Metformin Use Associated with Decreased PTSD Symptoms in Diabetic Veterans

June 16, 2020
Patrick Campbell

An analysis of nearly 8k veterans has found use of metformin was associated with a reduction in PTSD symptoms.

New research presented at the American Diabetes Association’s (ADA) 80th Scientific Sessions from a Tulane University-led team suggests a first-line treatment for diabetic patients could also help reduce symptoms of post-traumatic stress disorder (PTSD) in veterans.

Results of the study, which included data from nearly 8000 veterans with PTSD and diabetes, indicated patients receiving metformin experienced a 31% reduction in PTSD symptoms compared to the study’s control cohort and also suggested patients with 2 or more years of exposure to metformin were 22% more likely to experience a reduction in symptoms than those with less than 2 years of exposure.

With both diabetes and PTSD common among veteran populations, investigators designed their study to assess whether metformin was associated with a clinically meaningful reduction in PTSD symptoms using data from the Veterans Health Administration. To compare the effects of metformin in this patient population, investigators identified 2 propensity-matched cohorts for analysis—a metformin cohort and a control cohort.

For inclusion in the metformin cohort, an individual’s first metformin use needed to occur after a recorded PTSD diagnosis with a PTSD checklist for DSM-5 (PCL-5) score of 33 or more. To create the control cohort, patients were matched in 1:1 ratio based on the likelihood of initiating metformin with the covariates of age, gender, race, duration of PTSD, diabetes, and depression, and baseline Charlson Comorbidity Index (CCI), PCL-5, and PHQ-9 scores.

The primary endpoint of the study was an improvement in PCL-5 score, which investigators defined as the first date of a reduction of 20 or more points from baseline. Investigators used multivariable Cox proportional regression to assess the association between metformin exposure and improvement in PCL-5 scores.

In total, 7950 veterans were identified for inclusion in the study with 3975 in each cohort. Of note, mean age of the cohort was 51.9±13.6 years, 85% were men, 61.4% were white and 28.8% were African American.

At baseline, the mean PCL-5 score was 56.2±11.6 in the metformin cohort and 56.5±11.7 in the control cohort. The mean follow-up time was 1.2 years for both cohorts.

Upon analysis, results indicated patients within the metformin-treated cohort were more likely to experience a clinically meaningful reduction of 31% (aHR, 1.31; 95% CI, 1.16-1.48) in PTSD symptoms compared to the those in the control cohort. Additionally, patients with 2 or more years of exposure to metformin were 22% (aHR, 1.22; 95% CI, 1.04-1.43) more likely to reduced PTSD symptoms than those with less than 2 years of exposure to metformin.

Investigators pointed out multiple limitations to keep in mind when interpreting the results of their study. Limitations included the retrospective nature of the study, inability to randomly assign treatment groups, use of ICD-9/10-CM codes could lead to inaccurate information, and because the study was performed using a veterans population findings cannot be generalized to other populations.

This study, titled “Association between Metformin Treatment and Improved Symptoms of Post-Traumatic Stress Disorder,” was presented at ADA 2020.


x