P0951 - Primary Care Visit Frequency in Patients with NAFLD Based on Diabetes Status

Stephen Fuller, MD, Andrew Schreiner, MD
Medical University of South Carolina, Charleston, SC

Introduction: NAFLD is an increasingly prevalent cause of cirrhosis and early diagnosis and risk factor modification in primary care remain key tools in preventing disease progression. Much attention has been paid to improving NAFLD detection in patients with diabetes and decreasing risk through better glycemic control; however, diagnosis and monitoring in patients without diabetes may be less robust. In this study, we aimed to determine how primary care (PC) visit frequency differed in patients with NAFLD by diabetes status.

Methods: We performed a retrospective cohort study of electronic health record data from 2012 to 2021 in a primary care clinic. We included primary care patients with NAFLD and followed them from the determination of NAFLD until the end of the study period. We evaluated the primary outcome of PC visits per year for each patient during follow-up. The primary predictor variable of interest was a diagnosis of diabetes mellitus at the time, or prior to, NAFLD diagnosis. Covariates included demographic and clinical variables at the time of NAFLD ascertainment. We compared patient-level PC visits/year (continuous variable) by diabetes status, and PC visits/year by each covariate using two sample t-tests. We developed unadjusted and adjusted gamma regression models for the outcome of PC visits/year with diabetes as the primary predictor variable using PROC GENMOD in SAS. SAS was used for all statistical analyses.

Results: The sample included 1,561 patients with NAFLD, of whom 730 (47%) were diagnosed with diabetes. The cohort was followed for a mean 4.9 years (± 2.7) and had a mean 1.9 (± 2.6) PC visits/year during follow-up. Patients with diabetes had a mean 2.5 (± 2.9) PC visits/year compared to a mean of 1.5 (± 2.2) PC visits/year in patients without diabetes (p< 0.001). In the unadjusted gamma regression model, diabetes was associated with more PC visits/year (0.54; 95%CI 0.35-0.72; Table 1). After adjusting for other potentially confounding covariates, diabetes remained associated with a significantly greater number of PC visits/year compared to patients with NAFLD and no diabetes (0.26; 95%CI 0.05-0.46).

Discussion: In this study, patients with NAFLD and diabetes had significantly more contact with primary care than patients with NAFLD and no coexisting diabetes. Future NAFLD efforts will need to carefully include non-diabetic patients that may not be receiving the same PC care intensity.



Disclosures:


Stephen Fuller, MD, Andrew Schreiner, MD. P0951 - Primary Care Visit Frequency in Patients with NAFLD Based on Diabetes Status, ACG 2023 Annual Scientific Meeting Abstracts. Vancouver, BC, Canada: American College of Gastroenterology.