Colonoscopy Screening: Setting Epic Follow-up Times

1. Background and Significance Routine cancer screening reduces cancer-specific mortality and is recommended by the US Preventive Services Task Force (USPSTF) and other national guidelines for breast, cervical, colorectal, and lung cancer. Many research and quality improvement studies focus on the timely completion of cancer screening as a discrete event, but it is a process that requires repetition of screening over the life course and when surveillance is required because of prior abnormalities. Though much progress has been made in advancing the use of cancer screening, timely screening and surveillance of any abnormal screening result is critical to realizing the maximal benefits of screening but often is not achieved. These challenges are especially true for colorectal cancer (CRC) screening -the second leading cause of cancer-related deaths in the US. Responsibility for cancer screening and follow-up falls to the ordering provider, typically a primary care practitioner (PCP). Though electronic health records (EHRs) may have reminders for routine screening results with an option to manually edit modifiers for short interval screening, few PCPs or their practices have integrated management systems to track abnormalities and manage follow-up. The transition from routine screening to follow-up surveillance colonoscopy (FSC) includes responsibility for both PCPs and gastrointestinal (GI) specialists. Failure to receive an appropriate FSC undermines the benefits of screening, violates the trust that patients place in their providers and health systems, and increases medicolegal risk. This proposal seeks to address two key components of ensuring appropriate FSC needed for any systematic safety program - improving the recommended follow-up interval that is based upon current guidelines and ensuring that the follow-up interval specified in the EHR is kept up to date using the most recent colonoscopy results. The current system requires GI specialists to perform these tasks without any decision support. This proposal seeks to automate updating the follow-up interval specified in the EHR after a colonoscopy and to develop the framework of agreement among GI specialists for determining a follow-up interval based upon all patient factors.
2. General Description of Study Design Working with MGB digital health, the investigators will implement a new colonoscopy results letter with a structured field for recommended interval follow-up that automatically updates the Epic EHR reminder used by patients and PCPs. The investigators will then conduct a trial of MGH primary care patients aged 45-75 undergoing screening and surveillance colonoscopy and randomize the performing GI specialist to use the new protocol to set the FSC interval (intervention) or to usual care (control.)
3. Subject Selection Subjects are GI specialists who perform colonoscopies for colorectal cancer screening at MGH. Patients of participating GI specialists meeting the following criteria will be included. Patients seen in an MGH-affiliated primary care practice who are English or Spanish-speaking patients aged 45-75 and have had a colonoscopy at MGH. Patients undergoing a diagnostic colonoscopy, who have a history of inflammatory bowel disease, a hereditary colon condition, or already have a history of colon cancer will be excluded. Primary care patients at MGH will be retrospectively identified after undergoing a colonoscopy by a participating GI specialist. Patients will be identified using the Research Patient Data Repository (RPDR) and the MGB Enterprise Data Warehouse (EDW) which maintains a prospective cohort of current patients, their primary care doctor (staff and resident physicians), and the practice they are seen in.
4. Subject Enrollment Enrollment will involve recruiting GI specialists at MGH. GI specialists will be informed about the study by GI Leadership. GI specialists who perform screening colonoscopies will be invited to participate. GI specialists who respond that they are interested in participating will be included. Patients will not be directly enrolled in the trial since the intervention is directed at the provider performing the colonoscopy. Specifically, the investigators will enroll interested GI specialists at MGH who perform screening and surveillance colonoscopies. GI specialists will perform colonoscopies as part of routine care and are not required to recruit individual patients for the trial.
5. Study Procedures

This specific aim will develop, implement, and evaluate a new structured result letter template to be used by intervention specialists as part of routine care in sending patient result letters following a screening or surveillance colonoscopy. Currently, the GI specialist includes in the letter a recommended follow-up interval that may range from a few months to 10 years. This recommendation is based upon the GI specialist's understanding of a patient's personal and family history of colon polyps or cancer, the quality of the cleanout preparation, and the results of the colonoscopy and any pathological findings. The GI specialist is also supposed to separately update the patient's colonoscopy Health Maintenance (HM) topic modifier to reflect the recommended follow-up interval specified in the letter. The innovation will be the development of a structured result field that will take the recommendation embedded in the result letter and compare it with that of the current HM modifier. If the recommended follow-up is identical, no changes will be made. However, if there are differences, the HM topic modifier will be updated using the follow-up interval selected in the result letter.

Development of Structured Result Template Letter: The investigators will create new functionality in Epic that will allow a user to choose an option from a SmartList and simultaneously update the relevant Health Maintenance (HM) topic modifiers for the colonoscopy topic. This SmartList will be made available to clinicians to embed in their result letter templates. For example, a GI specialist writing a letter to their patient after viewing biopsy results would be able to open their standard letter template and at the bottom indicate that the patient should have a repeat colonoscopy in 3 years. When they select "3 Years" from the SmartList, this will simultaneously update the patient's HM topic modifier to let all clinicians (and the patient) know that they are again due for colonoscopy in three years' time. This HM topic modifier feeds into other clinical decision support in the EHR to help ensure these recommendations are followed up on.

Trial Design: Cluster randomized trial with the MGH GI specialist performing screening as the unit of randomization.

Provider Randomization: GI specialists will be randomized in a 1:1 ratio to intervention or usual care groups with stratification based upon the number of colonoscopies performed, match rate of follow up recommendations noted in the letter and health maintenance topic modifier, and gender.

Intervention Protocol: GI specialists randomly allocated within the intervention will each have an individual meeting either in person or by teleconference with study personnel. This meeting is to review the use of the new electronic link (SmartLink) as part of the structured result template letter. As part of this meeting, the GI specialist's current result letter templates will be identified that need to be updated with the SmartLink). If the GI specialist is interested, standardized letters that have been previously developed and disseminated by the MGH gastroenterology service will be offered to each GI specialist if interested.

Once the GI specialist has agreed to the letter templates they wish to use, the new letter templates will be loaded into each specialist's Epic SmartPhrase Manager. Note that the content of the result letter is not dictated by the study and is at the discretion of the GI specialist as part of her/his usual care. The use of the new structured result template letter will be reviewed and demonstrated with a test patient within the Epic test environment. Questions will be answered, and written instructions will be provided for review. Any handouts provided to the GI specialist will be submitted to the IRB for review and approval prior to use. The GI specialist can also contact research staff for assistance. Following the instructional meeting, the GI specialist will be advised that she/he can begin using the template immediately as part of routine care. All further research-related activities will occur electronically and involve data extraction as described below.

Data sources: Primary care patients at MGH will be identified using the MGB Enterprise Data Warehouse (EDW) that maintains a prospective cohort of current patients, their primary care doctor (staff and resident physicians), and the practice they are seen in.

Primary Outcome Measures:

Primary: The primary analysis is intention-to-treat (ITT), including all eligible patients. Agreement (yes/no) will be assessed between the follow-up interval specified in the patient result letter and the colonoscopy health maintenance topic modifier.