(130) Baseline characteristics of patients with chronic kidney disease and type 2 diabetes (CKD-T2D) initiating anti-hyperglycemic medication: differences before and during the outbreak of COVID-19 in US and UK
Background: COVID-19 has influenced delivery of healthcare to patients including but not limited to treatment choices. Concurrently with the pandemic, evidence on new treatments for CKD-T2D was published.
Objectives: Evaluate differences in baseline characteristics of patients with CKD-T2D receiving anti-hyperglycemic drugs before and during the COVID-19 pandemic.
Methods: This study included adults from 2 claims and 3 electronic medical records databases converted to the OMOP CDM: Marketscan, Optum Clinformatics and EHR (from US), and CPRD GOLD and AURUM (from UK). The study spanned from Jan-1-2012, to Dec-31-2020. We generated six cohorts of patients with CKD-T2D who initiated either: Dipeptidyl Peptidase-4 Inhibitors (DPP4i), Glucagon-like peptide-1 receptor agonists (GLP1ra), insulin, metformin, Sodium-glucose cotransporter-2 inhibitors (SGLT2i), or sulfonylurea. The index event was initiation of the treatment. Cohorts were stratified according to their index date in a pre-COVID period, before or on Mar-1-2020, and a COVID period after March-1-2020 since lockdowns in the US and UK occurred during Mar-2020. Demographics and comorbidities at baseline (1-year prior to index) were characterized.
Results: A total of 167.8 million patients across all databases were assessed. Our study cohorts’ size ranged from 14,620 for DPP4i to 178,527 for sulfonylureas, across all databases in pre-COVID period, and 358 for insulin to 65,457 for DPP4i in COVID. Overall, demographics were similar across cohorts, however, SGLT2i and GLP1ra initiators during COVID were older than pre-COVID. Likewise, we observed a consistent decrease in coding of diabetic retinopathy by 50-80% across cohorts and databases. We also observed a less consistent lower proportion of diabetic neuropathy coding. During COVID, the proportion of patients with baseline exposure to SGLT2i was at least twice as much as in pre-COVID across cohorts except among initiators of metformin or GLP1ra. Baseline use of spironolactone appeared less frequent during COVID than in pre-COVID. However, for SGLT2i initiators we observed an 2-fold increase in the proportion of patients with baseline use of spironolactone during COVID compared to pre-COVID. Records for acute diseases of the CV system, depressive disorders and potassium disorders were more frequent in COVID than in pre-COVID in US databases.
Conclusions: The COVID-19 pandemic substantially influenced healthcare delivery in patients with CKD-T2D. Complications of diabetes requiring monitoring, such as retinopathy were less frequently coded in claims databases and electronic medical records.
