S628 Disparities in Colonoscopy Utilization After the COVID-19 Pandemic

Abstract

Introduction: In the first 3 months of the COVID-19 pandemic, our institution in San Francisco like many others limited routine elective procedures which resulted in a stark decline in endoscopy volume. There is concern that despite returning to pre-pandemic colonoscopy rates in the recovery period, these procedural delays may have had a lasting impact and worsened preexisting disparities in access to endoscopy. We aim to examine disparities in colonoscopy utilization in the post COVID-19 era as compared to the pre-pandemic period. Methods: We included all adult patients who underwent outpatient colonoscopy at our institution in the following time periods: (1) 7/1/17 to 3/15/20 (“pre-COVID”) and (2) 3/16/20 to 11/30/21 (“post-COVID”). We abstracted age, gender, race, language preference, insurance type and income quartile by zip code from the medical record. We identified those that were “high risk” for CRC based on personal or family history. Multivariable logistic regression analysis was performed to assess the association between subject characteristics and colonoscopy exposure in the post-COVID era. (Table) Results: There were 14,381 and 10,376 colonoscopies included in pre and post-COVID periods respectively. English-speaking patients comprised 92.8% of the pre-COVID cohort which increased to 94.2% in the post-COVID era (p < .001). This difference was statistically significant on multivariable analysis (OR 1.63, 95% CI 1.09-2.42). While there was also a statistically significant increase in colonoscopies of patients who identify as LatinX compared to white race (OR 1.14, 95% CI 1.03-1.26), the magnitude of this change was small – 8.3% versus 9.2% in the pre and post-COVID cohorts. We found an inverse association between the post-COVID era and Medicare insurance status as compared to private insurance (OR 0.87, 95% CI 0.81-0.95). There were no significant associations on multivariable analysis between colonoscopy in the post COVID phase and age, gender, race, income quartile by zip code nor high-risk status. Conclusion: In the aftermath of COVID-19 and after colonoscopy volume had returned to pre-pandemic levels, there was a reduction in colonoscopy utilization by non-English speakers as well as patients with Medicare insurance. Reassuringly, it does not appear that the well-established pre-existing racial gap in colonoscopy access was worsened by the COVID 19 pandemic in our population nor was there a material difference in colonoscopy use by income quartile – a common measure of socioeconomic disparity. Table 1. - Multivariable Logistic Regression of Attributes Associated with Colonoscopy Utilization in the post-COVID era (March 16, 2020 - November 30, 2021) compared to pre-COVID (July 1, 2017 - March 15, 2020) Odds Ratio 95% CI p-value Age < 45 1.05 0.98-1.12 0.21 45-59 1.0 - - 60-69 0.92 0.86-0.99 0.03 70-79 1.03 0.93-1.14 0.58 >80 1.22 1.01-1.47 0.04 Female 1.05 1.00-1.11 0.05 Race White 1.0 - - Black 1.11 0.99-1.25 0.07 Asian 1.01 0.94-1.09 0.77 Latinx 1.14 1.03-1.26 < 0.01 Other 1.004 0.90-1.12 0.95 Language English 1.0 - - Spanish 0.70 0.44-1.12 0.14 Russian 0.59 0.38-0.93 0.02 Chinese Dialects 0.55 0.36-0.84 < 0.01 Other 0.66 0.43-1.01 0.054 English v. All Others 1.63 1.09-2.42 0.02 Insurance Private 1.0 - - Medicare 0.87 0.81-0.95 < 0.01 Medical/Covered CA 0.95 0.87-1.02 0.16 Income Highest Quartile 1 - - 2nd Quartile 0.88 0.39-1.97 0.76 3rd Quartile 0.83 0.37-1.85 0.65 4th Quartile 0.80 0.36-1.80 0.59 High Risk 0.95 0.90-1.00 0.05