Health Care Systems for Tracking Colorectal Cancer Screening Tests
According to the Centers for Disease Control and Prevention (CDC) and United States Cancer Statistics (USCS) data, colorectal cancer (CRC) is the "second leading cancer killer" in the United States among cancers affecting both men and women. It is also one of the most commonly diagnosed cancers. In 2006, 139,127 people (70,270 men and 68,857 women) were diagnosed with CRC, and 53,196 people (26,801 men and 26,395 women) died from it (USCS, 2010). According to CDC, when CRC is found and treated early, survival is high (90 percent). However, many colorectal cancers are not found early due to low screening rates.
This project sought to assess whether, to what extent, and how easily a health system redesign intervention could increase CRC screening and followup. The intervention was called the System Approach to Tracking and Increasing Screening for Public Health Improvement of Colorectal Cancer (SATIS-PHI/CRC) and was implemented in a network of primary care practices. The project was funded by the Centers for Disease Control and Prevention (CDC). It was carried out as a task order under the ACTION (Accelerating Change and Transformation In Organizations and Networks) program of the Agency for Healthcare Research and Quality (AHRQ) between October 2007 and July 2010. We implemented the intervention in early 2009, and it ran through February 2010.
SATIS-PHI/CRC is a population-based system-redesign intervention designed to improve CRC screening rates and rates of diagnostic followup for positive screens. We based the major components of SATIS-PHI/CRC on prior studies conducted by project staff at Thomas Jefferson University (TJU) (Myers, et al., 2007; Myers, et al., 2004; Myers, et al., 2001). Those studies showed that a targeted outreach intervention to patients in a large urban academic practice improved CRC screening rates. They also indicated that a feedback intervention to providers in practices affiliated with a large, for-profit managed care organization improved diagnostic followup for positive screens. We used a case study approach, informed by the PRISM (Practical, Robust Implementation and Sustainability Model) framework, to determine whether we could:
- Implement SATIS-PHI/CRC in a different setting.
- Increase screening and followup rates.
- Achieve rate improvements similar to those previously achieved by the TJU research team.
The health system setting for this project is the Lehigh Valley Physician-Hospital Organization (LVPHO) affiliated with the Lehigh Valley Health Network (LVHN) and the Greater Lehigh Valley Independent Practice Association (GLVIPA). The PHO, which offers a preferred provider organization health insurance plan, has an interest in value-based health care and sees preventive care, including CRC screening, as a means to that end. This project builds on the cited prior studies and examines both the process of implementing the SATIS-PHI/CRC intervention in the LVPHO network of practices and the outcome of the intervention.
The SATIS-PHI/CRC intervention has the following features:
- It is a population-based, system-level redesign of the way CRC screening and followup are conducted in a network of primary care practices.
- It is intended to assist the practices to better provide guideline-based preventive health care to their patients ages 50 through 79 years old who are at average risk for CRC.
- It assists practices to provide population-based CRC screening that follows recommendations and guidelines jointly issued in 2008 by the American Cancer Society, the U.S. Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology (Levin, et al., 2008) and also in 2008 by the U.S. Preventive Services Task Force (USPSTF, 2008).
- It provides a mechanism for identifying patients who are eligible for but not up to date in their CRC screening, contacting such patients on behalf of their physician's practice to encourage recommended screening, tracking screening results, and facilitating patient notification and appropriate followup through feedback to providers.
- It also provides a facilitating mechanism for patients to undergo screening. Although the Multi-Society and USPSTF guidelines identify a range of acceptable screening modalities, in an effort to avoid possibly confusing patients with too many choices, SATIS-PHI/CRC limits the choice to only two: (1) a less invasive modality patients can perform themselves at home (stool test) and (2) a more invasive modality requiring a physician-performed procedure (colonoscopy).
- Using academic detailing and performance feedback forms, it seeks to educate clinical providers and other staff in participating practices about recommended CRC screening and followup procedures.
- Using mailed information, it seeks to educate targeted patients of participating practices about the importance of and need for CRC screening and about the various types of recommended screening modalities.
We designed the SATIS-PHI/CRC intervention between October 2007 and early July 2008. Our design effort included updating our initial environmental scan of CRC screening interventions, developing goals and objectives for the intervention, designing an implementation plan, and planning for the assessment of the intervention. The intervention we designed includes six component steps:
- Recruit primary practices to participate in the intervention.
- Conduct academic detailing in these practices.
- Identify patients of these practices who are eligible to receive the intervention materials.
- Mail the screening intervention materials to them.
- Track resulting patient screening and results.
- Provide feedback to the participating practices regarding screening results for their patients and recommended, guideline-supported followup.
Pilot and Intervention Protocol
Because this project included an assessment of the intervention that entailed human subjects research, it required Institutional Review Board (IRB) approval. In addition, the assessment and several steps in the intervention process required data collection. The project was conducted under a task order contract with the Federal Government, so we had to obtain Paperwork Reduction Act clearance from the Office of Management and Budget (OMB) prior to any data collection. Once we received approval from the LVHN and TJU IRBs for the assessment study, we worked with AHRQ to obtain OMB clearance to collect data for this project.
