Abstract C79: Use of social network analysis to examine racial disparities in screening colonoscopy referral

Background: In Illinois for the year 2016, colorectal cancer (CRC) is projected to cause 2,030 deaths, making it the leading cause of non-tobacco related cancer mortality. African American (AA) Illinois residents have an approximately 7% greater incidence and a 30% higher mortality rate when compared to white residents. Guideline consistent routine screening is known to reduce death related to CRC. Screening colonoscopy is the most commonly performed CRC screening test. However, African Americans are known to have lower rates of both screening and diagnostic colonoscopy completion compared to whites. Chicago9s South Side includes 34 of the city9s 77 recognized community areas and is home to a combined population of more than 800,000 residents, of whom greater than 75% are African American. The contiguous communities that make up the South Side cover approximately 60% of Chicago9s land area. There are 41 facilities throughout Chicago that perform outpatient screening colonoscopy, of with 13 (32%) are located on Chicago9s South Side. The University of Chicago Medical Center (UCMC) is the largest healthcare provider on the South Side and represents 30% of the local capacity for screening colonoscopy services. Objective: The objectives of this study were to use social network analysis (SNA) to explore the structure of the screening colonoscopy referral network utilized by community-based primary care physicians for colonoscopies that were completed at the UCMC during calendar years 2013 and 2014. Methods: A retrospective chart review was performed for all individuals who completed a screening colonoscopy during years 2013 and 2014 at the UCMC. Screening colonoscopies were identified using CPT codes: G0105 and G0121. For each screening colonoscopy procedure, the referring PCP and colonoscopy provider (colonoscopist) were identified from screening colonoscopy procedure reports. A referral-tie was assigned between a PCP and colonoscopist, if the PCP9s name was listed as the referring physician on the procedure report and the colonoscopist9s name was listed as the performing physician. SNA was performed using UCINet for windows and network mapping was performed using Netdraw. Descriptive statistics were performed using STATA 13.1. Results: 405 outpatient screening colonoscopies were completed during the study interval (mean age, 58.4 ± 10 years; 62% female; 40% AA, 44% white). 60% of patients were privately insured, 31% Medicare, and 10% were public insurance or uninsured. A total of 261 individual community PCPs were identified as referring physicians, with a mean frequency of referrals/PCP of 1.18 ± 0.85 and a range of 1 referral up to 25 referrals. Following a SNA of PCP to colonoscopist referral ties, the resulting referral network exhibited a core/periphery structure. 13 PCPs (5%) belonged to the network core and 248 (95%) belonged to the network periphery. PCPs who belonged to the network core had a greater mean referral rate compared to the periphery (3.77 vs. 1.05 referrals; p Conclusions: The screening colonoscopy referral network at the UCMC exhibits a core/periphery structure. Amongst AA patients, screening colonoscopy referrals were disproportionately clustered within the network core amongst a highly inter-connected group of community-based PCPs. This variation in referral patterns indicates that patient characteristics, such as race, may affect the structure of the screening colonoscopy referral network. These variations in the structure of the referral network may impact PCP referral behavior and subsequently patient recommendation for and utilization of CRC screening. Screening colonoscopy referral networks may represent an under-recognized determinant of racial disparities in CRC screening. Citation Format: Keith B. Naylor, Olufemi Kassim, Karen E. Kim, John Schneider. Use of social network analysis to examine racial disparities in screening colonoscopy referral. [abstract]. In: Proceedings of the Ninth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2016 Sep 25-28; Fort Lauderdale, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(2 Suppl):Abstract nr C79.