Impact of antidiabetic medication type on a new episode of depression: a retrospective cohort study in Texas, USA.

OBJECTIVES

To examine the associations between antidiabetic medication type and a new episode of depression using 100% Texas Medicare database during 2009 and 2018.

DESIGN

A retrospective cohort study.

SETTING

A population-based study using the Texas Medicare data.

INTERVENTIONS

11 common antihyperglycaemic medication types, alone and in combinations: metformin-only, five non-metformin-containing regimens (dipeptidyl peptidase-4 inhibitor (DPP4i) only, sulfonylureas (SU) only, thiazolidinediones (TZD) only, SU/DPP4i and SU/TZD) and five metformin-containing combination treatments (metformin/DPP4i, metformin/SU, metformin/TZD, metformin/SU/DPP4i and metformin/SU/TZD).

PARTICIPANTS

This study included 59 057 type 2 diabetes (T2D) patients from a cohort of Texas Medicare beneficiaries who were aged ≥66 years, had consistent diabetes medication intake, were not diagnosed with depression or prescribed antidepressants during the 2-year look-back period and received regular care from Medicare providers.

MAIN OUTCOMES AND MEASURES

The main outcome was a new episode of depression, identified by a new depression diagnosis during the follow-up period.

RESULTS

A total of 59 057 T2D patients (mean (SD) age, 75.4 (6.4) years; 30 798 (52.1%) female) were followed up to 96 months. Of these, 22.5% patients had a new episode of depression at the 5-year follow-up. Compared with the metformin-only group, patients in the non-metformin-containing regimens had a higher risk of new episode depression (HR: 1.17, 95% CI 1.05 to 1.30 for DPP4i-only; HR: 1.06, 95% CI 1.01 to 1.12 for SU-only), but there was no significant difference among patients receiving metformin-containing combination therapy. Metformin/TZD and metformin/SU/DPP4i combination treatments had a lower risk of new episodes of depression than metformin-only (HR: 0.88, 95% CI 0.78 to 0.99 and HR: 0.83, 95% CI 0.71 to 0.98 separately). The same direction of association was observed in sensitivity analyses.

CONCLUSIONS

This retrospective cohort study found that T2D patients treated with metformin/TZD and metformin/SU/DPP4i had the lowest risk of new episodes of depression. These findings suggest that certain combinations of metformin with other antidiabetic medications may be associated with a reduced risk of new-onset depression. Therefore, it could be beneficial to incorporate depression risk evaluation into routine diabetes care and consider it in the decision-making process for diabetes medication types, especially when deprescribing metformin.