Colorectal cancer (CRC) screening is instrumental in reducing CRC-related mortality [1]. Until 2021, screening for CRC was recommended starting at age 50; however, the United States Preventive Services Task Force (USPSTF) lowered the recommendation to age 45 [2]. For people ages 45–75, the most recent data from the 2021 National Health Interview Survey show that 58.7% were up to date with CRC screening [3]. CRC screening rates vary by race and ethnicity, age group, insurance status, as well as education, income, and location of care [3, 4].

Stool-based tests are one type of CRC screening modality. These include fecal occult blood tests (FOBT) and fecal immunochemical tests (FIT), both of which need to be completed annually according to the USPSTF guidelines [2]. The multi-target stool DNA test (Cologuard; FIT-DNA) is recommended to be completed every 1–3 years [5]. Because of barriers, such as long wait times for screening colonoscopy appointments, travel, distance to providers, and bowel prep [6-8], stool-based tests may be a more readily available screening option. Mailed FIT programs have also been shown to be effective in overcoming CRC screening barriers [9].

Stool-based tests require timely follow-up with colonoscopy (“follow-up colonoscopy”) to complete the screening process if the initial stool test is abnormal [2]. A recent systematic review showed that the time interval should be no longer than 9 months, as the incidence of CRC and advanced stage diagnosis increases with time between stool test and follow-up colonoscopy [10]. However, the percentage of people completing follow-up ranges widely; studies have indicated that between 15.4 and 95.2% of patients complete a follow-up colonoscopy within 6 months [11-15].

Many health systems and clinics, including FQHCs, have challenges in identifying patients eligible for CRC screening and tracking results. Systems and clinics may be unable to track patients who need follow-up colonoscopies because their electronic medical record system (EMR) is not configured to track or link abnormal stool-based tests in a systematic way that triggers a clinician’s referral for a follow-up colonoscopy [16]. Patients need to visit either a hospital or endoscopy center for the follow-up appointment, which facility (i.e., clinic or procedure center) takes on the responsibility for making sure the patient has an appointment is often unclear. Further, these external facilities may not use the same EMR as the clinic or have a health information exchange in place, so there is no seamless electronic linkage between systems [17]. When any of these happen, it becomes much more labor intensive for clinics to monitor status of appointments, as well as obtain and document results of follow-up colonoscopies.

The Centers for Disease Control and Prevention funds the Colorectal Cancer Program (CRCCP). The CRCCP currently consists of 35 award recipients across the USA to promote CRC screening for age-eligible and uninsured or underinsured patients through the implementation of evidence-based interventions (e.g., patient and provider reminders, provider assessment and feedback, activities to reduce structural barriers, and patient navigation) [18]. The award recipients are made up of 20 state health departments, 8 universities, 2 tribal organizations, and 5 other types of organizations (e.g., an FQHC, a hospital), who often partner with FQHCs. In this paper, we report on the extent to which a sample of health systems, that award recipients’ partner with, provide support along the continuum of CRC screening to facilitate the completion of the screening episode. We also summarize the challenges and identify strategies that health systems can use to mitigate barriers in tracking data and improving completion of recommended follow-up colonoscopies.

The eight participating award recipients in this study are a part of the CRCCP Learning Collaborative, a subset of 21 CRCCP award recipients that assess the implementation, effectiveness, and cost of interventions [19]. Members of the Learning Collaborative participate in CDC CRCCP special studies and provide data to generate evidence based on their real-world practices. The eight participating award recipients in this study were selected through a purposive sampling process, using elements that included geographic diversity (and included all U.S. Census Regions) and ability to report data on tracking stool-based test screenings and their outcomes.

We introduced the study to the eight award recipients on videoconference calls in 2023 to gauge their interest in taking part. We provided data collection spreadsheets that the award recipients completed on behalf of their health systems and returned via email. Award recipients were asked to select representative health systems from among those who were participating in the CRCCP. The one inclusion criterion was that the health system had to have the ability to track colorectal cancer screening at some level and provide the requested data. Award recipients provided the sociodemographic characteristics of the CRC screening-eligible populations at the health systems. We report details on sex, race/ethnicity, and the proportion uninsured. We also report the overall screening uptake. Recipients also provided screening and follow-up colonoscopy outcomes data, and we calculated stool test return rates, percentage of abnormal FIT results, CRC screening uptake, and percentage of follow-up colonoscopies completed. We defined stool tests as the sum of FOBT, FIT, and FIT-DNA tests. The data were analyzed and managed in Microsoft Excel.

