The goal of this interventional study is to create and test a comprehensive and low burden text message program within existing Community Health Centers (CHC) electronic records system to encourage patients to complete at home colorectal cancer (CRC) screening and to make sure they get follow-up care if their results are abnormal.

First, the investigators will learn from clinic staff and patients what their needs and preferences are in terms of use of technology. This information will be used to design the text messages program.

• Clinic patients will participate in focus groups
• Clinic staff will participate in interviews

Second, the investigators will test a series of different message versions in two batches (experiments):

• Clinic patients with orders for an at-home colorectal cancer screening kit will receive the different message versions.
• In the first batch, the messages that get the most engagement from patients will be selected to be used in the second experiment.
• In the second batch, the investigators will test which messages lead to the most colorectal cancer screening completion.
• This will be rolled-out within the clinics existing electronic record system. The study team will not receive any information that will identify individual patients.

Lastly, the investigators will check again with clinic staff to learn how the program performed, and what would be needed to continue using the text message program in the long run.

-Clinic staff will participate in interviews and surveys.

This trial is studying optimization of text messages and outreach approaches by community health centers to increase patients' completion of FIT colon cancer screening tests. Text message reminders are a low cost, scalable way to deliver patient reminders that are low burden for staff. The proposed work uses a creative, community-engaged, systems-level approach to improve completion of FIT testing and abnormal follow-up and program sustainment in CHCs.

The study design includes three parts. First, the text messages will be optimized to maximize patient engagement and outreach (Aims 1 & 2). The study will enroll patients from three partnering CHCs to participate in six focus groups to inform message design. The investigators will also conduct key informant interviews with 5-15 staff per CHC regarding the logistics of ensuring seamless integration into routine clinical operations and perspectives on the intervention activities. Then, a factorial experiment will be conducted, assigning 2,000 patients who have received a stool-based screening kit to different reminder messages. Engagement will be measured as replies or use of ≥1 links, to indicate interaction with the message. The experiment will compare: (1) personalization; (2) concern or negative emotions about screening; and (3) gain framing. In the factorial design patients will not be individually enrolled, this intervention will be tested within the workflows of usual practice. All data for the factorial experiment will be de-identified.

With those tools in place, the investigators will optimize outreach strategies for CRC screening completion. In a second factorial experiment with 5,100 patients, the investigators will optimize outreach strategy for primary screening outcomes of home-based stool screening test completion and screening completion by any guideline-recommended CRC screening test. Secondary outcomes include follow-up care for abnormal stool test results. In this factorial experiment, the investigators will test the impact of three outreach conditions on screening completion: increasing patient autonomy, providing information for accessing resources for impediments to care, and provision of live support. As in the first factorial test, for the second factorial test patients will not be individually enrolled, as the intervention will be tested within the workflows of usual practice. All data will be de-identified.

Finally, the study will assess implementation outcomes and sustainment of intervention components and effects. The investigators will conduct staff interviews and surveys to assess organizational capacity for sustaining the intervention elements being tested as well as the acceptability and feasibility of the text messages and screening tools. Sustainment of the text message outreach intervention elements will be assessed at post intervention and two 6-month intervals post-intervention.

Eligibility Criteria for CHC Patients Cohorts

Inclusion Criteria (CHC Patients):

• Adults 45-75 years old
• Receive care in partner CHCs
• Are of average risk for colorectal cancer
• Received a stool-based testing order at their CHC
• Have a mobile phone and can receive text messages (Aim 1 quantitative cohort only)
• Fluent in English, Spanish, or another predominant language to be identified by the PI (applicable to patients focus groups cohort only)

Exclusion Criteria (CHC Patients):

• Personal or family history of CRC or colorectal polyps
• History of inflammatory bowel disease
• Heritable conditions that put them at above average risk for colorectal cancer (e.g., familial adenomatous polyposis)

Eligibility Criteria for CHC Staff Cohorts Inclusion Criteria (CHC Staff)

• Adult employees of participating CHCs
• Job responsibilities include activities related to CRC screening

Exclusion Criteria (CHC Staff)

• None

All EHR derived data or data on text message intervention interactions, and any descriptive quality improvement data that we receive from the CHCs are de-identified. If qualitative data is shared, transcripts would be stripped of all identifiers.

The data will be available when the project is complete. The data will be maintained for 7 years on our servers, per our standard practice.

The primary restriction will be approval of data use requests by the participating community health centers' leadership. We will share the data with potential users who have a legitimate and appropriate use for data of this nature. Because the data is owned by the participating health centers, data use requests will be reviewed and approved by the health centers' leadership committee. Since the data is owned by the participating community health centers, a data use agreement will be created to allow for use and analysis by the collaborating researchers. All data use requests will be made to Dr. Kruse, who will discuss them with the study team and with CHC leadership, before any data use agreements are made and data is released.