P0277 - Family History Screening Tool Reveals Gaps in Colorectal Cancer Screening and Surveillance

Erin J. Song, MD¹, Brendon O'Connell, MD¹, Brian Sullivan, MD, MHS¹, Rachel A. Myers, PhD², Tejinder K. Rakhra-Burris, ², Alice Parish, MSPH¹, Donna Niedzwiecki, PhD³, Ryanne Wu, MD¹, Lori A. Orlando, MD, MHS, MMCI⁴, Katherine Garman, MD, MHS^1
1Duke University, Durham, NC; ²Duke Center for Applied Genomics & Precision Medicine, Durham, NC; ³Duke University School of Medicine, Durham, NC; ⁴Duke University Medical Center, Durham, NC

Introduction: Colorectal cancer (CRC) screening guidelines rely on family history (FH) to determine the age to begin screening. Yet, many health systems lack validated FH tools to help identify these high risk patients. We compared rates of colonoscopy, age of index colonoscopy, and surveillance recommendations in patients with a FH of GI malignancy. Our aim was to evaluate whether implementation of an electronic FH tool in primary care clinics could improve identification of patients who would benefit from referral for earlier screening colonoscopy.


Methods: We performed a secondary analysis of the IGNITE trial, a multi-site study designed to evaluate MeTree (a patient-facing, web-based, FH-driven clinical decision support tool). Age, sex, race, ethnicity, insurance status, tobacco use, number of affected family members, and age at CRC diagnosis were collected. In patients with and without a FH of CRC in a first degree relative (FDR), we compared rates of colonoscopy, age at initial colonoscopy, and post-colonoscopy surveillance recommendations; analysis included t-test and chi-square test.


Results: Of 1,416 IGNITE participants, 276 reported FH of CRC, including 124 with a FDR with CRC. Participants were predominately white (89%), female (73%), and insured (95%). Colonoscopy rates for those with a FDR with CRC were higher compared to those without (92.7% vs. 77.1%, p< 0.01, Table 1). Of those who underwent colonoscopy, mean age at first recorded colonoscopy for those with a FDR with CRC was notably older than age 40 at 54.5 years. Importantly, there was no significant difference compared to those without a FDR with CRC at 53.2 years (p=0.24). After the initial colonoscopy, surveillance recommendations were documented (94.8%) and consistent with guidelines (97.0%) in those with or without a FDR with CRC.


Discussion: In participants of the IGNITE trial, 92.7% of those with a FDR with CRC had been referred for colonoscopy, but screening was often done later than recommended age of 40 or younger. However, after completion of the initial colonoscopy, accurate risk-based surveillance intervals were provided in most cases. Our data highlight the challenge in primary care of identifying high risk patients early enough to start CRC screening at the appropriate age. Successful integration of clinical decision support tools with an emphasis on FH in the primary care setting may help close this gap, which would be an important advance given the rising incidence of young onset CRC.

	First degree relative with CRC	No first degree relative with CRC	p-value	Total
Ever had colonoscopy	n=124	n=284	< 0.01(1)	n=408
No	9 (7.3%)	65 (22.9%)		74 (18.2%)
Yes	115 (92.7%)	219 (77.1%)		334 (81.8%)
Mean age at first colonoscopy	n=112	n=214	0.24 (2)	n=326
Mean	54.5	53.2		53.7
Range	25.5-82.7	16.7-77.0		16.7-82.8
Median (Q1, Q3)	54.3 (49.8,60.0)	52.5 (50.1,59.6)		53.1 (50.0,59.9)