Abstract
Objective:
To implement a colon cancer screening program for uninsured or underinsured Iowans.
Methods:
All 1995 uninsured patients or patients with Iowa Care insurance aged 50 to 64 years attending the University of Iowa Clinic or the Iowa City Free Medical Clinic were mailed information about the project. Recruitment also took place in person, by having the clinic receptionist hand subjects a research packet, and through community posters. Individuals with colonic symptoms or who were up to date with screening were ineligible. Eligible subjects received a free fecal immunochemical test (FIT), and those with positive FITs were provided with a colonoscopy at no cost to them.
Results:
Of 449 individuals who completed eligibility forms (23% of the study population), 297 (66%) were eligible and were provided with an FIT. Two-hundred thirty-five (79%) returned a stool sample, with 49 (21%) testing positive. Thirty of the 49 (61%) individuals had a colonoscopy, and 20 individuals had at least 1 polyp biopsied. Thirteen individuals had at least 1 tubular adenoma; 2 had adenomas more than 1 cm in diameter, with no colon cancers identified. Face-to-face recruitment had the highest rate of returned FITs (72%) compared with handing the subject a research packet (3%) or a mailing only (9%) (Chi-square, P < .001).
Conclusion:
There was high interest in and compliance with colon cancer screening using a FIT among underinsured individuals. Although the FIT positivity rate was higher than expected, many individuals did not complete recommended follow-up colonoscopies. Population-based strategies for offering FIT could significantly increase colon cancer screening among disadvantaged individuals, but programs will have to develop sustainable mechanisms to include the necessary organization and address substantial costs of providing mass screening, as well as facilitating and providing colonoscopies for those who test positive.
Keywords
Introduction
Colorectal cancer (CRC) is the second leading cause of cancer death in the United States,1,2 yet it is estimated that 76% to 90% of cases could be prevented with appropriate screening. 3 National clinical guidelines recommend colorectal cancer (CRC) screening for average-risk individuals beginning at age 50, yet fewer than half of all eligible Americans are in compliance.4-6 Individuals without health insurance or a usual source of health care are less likely than those with insurance to be screened for colon cancer.7-13 Lack of recommended screening for colon cancer leads to later stage at diagnosis, when cancers are less curable.14-16 A recent decision analysis concluded that for an average risk population, annual fecal occult blood testing or colonoscopy every 10 years leads to similar life-years gained. 17 Given that socioeconomically disadvantaged populations have less access to screening colonoscopies, a population-based strategy of annual testing with a fecal immunochemical test (FIT) may be worthwhile. 18
This is the first published report of a screening program using FIT in a population without adequate health insurance. The purpose of this project was to implement a colon cancer screening program for uninsured or underinsured Iowans using an FIT, to provide colonoscopies at no cost for individuals with a positive screening result, and to evaluate the program.
Methods
Funding was received from the Iowa Department of Public Health, “Iowa Get Screened” contract. The University of Iowa Institutional Review Board approved this study. The study was conducted from December 2007 through December 2008.
Study Population
A total of 1995 potential subjects 50 to 64 years of age were identified; 1883 were from University of Iowa Healthcare (UIHC) medical records, and an additional 112 individuals were patients at the Iowa City Free Medical Clinic (FMC). The case manager at the FMC reviewed all medical records to determine who met inclusion criteria. For the UIHC subjects, we developed a computer algorithm to electronically determine whether individuals were up to date with CRC screening by any of the 4 accepted methods.19,20 Potential subjects had to be aged 50 to 64 years, have either no insurance or Iowa Care insurance, and have had at least 2 visits to the clinic in the past year or 1 visit that was a physical exam to ensure that they were “active” patients in the practice. Iowa Care insurance is for Iowa residents who meet specific poverty guidelines.
