Problem Addressed

• A common, often deadly condition: In 2008, more than 108,000 new cases of colon cancer and more than 40,000 new cases of rectal cancer occurred, making colorectal cancer the third most common form of cancer in the United States. It is also the third leading cause of cancer deaths among U.S. adults, with almost 50,000 individuals dying from the disease in 2008.¹
• Higher risk for African Americans: African Americans have lower colorectal cancer screening rates and higher death rates than whites. In 2008, 62.9 percent of African Americans age 50 or older reported having had a fecal occult blood test (FOBT) within the past year or lower endoscopy within 10 years, compared to 66.2 percent of whites.² The annual death rate from colon cancer per 100,000 individuals was 29.8 for African-American men (compared to 19.5 for white men) and 19.8 for African-American women (compared to 13.6 for white women).³ (Nationally, Hispanics have lower screening and death rates than do African Americans or whites.^2,3) At Kaiser Permanente Colorado, African-American and Hispanic members both have slightly lower screening rates and higher death rates than whites.
• Many preventable deaths: Many colorectal cancer deaths can be prevented with appropriate screening as recommended by the United States Preventive Services Task Force (USPSTF) guidelines; such screening can identify early treatable cancers and enables the detection and removal of polyps before they progress to cancer.¹ However, despite a growing awareness of its necessity, screening remains underutilized, with only 57 percent of eligible U.S. adults being current with recommended screenings.⁴ Lower screening rates among African Americans are believed to be a major contributor to their higher prevalence and death rates.⁵ Research suggests that a major obstacle to improving screening rates is that a subset of the population, which includes many underserved, low-income, and/or minority patients, does not see a primary care physician regularly and tends to avoid preventive medical care, including recommended screening tests.⁶ Outreach strategies that are not dependent on office-based care may have greater potential to reach this subset of unengaged members.
• Unrealized potential of FIT: The relatively new FIT, a stool testing kit that can be completed at home and mailed in for laboratory analysis, offers several advantages over the standard guaiac FOBT. There are several versions of FIT; this program used a single-sample version in which the stool sample is collected at home and mailed in for laboratory analysis, and the test is then processed by machine. It is more accurate and convenient than standard FOBT.⁷ In addition, it does not require an office visit, thus appealing to patients without a primary care doctor and/or who have few office contacts. FIT provides a low-cost, convenient, noninvasive option, which can be used to reach large populations and combined with other screening modalities, as in the Kaiser Permanente Colorado program.

Description of the Innovative Activity

• Target population: The program targets men and women ages 50 to 74 who have an average risk of colorectal cancer and have not completed an FOBT or FIT in the past year or have not had a colonoscopy in the past 10 years (as USPSTF guidelines recommend). Those considered to be at average risk do not have a personal history of colorectal cancer or polyps, or a family history of colorectal cancer, inflammatory bowel disease, or certain genetic syndromes. (People with such a background are considered "high risk" and instead receive a personal telephone call or a letter urging them to schedule a colonoscopy.) Kaiser Permanente uses a registry that includes patients' family history, diagnoses, procedures, and pathology dating back to 1994 to identify who should participate in the program.
• Automated telephone call: As members become eligible, they receive an automated, interactive 1- to 3-minute telephone call in which they respond to a series of questions by punching the number keys. At the outset, members choose whether they want to hear the call in English or Spanish (an option added in 2012). If no one picks up, the system calls three times, leaving a message with a callback telephone number on the third call. The call covers the following areas:
  • Screening history: Members confirm or clarify their screening history and need for screening by indicating that they have never been screened before, been screened previously but are now due for a subsequent test, are not due for screening because they have had the appropriate test within the required time frame, or belong in the "high-risk" category because of their personal or family history. In the latter case, the registry is updated with new information regarding their high-risk status, and the member's primary care provider is notified of the need for a colonoscopy referral through Kaiser's electronic medical record (EMR) system.
  • Education: Members receive a brief education about the option of FIT or colonoscopy testing, including an overview of their respective benefits, limitations, and risks. For example, members learn that whereas FIT is a recommended screening test, it does not find all cancers, and that colonoscopy will be required if the FIT comes back positive.
  • Ranking of barriers to screening: During 2008 and 2009, members were also presented with four common barriers to screening (lack of knowledge of the importance of screening, copay/cost issues, fear of pain, and inconvenience), and asked to rank them in order from most burdensome to least. Lack of knowledge was the most common barrier identified, followed by fear of pain.
  • Opportunity to request screening: Members can request that a FIT kit be mailed to them, or can request to schedule a colonoscopy. Members who choose this latter option are asked to contact their provider or Kaiser's Colorectal Cancer Prevention Program for a referral.
• Mailing FIT to all who do not request colonoscopy: To maximize the screening rate, all average-risk members who receive a call and do not request a colonoscopy are subsequently mailed a FIT kit. Those who complete the call and request a FIT kit are mailed one within 2 weeks. Those who do not answer or complete the call are mailed one 30 days later. The kit is accompanied by a one-page letter with information in English and Spanish about the benefits and limitations of stool testing, a one-page instruction sheet, and a prepaid return envelope. During the program's first year, roughly 64,000 members received FIT kits, and about 67,000 now receive them annually.
• Followup with nonresponders: Patients who do not send in their kits receive a reminder letter 1 month later.
• Notification of results and followup: Members with negative FIT results receive notification by mail. For those who test positive, the patient's primary care provider receives notice through the EMR system, and the provider's office contacts the patient to schedule a followup diagnostic evaluation (typically a colonoscopy). Members who do not complete the colonoscopy after a positive FIT test receive a reminder letter 8 weeks later. If necessary, at 16 weeks the provider is sent an e-mail notification through the EMR that the member still has not completed the colonoscopy, and the provider follows up directly with the member. In addition, the program sends a FIT kit each year on the anniversary date of the initial testing for members who initially chose to screen by FIT and remain eligible for screening.

