ABSTRACT
Introduction: In gastroenterology, telemedicine has emerged as a means for improving patient access to medical care while limiting viral transmission during the pandemic. The purpose of this study was to assess patient order compliance in the virtual vs. in-person setting during the COVID-19 pandemic for the initial evaluation of constipation. Method(s): Using natural language processing, we identified outpatient gastroenterology visits (virtual and in-person) for constipation from March 2020 through December 2021. We assessed the number of orders placed for patients during these encounters and determined compliance based on order completion. A generalized linear mixed effects model with fixed effects for visit type and random intercepts for intra-patient correlation was used. A multivariable model was built controlling for age, socioeconomic status, BMI, dementia, stroke, and congestive heart failure. Result(s): Among 4,930 patients who presented for initial constipation evaluation since the start of the pandemic, 3,515 patients were evaluated in-person and 1,415 patients were evaluated virtually. Comparing order compliance in patients seen during the pandemic, patients seen virtually were 66% less likely to complete orders in comparison to patients seen in-person (p < .001). Patients seen in a pandemic virtual setting were 43% less likely to complete imaging orders (p < 0.001), 78% less likely to complete procedure orders (p < 0.001), and 90% less likely to complete lab orders (p < 0.001) (Table). Increased lab compliance was associated with the highest socioeconomic status ($75,000-$200,000) with patients eight times more likely to complete lab orders (p=0.049) and three times more likely to complete orders overall (p=0.021). Conclusion(s): Compared with in-person visits, patients seen virtually for their first presentation of constipation were less likely to complete labs, imaging and procedure evaluation ordered. In-person visits were more successful in leading to patient order completion during the pandemic. These findings suggest that virtual visits for constipation, despite convenience, may compromise care delivery;such visits may thus require additional care coordination to achieve compliance with medical recommendations. (Figure Presented).
ABSTRACT
Introduction: Telemedicine has been underscored during the COVID-19 pandemic as a way to improve healthcare access, efficiency, and reduce cost. In gastroenterology, telemedicine has been proposed as a means to optimize patient care. The purpose of this study was to assess patient order compliance in the virtual vs. in-person setting, prior to and during the COVID-19 pandemic, in the initial evaluation of abdominal pain (AP). Methods: Using natural language processing, we identified outpatient gastroenterology visits (virtual and inperson) for AP up to 1 year prior and after March 2020, labeled as pre-pandemic and pandemic visits, respectively. We assessed the number of orders placed for patients during these encounters and determined compliance based off order completion. A generalized linear mixed effects model with fixed effects for visit type, and random intercepts for intra-patient correlation was used. A multivariable model was built controlling for age, socioeconomic status, BMI, dementia, stroke and congestive heart failure. Results: Among 20,356 patients with AP at their initial visit, 16,089 patients (79%) had orders placed, 7,569 patients (47%) had pandemic in-person visits, 2,109 patients (13.1%) had pandemic virtual visits, and 6,409 patients (39.8%) had pre-pandemic in-person visits (Table 1). Comparing order compliance in patients seen during the pandemic, patients seen virtually were 38.3% less likely to complete orders in comparison to patients seen in-person (p < 0.0001). Patients seen in a pandemic virtual setting were 58.1% less likely to complete imaging orders (p < 0.0001), 73.9% less likely to complete procedure orders (p < 0.0001), and 47.8% less likely to complete lab orders (p = 0.0086) (Table 2). Lack of procedure compliance was compounded by low socioeconomic status ($25,000-$50,000) with patients 48.7% less likely to complete procedure orders (p <0.0264). Comparing compliance between pre-pandemic in-person visits with pandemic in-person visits, patients were 10% less likely to complete orders prior to the pandemic (p = 0.0151). Patients were 63.6% less likely to complete imaging orders (p < 0.0001), 68.8% less likely to complete procedures (p < 0.0001), and 38.7% less likely to complete lab orders (p=0.0002). Conclusion: Compared with in-person visits, patients seen virtually for their first presentation of abdominal pain were less likely to complete labs, imaging, and endoscopic evaluations. In-person visits were more successful at getting patients to complete orders prior to, and during, the pandemic. These findings highlight that virtual visits for abdominal pain, despite convenience, may be detrimental to care coordination and compliance. (Table Presented) (Table Presented)