Subject Recruitment
All of the potential subjects received mailed study information using a modified Dillman protocol. 21 Subjects were mailed a letter explaining the project, followed 2 weeks later by a study packet that included a cover letter, informed consent documents, an eligibility screening questionnaire, a pamphlet entitled “Colorectal Cancer Screening Saves Lives” published by the Centers for Disease Control (publication no. 99-6948), and a stamped, addressed return envelope. Eligibility criteria included the following: (1) income less than 250% of the federal poverty level; (2) uninsured or underinsured for CRC screening; (3) not up to date with CRC screening according to national guidelines; (4) no colonic symptoms such as blood in the stool, significant changes in bowel habits in the past 2 months, difficulty passing stools, or pencil-like stools; and (5) no personal history of colon cancer, adenomatous polyps, or inflammatory bowel disease. A family history of colon cancer did not disqualify an individual from the study. Individuals who returned a signed consent form and who met the specified eligibility criteria were mailed an FIT kit (Clearview ULTRA FOB, Inverness), directions for completion, and a postage-paid cardboard mailer.
In addition to the mailings, subjects were approached in the UIHC Family and Internal Medicine primary care clinics by having the receptionist or nursing staff hand pre-identified patients a research packet. For Primary Care North (PCN) (which sees Iowa Care patients), interested individuals were referred to a research staff member in the clinic for a face-to-face discussion. Project team members spent a total of 9 half-days each week for 3 months for recruitment in PCN.
Posters were also hung in various community sites in the county in which the study took place, as well as at IDPH Breast and Cervical Cancer Early Detection Programs in 8 adjoining counties. Since those recruited in the UIHC clinics also received mailings and those who responded to community posters were relatively few, we considered our study population to be those identified at UIHC and at the FMC.
Fecal immunochemical test
The stool specimens returned from the FIT kits were developed in the Department of Family Medicine at the University of Iowa by research staff using the Clearview ULTRA FOB (Inverness) kits and reagents.
Each individual was informed of their result by mail. In addition, all subjects with positive tests were called by the project director (BTL) and advised to have a colonoscopy. Individuals with a positive FIT were also telephoned by the UIHC Division of Gastroenterology scheduler and given colonoscopy prep instructions and materials according to their usual protocol.
Program evaluation
Postage-paid envelopes and program evaluation forms with no identifiers were included with the results letters sent to the individuals who returned a FIT kit.
Data Analysis
Data were loaded into SAS, version 9.13 (SAS Institute Inc., Cary, NC, 2008) for analysis. Means and frequencies were examined. To compare differences between groups, the Student t test was used for continuous variables and chi-square was used for categorical variables. The main outcome variable was whether a patient returned the FIT. The groups compared were: (1) among interested individuals, eligible versus noneligible individuals; and (2) among eligible individuals, those who returned the FIT versus those who did not. Another outcome variable was the FIT results, and if positive, whether a colonoscopy was completed. Results of the colonoscopy were summarized according to the types of polyps found and complications.
Results
Of the 1995 individuals in the study population, 23% (449) returned documents indicating an interest in the screening. Two-hundred ninety seven of 449 (66%) met specified eligibility criteria. Those ineligible reported being up to date with screening (n = 95), had symptoms potentially related to CRC (n = 97), and/or had insurance that would cover colon cancer screening (n = 6), which disqualified them from the study. We achieved a 79% return rate for the FIT (235/297). Of those who returned kits, 49 (21%) tested positive for blood in the stool.
Table 1 compares the demographic characteristics of interested individuals according to eligibility criteria, and there were no differences found. In addition, Table 1 shows the characteristics of those who returned a FIT kit with those who did not, and the mean household income was significantly higher for those who returned a kit versus those who did not ($12,042 vs. $7,323; t test P < .001).
Demographic Comparisons by Interest and Eligibility
Abbreviation: FIT, fecal immunochemical test.
The project director telephoned and reached each of the 49 individuals with a positive FIT to inform them of their result and to let them know that appropriate follow-up included an optical colonoscopy. Thirty of the 49 (61%) individuals had a colonoscopy. The remaining 19 individuals either did not respond to numerous phone contacts by the colonoscopy scheduler, chose not to have a colonoscopy, or failed to show up for their appointment for the colonoscopy. Despite substantial phone effort, several of these individuals failed multiple times to show up for their scheduled colonoscopy.