References/Related Articles

Contact the Innovator

Innovator Disclosures

Results

• Fourfold increase in screening rates: The intervention of an initial educational call followed by a mailed FIT resulted in an almost fourfold increase in the rate of screening completion by either FIT or colonoscopy in those who had just received the outreach compared with those who had not yet received it and were exposed to usual care. The increase in screening was especially effective among those without recent primary or specialty care visits, suggesting that the program has the potential to facilitate screening independent of the office visit.
• Increased screening rates for racial and ethnic minorities: During the 2008 study, 45.1 percent of Hispanic participants and 44.4 percent of African-American participants were screened. These percentages appear to be rising, but specific data are not yet available.
• Significant increase in overall screening rate: The screening rate among all eligible members rose from 47 percent in 2006 (1 year before program launch) to 72 percent in 2008 (1 year after implementation), an increase of 25 percentage points. By contrast, the screening rate among the entire insured population in Colorado improved only 3 percentage points over the same period, from 62 to 65 percent. Since 2008, screening rates at Kaiser have continued to improve but at a slower pace, reaching 76 percent by the end of 2011.

Evidence Rating (What is this?)

Context of the Innovation

Planning and Development Process

• Task force meetings to formulate program: Once Kaiser Colorado began phasing out FOBT in favor of FIT in 2007, the colon cancer screening task force (which includes a medical director and physician and staff representatives from primary care, prevention, laboratory, gastroenterology, surgery, and radiology) began discussing the feasibility of mailing FIT kits to eligible members. To enhance member awareness of the available screening options and help them make an informed decision, the task force decided to combine automated telephone calls with mailed kits. Analytic modeling assisted the team in gauging the impact on staff, equipment, and systems.
• Creation of materials: In 2007, task force members modified written materials that accompany the FIT kit from those used in the Northern California Kaiser program and developed the initial scripts for the automated calls.⁸ Information technology (IT) staff took the scripts and created test versions of the calls.
• Pilot testing: In early 2008, the task force tested the telephone calls on a small sample of eligible members to ensure that users could successfully navigate the menu of options. Test kits were then sent out and returned to confirm that the mail and results management processes functioned correctly.
• Preparation for launch: In anticipation of the program's rollout in April 2008, the task force developed a communication campaign that included regular updates on key steps in the project's rollout to create awareness and enthusiasm among physicians, staff, and leadership. The task force also coordinated with the health system's affiliated central laboratory and gastroenterologists to explain how the program would work and ensure they had sufficient staff and equipment to handle the expected increase in laboratory tests and colonoscopies. These discussions led to the decision to hire several additional staff, conduct relevant staff training, install new equipment, and make various laboratory modifications. To avoid overwhelming these departments, the initial round of calls and mailings were staggered from April through September.

Resources Used and Skills Needed

• Staffing: Creating the colon cancer screening program required significant investments of time by task force members and IT staff. Its implementation and ongoing use primarily employs the services of laboratory technicians, gastroenterologists, and clinical and administrative support staff.
• Costs: It is hard to calculate the incremental costs associated with this program in the integrated delivery system. Direct program-related expenses include the upfront costs for IT reprogramming along with the ongoing costs of FIT kits and mailings. However, some members would have been screened even if the program did not exist. In addition, the increase in screening rates has likely resulted in cancers being caught earlier, when they are less expensive to treat.

Getting Started with This Innovation

• Design calls to be concise: Members may not complete automated calls if they go on for too long. As a result, the call script should be as concise as possible, striking the right balance between including critical information and not overwhelming callers. Since the first year, program leaders have shaved several minutes off the total length of the call to make it easier for recipients to complete. Calls now average 1 to 3 minutes.
• Plan for increase in number of colonoscopies: Early on, the program is likely to significantly increase the number of colonoscopies performed, since some patients will request them when they receive the call and others will need them after testing positive on the FIT. Those adopting this program need to ensure that sufficient staff and facilities are available to accommodate the increase. Modeling the number of expected increased colonoscopies for initial screening and ongoing followup is critical to gauge the impact on gastroenterology capacity.

Sustaining This Innovation

• Maintain accurate registry: Whether new or existing, the registry used for this program must be kept as accurate and up to date as possible. To that end, new members should routinely be asked whether they have previously completed a screening, when it occurred, what type of screening they had, and what the results were. An accurate registry ensures, for example, that a member who had a colonoscopy 5 years ago does not receive a FIT kit. Note, too, that it is possible to implement a successful screening program without a registry. If the outreach group is not already stratified by risk, patients must be educated on criteria for high-risk status and receive a clear recommendation for colonoscopy screening if high risk along with instructions on how to schedule. FIT testing is only appropriate for average-risk individuals.
• Analyze screening rates and respond accordingly: Ongoing monitoring of the registry may reveal important trends, such as a leveling off in the proportion of eligible members screened and/or lower-than-average screening rates among certain subpopulations (e.g., those in a particular age range, of a certain race or ethnicity, living in a specific geographic region, or with a certain medical history). If screening rates in particular groups are lagging, develop targeted strategies for these groups.
• Consider additional outreach to nonresponders: If sufficient staffing is available, consider making personal calls to patients who repeatedly fail to complete the automated calls or mail in their kit.