The reporting periods used by the award recipients in most instances were the same for demographics and outcome measures. However, three award recipients reported different time periods for demographics and outcome measures because of the lag in implementation. The years for reporting demographics ranged from 2019 to 2022 and for outcomes ranged from 2020 to 2022. The outcome measures were reported for their baseline periods, which we defined as the time immediately prior to the introduction of any strategies to improve the reporting of screening measures.

We also examined whether award recipients were able to track patients at various stages along the CRC screening continuum and what processes health systems had in place to track patients with abnormal stool tests through follow-up colonoscopies. Although the main focus of our assessment was on follow-up colonoscopy, we included the full screening episode to assess the overall ability of health systems to track screening data. First, if the data collection had flaws in the initial screening steps, then the number of patients identified with abnormal stool test findings may not be accurate. Second, similar data issues may impact quality of information available along more than one step in the screening continuum and, thus, these issues may have to be addressed in tandem. Data are reported at the award recipient level and an average was calculated when multiple health systems were included for an award recipient.

Additionally, we conducted in-depth interviews by videoconference with stakeholders from each award recipient to learn more about their current processes as well as the barriers and challenges the health systems faced in tracking follow-up colonoscopies. The stakeholders were the CRCCP program directors or program managers for award recipients and their backgrounds included nursing, epidemiology, and evaluation. Questions included

The calls lasted approximately 30 min. We later conducted site visits during the spring of 2023 to five award recipients and some of their partner health systems, which were a convenience sample, where we further discussed the process the health systems used to track colonoscopies after abnormal stool test results. We maintained detailed notes of the interviews and site visit conversations that were maintained in Microsoft Word. These notes were reviewed by staff at Implenomics to derive key themes pertaining to barriers and potential solutions. The data gathered through these sources also provided additional contextual knowledge to interpret the results. Institutional Review Board approval was not required because it was determined this study was not research involving human subjects.

In this study, we report on the experiences of selected CRCCP award recipients and health systems in tracking stool-based screening (defined as FIT, FOBT, and FIT-DNA), with specific focus on follow-up colonoscopies after abnormal stool testing. During this baseline assessment, we found that only two of the award recipients reported that all their participating health systems could consistently track the number of stool-based tests distributed. In general, most of the health systems were able to document the stool tests that were returned, although three award recipients reported that some of their health systems were not able to track consistently. Overall, 75% (6 out of 8) of the award recipients reported that the health systems were able to document abnormal findings. These gaps may hinder the ability of health systems to monitor screening to maximize stool test returns and to accurately generate the proportion of abnormal findings, which is required for assessing test performance and informing quality assurance processes [20, 21]. Furthermore, only six of the eight award recipients reported that all participating health systems were able to report referrals and overall completion of follow-up colonoscopies. Importantly, only two award recipients indicated that their partners could report follow-up colonoscopy completion within specified timeframes, such as 6 months from date of abnormal test results. The ability to track timely completion of follow-up colonoscopy is essential, as studies have shown that longer time intervals can result in more advanced disease [10, 22, 23].

These findings highlight the importance of implementing approaches to improve tracking along the screening continuum. Health systems participating in the CRCCP are implementing and testing a range of approaches to improve tracking and follow-up of abnormal stool tests. There is limited evidence on the optimal approach to enhance or supplement EMRs [24], and we found that most health systems are incorporating approaches to track stool-based screenings as well as follow-up colonoscopies using additional data tracking tools, such as REDCap and Excel. Participants at CDC’s 2019 Mailed FIT summit also spoke to the importance of data infrastructure [9] and developed a mailed FIT implementation guide [25] with information on managing and tracking mailed FITs, which is also instructive for tracking abnormal FITs. Furthermore, the strategies highlighted by participants involve both improvement in data entry and the ability to capture the existing information for decision-making. Health systems are therefore looking beyond technological enhancements. Importantly, training and dedicated staff time were reported as essential to ensure accurate and complete data to track stool-based screening and follow-up colonoscopies.