Of the 30 completed colonoscopies, 20 individuals required at least 1 biopsy. Thirteen individuals had at least 1 tubular adenoma (considered a potentially precancerous type of adenomatous polyp), and 2 of these individuals had one that measured at least 10 mm in diameter. No individuals had cancer, and there were no complications from any of the colonoscopies.
Two individuals had incomplete colonoscopies owing to narrowing of the colon. One of these individuals had a follow-up air contrast barium enema, which was normal. A second subject’s procedure was incomplete owing to sigmoid tortuosity, and this individual has been advised to have further colon visualization with either a CT colonography or barium enema, since this individual had a tubular adenoma.
Twenty-five individuals telephoned in response to the posters; 18 were eligible, and all 18 returned the FIT. Table 2 shows the percentage of returned kits by recruitment strategy for the 1995 in the study population. Face-to-face recruitment worked best when compared with the other 2 methods (72% kit return rate compared with 3% handing out research packets and 9% mailing only [chi-square, P < .001), but research staff spent 9 half-days each week for 3 months to accomplish this result.
Recruitment Numbers by Recruitment Strategy (N = 1995)
Participant Evaluation of the Program
Evaluations were received from 51% of the individuals who completed the FIT. Ninety-seven percent reported they understood how the program worked, and 94% said the test was not difficult to complete. Those persons who had some difficulty wrote: “sample got wet,” “unsure if paper used to catch sample went in toilet or on top of seat,” “the paper actually did not stay on top of the water very well,” “need to explain just how much fecal matter is needed so it isn’t messy getting swab in container,” and “it was hard to get motivated, hard to have bowel movement on paper to secure sample, worried paper would get too wet.”
Ninety-three percent said they understood their screening test result, and 91% felt their results were received in a timely manner. Participants made the following suggestions for improvement: “doesn’t need improving, easier test than colonoscopy,” “great program, thanks so much,” “not sure, but I was thankful to be involved with the program, my best friend passed away with colon cancer,” “more follow-up on people with colon problems, such as colonoscopy for them even if their test is okay,” “I’m happy with the results. I just don’t feel 100% unless I have a colonoscopy,” and “the fact that this is a free test to help save lives is highly commendable.”
Discussion
This is the first study of which we are aware in which fecal immunochemical testing was used to screen an underinsured or uninsured population for colon cancer and colonoscopies were provided at no cost for those with positive results. Among eligible individuals, there was high interest and compliance with colon cancer screening, with a 79% return rate for the FIT from a group of low-income patients. The vast majority found the FIT very simple to complete. Face-to-face recruitment worked best, but it required significant time compared with mailing information to potential subjects.
The 21% rate of positive FIT that we found is much higher than that found in other studies of asymptomatic screening populations, which has been 5.5%, 22 5.6%, 23 6.3%, 24 and 6.7%. 25 The higher positivity rate found in our study population may relate to the relative lack of financial and other resources in our population of very low-income individuals who may not have had access to CRC screening until the Iowa Department of Public Health provided the resources for screening.
