based on that observed in 2010 by the NCSP, which was 30.8%, and the compliance rate of participants receiving stool devices by mail is estimated at 36% based on the assumption that the rate will be at least 5% greater.

Based on the above formula, 1,398 people will be needed in each group to detect the difference p ₁ -p ₂ in the proportions using a two-tailed test and 80% power (1-β) at a 5% significance level. However, continuity correction tests will be needed to estimate very closely approximate values, and thereby a sample size of 1,438 in each group will be required. Finally, to allow for a 20% loss to follow-up, we will enroll an additional 350 participants per group, resulting in a final total for each group of 1,798 participants; thus, a total of 3,596 participants will be enrolled in the study.

After providing verbal consent by telephone, those who are eligible will be allocated to one of four groups by a computer-generated randomization program using the SAS 9.2 statistical software (SAS Institute, Inc., Cary, NC, USA). The sequence will be generated by simple randomization. Participants will be subjected to different trial conditions depending on the number they are allocated.

Compliance rates for FOBT screening will be compared among the four groups by a chi-square test. The satisfaction and intent to rescreen using a FOBT will be assessed by analysis of variance (ANOVA). The rates of positive FOBTs and the rates of advanced adenoma or cancer detected by colonoscopy using the different stool devices will also be analyzed by a chi-square test. The adjusted odds ratios (ORs) and 95% confidence intervals (CIs) of each outcome will be calculated using multiple logistic regression. All data will be analyzed using the SAS software, version 9.2 (SAS Institute, Inc., Cary, NC, USA).

This study will use a randomized study design to evaluate whether distribution using mail delivery and an improved stool-collection device increase FOBT screening compliance. Participants will be recruited through letters of invitation and randomly assigned to one of four groups (three interventions and one control group). Participants will be included only if they submit a stool sample to the NCC in person within two weeks after receiving a stool device, and these participants will be rewarded with small gifts. To maintain consistency in data collection, we will assign two well-trained research staff members to oversee data management, supervise all steps, and train assistant staff throughout the study duration.

All information collected will be covered by the Personal Information Protection Act. Participants’ personal information such as name, identification number, and contact number will be not used until they have signed a consent form. Research staff will manage all information to ensure confidentiality under the personal data regulations. Participants’ identifiable data will be retained only until the end of the study. Any information will be eliminated upon participant request.