Steps 2 and 3 of the intervention process included surveying practice staff to ascertain baseline screening knowledge, attitudes, and practices. These steps also involved accessing and reviewing electronic records to identify patients eligible to receive the intervention. Thus, we needed to obtain OMB clearance before we could implement those steps. We began developing material for the OMB submission in late January 2008 and received written clearance in early December of that year.
In addition to developing, implementing, and then assessing the implementation and outcomes of the SATIS-PHI/CRC intervention, this project required us to develop materials for dissemination based on our experiences with the intervention. The dissemination materials were to include our findings, lessons learned, and a toolkit that could all be used by other health care systems interested in adopting this intervention to improve CRC screening and followup rates. Figure 1.1 presents a timeline for the implementation, assessment, and dissemination phases of this project.
To allow us to move forward on gaining experience with the intervention and its assessment while we awaited OMB clearance, the project's Task Order Officer granted permission for us to pilot test the IRB-approved intervention and assessment protocol in one primary care practice. We got permission on the condition that we not use any data collected during the pilot in any publication, presentation, or external report, whether separately or combined with data collected during the main intervention. We could, however, use lessons learned during the pilot to revise and refine the main intervention, and we were encouraged to do so by the Task Order Officer and the project's Technical Advisors from CDC.
We conducted the pilot test of the intervention between late June 2008 and March 2009. We recruited all of the practices for the pilot (1 practice) and the full intervention (25 practices) concurrently at the beginning of the pilot test period. We then began the pilot while waiting for OMB clearance. The pilot was well underway when we received OMB clearance in December 2008.
The pilot and the full intervention overlapped during the first few months of 2009, allowing us to begin the full intervention with lessons learned from the early steps in the pilot. Later, we incorporate further lessons learned from later steps in the pilot. This approach avoided having to wait until the pilot was fully completed before beginning the full intervention and helped us compensate for unanticipated delays due to the OMB clearance process.
To further expedite the intervention, we divided participating intervention practices into two waves for Steps 3 and 4. That way, we did not have to wait until we could access and review electronic records from practices whose records were difficult to work with before we moved ahead with nonproblematic practices. We also selected two Wave 2 practices as sites to introduce a variation of the intervention. The intervention period ended in February 2010.
The assessment began with preintervention surveys, focus groups, and key informant interviews with intervention practices concurrently with implementing step 2 of the intervention (academic detailing). Tracking screening and followup for outcome assessment purposes began and was conducted concurrently with the tracking performed for step 5 of the intervention. However, it continued past the end of the intervention period to allow us to identify screening and followup that occurred during the intervention period but did not show up in electronic records until after the close of that period. Assessment data collection also included postintervention surveys and focus groups, as well as chart audits.
We developed a dissemination plan between the beginning of June and the end of August 2009. We developed draft dissemination material between the beginning of August and the end of October 2009 and finalized it between May and July 2010. We conducted some dissemination activities before preparing this report in July 2010 and planned to continue these activities for several months past the formal period of performance of the task order.
The System Approach to Tracking and Increasing Screening for Population Health Improvement of Colorectal Cancer (SATIS-PHI/CRC) intervention seeks to (1) influence the behavior of primary care providers and their patients regarding CRC screening and followup through targeted communications and (2) facilitate the screening and followup process through improved eligibility identification and screening tracking systems. Figure 1.2 presents the framework for the six steps of SATIS-PHI/CRC.
Step 1 brings primary care practices and their patients into the intervention. Step 2 seeks to influence the screening knowledge and behavior of providers within those practices and, along with Step 6, to influence followup knowledge and behavior as well. By educating and influencing providers, Step 2 also seeks to ensure that providers will influence the screening behavior of their patients. Step 4 more directly seeks to influence patient screening.
The remaining steps facilitate the process. Step 3 identifies patients who are eligible (based on prevailing screening guidelines) to receive the Step 4 screening materials. Step 5 tracks patient screening and results. Those patients with no evidence of being screened receive a reminder Step 4 mailing whereas the practices of patients with evidence of screening are notified of screening results and receive feedback regarding recommended followup.
The intervention is intended to be conducted by a central entity, such as a health care delivery system or insurer, affiliated with a network of primary care practices on behalf of and in conjunction with those practices. Because the central entity will contact patients on behalf of practices participating in the intervention, the first step is to recruit practices to participate in the intervention and obtain the consent of all clinicians in each practice to represent them and to contact their patients.
The central entity then conducts academic detailing at each participating practice to bring clinicians and staff up to date on current screening and followup guidelines, inform them about the screening information and materials the central entity will send to their patients, and ask them to support the intervention effort by encouraging their patients to respond positively to the invitation to be screened. (The central entity can also conduct an optional survey of the practices to ascertain baseline knowledge and behavior prior to the academic detailing session in order to better tailor the session to the practices.)