The United States Preventive Services Task Force has recommended screening for colon cancer in individuals aged 50 to 75 years using fecal occult blood testing (with either a guaiac-based test or FIT), sigmoidoscopy, or colonoscopy. 26 A decision analysis has found that assuming equally high adherence, a strategy of either colonoscopy every 10 years or annual FIT provides similar life-years gained. 17 Depending on the simulation model used, about 1200 to 1400 fewer colonoscopies are performed in a population of 1000 individuals screened between ages 50 and 75 using the annual FIT compared with a strategy of colonoscopy every 10 years. 17
It has been well established that those without insurance are less likely to be screened for colon cancer and are more likely to be diagnosed at a later stage.9,12,16,27,28 Given that there are 45.7 million Americans without health insurance, 29 broader use of FITs could significantly increase screening among individuals who do not have the resources to access colonoscopy services. Fecal immunochemical tests have superior performance characteristics for detection of adenomas and colorectal cancer compared with stool guaiac tests,22,30-32 and the public often prefers a fecal test to a colonoscopy.33,34 Fecal immunochemical tests remove the difficulties created by dietary restrictions; the results are not affected by vitamin C ingestion; nonsteroidals can be continued; and FITs are relatively inexpensive, at approximately $30.35-38
Any national population-based strategy for CRC screening should allow for testing of the significant numbers of individuals who may have colonic symptoms and who do not meet traditional definitions for asymptomatic screening. In this study, there were many individuals interested in being tested who did not meet the specified eligibility criteria, with just over half of those ineligible having colonic symptoms. Our previous study of colon cancer screening in family physician practices found that 46% of eligible individuals were up to date with colon cancer screening guidelines, but only 17% received asymptomatic screening. 7
Study Limitations and Strengths
To our knowledge, this is the first reported study where an FIT has been used in a screening program for an uninsured/underinsured population. Although our uptake was only about 23% and it was difficult to precisely specify our study population, among those eligible who received an FIT, there was very high compliance with return of an acceptable sample. This study was conducted in a rural state, and the results may not be generalizeable to other uninsured populations. We do not know how many individuals had false-negative FITs, since colonoscopy was recommended only for those with positive tests.
Eligibility was determined by self-report of past CRC screening. Self-report is not always accurate; individuals tend to overestimate the recency of their screening, but they have fairly good recall for colonsocopy.39-41
Despite substantial effort, we were unable to convince many subjects with a positive FIT to complete a colonoscopy, which has been noted in other studies.22,42,43 Thirty-nine percent of those with a positive FIT who were recommended to undergo a colonoscopy failed to do so, despite receiving multiple telephone calls including a telephone call from the project director informing them of their results, multiple calls from schedulers and/or appointments for colonoscopy, as well as knowing from the outset that if they had a positive FIT, a colonoscopy would be recommended. We did not explore reasons for the lack of follow-through, but the population with which we were working had significant health and financial issues. Our rate of non-follow-through for colonoscopy is higher than the 16% reported by Van Rossum 22 and the 20% reported by Fisher et al, 42 whereas it is lower than the rate reported by Denburg, who found that only about 50% of patients referred for screening colonoscopy completed the procedure. 43 The exact number of individuals in our study population is not known, but we used the assumption that all of those who were mailed information composed the study population, since so few were recruited via posters.
Conclusions
There was significant interest in colorectal cancer screening using an FIT in a low-income population, with many of those interested ineligible due to symptoms. Close to 80% of those who received an FIT returned it. Twenty-one percent of individuals had a positive FIT, but a significant percentage of these individuals failed to follow through with a colonoscopy despite substantial efforts to educate subjects about the need for and to schedule the colonoscopy. Among those who received a colonoscopy, 43% had at least 1 tubular adenoma. Compared with direct patient mailings or handing out research packets, face-to-face recruitment required a substantial amount of personnel time but resulted in a much higher return rate for the FIT. Population-based strategies for offering FIT could significantly increase colon cancer screening among disadvantaged Americans, but programs will have to develop sustainable mechanisms to include the necessary organization and address substantial costs of mass screening and facilitating and providing colonoscopies for those who test positive.
Footnotes
The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
This study was supported by a contract from the Iowa Department of Public Health, the Iowa Get Screened: Colorectal Cancer Screening Program, the Departments of Family Medicine and Internal Medicine at the University of Iowa, the Iowa City Free Medical Clinic, the Primary Care North Clinic at the University of Iowa Healthcare, and the Holden Comprehensive Cancer Center at the University of Iowa.
Human participation protection: All study procedures were approved by the University of Iowa Institutional Review Board. All participants provided written informed consent.