Relationship between Social Risks and Colorectal Cancer Screening/Surveillance in a Large US Health System

<h3>Context</h3> Social determinants of health (SDOH) such as access to care and economic stability negatively impact colorectal cancer (CRC) screening rates. Implementation of electronic health record SDOH questionnaires is more common in health care institutions and allows for individual-level assessment of social risk in a primary care setting. <h3>Objective</h3> To determine the association of patient reported social risk on CRC screening/surveillance rates in a large US primary care population <h3>Study Design and Analysis</h3> Cross-sectional analysis comparing adult patients age 50-75 who have completed colorectal screening/surveillance to those who have not <h3>Setting</h3> Mayo Clinic Midwest adult patients with primary care clinician in Department of Family Medicine in 2022 <h3>Population Studied</h3> Mayo Midwest patients eligible for CRC screening/surveillance age 50-75 <h3>Instrument</h3> Demographic data including age, gender, and race and social risk questionnaire data (housing risk, food insecurity, financial resource strain, transportation risk, and intimate partner violence) were collected from electronic health record. <h3>Outcome Measures</h3> CRC screening metric is met if colonoscopy in last 10 years, flexible sigmoidoscopy in last 5 years, Cologuard in last 3 years or FIT testing in last year; CRC surveillance metric is met if colonoscopy performed in recommended time interval for patients with family history colon cancer or previous precancerous polyp. <h3>Results</h3> Among 159,491 adults age 50-75 eligible for CRC screening/surveillance, younger, male, non-white patients were less likely to be up to date on CRC testing. Using a multiple logistic regression model controlling for demographic factors, patients who gave high risk answers to questionnaires about housing risk (OR 0.58), financial risk (OR 0.46), food insecurity (OR 0.50), transportation risk (OR 0.42) and intimate partner violence (OR 0.70) were all less likely to be screened or surveilled for CRC compared to patients giving low risk answers (all p-values &lt;0.001). <h3>Conclusions</h3> All 5 patient reported social risk domains collected in electronic health record questionnaires were associated with lower likelihood of completion of CRC screening or surveillance. Significant differences may exist among screening and surveillance populations. Further study regarding the use of social risk questionnaires and interventions to improve CRC screening/surveillance rates in primary care is needed.