During Step 3, the central entity accesses and reviews electronic billing, claims, and medical records available through participating practices or cooperating insurance plans. This review is used to identify patients who appear to be eligible (meet the guideline-based criteria) for screening by age, prior screenings, and personal and family medical history. The central entity then mails those who appear to be eligible an introductory letter with a screening eligibility assessment (SEA) form for patients to fill out to confirm their eligibility or identify themselves as not eligible.
At the central entity's discretion, the SEA form can include an option for patients to opt out of receiving further information or materials. Those who continue to be eligible (and do not opt out) receive a second mailing with information about CRC and various screening modalities supported by the intervention at the central entity's discretion (we elected to support at-home stool testing by fecal immunochemical testing [FIT] and colonoscopy). If an at-home screening test is part of the intervention, this mailing also includes either a test kit or a mechanism, such as a mail-in request card, to request one.
The central entity then tracks electronic records to identify who does and does not get screened (Step 5). After allowing an appropriate time to elapse, the central entity mails reminders to nonscreeners (this cycle can be repeated more than once at the central entity's discretion). Step 5 tracking continues to identify results of screening tests. The central entity then provides feedback (Step 6) to practices and their clinicians regarding results for their patients and recommended followup procedures for negative (normal) and positive (abnormal) or inconclusive results.
Scope and Outline of the Report
The stated purpose of this ACTION task order project was "to design, implement, assess, and disseminate a redesign of important health care delivery system processes in CRC screening in order to increase their efficiency while sustaining or improving their value to patients" (quoted from the Request for Task Order for this project released on July 3, 2007). The scope of this report covers all aspects of this purpose. In it, we describe (1) the SATIS-PHI/CRC intervention we designed and implemented to improve CRC screening and followup, (2) our experience and the lessons learned from implementing it, (3) our assessment of it, and (4) our current and planned dissemination activities to encourage the spread and uptake of the intervention.
A major emphasis of this report is the intervention assessment. We devote a section to describing our assessment approach and design and another section to reporting our assessment findings. The scope of the assessment is broader than simply evaluating the effect or outcome of the intervention. It also encompasses an assessment of the implementation of the intervention. Since the intent of this project is to learn about the intervention implementation process as well as the intervention's effect, the assessment reflects this intent by assessing both process and outcome.
We previously submitted a series of deliverables under this ACTION task order contract that (1) updated the environmental scan we provided in our proposal for this project, (2) outlined our goals and objectives for our health system intervention approach to improve CRC screening and tracking, (3) presented our plan for implementing this intervention for primary care practices affiliated with the LVPHO, (4) described our plan for assessing the implementation process and its outcomes, (5) delineated our dissemination plan for facilitating the spread and uptake of the intervention, (6) presented a preliminary report of our work under this contract, and (7) provided our draft dissemination products and tools. This current report further documents our work and presents our overall experience and findings related to our implementation of the intervention and our assessment of its outcome. It draws on, refers to, and occasionally summarizes information contained in the previous deliverables but primarily provides information on our more recent work and findings.
At the time of our preliminary report in early October 2009, we had (1) fully completed Steps 1 through 3 of the intervention, (2) completed the initial and followup mailings of Step 4 but were continuing to mail stool test kits to patients who requested one, (3) completed an initial round of Step 5 tracking and were continuing to track screening and followup, and (4) had just begun Step 6 feedback to practices (see Figure 1.1 for a timeline of the implementation of the intervention and its assessment). As Figure 1.1 indicates, we continued to send requested stool test kits, track screening and followup, and provide feedback through February 2010. We began our postintervention data collection for the assessment in February 2010 and continued to collect assessment data through the end of April. The scope of this report, then, describes our experience implementing the full intervention, our methods for assessing the intervention, our assessment findings, and our dissemination plan and activities.
The following section of this report presents a detailed description of the SATIS-PHI/CRC intervention, including a rationale for the intervention, the role of a central entity to implement it, and descriptions of its components, our experience implementing them, and lessons learned from that experience that could help others to adopt and implement SATIS-PHI/CRC. The next section then presents our assessment plan and methodology. It first introduces the assessment framework we adopted (the Practical, Robust Implementation and Sustainability Model, or PRISM) and then describes our assessment research design. Finally, it describes our sources of data for the assessment, our assessment outcome measures, and the patient and provider attribute data we used in our assessment of the intervention. That section is followed by our assessment findings. Following the PRISM framework, these findings include an assessment of the context in which we implemented the intervention as well as an evaluation of the implementation process and the intervention's outcome.
We then turn to a discussion of our dissemination activities, including (1) a review of our dissemination plan, (2) a description of the contents of our intervention toolkit, and (3) a discussion of our recent, ongoing, and planned dissemination activities. We end this report with a Conclusions section in which we summarize our assessment results and our lessons learned regarding implementing the intervention and then discuss the transferability of the intervention to other system settings. In particular, based on our assessment and lessons learned, we identify the conditions and attributes of central entities and practices that we believe are needed for a successful adoption and implementation of the SATIS-PHI/CRC intervention.