Abstract
BACKGROUND: In May 2020, the US Food and Drug Administration (FDA) asked 5 pharmaceutical companies to voluntarily recall some formulations of metformin due to contamination. This observational study sought to provide insight changes in hemoglobin A(1c) (HbA(1c)) levels when veterans switched to alternative antihyperglycemic agents following the recall. METHODS: This study included veterans aged ≥ 18 years with type 2 diabetes who were receiving health care from Veterans Integrated Service Network 6 and had an active metformin sustained-action (SA) prescription as of June 1, 2020. This observational study used a complex random-effects within-between model to evaluate the impact that the recall had on HbA(1c) levels as patients transitioned from metformin SA to an alternative antihyperglycemic agent (dipeptidyl-peptidase-4 inhibitor; glitazone; glucagon-like peptide-1 [GLP-1] agonist; sodium-glucose cotransporter-2 [SGLT-2] inhibitor; long-acting, rapid-acting, and mixed insulin formulations; immediate-release metformin, or sulfonylurea). This model identified individual-level (within patient) changes and changes between groups of patients that occurred during the year following the recall. RESULTS: A total of 9130 veterans were included. GLP-1 agonists were associated with a substantial decrease in HbA(1c) levels for patients and a moderate increase between patients (P < .001). SGLT-2 inhibitors were associated with a notable decrease in HbA(1c) levels for patients (P < .001). Insulin use was associated with increased HbA(1c) levels, but only between patients. Long-acting insulin and mixed insulin demonstrated marked increases between patients (P < .001). CONCLUSIONS: This study demonstrated that following an FDA recall, newer diabetes medications lowered HbA(1c) levels compared with metformin SA. Additional registry research is needed to examine HbA(1c) trends over time as related to medication therapy and determine long-term complications within the registry population.