Assessment of Metformin Intolerance: A Retrospective Chart Review.

Abstract

Objective: The aim of the present study is to determine similarities between patients with type 2 diabetes not on metformin therapy compared to patients on metformin therapy at a resident-led primary care clinic. Methods: An exploratory, single-center retrospective chart review was performed on patients 18 years and older with a documented diagnosis of type 2 diabetes seen at the University of Arkansas for Medical Sciences Family Medicine Clinic in Little Rock, Arkansas. Of the 2452 patients who met criteria for the study, 1085 patients did not have a documented metformin allergy. A subset of 216 patients who were not currently prescribed metformin and had no documented metformin allergy were further examined and compared to the 869 patients who were prescribed metformin. We sought to determine reasons for nonuse by evaluating their EPIC electronic health record. Information on these patients such as race, gender, hemoglobin A1c (A1c), kidney function, stated metformin intolerance, and comorbid disease states such as neuropathy, chronic kidney disease (CKD), ulcerative colitis, and irritable bowel syndrome were collected. Further examination was performed to determine why patients were not on metformin therapy and potential similarities between metformin intolerant patients. Results: The results of the study indicated a significant difference between metformin users and non-users in relation to body mass index (BMI) and diagnosis of CKD. Metformin non-users were found to have significantly lower mean BMI (30.87 vs. 35.43; p-value <0.0001), and significantly higher rates of CKD (25.93% vs 14.73%; p-value <0.0001) as compared to metformin users. BMI value of patients (coefficient: 0.2033, p value: <0.0001) was found to be significantly and positively correlated with metformin use, and CKD (coefficient: - 0.1191, p-value: <0.0001) was found to be significantly and negatively correlated with metformin use. A1c levels for patients not on metformin therapy were evaluated. Most non-metformin patients fell in prediabetic A1c levels ranging from 5-6.4% (84 patients; 38.89%), and 31 patients (14.35%) should be on insulin therapy according to guidelines. Conclusion: The results demonstrated that patients with lower BMI, CKD, or A1c in the prediabetic range were less likely to be prescribed metformin.