We found wide variation in the screening process measures among the award recipients in this study whose partner health systems were able to report details on the steps along the screening continuum. On average, 12.3% (6.9 to 16.8%) had abnormal findings and the referral rate was an average of 80.9% (70.9 to 89.8%). Post-analysis conversations with study participants indicated that potential reasons why referrals were not provided may include evidence of a recent colonoscopy in the patients’ records and patient health status. The uptake of follow-up colonoscopy ranged from 24.2 to 75.5% with an average of 47.9%. Colonoscopy completion within 6 months of referral was lower with an average of 39.8%. Similarly, low follow-up colonoscopy uptake from 18 to 57% have been reported at FQHCs [12, 13], which are below the 80% target recommended by the Multi-Society Task Force on Colorectal Cancer [26]. Only one health system in our study was able to report a follow-up uptake of 75.5%. However, an evaluation of the CRCCP when the program first began indicated that, when quality measures were in place with systems in place to track and monitor follow-up, 82.9% of people with an abnormal blood stool test completed colonoscopy [11].

Additional research is required to explore the variability reported in this study and to identify optimal strategies to track follow-up colonoscopy completion rates as well as individual, provider, health system, and community-level strategies to support completion of follow-up colonoscopies. As shown above, individual health systems have implemented strategies regarding use of EMR and health population tools; however, there is no one system, as reported by respondents, which tracks individuals along the CRC screening spectrum. Further, the volume of patients served could be an important factor. On the one hand, large health systems may have more resources to implement strategies to improve data tracking and follow-up colonoscopy completion. On the other hand, large health systems may have a greater number of uninsured patients with abnormal stool test findings for whom they need to identify providers who can offer free colonoscopies or identify a payment source for the colonoscopies. And, rural locations and geographic distribution of colonoscopy capacity could also impact follow-up colonoscopy completion rates [27, 28]. Prior studies have indicated that multi-level interventions can likely support individual, provider, health system, and community-level strategies to improve adherence to follow-up colonoscopy [29-35]. In addition, research is required to expand the age range of these analyses to include the age group 45–49.

There are a few limitations that should be considered in interpreting the study findings. First, we included only a limited number of health systems, which were purposively sampled. Therefore, the findings may not be generalizable to community health centers with different context factors, including patient language preferences, payer mix, and relationships with providers to perform colonoscopies. There may also be selection bias in the health systems chosen by the award recipients. Second, as we did not conduct detailed chart reviews, there could potentially be inaccuracies in the baseline data reported, as health systems were just initiating data quality improvement activities via their participation in CRCCP. Third, there were differences in the annual periods for which the data were reported and, in some instances, the COVID-19 pandemic may have impacted the stool-based screening process and follow-up colonoscopy completion. Fourth, there was substantial variability in the tracking process and outcomes reported; hence, the means reported may be skewed. We have reported the range for each measure to accurately reflect the variation among the participants. Lastly, only a few award recipients reported on FIT-DNA and therefore, we combined all stool test return rates. The return rates may vary as FIT-DNA patients receive patient navigation services from Cologuard, but many health systems also offered various interventions to promote stool test returns. Any patient with abnormal findings though received similar support services at a given health system for completion of follow-up colonoscopies.

Findings from this study offer important lessons for tracking and timely completion of follow-up colonoscopies after abnormal stool tests to complete the screening episode. The health systems in this study that the CDC award recipients partnered with were often unable to track the entire stool-based screening episode accurately and therefore could use support to enhance their EMRs, implement additional tracking tools, and train providers. Furthermore, these health systems could be aided by guidance on evidence-based strategies that can be implemented to improve tracking and increase completion of follow-up colonoscopies after abnormal stool tests. More active engagement by health centers providing follow-up colonoscopies in communicating appointments and test results could support patients’ adherence to receiving follow-up colonoscopies. Improvements in data tracking and screening completion may also help to improve health equity and reduce mortality from CRC.